stacks-1.44/000755 000765 000024 00000000000 13001012452 013466 5ustar00catchenstaff000000 000000 stacks-1.44/acinclude.m4000644 000765 000024 00000033000 12774470747 015712 0ustar00catchenstaff000000 000000 # =========================================================================== # http://www.gnu.org/software/autoconf-archive/ax_cxx_compile_stdcxx.html # =========================================================================== # # SYNOPSIS # # AX_CXX_COMPILE_STDCXX(VERSION, [ext|noext], [mandatory|optional]) # # DESCRIPTION # # Check for baseline language coverage in the compiler for the specified # version of the C++ standard. If necessary, add switches to CXX and # CXXCPP to enable support. VERSION may be '11' (for the C++11 standard) # or '14' (for the C++14 standard). # # The second argument, if specified, indicates whether you insist on an # extended mode (e.g. -std=gnu++11) or a strict conformance mode (e.g. # -std=c++11). If neither is specified, you get whatever works, with # preference for an extended mode. # # The third argument, if specified 'mandatory' or if left unspecified, # indicates that baseline support for the specified C++ standard is # required and that the macro should error out if no mode with that # support is found. If specified 'optional', then configuration proceeds # regardless, after defining HAVE_CXX${VERSION} if and only if a # supporting mode is found. # # LICENSE # # Copyright (c) 2008 Benjamin Kosnik # Copyright (c) 2012 Zack Weinberg # Copyright (c) 2013 Roy Stogner # Copyright (c) 2014, 2015 Google Inc.; contributed by Alexey Sokolov # Copyright (c) 2015 Paul Norman # Copyright (c) 2015 Moritz Klammler # # Copying and distribution of this file, with or without modification, are # permitted in any medium without royalty provided the copyright notice # and this notice are preserved. This file is offered as-is, without any # warranty. #serial 4 dnl This macro is based on the code from the AX_CXX_COMPILE_STDCXX_11 macro dnl (serial version number 13). AC_DEFUN([AX_CXX_COMPILE_STDCXX], [dnl m4_if([$1], [11], [], [$1], [14], [], [$1], [17], [m4_fatal([support for C++17 not yet implemented in AX_CXX_COMPILE_STDCXX])], [m4_fatal([invalid first argument `$1' to AX_CXX_COMPILE_STDCXX])])dnl m4_if([$2], [], [], [$2], [ext], [], [$2], [noext], [], [m4_fatal([invalid second argument `$2' to AX_CXX_COMPILE_STDCXX])])dnl m4_if([$3], [], [ax_cxx_compile_cxx$1_required=true], [$3], [mandatory], [ax_cxx_compile_cxx$1_required=true], [$3], [optional], [ax_cxx_compile_cxx$1_required=false], [m4_fatal([invalid third argument `$3' to AX_CXX_COMPILE_STDCXX])]) AC_LANG_PUSH([C++])dnl ac_success=no AC_CACHE_CHECK(whether $CXX supports C++$1 features by default, ax_cv_cxx_compile_cxx$1, [AC_COMPILE_IFELSE([AC_LANG_SOURCE([_AX_CXX_COMPILE_STDCXX_testbody_$1])], [ax_cv_cxx_compile_cxx$1=yes], [ax_cv_cxx_compile_cxx$1=no])]) if test x$ax_cv_cxx_compile_cxx$1 = xyes; then ac_success=yes fi m4_if([$2], [noext], [], [dnl if test x$ac_success = xno; then for switch in -std=gnu++$1 -std=gnu++0x; do cachevar=AS_TR_SH([ax_cv_cxx_compile_cxx$1_$switch]) AC_CACHE_CHECK(whether $CXX supports C++$1 features with $switch, $cachevar, [ac_save_CXX="$CXX" CXX="$CXX $switch" AC_COMPILE_IFELSE([AC_LANG_SOURCE([_AX_CXX_COMPILE_STDCXX_testbody_$1])], [eval $cachevar=yes], [eval $cachevar=no]) CXX="$ac_save_CXX"]) if eval test x\$$cachevar = xyes; then CXX="$CXX $switch" if test -n "$CXXCPP" ; then CXXCPP="$CXXCPP $switch" fi ac_success=yes break fi done fi]) m4_if([$2], [ext], [], [dnl if test x$ac_success = xno; then dnl HP's aCC needs +std=c++11 according to: dnl http://h21007.www2.hp.com/portal/download/files/unprot/aCxx/PDF_Release_Notes/769149-001.pdf dnl Cray's crayCC needs "-h std=c++11" for switch in -std=c++$1 -std=c++0x +std=c++$1 "-h std=c++$1"; do cachevar=AS_TR_SH([ax_cv_cxx_compile_cxx$1_$switch]) AC_CACHE_CHECK(whether $CXX supports C++$1 features with $switch, $cachevar, [ac_save_CXX="$CXX" CXX="$CXX $switch" AC_COMPILE_IFELSE([AC_LANG_SOURCE([_AX_CXX_COMPILE_STDCXX_testbody_$1])], [eval $cachevar=yes], [eval $cachevar=no]) CXX="$ac_save_CXX"]) if eval test x\$$cachevar = xyes; then CXX="$CXX $switch" if test -n "$CXXCPP" ; then CXXCPP="$CXXCPP $switch" fi ac_success=yes break fi done fi]) AC_LANG_POP([C++]) if test x$ax_cxx_compile_cxx$1_required = xtrue; then if test x$ac_success = xno; then AC_MSG_ERROR([*** A compiler with support for C++$1 language features is required.]) fi fi if test x$ac_success = xno; then HAVE_CXX$1=0 AC_MSG_NOTICE([No compiler with C++$1 support was found]) else HAVE_CXX$1=1 AC_DEFINE(HAVE_CXX$1,1, [define if the compiler supports basic C++$1 syntax]) fi AC_SUBST(HAVE_CXX$1) ]) dnl Test body for checking C++11 support m4_define([_AX_CXX_COMPILE_STDCXX_testbody_11], _AX_CXX_COMPILE_STDCXX_testbody_new_in_11 ) dnl Test body for checking C++14 support m4_define([_AX_CXX_COMPILE_STDCXX_testbody_14], _AX_CXX_COMPILE_STDCXX_testbody_new_in_11 _AX_CXX_COMPILE_STDCXX_testbody_new_in_14 ) dnl Tests for new features in C++11 m4_define([_AX_CXX_COMPILE_STDCXX_testbody_new_in_11], [[ // If the compiler admits that it is not ready for C++11, why torture it? // Hopefully, this will speed up the test. #ifndef __cplusplus #error "This is not a C++ compiler" #elif __cplusplus < 201103L #error "This is not a C++11 compiler" #else namespace cxx11 { namespace test_static_assert { template struct check { static_assert(sizeof(int) <= sizeof(T), "not big enough"); }; } namespace test_final_override { struct Base { virtual void f() {} }; struct Derived : public Base { virtual void f() override {} }; } namespace test_double_right_angle_brackets { template < typename T > struct check {}; typedef check single_type; typedef check> double_type; typedef check>> triple_type; typedef check>>> quadruple_type; } namespace test_decltype { int f() { int a = 1; decltype(a) b = 2; return a + b; } } namespace test_type_deduction { template < typename T1, typename T2 > struct is_same { static const bool value = false; }; template < typename T > struct is_same { static const bool value = true; }; template < typename T1, typename T2 > auto add(T1 a1, T2 a2) -> decltype(a1 + a2) { return a1 + a2; } int test(const int c, volatile int v) { static_assert(is_same::value == true, ""); static_assert(is_same::value == false, ""); static_assert(is_same::value == false, ""); auto ac = c; auto av = v; auto sumi = ac + av + 'x'; auto sumf = ac + av + 1.0; static_assert(is_same::value == true, ""); static_assert(is_same::value == true, ""); static_assert(is_same::value == true, ""); static_assert(is_same::value == false, ""); static_assert(is_same::value == true, ""); return (sumf > 0.0) ? sumi : add(c, v); } } namespace test_noexcept { int f() { return 0; } int g() noexcept { return 0; } static_assert(noexcept(f()) == false, ""); static_assert(noexcept(g()) == true, ""); } namespace test_constexpr { template < typename CharT > unsigned long constexpr strlen_c_r(const CharT *const s, const unsigned long acc) noexcept { return *s ? strlen_c_r(s + 1, acc + 1) : acc; } template < typename CharT > unsigned long constexpr strlen_c(const CharT *const s) noexcept { return strlen_c_r(s, 0UL); } static_assert(strlen_c("") == 0UL, ""); static_assert(strlen_c("1") == 1UL, ""); static_assert(strlen_c("example") == 7UL, ""); static_assert(strlen_c("another\0example") == 7UL, ""); } namespace test_rvalue_references { template < int N > struct answer { static constexpr int value = N; }; answer<1> f(int&) { return answer<1>(); } answer<2> f(const int&) { return answer<2>(); } answer<3> f(int&&) { return answer<3>(); } void test() { int i = 0; const int c = 0; static_assert(decltype(f(i))::value == 1, ""); static_assert(decltype(f(c))::value == 2, ""); static_assert(decltype(f(0))::value == 3, ""); } } namespace test_uniform_initialization { struct test { static const int zero {}; static const int one {1}; }; static_assert(test::zero == 0, ""); static_assert(test::one == 1, ""); } namespace test_lambdas { void test1() { auto lambda1 = [](){}; auto lambda2 = lambda1; lambda1(); lambda2(); } int test2() { auto a = [](int i, int j){ return i + j; }(1, 2); auto b = []() -> int { return '0'; }(); auto c = [=](){ return a + b; }(); auto d = [&](){ return c; }(); auto e = [a, &b](int x) mutable { const auto identity = [](int y){ return y; }; for (auto i = 0; i < a; ++i) a += b--; return x + identity(a + b); }(0); return a + b + c + d + e; } int test3() { const auto nullary = [](){ return 0; }; const auto unary = [](int x){ return x; }; using nullary_t = decltype(nullary); using unary_t = decltype(unary); const auto higher1st = [](nullary_t f){ return f(); }; const auto higher2nd = [unary](nullary_t f1){ return [unary, f1](unary_t f2){ return f2(unary(f1())); }; }; return higher1st(nullary) + higher2nd(nullary)(unary); } } namespace test_variadic_templates { template struct sum; template struct sum { static constexpr auto value = N0 + sum::value; }; template <> struct sum<> { static constexpr auto value = 0; }; static_assert(sum<>::value == 0, ""); static_assert(sum<1>::value == 1, ""); static_assert(sum<23>::value == 23, ""); static_assert(sum<1, 2>::value == 3, ""); static_assert(sum<5, 5, 11>::value == 21, ""); static_assert(sum<2, 3, 5, 7, 11, 13>::value == 41, ""); } // http://stackoverflow.com/questions/13728184/template-aliases-and-sfinae // Clang 3.1 fails with headers of libstd++ 4.8.3 when using std::function // because of this. namespace test_template_alias_sfinae { struct foo {}; template using member = typename T::member_type; template void func(...) {} template void func(member*) {} void test(); void test() { func(0); } } } // namespace cxx11 #endif // __cplusplus >= 201103L ]]) dnl Tests for new features in C++14 m4_define([_AX_CXX_COMPILE_STDCXX_testbody_new_in_14], [[ // If the compiler admits that it is not ready for C++14, why torture it? // Hopefully, this will speed up the test. #ifndef __cplusplus #error "This is not a C++ compiler" #elif __cplusplus < 201402L #error "This is not a C++14 compiler" #else namespace cxx14 { namespace test_polymorphic_lambdas { int test() { const auto lambda = [](auto&&... args){ const auto istiny = [](auto x){ return (sizeof(x) == 1UL) ? 1 : 0; }; const int aretiny[] = { istiny(args)... }; return aretiny[0]; }; return lambda(1, 1L, 1.0f, '1'); } } namespace test_binary_literals { constexpr auto ivii = 0b0000000000101010; static_assert(ivii == 42, "wrong value"); } namespace test_generalized_constexpr { template < typename CharT > constexpr unsigned long strlen_c(const CharT *const s) noexcept { auto length = 0UL; for (auto p = s; *p; ++p) ++length; return length; } static_assert(strlen_c("") == 0UL, ""); static_assert(strlen_c("x") == 1UL, ""); static_assert(strlen_c("test") == 4UL, ""); static_assert(strlen_c("another\0test") == 7UL, ""); } namespace test_lambda_init_capture { int test() { auto x = 0; const auto lambda1 = [a = x](int b){ return a + b; }; const auto lambda2 = [a = lambda1(x)](){ return a; }; return lambda2(); } } namespace test_digit_seperators { constexpr auto ten_million = 100'000'000; static_assert(ten_million == 100000000, ""); } namespace test_return_type_deduction { auto f(int& x) { return x; } decltype(auto) g(int& x) { return x; } template < typename T1, typename T2 > struct is_same { static constexpr auto value = false; }; template < typename T > struct is_same { static constexpr auto value = true; }; int test() { auto x = 0; static_assert(is_same::value, ""); static_assert(is_same::value, ""); return x; } } } // namespace cxx14 #endif // __cplusplus >= 201402L ]]) stacks-1.44/aclocal.m4000644 000765 000024 00000122270 13001003444 015333 0ustar00catchenstaff000000 000000 # generated automatically by aclocal 1.15 -*- Autoconf -*- # Copyright (C) 1996-2014 Free Software Foundation, Inc. # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY, to the extent permitted by law; without # even the implied warranty of MERCHANTABILITY or FITNESS FOR A # PARTICULAR PURPOSE. m4_ifndef([AC_CONFIG_MACRO_DIRS], [m4_defun([_AM_CONFIG_MACRO_DIRS], [])m4_defun([AC_CONFIG_MACRO_DIRS], [_AM_CONFIG_MACRO_DIRS($@)])]) m4_ifndef([AC_AUTOCONF_VERSION], [m4_copy([m4_PACKAGE_VERSION], [AC_AUTOCONF_VERSION])])dnl m4_if(m4_defn([AC_AUTOCONF_VERSION]), [2.69],, [m4_warning([this file was generated for autoconf 2.69. You have another version of autoconf. It may work, but is not guaranteed to. If you have problems, you may need to regenerate the build system entirely. To do so, use the procedure documented by the package, typically 'autoreconf'.])]) # Copyright (C) 2002-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # AM_AUTOMAKE_VERSION(VERSION) # ---------------------------- # Automake X.Y traces this macro to ensure aclocal.m4 has been # generated from the m4 files accompanying Automake X.Y. # (This private macro should not be called outside this file.) AC_DEFUN([AM_AUTOMAKE_VERSION], [am__api_version='1.15' dnl Some users find AM_AUTOMAKE_VERSION and mistake it for a way to dnl require some minimum version. Point them to the right macro. m4_if([$1], [1.15], [], [AC_FATAL([Do not call $0, use AM_INIT_AUTOMAKE([$1]).])])dnl ]) # _AM_AUTOCONF_VERSION(VERSION) # ----------------------------- # aclocal traces this macro to find the Autoconf version. # This is a private macro too. Using m4_define simplifies # the logic in aclocal, which can simply ignore this definition. m4_define([_AM_AUTOCONF_VERSION], []) # AM_SET_CURRENT_AUTOMAKE_VERSION # ------------------------------- # Call AM_AUTOMAKE_VERSION and AM_AUTOMAKE_VERSION so they can be traced. # This function is AC_REQUIREd by AM_INIT_AUTOMAKE. AC_DEFUN([AM_SET_CURRENT_AUTOMAKE_VERSION], [AM_AUTOMAKE_VERSION([1.15])dnl m4_ifndef([AC_AUTOCONF_VERSION], [m4_copy([m4_PACKAGE_VERSION], [AC_AUTOCONF_VERSION])])dnl _AM_AUTOCONF_VERSION(m4_defn([AC_AUTOCONF_VERSION]))]) # AM_AUX_DIR_EXPAND -*- Autoconf -*- # Copyright (C) 2001-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # For projects using AC_CONFIG_AUX_DIR([foo]), Autoconf sets # $ac_aux_dir to '$srcdir/foo'. In other projects, it is set to # '$srcdir', '$srcdir/..', or '$srcdir/../..'. # # Of course, Automake must honor this variable whenever it calls a # tool from the auxiliary directory. The problem is that $srcdir (and # therefore $ac_aux_dir as well) can be either absolute or relative, # depending on how configure is run. This is pretty annoying, since # it makes $ac_aux_dir quite unusable in subdirectories: in the top # source directory, any form will work fine, but in subdirectories a # relative path needs to be adjusted first. # # $ac_aux_dir/missing # fails when called from a subdirectory if $ac_aux_dir is relative # $top_srcdir/$ac_aux_dir/missing # fails if $ac_aux_dir is absolute, # fails when called from a subdirectory in a VPATH build with # a relative $ac_aux_dir # # The reason of the latter failure is that $top_srcdir and $ac_aux_dir # are both prefixed by $srcdir. In an in-source build this is usually # harmless because $srcdir is '.', but things will broke when you # start a VPATH build or use an absolute $srcdir. # # So we could use something similar to $top_srcdir/$ac_aux_dir/missing, # iff we strip the leading $srcdir from $ac_aux_dir. That would be: # am_aux_dir='\$(top_srcdir)/'`expr "$ac_aux_dir" : "$srcdir//*\(.*\)"` # and then we would define $MISSING as # MISSING="\${SHELL} $am_aux_dir/missing" # This will work as long as MISSING is not called from configure, because # unfortunately $(top_srcdir) has no meaning in configure. # However there are other variables, like CC, which are often used in # configure, and could therefore not use this "fixed" $ac_aux_dir. # # Another solution, used here, is to always expand $ac_aux_dir to an # absolute PATH. The drawback is that using absolute paths prevent a # configured tree to be moved without reconfiguration. AC_DEFUN([AM_AUX_DIR_EXPAND], [AC_REQUIRE([AC_CONFIG_AUX_DIR_DEFAULT])dnl # Expand $ac_aux_dir to an absolute path. am_aux_dir=`cd "$ac_aux_dir" && pwd` ]) # AM_CONDITIONAL -*- Autoconf -*- # Copyright (C) 1997-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # AM_CONDITIONAL(NAME, SHELL-CONDITION) # ------------------------------------- # Define a conditional. AC_DEFUN([AM_CONDITIONAL], [AC_PREREQ([2.52])dnl m4_if([$1], [TRUE], [AC_FATAL([$0: invalid condition: $1])], [$1], [FALSE], [AC_FATAL([$0: invalid condition: $1])])dnl AC_SUBST([$1_TRUE])dnl AC_SUBST([$1_FALSE])dnl _AM_SUBST_NOTMAKE([$1_TRUE])dnl _AM_SUBST_NOTMAKE([$1_FALSE])dnl m4_define([_AM_COND_VALUE_$1], [$2])dnl if $2; then $1_TRUE= $1_FALSE='#' else $1_TRUE='#' $1_FALSE= fi AC_CONFIG_COMMANDS_PRE( [if test -z "${$1_TRUE}" && test -z "${$1_FALSE}"; then AC_MSG_ERROR([[conditional "$1" was never defined. Usually this means the macro was only invoked conditionally.]]) fi])]) # Copyright (C) 1999-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # There are a few dirty hacks below to avoid letting 'AC_PROG_CC' be # written in clear, in which case automake, when reading aclocal.m4, # will think it sees a *use*, and therefore will trigger all it's # C support machinery. Also note that it means that autoscan, seeing # CC etc. in the Makefile, will ask for an AC_PROG_CC use... # _AM_DEPENDENCIES(NAME) # ---------------------- # See how the compiler implements dependency checking. # NAME is "CC", "CXX", "OBJC", "OBJCXX", "UPC", or "GJC". # We try a few techniques and use that to set a single cache variable. # # We don't AC_REQUIRE the corresponding AC_PROG_CC since the latter was # modified to invoke _AM_DEPENDENCIES(CC); we would have a circular # dependency, and given that the user is not expected to run this macro, # just rely on AC_PROG_CC. AC_DEFUN([_AM_DEPENDENCIES], [AC_REQUIRE([AM_SET_DEPDIR])dnl AC_REQUIRE([AM_OUTPUT_DEPENDENCY_COMMANDS])dnl AC_REQUIRE([AM_MAKE_INCLUDE])dnl AC_REQUIRE([AM_DEP_TRACK])dnl m4_if([$1], [CC], [depcc="$CC" am_compiler_list=], [$1], [CXX], [depcc="$CXX" am_compiler_list=], [$1], [OBJC], [depcc="$OBJC" am_compiler_list='gcc3 gcc'], [$1], [OBJCXX], [depcc="$OBJCXX" am_compiler_list='gcc3 gcc'], [$1], [UPC], [depcc="$UPC" am_compiler_list=], [$1], [GCJ], [depcc="$GCJ" am_compiler_list='gcc3 gcc'], [depcc="$$1" am_compiler_list=]) AC_CACHE_CHECK([dependency style of $depcc], [am_cv_$1_dependencies_compiler_type], [if test -z "$AMDEP_TRUE" && test -f "$am_depcomp"; then # We make a subdir and do the tests there. Otherwise we can end up # making bogus files that we don't know about and never remove. For # instance it was reported that on HP-UX the gcc test will end up # making a dummy file named 'D' -- because '-MD' means "put the output # in D". rm -rf conftest.dir mkdir conftest.dir # Copy depcomp to subdir because otherwise we won't find it if we're # using a relative directory. cp "$am_depcomp" conftest.dir cd conftest.dir # We will build objects and dependencies in a subdirectory because # it helps to detect inapplicable dependency modes. For instance # both Tru64's cc and ICC support -MD to output dependencies as a # side effect of compilation, but ICC will put the dependencies in # the current directory while Tru64 will put them in the object # directory. mkdir sub am_cv_$1_dependencies_compiler_type=none if test "$am_compiler_list" = ""; then am_compiler_list=`sed -n ['s/^#*\([a-zA-Z0-9]*\))$/\1/p'] < ./depcomp` fi am__universal=false m4_case([$1], [CC], [case " $depcc " in #( *\ -arch\ *\ -arch\ *) am__universal=true ;; esac], [CXX], [case " $depcc " in #( *\ -arch\ *\ -arch\ *) am__universal=true ;; esac]) for depmode in $am_compiler_list; do # Setup a source with many dependencies, because some compilers # like to wrap large dependency lists on column 80 (with \), and # we should not choose a depcomp mode which is confused by this. # # We need to recreate these files for each test, as the compiler may # overwrite some of them when testing with obscure command lines. # This happens at least with the AIX C compiler. : > sub/conftest.c for i in 1 2 3 4 5 6; do echo '#include "conftst'$i'.h"' >> sub/conftest.c # Using ": > sub/conftst$i.h" creates only sub/conftst1.h with # Solaris 10 /bin/sh. echo '/* dummy */' > sub/conftst$i.h done echo "${am__include} ${am__quote}sub/conftest.Po${am__quote}" > confmf # We check with '-c' and '-o' for the sake of the "dashmstdout" # mode. It turns out that the SunPro C++ compiler does not properly # handle '-M -o', and we need to detect this. Also, some Intel # versions had trouble with output in subdirs. am__obj=sub/conftest.${OBJEXT-o} am__minus_obj="-o $am__obj" case $depmode in gcc) # This depmode causes a compiler race in universal mode. test "$am__universal" = false || continue ;; nosideeffect) # After this tag, mechanisms are not by side-effect, so they'll # only be used when explicitly requested. if test "x$enable_dependency_tracking" = xyes; then continue else break fi ;; msvc7 | msvc7msys | msvisualcpp | msvcmsys) # This compiler won't grok '-c -o', but also, the minuso test has # not run yet. These depmodes are late enough in the game, and # so weak that their functioning should not be impacted. am__obj=conftest.${OBJEXT-o} am__minus_obj= ;; none) break ;; esac if depmode=$depmode \ source=sub/conftest.c object=$am__obj \ depfile=sub/conftest.Po tmpdepfile=sub/conftest.TPo \ $SHELL ./depcomp $depcc -c $am__minus_obj sub/conftest.c \ >/dev/null 2>conftest.err && grep sub/conftst1.h sub/conftest.Po > /dev/null 2>&1 && grep sub/conftst6.h sub/conftest.Po > /dev/null 2>&1 && grep $am__obj sub/conftest.Po > /dev/null 2>&1 && ${MAKE-make} -s -f confmf > /dev/null 2>&1; then # icc doesn't choke on unknown options, it will just issue warnings # or remarks (even with -Werror). So we grep stderr for any message # that says an option was ignored or not supported. # When given -MP, icc 7.0 and 7.1 complain thusly: # icc: Command line warning: ignoring option '-M'; no argument required # The diagnosis changed in icc 8.0: # icc: Command line remark: option '-MP' not supported if (grep 'ignoring option' conftest.err || grep 'not supported' conftest.err) >/dev/null 2>&1; then :; else am_cv_$1_dependencies_compiler_type=$depmode break fi fi done cd .. rm -rf conftest.dir else am_cv_$1_dependencies_compiler_type=none fi ]) AC_SUBST([$1DEPMODE], [depmode=$am_cv_$1_dependencies_compiler_type]) AM_CONDITIONAL([am__fastdep$1], [ test "x$enable_dependency_tracking" != xno \ && test "$am_cv_$1_dependencies_compiler_type" = gcc3]) ]) # AM_SET_DEPDIR # ------------- # Choose a directory name for dependency files. # This macro is AC_REQUIREd in _AM_DEPENDENCIES. AC_DEFUN([AM_SET_DEPDIR], [AC_REQUIRE([AM_SET_LEADING_DOT])dnl AC_SUBST([DEPDIR], ["${am__leading_dot}deps"])dnl ]) # AM_DEP_TRACK # ------------ AC_DEFUN([AM_DEP_TRACK], [AC_ARG_ENABLE([dependency-tracking], [dnl AS_HELP_STRING( [--enable-dependency-tracking], [do not reject slow dependency extractors]) AS_HELP_STRING( [--disable-dependency-tracking], [speeds up one-time build])]) if test "x$enable_dependency_tracking" != xno; then am_depcomp="$ac_aux_dir/depcomp" AMDEPBACKSLASH='\' am__nodep='_no' fi AM_CONDITIONAL([AMDEP], [test "x$enable_dependency_tracking" != xno]) AC_SUBST([AMDEPBACKSLASH])dnl _AM_SUBST_NOTMAKE([AMDEPBACKSLASH])dnl AC_SUBST([am__nodep])dnl _AM_SUBST_NOTMAKE([am__nodep])dnl ]) # Generate code to set up dependency tracking. -*- Autoconf -*- # Copyright (C) 1999-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # _AM_OUTPUT_DEPENDENCY_COMMANDS # ------------------------------ AC_DEFUN([_AM_OUTPUT_DEPENDENCY_COMMANDS], [{ # Older Autoconf quotes --file arguments for eval, but not when files # are listed without --file. Let's play safe and only enable the eval # if we detect the quoting. case $CONFIG_FILES in *\'*) eval set x "$CONFIG_FILES" ;; *) set x $CONFIG_FILES ;; esac shift for mf do # Strip MF so we end up with the name of the file. mf=`echo "$mf" | sed -e 's/:.*$//'` # Check whether this is an Automake generated Makefile or not. # We used to match only the files named 'Makefile.in', but # some people rename them; so instead we look at the file content. # Grep'ing the first line is not enough: some people post-process # each Makefile.in and add a new line on top of each file to say so. # Grep'ing the whole file is not good either: AIX grep has a line # limit of 2048, but all sed's we know have understand at least 4000. if sed -n 's,^#.*generated by automake.*,X,p' "$mf" | grep X >/dev/null 2>&1; then dirpart=`AS_DIRNAME("$mf")` else continue fi # Extract the definition of DEPDIR, am__include, and am__quote # from the Makefile without running 'make'. DEPDIR=`sed -n 's/^DEPDIR = //p' < "$mf"` test -z "$DEPDIR" && continue am__include=`sed -n 's/^am__include = //p' < "$mf"` test -z "$am__include" && continue am__quote=`sed -n 's/^am__quote = //p' < "$mf"` # Find all dependency output files, they are included files with # $(DEPDIR) in their names. We invoke sed twice because it is the # simplest approach to changing $(DEPDIR) to its actual value in the # expansion. for file in `sed -n " s/^$am__include $am__quote\(.*(DEPDIR).*\)$am__quote"'$/\1/p' <"$mf" | \ sed -e 's/\$(DEPDIR)/'"$DEPDIR"'/g'`; do # Make sure the directory exists. test -f "$dirpart/$file" && continue fdir=`AS_DIRNAME(["$file"])` AS_MKDIR_P([$dirpart/$fdir]) # echo "creating $dirpart/$file" echo '# dummy' > "$dirpart/$file" done done } ])# _AM_OUTPUT_DEPENDENCY_COMMANDS # AM_OUTPUT_DEPENDENCY_COMMANDS # ----------------------------- # This macro should only be invoked once -- use via AC_REQUIRE. # # This code is only required when automatic dependency tracking # is enabled. FIXME. This creates each '.P' file that we will # need in order to bootstrap the dependency handling code. AC_DEFUN([AM_OUTPUT_DEPENDENCY_COMMANDS], [AC_CONFIG_COMMANDS([depfiles], [test x"$AMDEP_TRUE" != x"" || _AM_OUTPUT_DEPENDENCY_COMMANDS], [AMDEP_TRUE="$AMDEP_TRUE" ac_aux_dir="$ac_aux_dir"]) ]) # Do all the work for Automake. -*- Autoconf -*- # Copyright (C) 1996-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # This macro actually does too much. Some checks are only needed if # your package does certain things. But this isn't really a big deal. dnl Redefine AC_PROG_CC to automatically invoke _AM_PROG_CC_C_O. m4_define([AC_PROG_CC], m4_defn([AC_PROG_CC]) [_AM_PROG_CC_C_O ]) # AM_INIT_AUTOMAKE(PACKAGE, VERSION, [NO-DEFINE]) # AM_INIT_AUTOMAKE([OPTIONS]) # ----------------------------------------------- # The call with PACKAGE and VERSION arguments is the old style # call (pre autoconf-2.50), which is being phased out. PACKAGE # and VERSION should now be passed to AC_INIT and removed from # the call to AM_INIT_AUTOMAKE. # We support both call styles for the transition. After # the next Automake release, Autoconf can make the AC_INIT # arguments mandatory, and then we can depend on a new Autoconf # release and drop the old call support. AC_DEFUN([AM_INIT_AUTOMAKE], [AC_PREREQ([2.65])dnl dnl Autoconf wants to disallow AM_ names. We explicitly allow dnl the ones we care about. m4_pattern_allow([^AM_[A-Z]+FLAGS$])dnl AC_REQUIRE([AM_SET_CURRENT_AUTOMAKE_VERSION])dnl AC_REQUIRE([AC_PROG_INSTALL])dnl if test "`cd $srcdir && pwd`" != "`pwd`"; then # Use -I$(srcdir) only when $(srcdir) != ., so that make's output # is not polluted with repeated "-I." AC_SUBST([am__isrc], [' -I$(srcdir)'])_AM_SUBST_NOTMAKE([am__isrc])dnl # test to see if srcdir already configured if test -f $srcdir/config.status; then AC_MSG_ERROR([source directory already configured; run "make distclean" there first]) fi fi # test whether we have cygpath if test -z "$CYGPATH_W"; then if (cygpath --version) >/dev/null 2>/dev/null; then CYGPATH_W='cygpath -w' else CYGPATH_W=echo fi fi AC_SUBST([CYGPATH_W]) # Define the identity of the package. dnl Distinguish between old-style and new-style calls. m4_ifval([$2], [AC_DIAGNOSE([obsolete], [$0: two- and three-arguments forms are deprecated.]) m4_ifval([$3], [_AM_SET_OPTION([no-define])])dnl AC_SUBST([PACKAGE], [$1])dnl AC_SUBST([VERSION], [$2])], [_AM_SET_OPTIONS([$1])dnl dnl Diagnose old-style AC_INIT with new-style AM_AUTOMAKE_INIT. m4_if( m4_ifdef([AC_PACKAGE_NAME], [ok]):m4_ifdef([AC_PACKAGE_VERSION], [ok]), [ok:ok],, [m4_fatal([AC_INIT should be called with package and version arguments])])dnl AC_SUBST([PACKAGE], ['AC_PACKAGE_TARNAME'])dnl AC_SUBST([VERSION], ['AC_PACKAGE_VERSION'])])dnl _AM_IF_OPTION([no-define],, [AC_DEFINE_UNQUOTED([PACKAGE], ["$PACKAGE"], [Name of package]) AC_DEFINE_UNQUOTED([VERSION], ["$VERSION"], [Version number of package])])dnl # Some tools Automake needs. AC_REQUIRE([AM_SANITY_CHECK])dnl AC_REQUIRE([AC_ARG_PROGRAM])dnl AM_MISSING_PROG([ACLOCAL], [aclocal-${am__api_version}]) AM_MISSING_PROG([AUTOCONF], [autoconf]) AM_MISSING_PROG([AUTOMAKE], [automake-${am__api_version}]) AM_MISSING_PROG([AUTOHEADER], [autoheader]) AM_MISSING_PROG([MAKEINFO], [makeinfo]) AC_REQUIRE([AM_PROG_INSTALL_SH])dnl AC_REQUIRE([AM_PROG_INSTALL_STRIP])dnl AC_REQUIRE([AC_PROG_MKDIR_P])dnl # For better backward compatibility. To be removed once Automake 1.9.x # dies out for good. For more background, see: # # AC_SUBST([mkdir_p], ['$(MKDIR_P)']) # We need awk for the "check" target (and possibly the TAP driver). The # system "awk" is bad on some platforms. AC_REQUIRE([AC_PROG_AWK])dnl AC_REQUIRE([AC_PROG_MAKE_SET])dnl AC_REQUIRE([AM_SET_LEADING_DOT])dnl _AM_IF_OPTION([tar-ustar], [_AM_PROG_TAR([ustar])], [_AM_IF_OPTION([tar-pax], [_AM_PROG_TAR([pax])], [_AM_PROG_TAR([v7])])]) _AM_IF_OPTION([no-dependencies],, [AC_PROVIDE_IFELSE([AC_PROG_CC], [_AM_DEPENDENCIES([CC])], [m4_define([AC_PROG_CC], m4_defn([AC_PROG_CC])[_AM_DEPENDENCIES([CC])])])dnl AC_PROVIDE_IFELSE([AC_PROG_CXX], [_AM_DEPENDENCIES([CXX])], [m4_define([AC_PROG_CXX], m4_defn([AC_PROG_CXX])[_AM_DEPENDENCIES([CXX])])])dnl AC_PROVIDE_IFELSE([AC_PROG_OBJC], [_AM_DEPENDENCIES([OBJC])], [m4_define([AC_PROG_OBJC], m4_defn([AC_PROG_OBJC])[_AM_DEPENDENCIES([OBJC])])])dnl AC_PROVIDE_IFELSE([AC_PROG_OBJCXX], [_AM_DEPENDENCIES([OBJCXX])], [m4_define([AC_PROG_OBJCXX], m4_defn([AC_PROG_OBJCXX])[_AM_DEPENDENCIES([OBJCXX])])])dnl ]) AC_REQUIRE([AM_SILENT_RULES])dnl dnl The testsuite driver may need to know about EXEEXT, so add the dnl 'am__EXEEXT' conditional if _AM_COMPILER_EXEEXT was seen. This dnl macro is hooked onto _AC_COMPILER_EXEEXT early, see below. AC_CONFIG_COMMANDS_PRE(dnl [m4_provide_if([_AM_COMPILER_EXEEXT], [AM_CONDITIONAL([am__EXEEXT], [test -n "$EXEEXT"])])])dnl # POSIX will say in a future version that running "rm -f" with no argument # is OK; and we want to be able to make that assumption in our Makefile # recipes. So use an aggressive probe to check that the usage we want is # actually supported "in the wild" to an acceptable degree. # See automake bug#10828. # To make any issue more visible, cause the running configure to be aborted # by default if the 'rm' program in use doesn't match our expectations; the # user can still override this though. if rm -f && rm -fr && rm -rf; then : OK; else cat >&2 <<'END' Oops! Your 'rm' program seems unable to run without file operands specified on the command line, even when the '-f' option is present. This is contrary to the behaviour of most rm programs out there, and not conforming with the upcoming POSIX standard: Please tell bug-automake@gnu.org about your system, including the value of your $PATH and any error possibly output before this message. This can help us improve future automake versions. END if test x"$ACCEPT_INFERIOR_RM_PROGRAM" = x"yes"; then echo 'Configuration will proceed anyway, since you have set the' >&2 echo 'ACCEPT_INFERIOR_RM_PROGRAM variable to "yes"' >&2 echo >&2 else cat >&2 <<'END' Aborting the configuration process, to ensure you take notice of the issue. You can download and install GNU coreutils to get an 'rm' implementation that behaves properly: . If you want to complete the configuration process using your problematic 'rm' anyway, export the environment variable ACCEPT_INFERIOR_RM_PROGRAM to "yes", and re-run configure. END AC_MSG_ERROR([Your 'rm' program is bad, sorry.]) fi fi dnl The trailing newline in this macro's definition is deliberate, for dnl backward compatibility and to allow trailing 'dnl'-style comments dnl after the AM_INIT_AUTOMAKE invocation. See automake bug#16841. ]) dnl Hook into '_AC_COMPILER_EXEEXT' early to learn its expansion. Do not dnl add the conditional right here, as _AC_COMPILER_EXEEXT may be further dnl mangled by Autoconf and run in a shell conditional statement. m4_define([_AC_COMPILER_EXEEXT], m4_defn([_AC_COMPILER_EXEEXT])[m4_provide([_AM_COMPILER_EXEEXT])]) # When config.status generates a header, we must update the stamp-h file. # This file resides in the same directory as the config header # that is generated. The stamp files are numbered to have different names. # Autoconf calls _AC_AM_CONFIG_HEADER_HOOK (when defined) in the # loop where config.status creates the headers, so we can generate # our stamp files there. AC_DEFUN([_AC_AM_CONFIG_HEADER_HOOK], [# Compute $1's index in $config_headers. _am_arg=$1 _am_stamp_count=1 for _am_header in $config_headers :; do case $_am_header in $_am_arg | $_am_arg:* ) break ;; * ) _am_stamp_count=`expr $_am_stamp_count + 1` ;; esac done echo "timestamp for $_am_arg" >`AS_DIRNAME(["$_am_arg"])`/stamp-h[]$_am_stamp_count]) # Copyright (C) 2001-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # AM_PROG_INSTALL_SH # ------------------ # Define $install_sh. AC_DEFUN([AM_PROG_INSTALL_SH], [AC_REQUIRE([AM_AUX_DIR_EXPAND])dnl if test x"${install_sh+set}" != xset; then case $am_aux_dir in *\ * | *\ *) install_sh="\${SHELL} '$am_aux_dir/install-sh'" ;; *) install_sh="\${SHELL} $am_aux_dir/install-sh" esac fi AC_SUBST([install_sh])]) # Copyright (C) 2003-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # Check whether the underlying file-system supports filenames # with a leading dot. For instance MS-DOS doesn't. AC_DEFUN([AM_SET_LEADING_DOT], [rm -rf .tst 2>/dev/null mkdir .tst 2>/dev/null if test -d .tst; then am__leading_dot=. else am__leading_dot=_ fi rmdir .tst 2>/dev/null AC_SUBST([am__leading_dot])]) # Check to see how 'make' treats includes. -*- Autoconf -*- # Copyright (C) 2001-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # AM_MAKE_INCLUDE() # ----------------- # Check to see how make treats includes. AC_DEFUN([AM_MAKE_INCLUDE], [am_make=${MAKE-make} cat > confinc << 'END' am__doit: @echo this is the am__doit target .PHONY: am__doit END # If we don't find an include directive, just comment out the code. AC_MSG_CHECKING([for style of include used by $am_make]) am__include="#" am__quote= _am_result=none # First try GNU make style include. echo "include confinc" > confmf # Ignore all kinds of additional output from 'make'. case `$am_make -s -f confmf 2> /dev/null` in #( *the\ am__doit\ target*) am__include=include am__quote= _am_result=GNU ;; esac # Now try BSD make style include. if test "$am__include" = "#"; then echo '.include "confinc"' > confmf case `$am_make -s -f confmf 2> /dev/null` in #( *the\ am__doit\ target*) am__include=.include am__quote="\"" _am_result=BSD ;; esac fi AC_SUBST([am__include]) AC_SUBST([am__quote]) AC_MSG_RESULT([$_am_result]) rm -f confinc confmf ]) # Fake the existence of programs that GNU maintainers use. -*- Autoconf -*- # Copyright (C) 1997-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # AM_MISSING_PROG(NAME, PROGRAM) # ------------------------------ AC_DEFUN([AM_MISSING_PROG], [AC_REQUIRE([AM_MISSING_HAS_RUN]) $1=${$1-"${am_missing_run}$2"} AC_SUBST($1)]) # AM_MISSING_HAS_RUN # ------------------ # Define MISSING if not defined so far and test if it is modern enough. # If it is, set am_missing_run to use it, otherwise, to nothing. AC_DEFUN([AM_MISSING_HAS_RUN], [AC_REQUIRE([AM_AUX_DIR_EXPAND])dnl AC_REQUIRE_AUX_FILE([missing])dnl if test x"${MISSING+set}" != xset; then case $am_aux_dir in *\ * | *\ *) MISSING="\${SHELL} \"$am_aux_dir/missing\"" ;; *) MISSING="\${SHELL} $am_aux_dir/missing" ;; esac fi # Use eval to expand $SHELL if eval "$MISSING --is-lightweight"; then am_missing_run="$MISSING " else am_missing_run= AC_MSG_WARN(['missing' script is too old or missing]) fi ]) # Helper functions for option handling. -*- Autoconf -*- # Copyright (C) 2001-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # _AM_MANGLE_OPTION(NAME) # ----------------------- AC_DEFUN([_AM_MANGLE_OPTION], [[_AM_OPTION_]m4_bpatsubst($1, [[^a-zA-Z0-9_]], [_])]) # _AM_SET_OPTION(NAME) # -------------------- # Set option NAME. Presently that only means defining a flag for this option. AC_DEFUN([_AM_SET_OPTION], [m4_define(_AM_MANGLE_OPTION([$1]), [1])]) # _AM_SET_OPTIONS(OPTIONS) # ------------------------ # OPTIONS is a space-separated list of Automake options. AC_DEFUN([_AM_SET_OPTIONS], [m4_foreach_w([_AM_Option], [$1], [_AM_SET_OPTION(_AM_Option)])]) # _AM_IF_OPTION(OPTION, IF-SET, [IF-NOT-SET]) # ------------------------------------------- # Execute IF-SET if OPTION is set, IF-NOT-SET otherwise. AC_DEFUN([_AM_IF_OPTION], [m4_ifset(_AM_MANGLE_OPTION([$1]), [$2], [$3])]) # Copyright (C) 1999-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # _AM_PROG_CC_C_O # --------------- # Like AC_PROG_CC_C_O, but changed for automake. We rewrite AC_PROG_CC # to automatically call this. AC_DEFUN([_AM_PROG_CC_C_O], [AC_REQUIRE([AM_AUX_DIR_EXPAND])dnl AC_REQUIRE_AUX_FILE([compile])dnl AC_LANG_PUSH([C])dnl AC_CACHE_CHECK( [whether $CC understands -c and -o together], [am_cv_prog_cc_c_o], [AC_LANG_CONFTEST([AC_LANG_PROGRAM([])]) # Make sure it works both with $CC and with simple cc. # Following AC_PROG_CC_C_O, we do the test twice because some # compilers refuse to overwrite an existing .o file with -o, # though they will create one. am_cv_prog_cc_c_o=yes for am_i in 1 2; do if AM_RUN_LOG([$CC -c conftest.$ac_ext -o conftest2.$ac_objext]) \ && test -f conftest2.$ac_objext; then : OK else am_cv_prog_cc_c_o=no break fi done rm -f core conftest* unset am_i]) if test "$am_cv_prog_cc_c_o" != yes; then # Losing compiler, so override with the script. # FIXME: It is wrong to rewrite CC. # But if we don't then we get into trouble of one sort or another. # A longer-term fix would be to have automake use am__CC in this case, # and then we could set am__CC="\$(top_srcdir)/compile \$(CC)" CC="$am_aux_dir/compile $CC" fi AC_LANG_POP([C])]) # For backward compatibility. AC_DEFUN_ONCE([AM_PROG_CC_C_O], [AC_REQUIRE([AC_PROG_CC])]) # Copyright (C) 2001-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # AM_RUN_LOG(COMMAND) # ------------------- # Run COMMAND, save the exit status in ac_status, and log it. # (This has been adapted from Autoconf's _AC_RUN_LOG macro.) AC_DEFUN([AM_RUN_LOG], [{ echo "$as_me:$LINENO: $1" >&AS_MESSAGE_LOG_FD ($1) >&AS_MESSAGE_LOG_FD 2>&AS_MESSAGE_LOG_FD ac_status=$? echo "$as_me:$LINENO: \$? = $ac_status" >&AS_MESSAGE_LOG_FD (exit $ac_status); }]) # Check to make sure that the build environment is sane. -*- Autoconf -*- # Copyright (C) 1996-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # AM_SANITY_CHECK # --------------- AC_DEFUN([AM_SANITY_CHECK], [AC_MSG_CHECKING([whether build environment is sane]) # Reject unsafe characters in $srcdir or the absolute working directory # name. Accept space and tab only in the latter. am_lf=' ' case `pwd` in *[[\\\"\#\$\&\'\`$am_lf]]*) AC_MSG_ERROR([unsafe absolute working directory name]);; esac case $srcdir in *[[\\\"\#\$\&\'\`$am_lf\ \ ]]*) AC_MSG_ERROR([unsafe srcdir value: '$srcdir']);; esac # Do 'set' in a subshell so we don't clobber the current shell's # arguments. Must try -L first in case configure is actually a # symlink; some systems play weird games with the mod time of symlinks # (eg FreeBSD returns the mod time of the symlink's containing # directory). if ( am_has_slept=no for am_try in 1 2; do echo "timestamp, slept: $am_has_slept" > conftest.file set X `ls -Lt "$srcdir/configure" conftest.file 2> /dev/null` if test "$[*]" = "X"; then # -L didn't work. set X `ls -t "$srcdir/configure" conftest.file` fi if test "$[*]" != "X $srcdir/configure conftest.file" \ && test "$[*]" != "X conftest.file $srcdir/configure"; then # If neither matched, then we have a broken ls. This can happen # if, for instance, CONFIG_SHELL is bash and it inherits a # broken ls alias from the environment. This has actually # happened. Such a system could not be considered "sane". AC_MSG_ERROR([ls -t appears to fail. Make sure there is not a broken alias in your environment]) fi if test "$[2]" = conftest.file || test $am_try -eq 2; then break fi # Just in case. sleep 1 am_has_slept=yes done test "$[2]" = conftest.file ) then # Ok. : else AC_MSG_ERROR([newly created file is older than distributed files! Check your system clock]) fi AC_MSG_RESULT([yes]) # If we didn't sleep, we still need to ensure time stamps of config.status and # generated files are strictly newer. am_sleep_pid= if grep 'slept: no' conftest.file >/dev/null 2>&1; then ( sleep 1 ) & am_sleep_pid=$! fi AC_CONFIG_COMMANDS_PRE( [AC_MSG_CHECKING([that generated files are newer than configure]) if test -n "$am_sleep_pid"; then # Hide warnings about reused PIDs. wait $am_sleep_pid 2>/dev/null fi AC_MSG_RESULT([done])]) rm -f conftest.file ]) # Copyright (C) 2009-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # AM_SILENT_RULES([DEFAULT]) # -------------------------- # Enable less verbose build rules; with the default set to DEFAULT # ("yes" being less verbose, "no" or empty being verbose). AC_DEFUN([AM_SILENT_RULES], [AC_ARG_ENABLE([silent-rules], [dnl AS_HELP_STRING( [--enable-silent-rules], [less verbose build output (undo: "make V=1")]) AS_HELP_STRING( [--disable-silent-rules], [verbose build output (undo: "make V=0")])dnl ]) case $enable_silent_rules in @%:@ ((( yes) AM_DEFAULT_VERBOSITY=0;; no) AM_DEFAULT_VERBOSITY=1;; *) AM_DEFAULT_VERBOSITY=m4_if([$1], [yes], [0], [1]);; esac dnl dnl A few 'make' implementations (e.g., NonStop OS and NextStep) dnl do not support nested variable expansions. dnl See automake bug#9928 and bug#10237. am_make=${MAKE-make} AC_CACHE_CHECK([whether $am_make supports nested variables], [am_cv_make_support_nested_variables], [if AS_ECHO([['TRUE=$(BAR$(V)) BAR0=false BAR1=true V=1 am__doit: @$(TRUE) .PHONY: am__doit']]) | $am_make -f - >/dev/null 2>&1; then am_cv_make_support_nested_variables=yes else am_cv_make_support_nested_variables=no fi]) if test $am_cv_make_support_nested_variables = yes; then dnl Using '$V' instead of '$(V)' breaks IRIX make. AM_V='$(V)' AM_DEFAULT_V='$(AM_DEFAULT_VERBOSITY)' else AM_V=$AM_DEFAULT_VERBOSITY AM_DEFAULT_V=$AM_DEFAULT_VERBOSITY fi AC_SUBST([AM_V])dnl AM_SUBST_NOTMAKE([AM_V])dnl AC_SUBST([AM_DEFAULT_V])dnl AM_SUBST_NOTMAKE([AM_DEFAULT_V])dnl AC_SUBST([AM_DEFAULT_VERBOSITY])dnl AM_BACKSLASH='\' AC_SUBST([AM_BACKSLASH])dnl _AM_SUBST_NOTMAKE([AM_BACKSLASH])dnl ]) # Copyright (C) 2001-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # AM_PROG_INSTALL_STRIP # --------------------- # One issue with vendor 'install' (even GNU) is that you can't # specify the program used to strip binaries. This is especially # annoying in cross-compiling environments, where the build's strip # is unlikely to handle the host's binaries. # Fortunately install-sh will honor a STRIPPROG variable, so we # always use install-sh in "make install-strip", and initialize # STRIPPROG with the value of the STRIP variable (set by the user). AC_DEFUN([AM_PROG_INSTALL_STRIP], [AC_REQUIRE([AM_PROG_INSTALL_SH])dnl # Installed binaries are usually stripped using 'strip' when the user # run "make install-strip". However 'strip' might not be the right # tool to use in cross-compilation environments, therefore Automake # will honor the 'STRIP' environment variable to overrule this program. dnl Don't test for $cross_compiling = yes, because it might be 'maybe'. if test "$cross_compiling" != no; then AC_CHECK_TOOL([STRIP], [strip], :) fi INSTALL_STRIP_PROGRAM="\$(install_sh) -c -s" AC_SUBST([INSTALL_STRIP_PROGRAM])]) # Copyright (C) 2006-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # _AM_SUBST_NOTMAKE(VARIABLE) # --------------------------- # Prevent Automake from outputting VARIABLE = @VARIABLE@ in Makefile.in. # This macro is traced by Automake. AC_DEFUN([_AM_SUBST_NOTMAKE]) # AM_SUBST_NOTMAKE(VARIABLE) # -------------------------- # Public sister of _AM_SUBST_NOTMAKE. AC_DEFUN([AM_SUBST_NOTMAKE], [_AM_SUBST_NOTMAKE($@)]) # Check how to create a tarball. -*- Autoconf -*- # Copyright (C) 2004-2014 Free Software Foundation, Inc. # # This file is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # _AM_PROG_TAR(FORMAT) # -------------------- # Check how to create a tarball in format FORMAT. # FORMAT should be one of 'v7', 'ustar', or 'pax'. # # Substitute a variable $(am__tar) that is a command # writing to stdout a FORMAT-tarball containing the directory # $tardir. # tardir=directory && $(am__tar) > result.tar # # Substitute a variable $(am__untar) that extract such # a tarball read from stdin. # $(am__untar) < result.tar # AC_DEFUN([_AM_PROG_TAR], [# Always define AMTAR for backward compatibility. Yes, it's still used # in the wild :-( We should find a proper way to deprecate it ... AC_SUBST([AMTAR], ['$${TAR-tar}']) # We'll loop over all known methods to create a tar archive until one works. _am_tools='gnutar m4_if([$1], [ustar], [plaintar]) pax cpio none' m4_if([$1], [v7], [am__tar='$${TAR-tar} chof - "$$tardir"' am__untar='$${TAR-tar} xf -'], [m4_case([$1], [ustar], [# The POSIX 1988 'ustar' format is defined with fixed-size fields. # There is notably a 21 bits limit for the UID and the GID. In fact, # the 'pax' utility can hang on bigger UID/GID (see automake bug#8343 # and bug#13588). am_max_uid=2097151 # 2^21 - 1 am_max_gid=$am_max_uid # The $UID and $GID variables are not portable, so we need to resort # to the POSIX-mandated id(1) utility. Errors in the 'id' calls # below are definitely unexpected, so allow the users to see them # (that is, avoid stderr redirection). am_uid=`id -u || echo unknown` am_gid=`id -g || echo unknown` AC_MSG_CHECKING([whether UID '$am_uid' is supported by ustar format]) if test $am_uid -le $am_max_uid; then AC_MSG_RESULT([yes]) else AC_MSG_RESULT([no]) _am_tools=none fi AC_MSG_CHECKING([whether GID '$am_gid' is supported by ustar format]) if test $am_gid -le $am_max_gid; then AC_MSG_RESULT([yes]) else AC_MSG_RESULT([no]) _am_tools=none fi], [pax], [], [m4_fatal([Unknown tar format])]) AC_MSG_CHECKING([how to create a $1 tar archive]) # Go ahead even if we have the value already cached. We do so because we # need to set the values for the 'am__tar' and 'am__untar' variables. _am_tools=${am_cv_prog_tar_$1-$_am_tools} for _am_tool in $_am_tools; do case $_am_tool in gnutar) for _am_tar in tar gnutar gtar; do AM_RUN_LOG([$_am_tar --version]) && break done am__tar="$_am_tar --format=m4_if([$1], [pax], [posix], [$1]) -chf - "'"$$tardir"' am__tar_="$_am_tar --format=m4_if([$1], [pax], [posix], [$1]) -chf - "'"$tardir"' am__untar="$_am_tar -xf -" ;; plaintar) # Must skip GNU tar: if it does not support --format= it doesn't create # ustar tarball either. (tar --version) >/dev/null 2>&1 && continue am__tar='tar chf - "$$tardir"' am__tar_='tar chf - "$tardir"' am__untar='tar xf -' ;; pax) am__tar='pax -L -x $1 -w "$$tardir"' am__tar_='pax -L -x $1 -w "$tardir"' am__untar='pax -r' ;; cpio) am__tar='find "$$tardir" -print | cpio -o -H $1 -L' am__tar_='find "$tardir" -print | cpio -o -H $1 -L' am__untar='cpio -i -H $1 -d' ;; none) am__tar=false am__tar_=false am__untar=false ;; esac # If the value was cached, stop now. We just wanted to have am__tar # and am__untar set. test -n "${am_cv_prog_tar_$1}" && break # tar/untar a dummy directory, and stop if the command works. rm -rf conftest.dir mkdir conftest.dir echo GrepMe > conftest.dir/file AM_RUN_LOG([tardir=conftest.dir && eval $am__tar_ >conftest.tar]) rm -rf conftest.dir if test -s conftest.tar; then AM_RUN_LOG([$am__untar /dev/null 2>&1 && break fi done rm -rf conftest.dir AC_CACHE_VAL([am_cv_prog_tar_$1], [am_cv_prog_tar_$1=$_am_tool]) AC_MSG_RESULT([$am_cv_prog_tar_$1])]) AC_SUBST([am__tar]) AC_SUBST([am__untar]) ]) # _AM_PROG_TAR m4_include([acinclude.m4]) stacks-1.44/autogen.sh000755 000765 000024 00000000272 12335173442 015510 0ustar00catchenstaff000000 000000 #!/bin/sh -e test -n "$srcdir" || srcdir=`dirname "$0"` test -n "$srcdir" || srcdir=. autoreconf --force --install --verbose "$srcdir" test -n "$NOCONFIGURE" || "$srcdir/configure" "$@"stacks-1.44/ChangeLog000644 000765 000024 00000212102 13001006117 015237 0ustar00catchenstaff000000 000000 Stacks 1.44 - Oct 11, 2016 -------------------------- Bugfix: corrected an error in pstacks where '=' and 'X' symbols were not recognized properly in SAM/BAM CIGAR strings. Bugfix: corrected some typos in pstacks/populations help output. Stacks 1.43 - Oct 05, 2016 -------------------------- Feature: added alignment controls to pstacks, allowing the program to discard secondary alignments and to discard alignments where a significant portion of the read was not aligned (soft-masked). Bugfix: corrected a very small memory leak in the gapped alignment code, found by Valgrind. Feature: updated configure test to check if compiler can handle c++11 standard. Bugfix: rxstacks was not generating model files. Bugfix: corrected an uncaught exception in cstacks when processing gapped alignments. In some cases when a multiple alignment had to be recomputed the initial CIGAR string was not parsed properly leading to the catalog and query sequences coming out of sync in their length (which could throw the exception). Feature: reduced memory usage in ustacks and pstacks by not retaining all reads from a collapsed locus. Bugfix: corrected -V option for populations, which was causing a crash (although --in_vcf worked). Stacks 1.42 - Aug 05, 2016 -------------------------- Feature: Added Csp6I restriction enzyme. Feature: populations program is now able to calculate populations statistics using arbitrary VCF files as input. Feature: Upgraded to the latest release of HTSLib (1.3.1) for reading BAM files. Embedded the library in the Stacks distribution to remove previous libbam dependency. Feature: Added an output directory option to 'populations' (--out_path). Feature: Added restriction enzymes BsaHI, HpaII, NcoI; corrected NdeI. Bugfix: Made the VCF output by 'populations' more standard-compliant. Bugfix: Some output files included 0-based genomic coordinates, changed them to 1-based. Bugfix: Replaced populations IDs with populations names in 'populations' output. Bugfix: Corrected a bug affecting clone_filter when input was non-gzipped paired-end data. Stacks 1.41 - June 22, 2016 --------------------------- Bugfix: the kernel-smoothing procedure in populations (used for Fst, Pi, heterozygosity etc. smoothing) was not functioning at sizes larger than the default size. A bug was creating incorrect weights for the smoothing operation when the sliding window size was set to a large value causing the smoothing window to have a maximum size after which increasing the size did not change the smoothing. Bugfix: cstacks was reporting gapped alignments even when --gapped was not enabled. This affected a small number of (mostly) confounded catalog loci. Feature: Added the Csp6I restriction enzyme. Stacks 1.40 - May 04, 2016 -------------------------- Feature: Changed process_radtags and process_shortreads to print FASTQ/FASTA headers using "/1" and "/2" to represent the read number, instead of "_1" and "_2". Bugfix: fixed a regression where allele depths were not being loaded due to the use of the new *.models.tsv file. This file lacks the raw reads and therefore we can't count the raw stack depth when running sstacks. Bugfix: cstacks was calling errant SNPs in loci with a sample containing one gapped locus and one ungapped locus matching the same catalog locus. Stacks 1.39 - April 23, 2016 ---------------------------- Bugfix: rxstacks was not adjusting reads/SNPs to account for alignment gaps. There was also an bug in reading the input files. Bugfix: denovo_map.pl and ref_map.pl were not processing parents/progeny properly. Stacks 1.38 - April 18, 2016 ---------------------------- Feature: denovo_map.pl and ref_map.pl now print depth of coverage for each sample. The ustacks program now prints depth of coverage after each algorithm stage to see how each stage improves (or not) the depth of coverage. Feature: complete refactoring of denovo_map.pl and ref_map.pl. Separated computation from SQL loading. Added auto creation/deletion of database. Enabled samples to be read from population map instead of specifying them on the command line. Feature: added Needleman–Wunsch algorithm to ustacks, cstacks, sstacks to provide for gapped alignments. Includes --max_gaps and --min_aln_len parameters to contain crazy alignments. sstacks now includes a CIGAR string describing the alignment to the catalog. Feature: optimized ustacks for a 33% decrease in run time. Feature: added new file, sample_X.models.tsv.gz, produced by ustacks and pstacks. Contains a subset of the information in the sample_X.tags.tsv.gz file, allows for data to be loaded much faster in the later stages of the pipeline, greatly speeding up run times. Bugfix: added code to prevent populations from improperly reading SNP positions past the length of a particluar locus (that is shorter than the catalog locus). Bugfix: corrected bug in process_radtags when using inline barcodes on paired-end reads. The paired- end reads were not being truncated uniformly. Bugfix: corrected bug in populations where if enough empty files were fed into the program it could place files in the wrong population or segfault. Bugfix: corrected PHP files for exporting to include LnL filter. Bugfix: corrected mappable markers filter in web interface. Stacks 1.37 - Feb 24, 2016 -------------------------- Feature: converted PHP database code from MDB2 to MySQLi. MDB2 is no longer a prerequisite for installing Stacks. Stacks 1.36 - Feb 18, 2016 -------------------------- Feature: Added the BfaI, BspDI, AseI, and AciI restriction enzymes to process_radtags. Feature: Changed the way denovo_map.pl and ref_map.pl run sstacks. It is now set to run sstacks once for all samples, instead of one time per sample. Should provide a significant speed-up. Bugfix: corrected error in pstacks when handling long reads with complex SAM/BAM alignments. Bugfix: fixed memory leak in sstacks when more than one sample file was specified. Bugfix: corrected error in clone_filter causing it to fail when processing gzipped data without a random oligo attached. Bugfix: corrected error when reading gzipped FASTA files and the last sequence of the file was improperly doubled in length. Stacks 1.35 - Sept 09, 2015 --------------------------- Feature: Added --retain_header flag to process_radtags/process_shortreads which will keep the unmodified FASTQ header in the output. This allows clone_filter/process_radtags/ process_shortreads to be run in different sequences and more than one time. Feature: Added --treemix to the populations program, allowing SNPs to be output in TreeMix format. Feature: Added --phylip_var_all to the populations program. This option outputs the full sequence from each variable locus, encoding polymorphisms using IUPAC notation. -This option will also output a file containing the coordinates of each RAD locus so they can be input to phylogenetic software (such as RAxML) to partition each RAD locus out and then build the phylogenetic tree independently for each partitioned locus. Feature: Added the AgeI restriction enzyme. Feature: refactored clone_filter to handle random oligo sequences used as inline/indexed barcodes to identify and discard PCR duplicates. Bugfix: added code to process_radtags/process_shortreads to handle cases when data writes fail due to a filled disk or other error conditions. Bugfix: kmer_filter was not handling gzipped FASTQ files properly when filtering rare kmers. Stacks 1.34 - July 26, 2015 --------------------------- Bugfix: fixed phylip output to again include nucleotides from subsets of the full set of populations. Bugfix: private alleles were being associated to the incorrect population at a particular locus (the counts and summary statistics of private alleles were not affected). Stacks 1.33 - July 22, 2015 --------------------------- Bugfix: Corrected the second-stage filtering of the populations program to properly respect the -p flag. Bugfix: Corrected the display of individual samples in the web interface (tags.php file). Stacks 1.32 - June 18, 2015 --------------------------- Bugfix: Updated the Phylip output to reflect the changed meaning of 'fixed' as determined in the PopSum::tally() function. Stacks 1.31 - June 17, 2015 --------------------------- Bugfix: site-level filtering in the populations program was not working correctly when dealing with sites that were fixed within populations but variable among populations. The code in the PopSum::tally() function was not correctly identifying sites as not fixed in these cases causing them to be incorrectly filtered out. Bugfix: --write_random_snp was causing a segfault in the populations program in some cases. Feature: changed the default setting for the -n option of cstacks (number of fixed differences allowed between loci) to 1 (at the request of Josie Paris ). Bugfix: made some tweaks to improve layout in the web interface. Bugfix: single-end reads, with paired barcodes (inline/index) were not being handled properly, resulting in a segfault. Bugfix: process_radtags was allowing a non-null barcode type to be specified without specifying a barcode file, which caused a segfault. Feature: exposed kmer length setting in ustacks and cstacks. This allows the kmer length used for sequence matching to be set manually. While this can result in some missed matches (there is a trade off between kmer length and sequence length when searching for matches between the two) it also allows the algorithm to run at faster speeds. Feature: Changed default database engine type to be excplicitly MyISAM. Previously Stacks just used the default which at one time was MyISAM but has recently changed in many systems to be INNODB. Using MyISAM should provide much faster imports of data and ultimately use less disk space (as the space is reclaimed when databases are deleted). Stacks 1.30 - May 07, 2015 -------------------------- Feature: sstacks can now accept multiple sample files at a time, saving run time by only processing the catalog once. Feature: changed batch_X.sumstats.tsv file so the P/Q alleles are always presented in the same order in each local population (according to the overall frequency of the allele across all populations). This will sync results with the VCF exports but will sometimes cause the frequency of p in the local population to be less than 0.5 (up until now the frequency of p has always represented the most frequenct allele in the local population). Feature: added an maximum observed heterozygosity filter to populations program. Bugfix: Fis values in batch_X.sumstats_summary.tsv were incorrect (although raw values in batch_X.sumstats.tsv were correct). Bugfix: corrected the allele depth output in the VCF export to follow defacto standards used by other programs. Bugfix: in some cases loci were sneaking past the --write_single_snp directive in in populations (due to interactions with pruning out SNPs that fail the MAF filter). Feature: Updated the Stacks web interface. The web app is now almost 100% dynamic (parts of the page are draws on demand instead of fetching new, full pages from the server) using local javascript to draw the population view of genotypes, summary statistics, and the view of raw stacks. The web app uses asynchronous AJAX queries that trade data encoded in JSON to fetch the necessary data for dynamic display. Feature: added DdeI, RsaI, AluI restriction enzymes to process_radtags. Bugfix: sstacks could generate extra matching haplotpyes in a very small number of cases. Stacks 1.29 - Mar 21, 2015 -------------------------- Feature: added the --ordered_export option to the populations program. For the VCF, GenePop, and Structure exports, if this option is specified, only one copy of each SNP is exported in the case where one nucleotide position is covered by more than one RAD locus. Most useful for ddRAD data. Feature: VCF export now includes individual allele depths for each SNP call. Feature: improved the filtering logging code in populations, if the --verbose flag is specified, a reason is provided for each pruned site, or each removed locus. Bugfix: PHASE output was broken in the populations program. SNP pruning/filtering code did not update the catalog copies of the alleles after pruning which are needed by the PHASE output code. Bugfix: adjusted the filtering code in populations to not exclude fixed loci. Bugfix: removed extra tab from ID line for Structure export. Bugfix: fixed issue in genepop output that may have overfiltered some loci. Bugfix: fixed small problems with --write_single_snp/--write_random_snp in the populations program. Some polymorphic loci were erroneously being omitted. Stacks 1.28 - Mar 06, 2015 -------------------------- Feature: added a second barcode distance to process_radtags/process_shortreads. This allows you to specify two distances for recovering barcodes if you are using combinatorial barcodes (e.g. a 12bp barcode inline on the single-end read plus a 6bp index). I have changed the meaning of the parameter from "distance between barcodes" to "number of allowed mismatches when correcting barcodes." The --barcode_dist parameter is now --barcode_dist_1, and --barcode_dist_2 was added. Bugfix: the process_shortreads/process_radtags programs were trimming sequence as if an inline barcode was specified, even when it was an index barcode and no sequence should have been trimmed. Bugfix: the process_shortreads program was outputting FASTA even when FASTQ was requested due to not handling gzipped outputs properly. Bugfix: fixed segfault in populations that could occur when using a whitelist that contained loci that were being filtered out due to -p/-r constraints. Stacks 1.27 - Feb 25, 2015 -------------------------- Bugfix: the minor allele frequency filter and the proceny limit filter were not working properly in all cases with the other filters. Bugfix: barcode length (href->inline_bc_len) was not being correctly set for single-end, inline line barcodes of variable length. Stacks 1.26 - Feb 23, 2015 -------------------------- Bugfix: if you are running non-compressed data, then version 1.25 broke the parsing code. If your data were zipped (or a BAM file) when it went through pstacks/ustacks, then there was no bug. Feature: refactored the filtering code in the populations program to add a second filtering step. In previous versions the -r (sample limit) and -p (population limit) were applied on the basis of the entire RAD locus. This could lead to situations where a RAD locus remained in the data set while one or more of the individual SNPs on that locus were missing data and were below the -r or -p limits. Now, the filters are applied to individual SNPs after the filters are applied to the RAD loci. This greatly affects the -r (sample limit) filter with more SNPs being pruned out, as well as the -a (minor allele frequency filter) such that all SNPs below the MAF are pruned fully from the data set and will not appear in any statistical results or downstream exports. Feature: added restriction enzyme kpnI. Feature: added code to check for the existence of the loci and SNPs provided in a whitelist. Stacks 1.25 - Feb 17, 2015 -------------------------- Feature: added support for unaligned BAM files for process_radtaags and process_shortreads. The two programs can now read paired data that is interleaved in a single file (which is required to support paired-end data in BAM format). Feature: Haplotypes can now be output in VCF format from the populations program using the --vcf_haplotypes option. Feature: added --fasta_strict option to populations program. Will output full sequence for each individual at each haplotype at each locus, but only for biologically plausible loci. It won't output loci with more than two haplotypes and will output single haplotypes twice, once per allele. Feature: Changed the sumstats/hapstats files to output a one-based genome base pair position so it matches other export formats. Bugfix: fixed problem with gzipped files where last line of file was not read properly causing the program to output an erroneous error message. Bugfix: The FASTA output from the populations program was reporting the internal value (zero-based index) for the basepair position of each read (the first nucleotide of the cutsite) causing an off-by-one error for all reads and reads on the negative strand had the coordinate for the cutsite end of the read (right-most end) reported instead of the standard left-most end. Bugfix: the log likelihood filter was not working properly in export_sql.pl, causing many genotypes to be excluded during export. Bugfix: process_radtags was not looking for the paired-end RAD cutsite in the proper location when dealing with double-digest, inline/index barcoded reads. Feature: added initial, internal support for merging and phasing loci that overlap at a restriction enzyme cut site. Feature: code now prints program version and generation date to all internal Stacks files. Stacks 1.24 - Jan 07, 2015 -------------------------- Feature: added restriction enzyme ecoRV. Bugfix: fixed segmentation fault in process_radtags/process_shortreads when resizing sequence and phred internal buffer sizes. Stacks 1.23 - Dec 12, 2014 -------------------------- Bugfix: Fixed a segfault bug in process_radtags where the process_barcode function returned prematurely when one barcode was correct and one was incorrect in paired cases. Bugfix: fixed compiler warnings when building with CLANG. Stacks 1.22 - Dec 08, 2014 -------------------------- Feature: process_radtags and process_shortreads now support variable barcode lengths. In process_radtags sequences will automatically be trimmed to keep stacks a uniform length with the variable barcode lengths. Feature: a filename can now be specified in the barcodes file for process_radtags and process_shortreads. When a filename is specified, process_radtags will write data to this filename instead of a filename made up of the barcode. Feature: process_radtags and process_shortreads will now output gzipped files if provided gzipped inputs or if requested using the '-y' option. Feature: Added SacI and BgIII restriction enzymes. Bugfix: Tightened up parsing of FASTQ ID field to prevent a buffer overrun (and subsequent segfault) in FASTQ headers that look like the Illumina format but are malformed. Bugfix: Fixed GenePop output of populations program as last locus on second line was missing commas if more than one SNP was present at that locus. Bugfix: -R option to retain unused reads was not being recognized by ustacks. Bugfix: changed populations to record program run parameters and execution time to log file. Bugfix: corrected Makefile.am to include Sparsehash compile flags for process_radtags. Bugfix: corrected load_radtags.pl so as not to try and load the population ID as a number to the samples table (and instead as a string). Stacks 1.21 - Oct 02, 2014 -------------------------- Feature: Added the XbaI, BstYI, and XhoI restriction enzymes. Feature: Added ability to specify column position in whitelist along with locus ID in populations program. This allows for specific SNPs within specific loci to be processed. Feature: In populations program, changed implementation of --write_single_snp to create an internal whitelist from the first SNP in each catalog locus. Added a new command line option, --write_random_snp to select a single, random SNP per RAD locus using the same internal mechanism. Feature: Added HZAR, Hybrid Zone Analysis in R output to populations program. Bugfix: Added code in populations program to handle cases where a haplotype contains one or more uncalled bases (Ns). These haplotypes are now excluded from haplotype-based statistical calculations. Bugfix: In Phi_st/ct/sc calculations of populations program, total population count was not adjusted downward when one of the populations dropped out of the analysis at a particular locus in the all-populations, haplotype-based AMOVA calculation (batch_X.phistats.tsv). Bugfix: "All positions" Fis measure in batch_X.sumstats_summary.tsv file too negative due to internal logic error. Bugfix: updated queries in index_radtags.pl to account for new 'type' variable in SNPs tables. Stacks 1.20 - Jul 29, 2014 -------------------------- Synced corrections module branch with main Stacks branch. *** The internal formats of the *.tags.tsv, *.snps.tsv, and *.matches.tsv files have changed and therefore version 1.20 programs cannot be used on earlier generated data sets. However, the convert_stacks.pl script is included in this release to convert an older data set into the new formats. *** Feature: Implemented new haplotype trimming algorithm for rxstacks. Feature: new script, convert_stacks.pl, to convert previous Stacks files to new format. Feature: Modified VCF output to include likelihood values from heterozygous and homozygous SNP model calls. Feature: added log likelihood filter to genotypes and populations programs and to web interface. Feature: Added SpeI restriction enzyme to process_radtags. Feature: Modified Beagle output formats in populations program to be population-specific and not to include monomorphic nucleotide positions. Stacks 1.19 - Apr 23, 2014 -------------------------- Feature: the populations program now calculates Fst' and D_est on haplotypes between all pairwise populations. Our implementations are based on: Bird, Karl, Smouse & Toonen. (2011) Detecting and measuring genetic differentiation. D_est: Jost. (2008) Gst and its relatives do not measure differentiation. Fst': based on modifying the AMOVA implementation from Excoffier, Smouse, & Quattro (1992). Feature: we have refactored the populations program to use a common framework for kernel smoothing and bootstrapping. This has allowed us to add smoothing and bootstrapping to all statistics calculated by the populations program: pi, Fis, Fst, Fst', D_est, Phi_st, Phi_ct, Phi_sc, Haplotype diversity, gene diversity. Feature: we have implemented fine-grained control of bootstrapping by providing flags to turn on bootstrapping for each group of population statistics, as well as providing a bootstrapping whitelist allowing only certain loci to be included in the bootstrapping calculations. Stacks 1.18 - Apr 04, 2014 -------------------------- Feature: we now use chi squared segregation ratios to detect missing alleles in parental mapping markers. in F1 crosses (CP map type). We can now map ab/a- and -a/ab as ab/--, and --/ab markers; we can map ab/c- and -c/ab markers as ab/cd markers; we can map aa/b- and -a/bb markers as ab/-- and --/ab markers. Feature: in F1 crosses we are now mapping ab/cc and cc/ab markers as ab/-- and --/ab markers. Feature: reworked genetic map display of web interface. Included chisq p-value from segregation distortion test as a filter. Feature: implemented measure of segregation distortion in genotypes program based on chi square test of genotype counts. Removed deprecated measure of F, inbreeding coefficient, replaced it with segregation distortion. Bugfix: corrected calling of markers in genotypes program. When a whitelist with a small number of markers is specified, some of the parental IDs could be missed, causing markers not to be called and hence dropped from the analysis. Bugfix: changed genotype mappings for generic map types to make certain non-biologically plausible genotype combinations illegal. Bugfix: fixed compilation issues when using Google's SparseHash (thanks to khuck@cs.uoregon.edu for the patch). Stacks 1.17 - Mar 26, 2014 -------------------------- Bugfix: Added #ifdefs to deal with missing functions in older versions of zlib. Stacks 1.16 - Mar 25, 2014 -------------------------- Feature: added haplotype counts for each population and locus to the batch_X.hapstats.tsv file. Feature: haplotype F statistics are now calculated for the whole set of populations (one analysis of variance calculation for all populations), and also as a set of pairwise calculations to mirror the existing Fst calculations. Bugfix: fixed small bug in calculation of MSD(Total) component of Phi_st (haplotype F statistics). Bugfix: fixed bug in parsing of populations maps when using strings for population identifiers. Bugfix: kernel-smoothing not correct for haplotype/gene diversity. Stacks 1.15 - Mar 15, 2014 -------------------------- Bugfix: fix various bugs related to gzip support. Stacks 1.14 - Mar 14, 2014 -------------------------- Feature: Stacks files are now kept in gzipped format if FASTQ data is fed into pipeline gzipped or as a BAM. Bugfix: fixed some compile bugs on OSX Mavericks. Stacks 1.13 - Feb 24, 2014 -------------------------- Feature: We have implemented the first set of haplotype-level population genetics statistics. Specifically, we are now calculating gene diversity and haplotype diversity (pi) for each locus, as well as F statistics for haplotypes including, Phi_st, Phi_ct, and Phi_sc, which are calculated using Analysis of Molecular Variance (AMOVA): Excoffier, Smouse, & Quattro, (1992). Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics. Data can be analyzed as populations of individuals (the previous default) and now using populations of individuals, and groups of populations. Feature: If a reference genome is available, haplotype F statistics can also be kernel-smoothed. Feature: populations in the population map can now be specified as text strings or numbers. Groups of populations can now be specified by adding a third column to the population map for each individual and listing the group they belong to (again as a text string or number). Bugfix: allow batch IDs of 0 in populations and genotypes. Bugfix: in populations, changed VCF output to be ordered by basepair. Bugfix: in populations, change value of expected homozygosity to be set to 1 - expected heterozygosity instead of 1 - Pi. Pi (computed as [1 - ((p choose 2) + (q choose 2) / (n choose 2))] and expected heterozygosity (2pq) can produce sligthly different estimates resulting in exp het + exp hom != 1. Stacks 1.12 - Jan 21, 2014 -------------------------- Bugfix: accidentally broke gzipped FASTQ support through a typo in gzFastq.h. Stacks 1.11 - Jan 09, 2014 -------------------------- Feature: changed build to work properly with g++ and clang, which is the native compiler on Apple's OS X. Feature: Added NheI restriction enzyme. Bugfix: changed logging in denovo_map.pl/ref_map.pl to write outputs from Stacks programs continuously instead of waiting until the program completed to write output to log file. Bugfix: corrected parsing of population map for gzipped input files for denovo_map.pl. Stacks 1.10 - Dec 10, 2013 -------------------------- Feature: Added phased output for PHASE and Beagle. The phased output writes multiple SNPs in a single RAD locus as an already phased haplotype, leaving PHASE and Beagle to only phase between these haplotypes, instead of having to re-phase SNPs from within the same RAD site. Bugfix: corrected the SNP genotype output for Beagle. Bugfix: Corrected PHP warnings; enabled scrolling in catalog.php for iframes. Bugfix: allele percentages from ustacks were off since ustacks was changed to load/unload read IDs from disk (Stacks 0.99995). Only the calculation of the percentages was affected, not the underlying algorithms. Stacks 1.09 - Oct 30, 2013 -------------------------- Feature: added export support for F2 and backcross map types for Onemap to genotypes. Feature: added EaeI, ClaI, and TaqI restriction enzymes to process_radtags. Feature: changed populations bootstrap to use AMOVA Fst. Feature: added bootstrap whitelist to populations, so users can restrict the loci that are bootstrapped to a particular set (e.g. on a single chromosome). Bugfix: modified PHASE output so that SNPs are ordered properly. Previously, although RAD loci are ordered properly, some individual SNPs between RAD loci could still be output out of order. Bugfix: corrected onemap CP output so that B3.7 markers are output as "ab", not "2ab". Stacks 1.10.Beta1 - Sept 30, 2013 --------------------------------- Feature: completed implementation of rxstacks. Bugfix: when merging a homozygous locus into the catalog, if homozygous allele conflicted with existing catalog SNP alleles, new allele was not added to SNP object (but was added to the allele list). Bugfix: found small memory leak in cstacks - old SNP objects were not being freed when new SNPs were merged into the catalog. Bugfix: empty alleles were being output to the batch_X.catalog.alleles file by cstacks. Did not affect the function of the program. Stacks 1.08 - Sept 24, 2013 --------------------------- Feature: added a FASTA output to populations to output the full locus sequence for each allele at each sample locus, applying any filters or whitelists supplied to populations. Stacks 1.07 - Sept 23, 2013 --------------------------- Bugfix: updated process_radtags to drop reads shorter than length limit when read trimming turned on. Bugfix: corrected build failures on Mac OS X due to Samtools' bam.h header conflicting with Stacks' Bam.h header when building on OS X's case insensitive file system. Feature: changed process_radtags to drop reads already shorter than limit if sequence truncation turned on. You can also specify the read length limit to drop reads if your data have already been trimmed. Bugfix: Updated VCF ouput, missing genotypes now reported as "./." instead of "." Bugfix: Updated VCF ouput, alleles reported on the negative strand are now complemented so their positive strand conterparts are reported and will align aginst a reference genome. Bugfix: Updated VCF ouput, "reference allele" is now always reported as most frequent allele. Stacks 1.06 - August 28, 2013 ----------------------------- Bugfix: Illumina FASTQ header specifying read pair could override internal enumeration of read pair if paired-end data was fed in as a single-end file. Bugfix: corrected locus starting base in reference-aligned data. Feature: refactored sort_read_pairs.pl to process input files one at a time, without retaining them in memory. The program should now be able to handle an arbitrary number of samples. Feature: sort_read_pairs.pl can now read gzipped files directly. Stacks 1.05 - August 17, 2013 ----------------------------- Bugfix: adapter filtering code in process_radtags/process_shortreads bit rotted and was not properly functioning. Switching from deprecated hash function to TR1 hash broke the expected hashing behavior for char *. Bugfix: modified process_radtags/process_shortreads to handle single adapters when processing paired-end data (previously you had to specify two adapters for paired data). Bugfix: corrected barcode-specific counters in process_radtags/process_shortreads. Overall counts were correct but counts for barcodes were off due to shuffling of code that happened with support of combinatorial barcodes. Stacks 1.04 - July 25, 2013 --------------------------- Bugfix: process_radtags was not properly handling index_index and inline_inline barcode types. Bugfix: the hindIII restriction enzyme sequence was incorrectly specified in renz.h. Bugfix: ustacks wasn't properly removing file suffix when gzip files are processed. Stacks 1.03 - June 28, 2013 --------------------------- Bugfix: non-barcoded data were not being handled properly by process_radtags/process_shortreads. Stacks 1.02 - June 24, 2013 --------------------------- Bugfix: single-end barcode, double-digested data were not being handled properly by process_radtags causing a crash. Feature: added support for PLINK and Beagle output files from the populations program. Feature: Modified the minor allele frequency (MAF) filter to remove polymorphic nucleotide SNPs from Stacks output on a per-population basis. So, if a second allele is present at a frequency below the MAF, that nucleotide site is not output (although other sites at the same RAD locus could still be output). Bugfix: Tri-allelic loci were being output into the STRUCTURE, GENEPOP and PHASE output (but not in sumstats or Fst). Stacks 1.01 - June 07, 2013 --------------------------- Bugfix: an off-by-one error was preventing haplotypes from being verified by sstacks if a SNP occurred in the last position of the read. This could cause tags to fail to match to the catalog if there is a SNP in the final position. Stacks 1.0 - June 06, 2013 -------------------------- Feature: added XbaI and BamHI restriction enzymes to process_radtags. Feature: added code to output genotypes in PHASE/fastPHASE format. Feature: extended combinatorial barcodes support so one can process single-end data that contains both an inline and indexed barcode. Feature: added command line option and supporting code to cstacks to allow samples to be added to an existing catalog. Feature: refactored command line handling in denovo_map.pl and ref_map.pl to be much more flexible. Arbitrary command line options can now be passed to particular pipeline programs using the -X flag. Feature: for genetic maps, catalog may now be constructed out of mulitple parents, genotypes is smart enough to cross check the parents used to construct the catalog against those submitted to genotypes for producing a map. Will allow for a single catalog to be used across a series of crosses so all maps share the same catalog IDs. Feature: added option to genotypes to import manual corrections exported from Stacks SQL database. Feature: added --log_fst_comp option to populations to log components of the Fst calculations to a file for debugging / testing purposes. Bugfix: corrected handling of files in kmer_filter. Adding support for gzipped files broke file handling in some cases. Stacks 0.999991 - May 14, 2013 ------------------------------ Feature: changed populations to use AMOVA Fst for batch_1.fst_summary.tsv file. Previously it used the Binomial Fst. Bugfix: If --write_single_snp not specified, Structure output was not naming loci properly (it was naming each SNP from the same RAD locus using the same ID, instead of differentiating each SNP in each RAD locus). Feature: Added Sau3AI and SexAI restriction enzymes. Fixed bug in specificaion of MseI, MspI enzymes. Bugfix: changed VCF and Fst code in populations to output SNPs from reads aligned to the negative strand on a reference genome correctly. Stacks 0.99999 - May 06, 2013 ----------------------------- Bugfix: process_shortreads/process_radtags not working with non-barcoded data. Stacks 0.99998 - May 01, 2013 ----------------------------- Feature: Added option to sort_read_pairs.pl to output FASTQ if desired. Bugfix: make sort_read_pairs.pl understand new file naming scheme. Feature: added mseI, mspI restriction enzymes to process_radtags. Bugfix: corrected sphI cutsite sequence in process_radtags. Bugfix: stopped "uninitialized value" errors in export_sql.pl when marker type is undefined for a particular map. Stacks 0.99997 - April 01, 2013 ------------------------------- Bugfix: paired barcode could become uninitialized on second pair of files in process_radtags/process_shortreads causing all barcodes to mismatch. Made Read class explicitly initialize everything. Stacks 0.99996 - March 24, 2013 ------------------------------- Feature: major overhaul of the process_radtags / process_shortreads programs to support combinatorial barcodes and double-digested data. Programs now support a mixture of barcodes from single-end inline or index barcodes, to mixtures of inline/index barcodes. 1) changed naming scheme for process_radtags/process_shortreads output files for paired reads. Changed file suffix to properly be ".fq" or ".fa", with paired-reads named sample_XXX.1.fq and sample_XXX.2.fq instead of the previous ".fq_1" and ".fq_2". 2) Paired-reads remain synced in output files, with sinlgetons written to sample_XXX.rem.1.fq and sample_XXX.rem.2.fq. 2) changed Phred+33 to be the default encoding scheme (previously was the now deprecated Phred+64) 3) Combinatorial barcdoes are specified as --inline_index or --inline_inline among a number of other supported possibilities. Barcodes are listed in the barcode file as either a single column or two, tab-separated columns. 4) Two restriction enzymes can now be specified via --renz_1 and --renz_2 to have the program check (and correct) the restriction enzyme cut site on the first and second read respectively. 5) programs now properly ignore files starting with "." which is required for Mac OS X's ".DS_Store" files and for "." and ".." on Linux. Bugfix: processing paired-end data with process_radtags could incorrectly alter the first few nucleotides of the paired-read when correcting barcodes. Bugfix: two regressions were fixed in process_shortreads causing all reads to be improperly trimmed. Bugfix: VCF output did not include sites fixed within and variable among populations. Bugfix: changed the parsing code to accept a wider range of Illumina named, paired-end files in process_radtags/shortreads. Bugfix: gzipped files were not read properly in process_radtags/shortreads when a directory was specified with -P. Bugfix: setting secondary read distance to 0 in ustacks (-N) was ineffective. Bugfix: changed the PHP code to remove 'Strict Standards' warnings and a few other warnings. Thanks to Yue Yu for tracking down the proper changes to avoid the warnings. Stacks 0.99995 - February 19, 2013 ---------------------------------- Feature: added support for using Google's Sparsehash Object: http://code.google.com/p/sparsehash/ If enabled at compile time, this object will replace all the hash maps with Google's sparsehash saving significant memory. Feature: removed the -S command line option from cstacks and sstacks. These programs now read this ID directly from the Stacks input files. Feature: altered ustacks to no longer store FASTQ/FASTA IDs from input files in memory to lower memory usage. Instead, an integer representing the read is stored and the IDs are read back in from disk just before results are written. Feature: added the '--write_single_snp' option to populations. When writing Genepop or Structure files this option will cause populations to write just the first SNP per locus to the file, avoiding potential problems with linked SNPs originating from the same locus. Feature: compressed the Hval/Stack/Rem objects to remove convenience integer variables to save memory. Feature: updated Stacks programs to use the newer TR1 unordered_map hash object instead of the deprecated SGI hash_map object. Bugfix: fixed a memory leak in cstacks in which not all of the Locus Class elements was being properly freed (only the SNP objects were being freed). Bugfix: Added code to denovo_map.pl/ref_map.pl to remove from the logfile the 'counter' lines that printed when initially loading radtags data. Stacks 0.99994 - February 12, 2013 ---------------------------------- Bugfix: process_radtags/process_shortreads, when adding support for reads of different length, I clobbered the sequence truncation option. Fixed this regression. Bugfix: the kernel smoothing algorithms for calculating Fst, Pi, and Fis could sometimes segfault as some RAD sites can overlap. Added code to find and describe overlapping RAD sites and report these to the user. Stacks 0.99993 - January 30, 2013 --------------------------------- Feature: process_radtags/process_shortreads/ustacks can now read gzipped Fasta/Fastq input files. Feature: ref_map.pl/pstacks now supports the use of BAM alignment files. This feature is optional and must be enabled during compilation. It requires the Samtools library to be installed. Bugfix: When using referenced aligned data, soft-masked alignments (Ns) were getting imporperly injected into the SNP models, which would call them as Homozygous Ns, and this data would eventually be passed to the summary statistics in populations, which would make errant Fst calculations. Bugfix: In rare cases, sequences aligned to the negative strand had their base pair positions slightly off, this could cause a segfault during populations' kernel-smoothed Fst calculations. Bugfix: In populations, fixed a rare, infinite loop condition in Fisher's exact test for Fst calculations. Could occur due to a floating point rounding error when calculating allele frequencies for Fst calculation. Stacks 0.99992 - January 8, 2013 -------------------------------- Bugfix: floating point command line options were not being processed correctly and may have been truncated. Stacks 0.99991 - December 17, 2012 ---------------------------------- Feature: process_shortreads and process_radtags can now filter for adapter sequence in raw data, trimming (process_shortreads) or discarding (process_shortreads/process_radtags) it. Mismatches to the adapter sequence are allowed to accomodate for sequencing error. Bugfix: added --merge flag to process_shortreads/process_radtags to handle regression where unbarcoded data should be merged together into single output files. Bugfix: code in cstacks to characterize differentially fixed SNPs was only running with -n > 0, but should also run by default if -g is specified. Feature: made automated correction thresholds for the genotypes program accessible from the command line, including --min_hom_seqs, --min_het_seqs, and --max_het_seqs options. Feature: refactored clone_filter to be more functional. Now can output sequences in FASTA or FASTQ (FASTA will save memory). Keeps sequence headers intact, can capture discarded reads, and prints a distribution of the number of cloned read pairs. Bugfix: Remainder reads weren't being written properly as the file handles weren't properly closed. Bugfix: Processing paired reads with process_radtags/process_shortreads was not functioning correctly, barcode was not being transferred properly from P1 to P2 read. Regression introduce Aug 21, 2012. Feature: added support for OneMap CP map export in genotypes. Bugfix: Fixed some bugs in pstacks/ustacks command line processing involving --alpha and --model_type. Bugfix: several bugs in the exact and approximate bootstrap algorithms were corrected. These algorithms are now robust. Bugfix: Added code to ensure command line IDs are in fact integers. Bugfix: fixed nucleotide positions were not being tallied across populations properly resulting in an incorrect value for number of sites and percent polymorphic sites in the sumstats_summary file. Bugfix: pstacks could identify a locus that despite having SNPs would have no haplotypes generated. This would late cause sstacks to segfault. Added code in pstacks to blacklist these loci and code in sstacks to catch this case and not segfault, now will print a warning. Stacks 0.9999 - October 03, 2012 -------------------------------- Feature: two bootstrapping procedures have been introduced into the populations program to determine the statistical significance of kernel smoothed windows. These algorithms are controlled by the --bootstrap and --bootstrap_reps command line options. Feature: summary summary statistics are now written for all populations, giving the mean, variance, and standard error for each of the population-specific summary stats. In addition, private alleles are identified and marked in the sumstats file, and summarized across populations. Number and percent of polymorphic loci are also reported. The actual variable nucleotides at each site are now reported in the sumstats file. Feature: the populations program can now generate kernel-smoothed values for Fis and Pi, in addition to the current support for Fst. Feature: the populations program can now output SNP data for use in the program Structure. Feature: various sections of the populations program have been parallelized. Feature: the populations program can now output SNP data in the Phylip file format. If --phylip is specified, the populations program will identify SNPs that are fixed within populations, but variable between populations and output these in a Phylip file. This file can then be fed into any phylogenetics program, such as PhyML. This feature is equivalent to the analysis done in Emerson, et al., 2010. In addition, if the --phylip_var flag is specified as well, variable sites within populations are encoded into the Phylip file using standard alternative nucleotide encodings. Feature: for ustacks/pstacks, the alpha significance level can now be specified on the command line. Specifying --alpha to ustacks or pstacks will set the chi square significance level to determine whether a heterozygous or homozygous model call is statistically significant. Legal values of alpha are 0.1, 0.05 (the previous default), 0.01, or 0.001. Feature: for ustacks/pstacks, a new bounded SNP calling model has been introduced, allowing limits to be set on the error rate. This model allows the calling of SNPs to be affected by prior knowledge as to how likely polymorphism is in the data set. This behavior is controlled by the --bound_low and --bound_high parameters to ustacks and pstacks. Feature: additional sections of ustacks has been parallelized. In addition, stack merging has been changed to occur in a single step (instead of in rounds as done previously). Feature: the deleveraging algorithm in ustacks has been replaced with a simple algorithm based on a minimum spanning tree. A new parameter has been introduced, --max_locus_stacks, which controls the number of stacks allowed to be merged together into a single locus. Loci that contain more than --max_locus_stacks stacks are set aside and not added to the catalog later on. Feature: export_sql.pl now has two depth parameters, allele and locus depth, allowing for the filtering of loci based on either one. Feature: added a 'dry run' flag (-d) to denovo_map.pl and ref_map.pl to allow the pipeline to be tested to see what it would execute, before actually executing any programs. Bugfix: problem with the FASTA parser fixed (it was introduced with fixes to handle windows-style files). Bugfix: sample counts where off in batch_*.haplotypes.tsv file generated by populations program. Stacks 0.9996 - August 24, 2012 ------------------------------- Bugfix: fixed significant memory leak in Kmer hashing for both ustacks and cstacks. Results in an approximately 3.4x reduction in memory use for cstacks, and an approximately 1.6x reduction in ustacks. Feature: process_radtags and process_shortreads can handle non-Illumina FASTQ headers (any generic FASTQ type). Feature: process_radtags can process data without barcodes. Feature: process_radtags and process_shortreads can handle Illumina barcodes, when the barcode is not inline but is instead provided in the FASTQ header. Bugfix: Corrected the behavior of the '-m' parameter to populations and genotypes. It is meant to apply to the total depth of a stack at a locus, but was instead being applied to the depth of each allele at each locus. Feature: process_radtags and process_shortreads can now automatically discard reads marked as 'failed' by Illumina's chastity/purity filter. Feature: added ecoT22I, mluCI, nlaIII, and sphI restriction enzymes to process_radtags Bugfix: modified Stacks programs to handle Windows-style line endings ('\r\n') from FASTQ, FASTA, and SAM files as well as population maps. Bugfix: changed populations' genepop output to only include loci that are variable in the populations specified. Previously, in some cases, additional fixed loci were included, which are not included in the VCF output, causing the two files to have different loci present. Bugfix: expected homozygosity and observed homozygosity were not being reported correctly in the sumstats files. The other population statistics were not affected by the bug. Feature: process_radtags and process_shortreads now print command and time executed to log file. Stacks 0.9995 - July 05, 2012 ----------------------------- Bugfix: Fst summary matrix was being incorrectly written. Stacks 0.9994 - July 01, 2012 ----------------------------- Feature: the populations program can now write a file in the GenePop format. GenePop files can be read by the GenePop program and converted for other population genetics programs such as Arlequin. Caution: you may not be able to include all loci from a Stacks run in the output as these programs aren't necessarily capable of handling such a volume of data. However, you can use populations' whitelist feature to only include certain loci in the output. Feature: the populations program now writes an Fst summary file providing a matrix of mean Fst measures for each pair of populations in the analysis. Feature: added two filters to populations to require a locus to be present in a certain percentage of individuals in a population, and requiring a locus to be present in a certain number of populations. If the former criteria is not reached, the locus is zeroed out only in the specific population, if the latter criteria is not met, the locus is discarded from the analysis. Feature: three Fst corrections are now provided by the populations program: requiring a locus to have a significant p-value (smaller than 0.05, although its configurable), applying a Bonferroni correction according to the number of data points in the sliding window, and applying a Bonferroni correction according to the number of data points in the genome. Loci that fail to reach statitical significance in each case are considered not different from zero and are set to zero. Feature: a filter can be specified to the populations program requiring a minimum allele frequency (MAF) at a locus to consider the locus variable. If an allele at a locus is below the MAF, the locus is considered fixed. Feature: when using a reference genome, Stacks can now work with samples of different sequence lengths. This means one can combine samples generated from different Illumina runs of different length. Each individual sample must be of the same length internally, however. Feature: pstacks can now handle gapped alignments properly. It parses the CIGAR string in the SAM file and inserts/removes Ns to accomodate indels and soft-masked alignment fragments. This prevents the SNP model from mistakenly calling polymorphisms due to indel frameshifts. Bugfix: Removed O(n^2) algorithm from Sliding window Fst calculation in populations program, significant speedup acheived. Bugfix: Updated load_radtags.pl to support population types and to import sumstats, fst, and genotypes files. Bugfix: fixed a small memory leak in DNANSeq. Stacks 0.9993 - June 07, 2012 -------------------------------- Feature: Added Fisher's Exact Test statistics to Fst estimates. This provides a p-value, an odds ratio along with a 95% confidence interval and a Log of Odds (LOD) score for each Fst estimate. These statistics allow one to decide if a particular Fst measurement is significant. Feature: denovo_map.pl and ref_map.pl now import population statistics files into the database (fst and sumstats files). Feature: Web interface now displays summary statistics and Fst values for every locus. Feature: population names can now be directly added through the web interface and they will be stored in the database and propogated. Stacks 0.9992 - May 22, 2012 -------------------------------- Bugfix: fixed massive memory leak in Fst calcuations in populations program. Bugfix: if using a population map to calculate Fst in the populations program, some individuals could be inadvertently attributed to the wrong populations, due to a mismatch between the indices of the population map (PopMap.h) and the indexes recorded for making the population summary (PopSum.h). Feature: population map can now be specified to denovo_map.pl and ref_map.pl. This data is populated into the database and samples are displayed according to their population in the web interface. Feature: improved denovo_map.pl and ref_map.pl to check for existence of input files. Bugfix: export_sql.pl wasn't properly using the new filters that use a lower and upper bound (snps, alle, pare). Feature: improved how values are generated for web-based filters, allowing for larger populations/maps. Improved HTML rendering for extremely long haplotype strings. Bugfix: corrected alleles to be output as "unphased" in VCF file; corrected homozygotes to be printed as diploid values, e.g. '0/0' or '1/1' instead of just '0'. Bugfix: changed reporting of SNPs on samples.php page to specify total number of SNPs and the number of polymorphic loci (containing one or more SNPs). Bugfix: an extra tab was being placed in the VCF output file. Feature: added flag to process_radtags to disable checking the integrity of the RAD site in each raw read. Added a flag to allow more nucleotide mismatches in the barcode when rescuing barcodes. Stacks 0.9991 - April 17, 2012 -------------------------------- Bugfix: replaced bit-rotted code causing all nucleotides to be masked as 'N' when fixed model engaged on ustacks. Stacks 0.999 - April 11, 2012 -------------------------------- Feature: Added support for the 1000 Genomes Project, Variant Call Format (VCF) in the populations program. (http://www.1000genomes.org/node/101). This file output includes the genotype calls for every individual for each locus, allele depth, and likelihood values for heterozygous SNP calls. Feature: implemented a three-bit compression scheme so that uncalled bases ('N's) can be stored in compressed format in pstacks. Other stacks programs currently use two-bit compression which is more compact, but can only store plain nucleotides ('A', 'C', 'G', 'T'). This restores earlier behavior that allowed Ns in pstacks prior to the implementation of the two-bit compression scheme. Bugfix: the populations program was only outputing sites to the summary statistics file (*.sumstats.tsv) if they were heterozygous in a population. This could give the impression that the same site may be absent in other populations when in reality it was simply fixed in the other populations. Now, if a site is heterozygous in any of the populations, it will be output for all populations. Bugfix: added lots of error checking code to populations so it properly handles poorly formatted population maps, missing files, and similar errors. Bugfix: added uncalled bases ('n', 'N', and '.') to the reverse complement function (reads aligned on the negative strand and processed by pstacks will be stored reverse complement. Bugfix: updated the PHP code as well as export_sql.pl to properly use the new filters for chromosome, basepair, as well as lower and upper ranges to various filters. Other: Removed the deprecated markers.pl, genotypes.pl, and process_radtags.pl programs from the distribution. Stacks 0.998 - January 06, 2012 -------------------------------- Feature: Pipeline is now aware if samples are submitted as a 'population' or a 'mapping cross'. A new command line option, -s, has been added to denovo_map.pl and ref_map.pl that will label the dataset as a population. The -p/-r flags continue to keep the samples as a mapping cross. Feature: The web interface has been updated to display more information specific to populations. The filtering code has been changed to include lower and upper limits for filter fields such as SNPs, alleles, and number of parents/samples. Feature: A new program, populations, has been written to be executed in place of the exisiting genotypes program when a population is being processed through the pipeline. A map specifiying which individuals belong to which population is submitted to the program and the program will then calculate population genetics statistics, expected/observed heterzygosity, Pi, and Fis at each nucleotide position. Feature: the populations program will compare all populations pairwise to compute Fst. If a set of data is reference aligned, then a kernel-smoothed Fst will also be calculated. These statistics were originally designed by Paul Hohenlohe and Bill Cresko, and are described in the paper: Population Genomics of Parallel Adaptation in Threespine Stickleback using Sequenced RAD Tags, http://www.plosgenetics.org/article/info%3Adoi%2F10.1371%2Fjournal.pgen.1000862 They have been implemented independently in Stacks. Feature: added the DpnII enzyme to the process_radtags program. Feature: Added new 'model' line to *.tags.tsv files. This line records the output of the SNP model at every position in the read as either Homozygous (O), Heterozygous (E), or unknown (U). Previously only polymorphic loci were recorded in the SNPs file (and this remains unchanged). The model output line is now also available in the web interface. Bugfix: fixed crasher bug in cstacks when parallel processing was enabled for genomic-aligned data. Bugfix: allele depths are now properly reported in reference-aligned data. Stacks 0.997 - November 22, 2011 -------------------------------- Feature: new program, called clone_filter, that will take a set of paired-end reads and reduce them according to PCR clones (a PCR clone is a pair of reads that match exactly, while paried-end reads from two different DNA molecules will nearly always be slightly different lengths). Feature: new program, called kmer_filter, that allows paired or single-end reads to be filtered according to the number or rare or abundant kmers they contain. Useful for both RAD datasets as well as randomly sheared genomic or transcriptomic data. Feature: new program, called process_shortreads, performs the same task as process_radtags for fast cleaning of randomly sheared genomic or transcriptomic data (a 'beta' version of this program has actually been distributed in the last few Stacks releases). Feature: the Stacks tags.tsv file format has a new column to record the DNA strand that a particular read is aligned to, currently only used in datasets aligned to a reference genome. Feature: pstacks now reverse complements all stacks aligned to the negative strand and stores them in this orientation in the output files and database. (All aligners always present these reads in the positive orientation.) This change allows one to align reads to a reference genome using a gapped aligner, such as Tophat or GSNAP and have the RAD site still align with genomic data. (One can then compare genomic RAD tags along with cDNA RAD tags.) Feature: added the '-d' flag to export_sql.pl to export allele depths from the database. Feature: altered process_radtags to store orphaned, paired-end reads in a remainder file, keeping paired-reads in frame. Bugfix: fixed the handling of the paired-end barcode in process_shortreads, added a check to make sure the barcodes from both pairs of a read match. Bugfix: genotypes was not capitalizing auto-corrected genotypes in the generic format (it was in joinmap/rqtl specific formats). Bugfix: corrected cut site sequence for ApeKI in process_radtags. Bugfix: process_radtags inadvertantly used newly initialized memory that had not been cleared, causing rare parsing errors when uncleared memory resembled portions of a FASTQ record. Bugfix: the default MySQL permissions were not being properly passed to index_radatags.pl. Bugfix: changed load_radtags.pl to extract parental IDs from directly catalog files, instead of relying on file names. Feature: added a 'dry run' option to load_radtags.pl so it will print what it intends to do without actually doing it. Stacks 0.996 - October 5, 2011 --------------------------------- Web interface updates: * If the RAD tags are aligned to a reference genome, a filter is now available to view markers from a particular genomic region. * The individual RAD tag viewer now scrolls while keeping the scale view and consensus sequence always visible. * The RAD tag viewer now highlights columns for which the catalog locus shows a SNP, but the RAD tag does not. * In the genotype viewer, the map between the haplotype and genotype is now available. * The depth of each RAD tag is now visible in the genotype viewer. * The genotype viewer has now been integrated with the observed haplotype viewer. You can make changes/corrections to genotypes directly now, no need to submit a form and wait for the page to reload. Bugfix: process_radtags wasn't properly parsing the names of v1 Illumnina BUSTARD files. Bugfix: process_radtags wasn't counting the total number of barceded paired-end reads correctly. Bugfix: sstacks' impute_haplotype() was causing spurious matching in some, error-based cases. Bugfix: build system was not properly replacing the _PKGDATADIR_ variable in denovo/ref_map.pl programs. Stacks 0.995 - September 23, 2011 --------------------------------- Feature: sstacks can now handle samples and catalogs that have different length reads. Each individual sample must be constructed from the same length reads (by ustacks and cstacks), but between samples there can be different lengths, e.g. a catalog of length 50bp and samples of length 100bp, or vice versa. Feature: Added the ApeKI restriction enzyme to process_radtags Feature: process_radtags can now capture discarded reads to a file. Bugfix: a coding limitation was removed that required polymorphic sites in the catalog to contain only two alleles. Now, all four alleles can be recorded at a single site in a locus in the catalog. Bugfix: Exporting results from the web interface was not including manual genotype corrections when requested. Stacks 0.994 - August 08, 2011 ------------------------------ Feature: added catalog index structure to cstacks to speed construction of catalog when using reference-aligned sequences. Feature: added a new output type, 'genomic' to genotypes. Outputs SNPs individually, encoded as a set of integers, for reference-aligned reads. Bugfix: pstacks was not writing individual stack sequences properly. Bugfix: process_radtags was still checking the quality of sequence that was destined to be truncated off the read. Bugfix: process_radtags segfault fixed, parsing stop position mis-specified in parse_input_record(). Stacks 0.993 - August 05, 2011 ------------------------------ Memory usage optimization: Individual sequence reads are now stored internally using a 2-bit encoding of DNA nucleotides. Some simple benchmarking of ustacks (previous version / new version): Sample size Elapsed Time Used Memory ------------- ----------------- ------------- 3.78m reads 3:16 / 3:23 4.64G / 1.86G 17.62m reads 1:31:21 / 1:43:54 55.55G / 45.42G Feature: Added the programs sort_read_pairs.pl, exec_velvet.pl, load_sequences.pl to facilitate the assembly of paired-end RAD-Tags into mini-contigs and allow them to be uploaded into and viewed from the web interface. Bugfix: made process_radtags emit an error when an unrecognized restriction enzyme is specified. Bugfix: made process_radtags accept barcodes with trailing whitespace, such as would be seen in a DOS text file or if errant tabs are specified. Stacks 0.992 - July 04, 2011 ---------------------------- Feature: process_radtags can now handle Phred+33 or Phred+64 encodings, Phred+33 is the new default encoding in Illumina's CASAVA software (v1.8). Bugfix: Changed the sql input parser to handle variable length input lines. Necessary if loading tens of individuals into a catalog. Bugfix: Added command line options to ustacks to better control the use of secondary reads in the stack-building procedure. Stacks 0.991 - June 06, 2011 ---------------------------- Bugfix: genotypes was failing to parse Stacks output files with unanticipated names. Bugfix: when using ref_map.pl, tags without SNPs were failing to match against the catalog. Stacks 0.99 - May 20, 2011 -------------------------- *A new C++ genotypes program has been added. This program works independently from the database allowing the pipeline to fully function without installing the database. The new program performs the tasks once completed by markers.pl and genotypes.pl. - The pipeline has been modified to now automatically execute the genotypes program as the last stage in an analysis. It will generate a file containing the observed haplotypes and an additional file containing a map-agnostic set of genotype calls. - If SQL interaction is enabled, the genotypes will be imported to the database and serve as a base to export genotypes directly from the web interface for a particular map and using the set of filters available online. - If a population is being examinined, the observed haplotypes file can be imported into Microsoft Excel or another tab-separated file viewer to immediately see the results. - By replacing the Perl version of genotypes.pl we also no longer need to install or worry about the caching mechanism for auto-correcting stacks, the C++ version can do this by directly reading the Stacks output files. *markers.pl and genotypes.pl are now deprecated and will no longer be supported. *Feature: When exporting observed haplotypes, you can now specify a minimum stack depth to include a particular individual at a locus. *Feature: map-specific genotypes can now be exported directly from the database/web server. *Bugfix: genotypes.pl: make script ignore parental genotypes based on the sample type from the MySQL table, not based on the file name. *Bugfix: genotypes.pl: some loci were sneaking in despite being under the progeny limit. *Bugfix: made process_radtags Bustard file parser check number of fields to prevent attempting to parse FASTQ (and segfaulting). Thanks to Maureen.Liu -at- nottingham.ac.uk for reporting it. *Bugfix: in sstacks, when matching to the catalog using reads aligned to a reference genome (-g), sstacks did not verify that haplotypes matched exactly, causing some spurious matching, which later translated into dropped genotypes. *Bugfix: in markers.pl, the ratio observed alleles in the progeny was not being tallied correctly for ab/ac markers. Stacks 0.984 - May 04, 2011 --------------------------- *Bugfix: renamed constants.php to constants.php.dist to avoid overwriting an existing file on reinstallation. *Feature: process_radtags has been converted to a C++ program increasing its speed by approximately 25x. The parameters were modified to be a little more intuitive and parameters were added to control the size and score limit of the sliding window. The program can process a GAII lane in about 5 minutes, a HiSeq lane in about 12 minutes, depending on the hardware used. Stacks 0.983 - Apr 30, 2011 --------------------------- *Bugfix: sstacks segfault when running parallelized. Improper insertion into map object when it should have only been checking for element presence/absence. Thanks to for first reporting it. *Feature: added code to impute the genotype of a missing, second parent for some map types. This code adds up all the observed haplotypes in the progeny and then compares their frequencies against those that would be expected for the marker under Hardy-Weinberg equilibrium, choosing the marker type that best fits the Hardy-Weinberg expectation. Stacks 0.982 - Mar 29, 2011 --------------------------- *Bugfix: process_radtags.pl was not properly parsing FASTQ formated, paired-end file names. *Bugfix: counts of matching parents/progeny were sometimes incorrect due to a slightly promiscuous SQL query in index_radtags.pl. Stacks 0.98 - Feb 25, 2011 --------------------------- Note: if you have pre-existing databases, you must rebuild the catalog index (index_radtags.pl -D db -c) so that they are compatible with the new elements of the web interface. *Added option to pstacks to require a minimum depth of coverage for a stack aligned to the refernce genome before reporting it. *Added double haploid (DH) and F2 export types to the genotypes.pl script. *Added option to output any map in R/QTL output in genotypes.pl *Added feature to filter by number of available genotypes in progeny *Added command line option to ustacks to capture and output unused reads. *Added display of chromosome/base pair to web interface for stacks aligned to a reference genome. *Bugfix: FASTA parser was missing records due to a bug introduced from a FASTQ parser fix. *Bugfix: process_radtags.pl was not properly checking the integrity of the RAD site after adding restriction enzymes with alternate nucleotides. *Bugfix: when constructing the catalog, some tags being added to the did not have their genomic location transferred over to a new catalog tag. *Modified sstacks to include an option to match stacks against the catalog based on the genomic location (assuming individuals were processed with pstacks). *Bugfix: Lots of clean-ups and command line option fixes, thanks to . Stacks 0.971 - Jan 30, 2011 --------------------------- *Illumina software version 1.3 produces Phred scores that can begin with a '@' character, throwing off the FASTQ parser. Added code to clear the read buffer in between records to solve the problem. Thanks to Aarti for finding the bug. Stacks 0.97 --------------------------- *ustacks now detects when there are uncalled nucleotides in FASTA or FASTQ input files, replaces those bases with 'A'. *process_radtags.pl now detects barcode length automatically. Removed spurious error messages when no data is processed. Stacks 0.96 - Jan 7, 2011 --------------------------- *Fixed typo in README giving the wrong file path for the Apache configuration file. *Fixed several hard-coded paths in PHP files that referred to our local system. stacks-1.44/config/000755 000765 000024 00000000000 13001012452 014733 5ustar00catchenstaff000000 000000 stacks-1.44/config.h.in000644 000765 000024 00000006624 13001003451 015520 0ustar00catchenstaff000000 000000 /* config.h.in. Generated from configure.ac by autoheader. */ /* Define to 1 if the `closedir' function returns void instead of `int'. */ #undef CLOSEDIR_VOID /* Enable compilation with Samtools BAM library */ #undef HAVE_BAM /* define if the compiler supports basic C++11 syntax */ #undef HAVE_CXX11 /* Define to 1 if you have the header file, and it defines `DIR'. */ #undef HAVE_DIRENT_H /* Define to 1 if you have the header file. */ #undef HAVE_FLOAT_H /* Define to 1 if you have the `floor' function. */ #undef HAVE_FLOOR /* Define to 1 if you have the header file. */ #undef HAVE_INTTYPES_H /* Define to 1 if you have the `gomp' library (-lgomp). */ #undef HAVE_LIBGOMP /* Define to 1 if you have the `z' library (-lz). */ #undef HAVE_LIBZ /* Define to 1 if you have the header file. */ #undef HAVE_LIMITS_H /* Define to 1 if your system has a GNU libc compatible `malloc' function, and to 0 otherwise. */ #undef HAVE_MALLOC /* Define to 1 if you have the header file. */ #undef HAVE_MEMORY_H /* Define to 1 if you have the `memset' function. */ #undef HAVE_MEMSET /* Define to 1 if you have the header file, and it defines `DIR'. */ #undef HAVE_NDIR_H /* Define to 1 if you have the `pow' function. */ #undef HAVE_POW /* Define to 1 if your system has a GNU libc compatible `realloc' function, and to 0 otherwise. */ #undef HAVE_REALLOC /* Enable compilation with Google Sparsehash */ #undef HAVE_SPARSEHASH /* Define to 1 if you have the `sqrt' function. */ #undef HAVE_SQRT /* Define to 1 if stdbool.h conforms to C99. */ #undef HAVE_STDBOOL_H /* Define to 1 if you have the header file. */ #undef HAVE_STDINT_H /* Define to 1 if you have the header file. */ #undef HAVE_STDLIB_H /* Define to 1 if you have the header file. */ #undef HAVE_STRINGS_H /* Define to 1 if you have the header file. */ #undef HAVE_STRING_H /* Define to 1 if you have the header file, and it defines `DIR'. */ #undef HAVE_SYS_DIR_H /* Define to 1 if you have the header file, and it defines `DIR'. */ #undef HAVE_SYS_NDIR_H /* Define to 1 if you have the header file. */ #undef HAVE_SYS_STAT_H /* Define to 1 if you have the header file. */ #undef HAVE_SYS_TYPES_H /* Define to 1 if you have the header file. */ #undef HAVE_UNISTD_H /* Define to 1 if the system has the type `_Bool'. */ #undef HAVE__BOOL /* Name of package */ #undef PACKAGE /* Define to the address where bug reports for this package should be sent. */ #undef PACKAGE_BUGREPORT /* Define to the full name of this package. */ #undef PACKAGE_NAME /* Define to the full name and version of this package. */ #undef PACKAGE_STRING /* Define to the one symbol short name of this package. */ #undef PACKAGE_TARNAME /* Define to the home page for this package. */ #undef PACKAGE_URL /* Define to the version of this package. */ #undef PACKAGE_VERSION /* Define to 1 if you have the ANSI C header files. */ #undef STDC_HEADERS /* Version number of package */ #undef VERSION /* Define to empty if `const' does not conform to ANSI C. */ #undef const /* Define to rpl_malloc if the replacement function should be used. */ #undef malloc /* Define to rpl_realloc if the replacement function should be used. */ #undef realloc /* Define to `unsigned int' if does not define. */ #undef size_t stacks-1.44/configure000755 000765 000024 00000665727 13001003444 015425 0ustar00catchenstaff000000 000000 #! /bin/sh # Guess values for system-dependent variables and create Makefiles. # Generated by GNU Autoconf 2.69 for Stacks 1.44. # # # Copyright (C) 1992-1996, 1998-2012 Free Software Foundation, Inc. # # # This configure script is free software; the Free Software Foundation # gives unlimited permission to copy, distribute and modify it. ## -------------------- ## ## M4sh Initialization. ## ## -------------------- ## # Be more Bourne compatible DUALCASE=1; export DUALCASE # for MKS sh if test -n "${ZSH_VERSION+set}" && (emulate sh) >/dev/null 2>&1; then : emulate sh NULLCMD=: # Pre-4.2 versions of Zsh do word splitting on ${1+"$@"}, which # is contrary to our usage. Disable this feature. alias -g '${1+"$@"}'='"$@"' setopt NO_GLOB_SUBST else case `(set -o) 2>/dev/null` in #( *posix*) : set -o posix ;; #( *) : ;; esac fi as_nl=' ' export as_nl # Printing a long string crashes Solaris 7 /usr/bin/printf. as_echo='\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\' as_echo=$as_echo$as_echo$as_echo$as_echo$as_echo as_echo=$as_echo$as_echo$as_echo$as_echo$as_echo$as_echo # Prefer a ksh shell builtin over an external printf program on Solaris, # but without wasting forks for bash or zsh. if test -z "$BASH_VERSION$ZSH_VERSION" \ && (test "X`print -r -- $as_echo`" = "X$as_echo") 2>/dev/null; then as_echo='print -r --' as_echo_n='print -rn --' elif (test "X`printf %s $as_echo`" = "X$as_echo") 2>/dev/null; then as_echo='printf %s\n' as_echo_n='printf %s' else if test "X`(/usr/ucb/echo -n -n $as_echo) 2>/dev/null`" = "X-n $as_echo"; then as_echo_body='eval /usr/ucb/echo -n "$1$as_nl"' as_echo_n='/usr/ucb/echo -n' else as_echo_body='eval expr "X$1" : "X\\(.*\\)"' as_echo_n_body='eval arg=$1; case $arg in #( *"$as_nl"*) expr "X$arg" : "X\\(.*\\)$as_nl"; arg=`expr "X$arg" : ".*$as_nl\\(.*\\)"`;; esac; expr "X$arg" : "X\\(.*\\)" | tr -d "$as_nl" ' export as_echo_n_body as_echo_n='sh -c $as_echo_n_body as_echo' fi export as_echo_body as_echo='sh -c $as_echo_body as_echo' fi # The user is always right. if test "${PATH_SEPARATOR+set}" != set; then PATH_SEPARATOR=: (PATH='/bin;/bin'; FPATH=$PATH; sh -c :) >/dev/null 2>&1 && { (PATH='/bin:/bin'; FPATH=$PATH; sh -c :) >/dev/null 2>&1 || PATH_SEPARATOR=';' } fi # IFS # We need space, tab and new line, in precisely that order. Quoting is # there to prevent editors from complaining about space-tab. # (If _AS_PATH_WALK were called with IFS unset, it would disable word # splitting by setting IFS to empty value.) IFS=" "" $as_nl" # Find who we are. Look in the path if we contain no directory separator. as_myself= case $0 in #(( *[\\/]* ) as_myself=$0 ;; *) as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. test -r "$as_dir/$0" && as_myself=$as_dir/$0 && break done IFS=$as_save_IFS ;; esac # We did not find ourselves, most probably we were run as `sh COMMAND' # in which case we are not to be found in the path. if test "x$as_myself" = x; then as_myself=$0 fi if test ! -f "$as_myself"; then $as_echo "$as_myself: error: cannot find myself; rerun with an absolute file name" >&2 exit 1 fi # Unset variables that we do not need and which cause bugs (e.g. in # pre-3.0 UWIN ksh). But do not cause bugs in bash 2.01; the "|| exit 1" # suppresses any "Segmentation fault" message there. '((' could # trigger a bug in pdksh 5.2.14. for as_var in BASH_ENV ENV MAIL MAILPATH do eval test x\${$as_var+set} = xset \ && ( (unset $as_var) || exit 1) >/dev/null 2>&1 && unset $as_var || : done PS1='$ ' PS2='> ' PS4='+ ' # NLS nuisances. LC_ALL=C export LC_ALL LANGUAGE=C export LANGUAGE # CDPATH. (unset CDPATH) >/dev/null 2>&1 && unset CDPATH # Use a proper internal environment variable to ensure we don't fall # into an infinite loop, continuously re-executing ourselves. if test x"${_as_can_reexec}" != xno && test "x$CONFIG_SHELL" != x; then _as_can_reexec=no; export _as_can_reexec; # We cannot yet assume a decent shell, so we have to provide a # neutralization value for shells without unset; and this also # works around shells that cannot unset nonexistent variables. # Preserve -v and -x to the replacement shell. BASH_ENV=/dev/null ENV=/dev/null (unset BASH_ENV) >/dev/null 2>&1 && unset BASH_ENV ENV case $- in # (((( *v*x* | *x*v* ) as_opts=-vx ;; *v* ) as_opts=-v ;; *x* ) as_opts=-x ;; * ) as_opts= ;; esac exec $CONFIG_SHELL $as_opts "$as_myself" ${1+"$@"} # Admittedly, this is quite paranoid, since all the known shells bail # out after a failed `exec'. $as_echo "$0: could not re-execute with $CONFIG_SHELL" >&2 as_fn_exit 255 fi # We don't want this to propagate to other subprocesses. { _as_can_reexec=; unset _as_can_reexec;} if test "x$CONFIG_SHELL" = x; then as_bourne_compatible="if test -n \"\${ZSH_VERSION+set}\" && (emulate sh) >/dev/null 2>&1; then : emulate sh NULLCMD=: # Pre-4.2 versions of Zsh do word splitting on \${1+\"\$@\"}, which # is contrary to our usage. Disable this feature. alias -g '\${1+\"\$@\"}'='\"\$@\"' setopt NO_GLOB_SUBST else case \`(set -o) 2>/dev/null\` in #( *posix*) : set -o posix ;; #( *) : ;; esac fi " as_required="as_fn_return () { (exit \$1); } as_fn_success () { as_fn_return 0; } as_fn_failure () { as_fn_return 1; } as_fn_ret_success () { return 0; } as_fn_ret_failure () { return 1; } exitcode=0 as_fn_success || { exitcode=1; echo as_fn_success failed.; } as_fn_failure && { exitcode=1; echo as_fn_failure succeeded.; } as_fn_ret_success || { exitcode=1; echo as_fn_ret_success failed.; } as_fn_ret_failure && { exitcode=1; echo as_fn_ret_failure succeeded.; } if ( set x; as_fn_ret_success y && test x = \"\$1\" ); then : else exitcode=1; echo positional parameters were not saved. fi test x\$exitcode = x0 || exit 1 test -x / || exit 1" as_suggested=" as_lineno_1=";as_suggested=$as_suggested$LINENO;as_suggested=$as_suggested" as_lineno_1a=\$LINENO as_lineno_2=";as_suggested=$as_suggested$LINENO;as_suggested=$as_suggested" as_lineno_2a=\$LINENO eval 'test \"x\$as_lineno_1'\$as_run'\" != \"x\$as_lineno_2'\$as_run'\" && test \"x\`expr \$as_lineno_1'\$as_run' + 1\`\" = \"x\$as_lineno_2'\$as_run'\"' || exit 1 test \$(( 1 + 1 )) = 2 || exit 1" if (eval "$as_required") 2>/dev/null; then : as_have_required=yes else as_have_required=no fi if test x$as_have_required = xyes && (eval "$as_suggested") 2>/dev/null; then : else as_save_IFS=$IFS; IFS=$PATH_SEPARATOR as_found=false for as_dir in /bin$PATH_SEPARATOR/usr/bin$PATH_SEPARATOR$PATH do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. as_found=: case $as_dir in #( /*) for as_base in sh bash ksh sh5; do # Try only shells that exist, to save several forks. as_shell=$as_dir/$as_base if { test -f "$as_shell" || test -f "$as_shell.exe"; } && { $as_echo "$as_bourne_compatible""$as_required" | as_run=a "$as_shell"; } 2>/dev/null; then : CONFIG_SHELL=$as_shell as_have_required=yes if { $as_echo "$as_bourne_compatible""$as_suggested" | as_run=a "$as_shell"; } 2>/dev/null; then : break 2 fi fi done;; esac as_found=false done $as_found || { if { test -f "$SHELL" || test -f "$SHELL.exe"; } && { $as_echo "$as_bourne_compatible""$as_required" | as_run=a "$SHELL"; } 2>/dev/null; then : CONFIG_SHELL=$SHELL as_have_required=yes fi; } IFS=$as_save_IFS if test "x$CONFIG_SHELL" != x; then : export CONFIG_SHELL # We cannot yet assume a decent shell, so we have to provide a # neutralization value for shells without unset; and this also # works around shells that cannot unset nonexistent variables. # Preserve -v and -x to the replacement shell. BASH_ENV=/dev/null ENV=/dev/null (unset BASH_ENV) >/dev/null 2>&1 && unset BASH_ENV ENV case $- in # (((( *v*x* | *x*v* ) as_opts=-vx ;; *v* ) as_opts=-v ;; *x* ) as_opts=-x ;; * ) as_opts= ;; esac exec $CONFIG_SHELL $as_opts "$as_myself" ${1+"$@"} # Admittedly, this is quite paranoid, since all the known shells bail # out after a failed `exec'. $as_echo "$0: could not re-execute with $CONFIG_SHELL" >&2 exit 255 fi if test x$as_have_required = xno; then : $as_echo "$0: This script requires a shell more modern than all" $as_echo "$0: the shells that I found on your system." if test x${ZSH_VERSION+set} = xset ; then $as_echo "$0: In particular, zsh $ZSH_VERSION has bugs and should" $as_echo "$0: be upgraded to zsh 4.3.4 or later." else $as_echo "$0: Please tell bug-autoconf@gnu.org about your system, $0: including any error possibly output before this $0: message. Then install a modern shell, or manually run $0: the script under such a shell if you do have one." fi exit 1 fi fi fi SHELL=${CONFIG_SHELL-/bin/sh} export SHELL # Unset more variables known to interfere with behavior of common tools. CLICOLOR_FORCE= GREP_OPTIONS= unset CLICOLOR_FORCE GREP_OPTIONS ## --------------------- ## ## M4sh Shell Functions. ## ## --------------------- ## # as_fn_unset VAR # --------------- # Portably unset VAR. as_fn_unset () { { eval $1=; unset $1;} } as_unset=as_fn_unset # as_fn_set_status STATUS # ----------------------- # Set $? to STATUS, without forking. as_fn_set_status () { return $1 } # as_fn_set_status # as_fn_exit STATUS # ----------------- # Exit the shell with STATUS, even in a "trap 0" or "set -e" context. as_fn_exit () { set +e as_fn_set_status $1 exit $1 } # as_fn_exit # as_fn_mkdir_p # ------------- # Create "$as_dir" as a directory, including parents if necessary. as_fn_mkdir_p () { case $as_dir in #( -*) as_dir=./$as_dir;; esac test -d "$as_dir" || eval $as_mkdir_p || { as_dirs= while :; do case $as_dir in #( *\'*) as_qdir=`$as_echo "$as_dir" | sed "s/'/'\\\\\\\\''/g"`;; #'( *) as_qdir=$as_dir;; esac as_dirs="'$as_qdir' $as_dirs" as_dir=`$as_dirname -- "$as_dir" || $as_expr X"$as_dir" : 'X\(.*[^/]\)//*[^/][^/]*/*$' \| \ X"$as_dir" : 'X\(//\)[^/]' \| \ X"$as_dir" : 'X\(//\)$' \| \ X"$as_dir" : 'X\(/\)' \| . 2>/dev/null || $as_echo X"$as_dir" | sed '/^X\(.*[^/]\)\/\/*[^/][^/]*\/*$/{ s//\1/ q } /^X\(\/\/\)[^/].*/{ s//\1/ q } /^X\(\/\/\)$/{ s//\1/ q } /^X\(\/\).*/{ s//\1/ q } s/.*/./; q'` test -d "$as_dir" && break done test -z "$as_dirs" || eval "mkdir $as_dirs" } || test -d "$as_dir" || as_fn_error $? "cannot create directory $as_dir" } # as_fn_mkdir_p # as_fn_executable_p FILE # ----------------------- # Test if FILE is an executable regular file. as_fn_executable_p () { test -f "$1" && test -x "$1" } # as_fn_executable_p # as_fn_append VAR VALUE # ---------------------- # Append the text in VALUE to the end of the definition contained in VAR. Take # advantage of any shell optimizations that allow amortized linear growth over # repeated appends, instead of the typical quadratic growth present in naive # implementations. if (eval "as_var=1; as_var+=2; test x\$as_var = x12") 2>/dev/null; then : eval 'as_fn_append () { eval $1+=\$2 }' else as_fn_append () { eval $1=\$$1\$2 } fi # as_fn_append # as_fn_arith ARG... # ------------------ # Perform arithmetic evaluation on the ARGs, and store the result in the # global $as_val. Take advantage of shells that can avoid forks. The arguments # must be portable across $(()) and expr. if (eval "test \$(( 1 + 1 )) = 2") 2>/dev/null; then : eval 'as_fn_arith () { as_val=$(( $* )) }' else as_fn_arith () { as_val=`expr "$@" || test $? -eq 1` } fi # as_fn_arith # as_fn_error STATUS ERROR [LINENO LOG_FD] # ---------------------------------------- # Output "`basename $0`: error: ERROR" to stderr. If LINENO and LOG_FD are # provided, also output the error to LOG_FD, referencing LINENO. Then exit the # script with STATUS, using 1 if that was 0. as_fn_error () { as_status=$1; test $as_status -eq 0 && as_status=1 if test "$4"; then as_lineno=${as_lineno-"$3"} as_lineno_stack=as_lineno_stack=$as_lineno_stack $as_echo "$as_me:${as_lineno-$LINENO}: error: $2" >&$4 fi $as_echo "$as_me: error: $2" >&2 as_fn_exit $as_status } # as_fn_error if expr a : '\(a\)' >/dev/null 2>&1 && test "X`expr 00001 : '.*\(...\)'`" = X001; then as_expr=expr else as_expr=false fi if (basename -- /) >/dev/null 2>&1 && test "X`basename -- / 2>&1`" = "X/"; then as_basename=basename else as_basename=false fi if (as_dir=`dirname -- /` && test "X$as_dir" = X/) >/dev/null 2>&1; then as_dirname=dirname else as_dirname=false fi as_me=`$as_basename -- "$0" || $as_expr X/"$0" : '.*/\([^/][^/]*\)/*$' \| \ X"$0" : 'X\(//\)$' \| \ X"$0" : 'X\(/\)' \| . 2>/dev/null || $as_echo X/"$0" | sed '/^.*\/\([^/][^/]*\)\/*$/{ s//\1/ q } /^X\/\(\/\/\)$/{ s//\1/ q } /^X\/\(\/\).*/{ s//\1/ q } s/.*/./; q'` # Avoid depending upon Character Ranges. as_cr_letters='abcdefghijklmnopqrstuvwxyz' as_cr_LETTERS='ABCDEFGHIJKLMNOPQRSTUVWXYZ' as_cr_Letters=$as_cr_letters$as_cr_LETTERS as_cr_digits='0123456789' as_cr_alnum=$as_cr_Letters$as_cr_digits as_lineno_1=$LINENO as_lineno_1a=$LINENO as_lineno_2=$LINENO as_lineno_2a=$LINENO eval 'test "x$as_lineno_1'$as_run'" != "x$as_lineno_2'$as_run'" && test "x`expr $as_lineno_1'$as_run' + 1`" = "x$as_lineno_2'$as_run'"' || { # Blame Lee E. McMahon (1931-1989) for sed's syntax. :-) sed -n ' p /[$]LINENO/= ' <$as_myself | sed ' s/[$]LINENO.*/&-/ t lineno b :lineno N :loop s/[$]LINENO\([^'$as_cr_alnum'_].*\n\)\(.*\)/\2\1\2/ t loop s/-\n.*// ' >$as_me.lineno && chmod +x "$as_me.lineno" || { $as_echo "$as_me: error: cannot create $as_me.lineno; rerun with a POSIX shell" >&2; as_fn_exit 1; } # If we had to re-execute with $CONFIG_SHELL, we're ensured to have # already done that, so ensure we don't try to do so again and fall # in an infinite loop. This has already happened in practice. _as_can_reexec=no; export _as_can_reexec # Don't try to exec as it changes $[0], causing all sort of problems # (the dirname of $[0] is not the place where we might find the # original and so on. Autoconf is especially sensitive to this). . "./$as_me.lineno" # Exit status is that of the last command. exit } ECHO_C= ECHO_N= ECHO_T= case `echo -n x` in #((((( -n*) case `echo 'xy\c'` in *c*) ECHO_T=' ';; # ECHO_T is single tab character. xy) ECHO_C='\c';; *) echo `echo ksh88 bug on AIX 6.1` > /dev/null ECHO_T=' ';; esac;; *) ECHO_N='-n';; esac rm -f conf$$ conf$$.exe conf$$.file if test -d conf$$.dir; then rm -f conf$$.dir/conf$$.file else rm -f conf$$.dir mkdir conf$$.dir 2>/dev/null fi if (echo >conf$$.file) 2>/dev/null; then if ln -s conf$$.file conf$$ 2>/dev/null; then as_ln_s='ln -s' # ... but there are two gotchas: # 1) On MSYS, both `ln -s file dir' and `ln file dir' fail. # 2) DJGPP < 2.04 has no symlinks; `ln -s' creates a wrapper executable. # In both cases, we have to default to `cp -pR'. ln -s conf$$.file conf$$.dir 2>/dev/null && test ! -f conf$$.exe || as_ln_s='cp -pR' elif ln conf$$.file conf$$ 2>/dev/null; then as_ln_s=ln else as_ln_s='cp -pR' fi else as_ln_s='cp -pR' fi rm -f conf$$ conf$$.exe conf$$.dir/conf$$.file conf$$.file rmdir conf$$.dir 2>/dev/null if mkdir -p . 2>/dev/null; then as_mkdir_p='mkdir -p "$as_dir"' else test -d ./-p && rmdir ./-p as_mkdir_p=false fi as_test_x='test -x' as_executable_p=as_fn_executable_p # Sed expression to map a string onto a valid CPP name. as_tr_cpp="eval sed 'y%*$as_cr_letters%P$as_cr_LETTERS%;s%[^_$as_cr_alnum]%_%g'" # Sed expression to map a string onto a valid variable name. as_tr_sh="eval sed 'y%*+%pp%;s%[^_$as_cr_alnum]%_%g'" test -n "$DJDIR" || exec 7<&0 &1 # Name of the host. # hostname on some systems (SVR3.2, old GNU/Linux) returns a bogus exit status, # so uname gets run too. ac_hostname=`(hostname || uname -n) 2>/dev/null | sed 1q` # # Initializations. # ac_default_prefix=/usr/local ac_clean_files= ac_config_libobj_dir=. LIBOBJS= cross_compiling=no subdirs= MFLAGS= MAKEFLAGS= # Identity of this package. PACKAGE_NAME='Stacks' PACKAGE_TARNAME='stacks' PACKAGE_VERSION='1.44' PACKAGE_STRING='Stacks 1.44' PACKAGE_BUGREPORT='' PACKAGE_URL='' ac_unique_file="src/ustacks.cc" # Factoring default headers for most tests. ac_includes_default="\ #include #ifdef HAVE_SYS_TYPES_H # include #endif #ifdef HAVE_SYS_STAT_H # include #endif #ifdef STDC_HEADERS # include # include #else # ifdef HAVE_STDLIB_H # include # endif #endif #ifdef HAVE_STRING_H # if !defined STDC_HEADERS && defined HAVE_MEMORY_H # include # endif # include #endif #ifdef HAVE_STRINGS_H # include #endif #ifdef HAVE_INTTYPES_H # include #endif #ifdef HAVE_STDINT_H # include #endif #ifdef HAVE_UNISTD_H # include #endif" ac_subst_vars='am__EXEEXT_FALSE am__EXEEXT_TRUE LTLIBOBJS LIBOBJS OPENMP_CFLAGS EGREP GREP CPP HAVE_CXX11 am__fastdepCC_FALSE am__fastdepCC_TRUE CCDEPMODE ac_ct_CC CFLAGS CC am__fastdepCXX_FALSE am__fastdepCXX_TRUE CXXDEPMODE am__nodep AMDEPBACKSLASH AMDEP_FALSE AMDEP_TRUE am__quote am__include DEPDIR OBJEXT EXEEXT ac_ct_CXX CPPFLAGS LDFLAGS CXXFLAGS CXX SPARSEHASH_CFLAGS BAM_LIBS BAM_CFLAGS AM_BACKSLASH AM_DEFAULT_VERBOSITY AM_DEFAULT_V AM_V am__untar am__tar AMTAR am__leading_dot SET_MAKE AWK mkdir_p MKDIR_P INSTALL_STRIP_PROGRAM STRIP install_sh MAKEINFO AUTOHEADER AUTOMAKE AUTOCONF ACLOCAL VERSION PACKAGE CYGPATH_W am__isrc INSTALL_DATA INSTALL_SCRIPT INSTALL_PROGRAM target_alias host_alias build_alias LIBS ECHO_T ECHO_N ECHO_C DEFS mandir localedir libdir psdir pdfdir dvidir htmldir infodir docdir oldincludedir includedir localstatedir sharedstatedir sysconfdir datadir datarootdir libexecdir sbindir bindir program_transform_name prefix exec_prefix PACKAGE_URL PACKAGE_BUGREPORT PACKAGE_STRING PACKAGE_VERSION PACKAGE_TARNAME PACKAGE_NAME PATH_SEPARATOR SHELL' ac_subst_files='' ac_user_opts=' enable_option_checking enable_silent_rules enable_bam enable_sparsehash with_sparsehash_include_path enable_dependency_tracking enable_openmp ' ac_precious_vars='build_alias host_alias target_alias CXX CXXFLAGS LDFLAGS LIBS CPPFLAGS CCC CC CFLAGS CPP' # Initialize some variables set by options. ac_init_help= ac_init_version=false ac_unrecognized_opts= ac_unrecognized_sep= # The variables have the same names as the options, with # dashes changed to underlines. cache_file=/dev/null exec_prefix=NONE no_create= no_recursion= prefix=NONE program_prefix=NONE program_suffix=NONE program_transform_name=s,x,x, silent= site= srcdir= verbose= x_includes=NONE x_libraries=NONE # Installation directory options. # These are left unexpanded so users can "make install exec_prefix=/foo" # and all the variables that are supposed to be based on exec_prefix # by default will actually change. # Use braces instead of parens because sh, perl, etc. also accept them. # (The list follows the same order as the GNU Coding Standards.) bindir='${exec_prefix}/bin' sbindir='${exec_prefix}/sbin' libexecdir='${exec_prefix}/libexec' datarootdir='${prefix}/share' datadir='${datarootdir}' sysconfdir='${prefix}/etc' sharedstatedir='${prefix}/com' localstatedir='${prefix}/var' includedir='${prefix}/include' oldincludedir='/usr/include' docdir='${datarootdir}/doc/${PACKAGE_TARNAME}' infodir='${datarootdir}/info' htmldir='${docdir}' dvidir='${docdir}' pdfdir='${docdir}' psdir='${docdir}' libdir='${exec_prefix}/lib' localedir='${datarootdir}/locale' mandir='${datarootdir}/man' ac_prev= ac_dashdash= for ac_option do # If the previous option needs an argument, assign it. if test -n "$ac_prev"; then eval $ac_prev=\$ac_option ac_prev= continue fi case $ac_option in *=?*) ac_optarg=`expr "X$ac_option" : '[^=]*=\(.*\)'` ;; *=) ac_optarg= ;; *) ac_optarg=yes ;; esac # Accept the important Cygnus configure options, so we can diagnose typos. case $ac_dashdash$ac_option in --) ac_dashdash=yes ;; -bindir | --bindir | --bindi | --bind | --bin | --bi) ac_prev=bindir ;; -bindir=* | --bindir=* | --bindi=* | --bind=* | --bin=* | --bi=*) bindir=$ac_optarg ;; -build | --build | --buil | --bui | --bu) ac_prev=build_alias ;; -build=* | --build=* | --buil=* | --bui=* | --bu=*) build_alias=$ac_optarg ;; -cache-file | --cache-file | --cache-fil | --cache-fi \ | --cache-f | --cache- | --cache | --cach | --cac | --ca | --c) ac_prev=cache_file ;; -cache-file=* | --cache-file=* | --cache-fil=* | --cache-fi=* \ | --cache-f=* | --cache-=* | --cache=* | --cach=* | --cac=* | --ca=* | --c=*) cache_file=$ac_optarg ;; --config-cache | -C) cache_file=config.cache ;; -datadir | --datadir | --datadi | --datad) ac_prev=datadir ;; -datadir=* | --datadir=* | --datadi=* | --datad=*) datadir=$ac_optarg ;; -datarootdir | --datarootdir | --datarootdi | --datarootd | --dataroot \ | --dataroo | --dataro | --datar) ac_prev=datarootdir ;; -datarootdir=* | --datarootdir=* | --datarootdi=* | --datarootd=* \ | --dataroot=* | --dataroo=* | --dataro=* | --datar=*) datarootdir=$ac_optarg ;; -disable-* | --disable-*) ac_useropt=`expr "x$ac_option" : 'x-*disable-\(.*\)'` # Reject names that are not valid shell variable names. expr "x$ac_useropt" : ".*[^-+._$as_cr_alnum]" >/dev/null && as_fn_error $? "invalid feature name: $ac_useropt" ac_useropt_orig=$ac_useropt ac_useropt=`$as_echo "$ac_useropt" | sed 's/[-+.]/_/g'` case $ac_user_opts in *" "enable_$ac_useropt" "*) ;; *) ac_unrecognized_opts="$ac_unrecognized_opts$ac_unrecognized_sep--disable-$ac_useropt_orig" ac_unrecognized_sep=', ';; esac eval enable_$ac_useropt=no ;; -docdir | --docdir | --docdi | --doc | --do) ac_prev=docdir ;; -docdir=* | --docdir=* | --docdi=* | --doc=* | --do=*) docdir=$ac_optarg ;; -dvidir | --dvidir | --dvidi | --dvid | --dvi | --dv) ac_prev=dvidir ;; -dvidir=* | --dvidir=* | --dvidi=* | --dvid=* | --dvi=* | --dv=*) dvidir=$ac_optarg ;; -enable-* | --enable-*) ac_useropt=`expr "x$ac_option" : 'x-*enable-\([^=]*\)'` # Reject names that are not valid shell variable names. expr "x$ac_useropt" : ".*[^-+._$as_cr_alnum]" >/dev/null && as_fn_error $? "invalid feature name: $ac_useropt" ac_useropt_orig=$ac_useropt ac_useropt=`$as_echo "$ac_useropt" | sed 's/[-+.]/_/g'` case $ac_user_opts in *" "enable_$ac_useropt" "*) ;; *) ac_unrecognized_opts="$ac_unrecognized_opts$ac_unrecognized_sep--enable-$ac_useropt_orig" ac_unrecognized_sep=', ';; esac eval enable_$ac_useropt=\$ac_optarg ;; -exec-prefix | --exec_prefix | --exec-prefix | --exec-prefi \ | --exec-pref | --exec-pre | --exec-pr | --exec-p | --exec- \ | --exec | --exe | --ex) ac_prev=exec_prefix ;; -exec-prefix=* | --exec_prefix=* | --exec-prefix=* | --exec-prefi=* \ | --exec-pref=* | --exec-pre=* | --exec-pr=* | --exec-p=* | --exec-=* \ | --exec=* | --exe=* | --ex=*) exec_prefix=$ac_optarg ;; -gas | --gas | --ga | --g) # Obsolete; use --with-gas. with_gas=yes ;; -help | --help | --hel | --he | -h) ac_init_help=long ;; -help=r* | --help=r* | --hel=r* | --he=r* | -hr*) ac_init_help=recursive ;; -help=s* | --help=s* | --hel=s* | --he=s* | -hs*) ac_init_help=short ;; -host | --host | --hos | --ho) ac_prev=host_alias ;; -host=* | --host=* | --hos=* | --ho=*) host_alias=$ac_optarg ;; -htmldir | --htmldir | --htmldi | --htmld | --html | --htm | --ht) ac_prev=htmldir ;; -htmldir=* | --htmldir=* | --htmldi=* | --htmld=* | --html=* | --htm=* \ | --ht=*) htmldir=$ac_optarg ;; -includedir | --includedir | --includedi | --included | --include \ | --includ | --inclu | --incl | --inc) ac_prev=includedir ;; -includedir=* | --includedir=* | --includedi=* | --included=* | --include=* \ | --includ=* | --inclu=* | --incl=* | --inc=*) includedir=$ac_optarg ;; -infodir | --infodir | --infodi | --infod | --info | --inf) ac_prev=infodir ;; -infodir=* | --infodir=* | --infodi=* | --infod=* | --info=* | --inf=*) infodir=$ac_optarg ;; -libdir | --libdir | --libdi | --libd) ac_prev=libdir ;; -libdir=* | --libdir=* | --libdi=* | --libd=*) libdir=$ac_optarg ;; -libexecdir | --libexecdir | --libexecdi | --libexecd | --libexec \ | --libexe | --libex | --libe) ac_prev=libexecdir ;; -libexecdir=* | --libexecdir=* | --libexecdi=* | --libexecd=* | --libexec=* \ | --libexe=* | --libex=* | --libe=*) libexecdir=$ac_optarg ;; -localedir | --localedir | --localedi | --localed | --locale) ac_prev=localedir ;; -localedir=* | --localedir=* | --localedi=* | --localed=* | --locale=*) localedir=$ac_optarg ;; -localstatedir | --localstatedir | --localstatedi | --localstated \ | --localstate | --localstat | --localsta | --localst | --locals) ac_prev=localstatedir ;; -localstatedir=* | --localstatedir=* | --localstatedi=* | --localstated=* \ | --localstate=* | --localstat=* | --localsta=* | --localst=* | --locals=*) localstatedir=$ac_optarg ;; -mandir | --mandir | --mandi | --mand | --man | --ma | --m) ac_prev=mandir ;; -mandir=* | --mandir=* | --mandi=* | --mand=* | --man=* | --ma=* | --m=*) mandir=$ac_optarg ;; -nfp | --nfp | --nf) # Obsolete; use --without-fp. with_fp=no ;; -no-create | --no-create | --no-creat | --no-crea | --no-cre \ | --no-cr | --no-c | -n) no_create=yes ;; -no-recursion | --no-recursion | --no-recursio | --no-recursi \ | --no-recurs | --no-recur | --no-recu | --no-rec | --no-re | --no-r) no_recursion=yes ;; -oldincludedir | --oldincludedir | --oldincludedi | --oldincluded \ | --oldinclude | --oldinclud | --oldinclu | --oldincl | --oldinc \ | --oldin | --oldi | --old | --ol | --o) ac_prev=oldincludedir ;; -oldincludedir=* | --oldincludedir=* | --oldincludedi=* | --oldincluded=* \ | --oldinclude=* | --oldinclud=* | --oldinclu=* | --oldincl=* | --oldinc=* \ | --oldin=* | --oldi=* | --old=* | --ol=* | --o=*) oldincludedir=$ac_optarg ;; -prefix | --prefix | --prefi | --pref | --pre | --pr | --p) ac_prev=prefix ;; -prefix=* | --prefix=* | --prefi=* | --pref=* | --pre=* | --pr=* | --p=*) prefix=$ac_optarg ;; -program-prefix | --program-prefix | --program-prefi | --program-pref \ | --program-pre | --program-pr | --program-p) ac_prev=program_prefix ;; -program-prefix=* | --program-prefix=* | --program-prefi=* \ | --program-pref=* | --program-pre=* | --program-pr=* | --program-p=*) program_prefix=$ac_optarg ;; -program-suffix | --program-suffix | --program-suffi | --program-suff \ | --program-suf | --program-su | --program-s) ac_prev=program_suffix ;; -program-suffix=* | --program-suffix=* | --program-suffi=* \ | --program-suff=* | --program-suf=* | --program-su=* | --program-s=*) program_suffix=$ac_optarg ;; -program-transform-name | --program-transform-name \ | --program-transform-nam | --program-transform-na \ | --program-transform-n | --program-transform- \ | --program-transform | --program-transfor \ | --program-transfo | --program-transf \ | --program-trans | --program-tran \ | --progr-tra | --program-tr | --program-t) ac_prev=program_transform_name ;; -program-transform-name=* | --program-transform-name=* \ | --program-transform-nam=* | --program-transform-na=* \ | --program-transform-n=* | --program-transform-=* \ | --program-transform=* | --program-transfor=* \ | --program-transfo=* | --program-transf=* \ | --program-trans=* | --program-tran=* \ | --progr-tra=* | --program-tr=* | --program-t=*) program_transform_name=$ac_optarg ;; -pdfdir | --pdfdir | --pdfdi | --pdfd | --pdf | --pd) ac_prev=pdfdir ;; -pdfdir=* | --pdfdir=* | --pdfdi=* | --pdfd=* | --pdf=* | --pd=*) pdfdir=$ac_optarg ;; -psdir | --psdir | --psdi | --psd | --ps) ac_prev=psdir ;; -psdir=* | --psdir=* | --psdi=* | --psd=* | --ps=*) psdir=$ac_optarg ;; -q | -quiet | --quiet | --quie | --qui | --qu | --q \ | -silent | --silent | --silen | --sile | --sil) silent=yes ;; -sbindir | --sbindir | --sbindi | --sbind | --sbin | --sbi | --sb) ac_prev=sbindir ;; -sbindir=* | --sbindir=* | --sbindi=* | --sbind=* | --sbin=* \ | --sbi=* | --sb=*) sbindir=$ac_optarg ;; -sharedstatedir | --sharedstatedir | --sharedstatedi \ | --sharedstated | --sharedstate | --sharedstat | --sharedsta \ | --sharedst | --shareds | --shared | --share | --shar \ | --sha | --sh) ac_prev=sharedstatedir ;; -sharedstatedir=* | --sharedstatedir=* | --sharedstatedi=* \ | --sharedstated=* | --sharedstate=* | --sharedstat=* | --sharedsta=* \ | --sharedst=* | --shareds=* | --shared=* | --share=* | --shar=* \ | --sha=* | --sh=*) sharedstatedir=$ac_optarg ;; -site | --site | --sit) ac_prev=site ;; -site=* | --site=* | --sit=*) site=$ac_optarg ;; -srcdir | --srcdir | --srcdi | --srcd | --src | --sr) ac_prev=srcdir ;; -srcdir=* | --srcdir=* | --srcdi=* | --srcd=* | --src=* | --sr=*) srcdir=$ac_optarg ;; -sysconfdir | --sysconfdir | --sysconfdi | --sysconfd | --sysconf \ | --syscon | --sysco | --sysc | --sys | --sy) ac_prev=sysconfdir ;; -sysconfdir=* | --sysconfdir=* | --sysconfdi=* | --sysconfd=* | --sysconf=* \ | --syscon=* | --sysco=* | --sysc=* | --sys=* | --sy=*) sysconfdir=$ac_optarg ;; -target | --target | --targe | --targ | --tar | --ta | --t) ac_prev=target_alias ;; -target=* | --target=* | --targe=* | --targ=* | --tar=* | --ta=* | --t=*) target_alias=$ac_optarg ;; -v | -verbose | --verbose | --verbos | --verbo | --verb) verbose=yes ;; -version | --version | --versio | --versi | --vers | -V) ac_init_version=: ;; -with-* | --with-*) ac_useropt=`expr "x$ac_option" : 'x-*with-\([^=]*\)'` # Reject names that are not valid shell variable names. expr "x$ac_useropt" : ".*[^-+._$as_cr_alnum]" >/dev/null && as_fn_error $? "invalid package name: $ac_useropt" ac_useropt_orig=$ac_useropt ac_useropt=`$as_echo "$ac_useropt" | sed 's/[-+.]/_/g'` case $ac_user_opts in *" "with_$ac_useropt" "*) ;; *) ac_unrecognized_opts="$ac_unrecognized_opts$ac_unrecognized_sep--with-$ac_useropt_orig" ac_unrecognized_sep=', ';; esac eval with_$ac_useropt=\$ac_optarg ;; -without-* | --without-*) ac_useropt=`expr "x$ac_option" : 'x-*without-\(.*\)'` # Reject names that are not valid shell variable names. expr "x$ac_useropt" : ".*[^-+._$as_cr_alnum]" >/dev/null && as_fn_error $? "invalid package name: $ac_useropt" ac_useropt_orig=$ac_useropt ac_useropt=`$as_echo "$ac_useropt" | sed 's/[-+.]/_/g'` case $ac_user_opts in *" "with_$ac_useropt" "*) ;; *) ac_unrecognized_opts="$ac_unrecognized_opts$ac_unrecognized_sep--without-$ac_useropt_orig" ac_unrecognized_sep=', ';; esac eval with_$ac_useropt=no ;; --x) # Obsolete; use --with-x. with_x=yes ;; -x-includes | --x-includes | --x-include | --x-includ | --x-inclu \ | --x-incl | --x-inc | --x-in | --x-i) ac_prev=x_includes ;; -x-includes=* | --x-includes=* | --x-include=* | --x-includ=* | --x-inclu=* \ | --x-incl=* | --x-inc=* | --x-in=* | --x-i=*) x_includes=$ac_optarg ;; -x-libraries | --x-libraries | --x-librarie | --x-librari \ | --x-librar | --x-libra | --x-libr | --x-lib | --x-li | --x-l) ac_prev=x_libraries ;; -x-libraries=* | --x-libraries=* | --x-librarie=* | --x-librari=* \ | --x-librar=* | --x-libra=* | --x-libr=* | --x-lib=* | --x-li=* | --x-l=*) x_libraries=$ac_optarg ;; -*) as_fn_error $? "unrecognized option: \`$ac_option' Try \`$0 --help' for more information" ;; *=*) ac_envvar=`expr "x$ac_option" : 'x\([^=]*\)='` # Reject names that are not valid shell variable names. case $ac_envvar in #( '' | [0-9]* | *[!_$as_cr_alnum]* ) as_fn_error $? "invalid variable name: \`$ac_envvar'" ;; esac eval $ac_envvar=\$ac_optarg export $ac_envvar ;; *) # FIXME: should be removed in autoconf 3.0. $as_echo "$as_me: WARNING: you should use --build, --host, --target" >&2 expr "x$ac_option" : ".*[^-._$as_cr_alnum]" >/dev/null && $as_echo "$as_me: WARNING: invalid host type: $ac_option" >&2 : "${build_alias=$ac_option} ${host_alias=$ac_option} ${target_alias=$ac_option}" ;; esac done if test -n "$ac_prev"; then ac_option=--`echo $ac_prev | sed 's/_/-/g'` as_fn_error $? "missing argument to $ac_option" fi if test -n "$ac_unrecognized_opts"; then case $enable_option_checking in no) ;; fatal) as_fn_error $? "unrecognized options: $ac_unrecognized_opts" ;; *) $as_echo "$as_me: WARNING: unrecognized options: $ac_unrecognized_opts" >&2 ;; esac fi # Check all directory arguments for consistency. for ac_var in exec_prefix prefix bindir sbindir libexecdir datarootdir \ datadir sysconfdir sharedstatedir localstatedir includedir \ oldincludedir docdir infodir htmldir dvidir pdfdir psdir \ libdir localedir mandir do eval ac_val=\$$ac_var # Remove trailing slashes. case $ac_val in */ ) ac_val=`expr "X$ac_val" : 'X\(.*[^/]\)' \| "X$ac_val" : 'X\(.*\)'` eval $ac_var=\$ac_val;; esac # Be sure to have absolute directory names. case $ac_val in [\\/$]* | ?:[\\/]* ) continue;; NONE | '' ) case $ac_var in *prefix ) continue;; esac;; esac as_fn_error $? "expected an absolute directory name for --$ac_var: $ac_val" done # There might be people who depend on the old broken behavior: `$host' # used to hold the argument of --host etc. # FIXME: To remove some day. build=$build_alias host=$host_alias target=$target_alias # FIXME: To remove some day. if test "x$host_alias" != x; then if test "x$build_alias" = x; then cross_compiling=maybe elif test "x$build_alias" != "x$host_alias"; then cross_compiling=yes fi fi ac_tool_prefix= test -n "$host_alias" && ac_tool_prefix=$host_alias- test "$silent" = yes && exec 6>/dev/null ac_pwd=`pwd` && test -n "$ac_pwd" && ac_ls_di=`ls -di .` && ac_pwd_ls_di=`cd "$ac_pwd" && ls -di .` || as_fn_error $? "working directory cannot be determined" test "X$ac_ls_di" = "X$ac_pwd_ls_di" || as_fn_error $? "pwd does not report name of working directory" # Find the source files, if location was not specified. if test -z "$srcdir"; then ac_srcdir_defaulted=yes # Try the directory containing this script, then the parent directory. ac_confdir=`$as_dirname -- "$as_myself" || $as_expr X"$as_myself" : 'X\(.*[^/]\)//*[^/][^/]*/*$' \| \ X"$as_myself" : 'X\(//\)[^/]' \| \ X"$as_myself" : 'X\(//\)$' \| \ X"$as_myself" : 'X\(/\)' \| . 2>/dev/null || $as_echo X"$as_myself" | sed '/^X\(.*[^/]\)\/\/*[^/][^/]*\/*$/{ s//\1/ q } /^X\(\/\/\)[^/].*/{ s//\1/ q } /^X\(\/\/\)$/{ s//\1/ q } /^X\(\/\).*/{ s//\1/ q } s/.*/./; q'` srcdir=$ac_confdir if test ! -r "$srcdir/$ac_unique_file"; then srcdir=.. fi else ac_srcdir_defaulted=no fi if test ! -r "$srcdir/$ac_unique_file"; then test "$ac_srcdir_defaulted" = yes && srcdir="$ac_confdir or .." as_fn_error $? "cannot find sources ($ac_unique_file) in $srcdir" fi ac_msg="sources are in $srcdir, but \`cd $srcdir' does not work" ac_abs_confdir=`( cd "$srcdir" && test -r "./$ac_unique_file" || as_fn_error $? "$ac_msg" pwd)` # When building in place, set srcdir=. if test "$ac_abs_confdir" = "$ac_pwd"; then srcdir=. fi # Remove unnecessary trailing slashes from srcdir. # Double slashes in file names in object file debugging info # mess up M-x gdb in Emacs. case $srcdir in */) srcdir=`expr "X$srcdir" : 'X\(.*[^/]\)' \| "X$srcdir" : 'X\(.*\)'`;; esac for ac_var in $ac_precious_vars; do eval ac_env_${ac_var}_set=\${${ac_var}+set} eval ac_env_${ac_var}_value=\$${ac_var} eval ac_cv_env_${ac_var}_set=\${${ac_var}+set} eval ac_cv_env_${ac_var}_value=\$${ac_var} done # # Report the --help message. # if test "$ac_init_help" = "long"; then # Omit some internal or obsolete options to make the list less imposing. # This message is too long to be a string in the A/UX 3.1 sh. cat <<_ACEOF \`configure' configures Stacks 1.44 to adapt to many kinds of systems. Usage: $0 [OPTION]... [VAR=VALUE]... To assign environment variables (e.g., CC, CFLAGS...), specify them as VAR=VALUE. See below for descriptions of some of the useful variables. Defaults for the options are specified in brackets. Configuration: -h, --help display this help and exit --help=short display options specific to this package --help=recursive display the short help of all the included packages -V, --version display version information and exit -q, --quiet, --silent do not print \`checking ...' messages --cache-file=FILE cache test results in FILE [disabled] -C, --config-cache alias for \`--cache-file=config.cache' -n, --no-create do not create output files --srcdir=DIR find the sources in DIR [configure dir or \`..'] Installation directories: --prefix=PREFIX install architecture-independent files in PREFIX [$ac_default_prefix] --exec-prefix=EPREFIX install architecture-dependent files in EPREFIX [PREFIX] By default, \`make install' will install all the files in \`$ac_default_prefix/bin', \`$ac_default_prefix/lib' etc. You can specify an installation prefix other than \`$ac_default_prefix' using \`--prefix', for instance \`--prefix=\$HOME'. For better control, use the options below. Fine tuning of the installation directories: --bindir=DIR user executables [EPREFIX/bin] --sbindir=DIR system admin executables [EPREFIX/sbin] --libexecdir=DIR program executables [EPREFIX/libexec] --sysconfdir=DIR read-only single-machine data [PREFIX/etc] --sharedstatedir=DIR modifiable architecture-independent data [PREFIX/com] --localstatedir=DIR modifiable single-machine data [PREFIX/var] --libdir=DIR object code libraries [EPREFIX/lib] --includedir=DIR C header files [PREFIX/include] --oldincludedir=DIR C header files for non-gcc [/usr/include] --datarootdir=DIR read-only arch.-independent data root [PREFIX/share] --datadir=DIR read-only architecture-independent data [DATAROOTDIR] --infodir=DIR info documentation [DATAROOTDIR/info] --localedir=DIR locale-dependent data [DATAROOTDIR/locale] --mandir=DIR man documentation [DATAROOTDIR/man] --docdir=DIR documentation root [DATAROOTDIR/doc/stacks] --htmldir=DIR html documentation [DOCDIR] --dvidir=DIR dvi documentation [DOCDIR] --pdfdir=DIR pdf documentation [DOCDIR] --psdir=DIR ps documentation [DOCDIR] _ACEOF cat <<\_ACEOF Program names: --program-prefix=PREFIX prepend PREFIX to installed program names --program-suffix=SUFFIX append SUFFIX to installed program names --program-transform-name=PROGRAM run sed PROGRAM on installed program names _ACEOF fi if test -n "$ac_init_help"; then case $ac_init_help in short | recursive ) echo "Configuration of Stacks 1.44:";; esac cat <<\_ACEOF Optional Features: --disable-option-checking ignore unrecognized --enable/--with options --disable-FEATURE do not include FEATURE (same as --enable-FEATURE=no) --enable-FEATURE[=ARG] include FEATURE [ARG=yes] --enable-silent-rules less verbose build output (undo: "make V=1") --disable-silent-rules verbose build output (undo: "make V=0") --disable-bam Disable use of BAM files through HTSLib. --enable-sparsehash Enable the use of Google Sparsehash (must be installed). --enable-dependency-tracking do not reject slow dependency extractors --disable-dependency-tracking speeds up one-time build --disable-openmp do not use OpenMP Optional Packages: --with-PACKAGE[=ARG] use PACKAGE [ARG=yes] --without-PACKAGE do not use PACKAGE (same as --with-PACKAGE=no) --with-sparsehash-include-path location of Google Sparsehash headers Some influential environment variables: CXX C++ compiler command CXXFLAGS C++ compiler flags LDFLAGS linker flags, e.g. -L if you have libraries in a nonstandard directory LIBS libraries to pass to the linker, e.g. -l CPPFLAGS (Objective) C/C++ preprocessor flags, e.g. -I if you have headers in a nonstandard directory CC C compiler command CFLAGS C compiler flags CPP C preprocessor Use these variables to override the choices made by `configure' or to help it to find libraries and programs with nonstandard names/locations. Report bugs to the package provider. _ACEOF ac_status=$? fi if test "$ac_init_help" = "recursive"; then # If there are subdirs, report their specific --help. for ac_dir in : $ac_subdirs_all; do test "x$ac_dir" = x: && continue test -d "$ac_dir" || { cd "$srcdir" && ac_pwd=`pwd` && srcdir=. && test -d "$ac_dir"; } || continue ac_builddir=. case "$ac_dir" in .) ac_dir_suffix= ac_top_builddir_sub=. ac_top_build_prefix= ;; *) ac_dir_suffix=/`$as_echo "$ac_dir" | sed 's|^\.[\\/]||'` # A ".." for each directory in $ac_dir_suffix. ac_top_builddir_sub=`$as_echo "$ac_dir_suffix" | sed 's|/[^\\/]*|/..|g;s|/||'` case $ac_top_builddir_sub in "") ac_top_builddir_sub=. ac_top_build_prefix= ;; *) ac_top_build_prefix=$ac_top_builddir_sub/ ;; esac ;; esac ac_abs_top_builddir=$ac_pwd ac_abs_builddir=$ac_pwd$ac_dir_suffix # for backward compatibility: ac_top_builddir=$ac_top_build_prefix case $srcdir in .) # We are building in place. ac_srcdir=. ac_top_srcdir=$ac_top_builddir_sub ac_abs_top_srcdir=$ac_pwd ;; [\\/]* | ?:[\\/]* ) # Absolute name. ac_srcdir=$srcdir$ac_dir_suffix; ac_top_srcdir=$srcdir ac_abs_top_srcdir=$srcdir ;; *) # Relative name. ac_srcdir=$ac_top_build_prefix$srcdir$ac_dir_suffix ac_top_srcdir=$ac_top_build_prefix$srcdir ac_abs_top_srcdir=$ac_pwd/$srcdir ;; esac ac_abs_srcdir=$ac_abs_top_srcdir$ac_dir_suffix cd "$ac_dir" || { ac_status=$?; continue; } # Check for guested configure. if test -f "$ac_srcdir/configure.gnu"; then echo && $SHELL "$ac_srcdir/configure.gnu" --help=recursive elif test -f "$ac_srcdir/configure"; then echo && $SHELL "$ac_srcdir/configure" --help=recursive else $as_echo "$as_me: WARNING: no configuration information is in $ac_dir" >&2 fi || ac_status=$? cd "$ac_pwd" || { ac_status=$?; break; } done fi test -n "$ac_init_help" && exit $ac_status if $ac_init_version; then cat <<\_ACEOF Stacks configure 1.44 generated by GNU Autoconf 2.69 Copyright (C) 2012 Free Software Foundation, Inc. This configure script is free software; the Free Software Foundation gives unlimited permission to copy, distribute and modify it. _ACEOF exit fi ## ------------------------ ## ## Autoconf initialization. ## ## ------------------------ ## # ac_fn_cxx_try_compile LINENO # ---------------------------- # Try to compile conftest.$ac_ext, and return whether this succeeded. ac_fn_cxx_try_compile () { as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack rm -f conftest.$ac_objext if { { ac_try="$ac_compile" case "(($ac_try" in *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; *) ac_try_echo=$ac_try;; esac eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" $as_echo "$ac_try_echo"; } >&5 (eval "$ac_compile") 2>conftest.err ac_status=$? if test -s conftest.err; then grep -v '^ *+' conftest.err >conftest.er1 cat conftest.er1 >&5 mv -f conftest.er1 conftest.err fi $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 test $ac_status = 0; } && { test -z "$ac_cxx_werror_flag" || test ! -s conftest.err } && test -s conftest.$ac_objext; then : ac_retval=0 else $as_echo "$as_me: failed program was:" >&5 sed 's/^/| /' conftest.$ac_ext >&5 ac_retval=1 fi eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno as_fn_set_status $ac_retval } # ac_fn_cxx_try_compile # ac_fn_c_try_compile LINENO # -------------------------- # Try to compile conftest.$ac_ext, and return whether this succeeded. ac_fn_c_try_compile () { as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack rm -f conftest.$ac_objext if { { ac_try="$ac_compile" case "(($ac_try" in *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; *) ac_try_echo=$ac_try;; esac eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" $as_echo "$ac_try_echo"; } >&5 (eval "$ac_compile") 2>conftest.err ac_status=$? if test -s conftest.err; then grep -v '^ *+' conftest.err >conftest.er1 cat conftest.er1 >&5 mv -f conftest.er1 conftest.err fi $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 test $ac_status = 0; } && { test -z "$ac_c_werror_flag" || test ! -s conftest.err } && test -s conftest.$ac_objext; then : ac_retval=0 else $as_echo "$as_me: failed program was:" >&5 sed 's/^/| /' conftest.$ac_ext >&5 ac_retval=1 fi eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno as_fn_set_status $ac_retval } # ac_fn_c_try_compile # ac_fn_c_try_link LINENO # ----------------------- # Try to link conftest.$ac_ext, and return whether this succeeded. ac_fn_c_try_link () { as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack rm -f conftest.$ac_objext conftest$ac_exeext if { { ac_try="$ac_link" case "(($ac_try" in *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; *) ac_try_echo=$ac_try;; esac eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" $as_echo "$ac_try_echo"; } >&5 (eval "$ac_link") 2>conftest.err ac_status=$? if test -s conftest.err; then grep -v '^ *+' conftest.err >conftest.er1 cat conftest.er1 >&5 mv -f conftest.er1 conftest.err fi $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 test $ac_status = 0; } && { test -z "$ac_c_werror_flag" || test ! -s conftest.err } && test -s conftest$ac_exeext && { test "$cross_compiling" = yes || test -x conftest$ac_exeext }; then : ac_retval=0 else $as_echo "$as_me: failed program was:" >&5 sed 's/^/| /' conftest.$ac_ext >&5 ac_retval=1 fi # Delete the IPA/IPO (Inter Procedural Analysis/Optimization) information # created by the PGI compiler (conftest_ipa8_conftest.oo), as it would # interfere with the next link command; also delete a directory that is # left behind by Apple's compiler. We do this before executing the actions. rm -rf conftest.dSYM conftest_ipa8_conftest.oo eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno as_fn_set_status $ac_retval } # ac_fn_c_try_link # ac_fn_c_try_cpp LINENO # ---------------------- # Try to preprocess conftest.$ac_ext, and return whether this succeeded. ac_fn_c_try_cpp () { as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack if { { ac_try="$ac_cpp conftest.$ac_ext" case "(($ac_try" in *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; *) ac_try_echo=$ac_try;; esac eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" $as_echo "$ac_try_echo"; } >&5 (eval "$ac_cpp conftest.$ac_ext") 2>conftest.err ac_status=$? if test -s conftest.err; then grep -v '^ *+' conftest.err >conftest.er1 cat conftest.er1 >&5 mv -f conftest.er1 conftest.err fi $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 test $ac_status = 0; } > conftest.i && { test -z "$ac_c_preproc_warn_flag$ac_c_werror_flag" || test ! -s conftest.err }; then : ac_retval=0 else $as_echo "$as_me: failed program was:" >&5 sed 's/^/| /' conftest.$ac_ext >&5 ac_retval=1 fi eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno as_fn_set_status $ac_retval } # ac_fn_c_try_cpp # ac_fn_c_try_run LINENO # ---------------------- # Try to link conftest.$ac_ext, and return whether this succeeded. Assumes # that executables *can* be run. ac_fn_c_try_run () { as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack if { { ac_try="$ac_link" case "(($ac_try" in *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; *) ac_try_echo=$ac_try;; esac eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" $as_echo "$ac_try_echo"; } >&5 (eval "$ac_link") 2>&5 ac_status=$? $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 test $ac_status = 0; } && { ac_try='./conftest$ac_exeext' { { case "(($ac_try" in *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; *) ac_try_echo=$ac_try;; esac eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" $as_echo "$ac_try_echo"; } >&5 (eval "$ac_try") 2>&5 ac_status=$? $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 test $ac_status = 0; }; }; then : ac_retval=0 else $as_echo "$as_me: program exited with status $ac_status" >&5 $as_echo "$as_me: failed program was:" >&5 sed 's/^/| /' conftest.$ac_ext >&5 ac_retval=$ac_status fi rm -rf conftest.dSYM conftest_ipa8_conftest.oo eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno as_fn_set_status $ac_retval } # ac_fn_c_try_run # ac_fn_c_check_header_mongrel LINENO HEADER VAR INCLUDES # ------------------------------------------------------- # Tests whether HEADER exists, giving a warning if it cannot be compiled using # the include files in INCLUDES and setting the cache variable VAR # accordingly. ac_fn_c_check_header_mongrel () { as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack if eval \${$3+:} false; then : { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $2" >&5 $as_echo_n "checking for $2... " >&6; } if eval \${$3+:} false; then : $as_echo_n "(cached) " >&6 fi eval ac_res=\$$3 { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_res" >&5 $as_echo "$ac_res" >&6; } else # Is the header compilable? { $as_echo "$as_me:${as_lineno-$LINENO}: checking $2 usability" >&5 $as_echo_n "checking $2 usability... " >&6; } cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ $4 #include <$2> _ACEOF if ac_fn_c_try_compile "$LINENO"; then : ac_header_compiler=yes else ac_header_compiler=no fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_header_compiler" >&5 $as_echo "$ac_header_compiler" >&6; } # Is the header present? { $as_echo "$as_me:${as_lineno-$LINENO}: checking $2 presence" >&5 $as_echo_n "checking $2 presence... " >&6; } cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ #include <$2> _ACEOF if ac_fn_c_try_cpp "$LINENO"; then : ac_header_preproc=yes else ac_header_preproc=no fi rm -f conftest.err conftest.i conftest.$ac_ext { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_header_preproc" >&5 $as_echo "$ac_header_preproc" >&6; } # So? What about this header? case $ac_header_compiler:$ac_header_preproc:$ac_c_preproc_warn_flag in #(( yes:no: ) { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $2: accepted by the compiler, rejected by the preprocessor!" >&5 $as_echo "$as_me: WARNING: $2: accepted by the compiler, rejected by the preprocessor!" >&2;} { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $2: proceeding with the compiler's result" >&5 $as_echo "$as_me: WARNING: $2: proceeding with the compiler's result" >&2;} ;; no:yes:* ) { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $2: present but cannot be compiled" >&5 $as_echo "$as_me: WARNING: $2: present but cannot be compiled" >&2;} { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $2: check for missing prerequisite headers?" >&5 $as_echo "$as_me: WARNING: $2: check for missing prerequisite headers?" >&2;} { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $2: see the Autoconf documentation" >&5 $as_echo "$as_me: WARNING: $2: see the Autoconf documentation" >&2;} { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $2: section \"Present But Cannot Be Compiled\"" >&5 $as_echo "$as_me: WARNING: $2: section \"Present But Cannot Be Compiled\"" >&2;} { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $2: proceeding with the compiler's result" >&5 $as_echo "$as_me: WARNING: $2: proceeding with the compiler's result" >&2;} ;; esac { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $2" >&5 $as_echo_n "checking for $2... " >&6; } if eval \${$3+:} false; then : $as_echo_n "(cached) " >&6 else eval "$3=\$ac_header_compiler" fi eval ac_res=\$$3 { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_res" >&5 $as_echo "$ac_res" >&6; } fi eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno } # ac_fn_c_check_header_mongrel # ac_fn_c_check_header_compile LINENO HEADER VAR INCLUDES # ------------------------------------------------------- # Tests whether HEADER exists and can be compiled using the include files in # INCLUDES, setting the cache variable VAR accordingly. ac_fn_c_check_header_compile () { as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $2" >&5 $as_echo_n "checking for $2... " >&6; } if eval \${$3+:} false; then : $as_echo_n "(cached) " >&6 else cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ $4 #include <$2> _ACEOF if ac_fn_c_try_compile "$LINENO"; then : eval "$3=yes" else eval "$3=no" fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext fi eval ac_res=\$$3 { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_res" >&5 $as_echo "$ac_res" >&6; } eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno } # ac_fn_c_check_header_compile # ac_fn_c_check_type LINENO TYPE VAR INCLUDES # ------------------------------------------- # Tests whether TYPE exists after having included INCLUDES, setting cache # variable VAR accordingly. ac_fn_c_check_type () { as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $2" >&5 $as_echo_n "checking for $2... " >&6; } if eval \${$3+:} false; then : $as_echo_n "(cached) " >&6 else eval "$3=no" cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ $4 int main () { if (sizeof ($2)) return 0; ; return 0; } _ACEOF if ac_fn_c_try_compile "$LINENO"; then : cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ $4 int main () { if (sizeof (($2))) return 0; ; return 0; } _ACEOF if ac_fn_c_try_compile "$LINENO"; then : else eval "$3=yes" fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext fi eval ac_res=\$$3 { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_res" >&5 $as_echo "$ac_res" >&6; } eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno } # ac_fn_c_check_type # ac_fn_c_check_func LINENO FUNC VAR # ---------------------------------- # Tests whether FUNC exists, setting the cache variable VAR accordingly ac_fn_c_check_func () { as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $2" >&5 $as_echo_n "checking for $2... " >&6; } if eval \${$3+:} false; then : $as_echo_n "(cached) " >&6 else cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ /* Define $2 to an innocuous variant, in case declares $2. For example, HP-UX 11i declares gettimeofday. */ #define $2 innocuous_$2 /* System header to define __stub macros and hopefully few prototypes, which can conflict with char $2 (); below. Prefer to if __STDC__ is defined, since exists even on freestanding compilers. */ #ifdef __STDC__ # include #else # include #endif #undef $2 /* Override any GCC internal prototype to avoid an error. Use char because int might match the return type of a GCC builtin and then its argument prototype would still apply. */ #ifdef __cplusplus extern "C" #endif char $2 (); /* The GNU C library defines this for functions which it implements to always fail with ENOSYS. Some functions are actually named something starting with __ and the normal name is an alias. */ #if defined __stub_$2 || defined __stub___$2 choke me #endif int main () { return $2 (); ; return 0; } _ACEOF if ac_fn_c_try_link "$LINENO"; then : eval "$3=yes" else eval "$3=no" fi rm -f core conftest.err conftest.$ac_objext \ conftest$ac_exeext conftest.$ac_ext fi eval ac_res=\$$3 { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_res" >&5 $as_echo "$ac_res" >&6; } eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno } # ac_fn_c_check_func cat >config.log <<_ACEOF This file contains any messages produced by compilers while running configure, to aid debugging if configure makes a mistake. It was created by Stacks $as_me 1.44, which was generated by GNU Autoconf 2.69. Invocation command line was $ $0 $@ _ACEOF exec 5>>config.log { cat <<_ASUNAME ## --------- ## ## Platform. ## ## --------- ## hostname = `(hostname || uname -n) 2>/dev/null | sed 1q` uname -m = `(uname -m) 2>/dev/null || echo unknown` uname -r = `(uname -r) 2>/dev/null || echo unknown` uname -s = `(uname -s) 2>/dev/null || echo unknown` uname -v = `(uname -v) 2>/dev/null || echo unknown` /usr/bin/uname -p = `(/usr/bin/uname -p) 2>/dev/null || echo unknown` /bin/uname -X = `(/bin/uname -X) 2>/dev/null || echo unknown` /bin/arch = `(/bin/arch) 2>/dev/null || echo unknown` /usr/bin/arch -k = `(/usr/bin/arch -k) 2>/dev/null || echo unknown` /usr/convex/getsysinfo = `(/usr/convex/getsysinfo) 2>/dev/null || echo unknown` /usr/bin/hostinfo = `(/usr/bin/hostinfo) 2>/dev/null || echo unknown` /bin/machine = `(/bin/machine) 2>/dev/null || echo unknown` /usr/bin/oslevel = `(/usr/bin/oslevel) 2>/dev/null || echo unknown` /bin/universe = `(/bin/universe) 2>/dev/null || echo unknown` _ASUNAME as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. $as_echo "PATH: $as_dir" done IFS=$as_save_IFS } >&5 cat >&5 <<_ACEOF ## ----------- ## ## Core tests. ## ## ----------- ## _ACEOF # Keep a trace of the command line. # Strip out --no-create and --no-recursion so they do not pile up. # Strip out --silent because we don't want to record it for future runs. # Also quote any args containing shell meta-characters. # Make two passes to allow for proper duplicate-argument suppression. ac_configure_args= ac_configure_args0= ac_configure_args1= ac_must_keep_next=false for ac_pass in 1 2 do for ac_arg do case $ac_arg in -no-create | --no-c* | -n | -no-recursion | --no-r*) continue ;; -q | -quiet | --quiet | --quie | --qui | --qu | --q \ | -silent | --silent | --silen | --sile | --sil) continue ;; *\'*) ac_arg=`$as_echo "$ac_arg" | sed "s/'/'\\\\\\\\''/g"` ;; esac case $ac_pass in 1) as_fn_append ac_configure_args0 " '$ac_arg'" ;; 2) as_fn_append ac_configure_args1 " '$ac_arg'" if test $ac_must_keep_next = true; then ac_must_keep_next=false # Got value, back to normal. else case $ac_arg in *=* | --config-cache | -C | -disable-* | --disable-* \ | -enable-* | --enable-* | -gas | --g* | -nfp | --nf* \ | -q | -quiet | --q* | -silent | --sil* | -v | -verb* \ | -with-* | --with-* | -without-* | --without-* | --x) case "$ac_configure_args0 " in "$ac_configure_args1"*" '$ac_arg' "* ) continue ;; esac ;; -* ) ac_must_keep_next=true ;; esac fi as_fn_append ac_configure_args " '$ac_arg'" ;; esac done done { ac_configure_args0=; unset ac_configure_args0;} { ac_configure_args1=; unset ac_configure_args1;} # When interrupted or exit'd, cleanup temporary files, and complete # config.log. We remove comments because anyway the quotes in there # would cause problems or look ugly. # WARNING: Use '\'' to represent an apostrophe within the trap. # WARNING: Do not start the trap code with a newline, due to a FreeBSD 4.0 bug. trap 'exit_status=$? # Save into config.log some information that might help in debugging. { echo $as_echo "## ---------------- ## ## Cache variables. ## ## ---------------- ##" echo # The following way of writing the cache mishandles newlines in values, ( for ac_var in `(set) 2>&1 | sed -n '\''s/^\([a-zA-Z_][a-zA-Z0-9_]*\)=.*/\1/p'\''`; do eval ac_val=\$$ac_var case $ac_val in #( *${as_nl}*) case $ac_var in #( *_cv_*) { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: cache variable $ac_var contains a newline" >&5 $as_echo "$as_me: WARNING: cache variable $ac_var contains a newline" >&2;} ;; esac case $ac_var in #( _ | IFS | as_nl) ;; #( BASH_ARGV | BASH_SOURCE) eval $ac_var= ;; #( *) { eval $ac_var=; unset $ac_var;} ;; esac ;; esac done (set) 2>&1 | case $as_nl`(ac_space='\'' '\''; set) 2>&1` in #( *${as_nl}ac_space=\ *) sed -n \ "s/'\''/'\''\\\\'\'''\''/g; s/^\\([_$as_cr_alnum]*_cv_[_$as_cr_alnum]*\\)=\\(.*\\)/\\1='\''\\2'\''/p" ;; #( *) sed -n "/^[_$as_cr_alnum]*_cv_[_$as_cr_alnum]*=/p" ;; esac | sort ) echo $as_echo "## ----------------- ## ## Output variables. ## ## ----------------- ##" echo for ac_var in $ac_subst_vars do eval ac_val=\$$ac_var case $ac_val in *\'\''*) ac_val=`$as_echo "$ac_val" | sed "s/'\''/'\''\\\\\\\\'\'''\''/g"`;; esac $as_echo "$ac_var='\''$ac_val'\''" done | sort echo if test -n "$ac_subst_files"; then $as_echo "## ------------------- ## ## File substitutions. ## ## ------------------- ##" echo for ac_var in $ac_subst_files do eval ac_val=\$$ac_var case $ac_val in *\'\''*) ac_val=`$as_echo "$ac_val" | sed "s/'\''/'\''\\\\\\\\'\'''\''/g"`;; esac $as_echo "$ac_var='\''$ac_val'\''" done | sort echo fi if test -s confdefs.h; then $as_echo "## ----------- ## ## confdefs.h. ## ## ----------- ##" echo cat confdefs.h echo fi test "$ac_signal" != 0 && $as_echo "$as_me: caught signal $ac_signal" $as_echo "$as_me: exit $exit_status" } >&5 rm -f core *.core core.conftest.* && rm -f -r conftest* confdefs* conf$$* $ac_clean_files && exit $exit_status ' 0 for ac_signal in 1 2 13 15; do trap 'ac_signal='$ac_signal'; as_fn_exit 1' $ac_signal done ac_signal=0 # confdefs.h avoids OS command line length limits that DEFS can exceed. rm -f -r conftest* confdefs.h $as_echo "/* confdefs.h */" > confdefs.h # Predefined preprocessor variables. cat >>confdefs.h <<_ACEOF #define PACKAGE_NAME "$PACKAGE_NAME" _ACEOF cat >>confdefs.h <<_ACEOF #define PACKAGE_TARNAME "$PACKAGE_TARNAME" _ACEOF cat >>confdefs.h <<_ACEOF #define PACKAGE_VERSION "$PACKAGE_VERSION" _ACEOF cat >>confdefs.h <<_ACEOF #define PACKAGE_STRING "$PACKAGE_STRING" _ACEOF cat >>confdefs.h <<_ACEOF #define PACKAGE_BUGREPORT "$PACKAGE_BUGREPORT" _ACEOF cat >>confdefs.h <<_ACEOF #define PACKAGE_URL "$PACKAGE_URL" _ACEOF # Let the site file select an alternate cache file if it wants to. # Prefer an explicitly selected file to automatically selected ones. ac_site_file1=NONE ac_site_file2=NONE if test -n "$CONFIG_SITE"; then # We do not want a PATH search for config.site. case $CONFIG_SITE in #(( -*) ac_site_file1=./$CONFIG_SITE;; */*) ac_site_file1=$CONFIG_SITE;; *) ac_site_file1=./$CONFIG_SITE;; esac elif test "x$prefix" != xNONE; then ac_site_file1=$prefix/share/config.site ac_site_file2=$prefix/etc/config.site else ac_site_file1=$ac_default_prefix/share/config.site ac_site_file2=$ac_default_prefix/etc/config.site fi for ac_site_file in "$ac_site_file1" "$ac_site_file2" do test "x$ac_site_file" = xNONE && continue if test /dev/null != "$ac_site_file" && test -r "$ac_site_file"; then { $as_echo "$as_me:${as_lineno-$LINENO}: loading site script $ac_site_file" >&5 $as_echo "$as_me: loading site script $ac_site_file" >&6;} sed 's/^/| /' "$ac_site_file" >&5 . "$ac_site_file" \ || { { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 $as_echo "$as_me: error: in \`$ac_pwd':" >&2;} as_fn_error $? "failed to load site script $ac_site_file See \`config.log' for more details" "$LINENO" 5; } fi done if test -r "$cache_file"; then # Some versions of bash will fail to source /dev/null (special files # actually), so we avoid doing that. DJGPP emulates it as a regular file. if test /dev/null != "$cache_file" && test -f "$cache_file"; then { $as_echo "$as_me:${as_lineno-$LINENO}: loading cache $cache_file" >&5 $as_echo "$as_me: loading cache $cache_file" >&6;} case $cache_file in [\\/]* | ?:[\\/]* ) . "$cache_file";; *) . "./$cache_file";; esac fi else { $as_echo "$as_me:${as_lineno-$LINENO}: creating cache $cache_file" >&5 $as_echo "$as_me: creating cache $cache_file" >&6;} >$cache_file fi # Check that the precious variables saved in the cache have kept the same # value. ac_cache_corrupted=false for ac_var in $ac_precious_vars; do eval ac_old_set=\$ac_cv_env_${ac_var}_set eval ac_new_set=\$ac_env_${ac_var}_set eval ac_old_val=\$ac_cv_env_${ac_var}_value eval ac_new_val=\$ac_env_${ac_var}_value case $ac_old_set,$ac_new_set in set,) { $as_echo "$as_me:${as_lineno-$LINENO}: error: \`$ac_var' was set to \`$ac_old_val' in the previous run" >&5 $as_echo "$as_me: error: \`$ac_var' was set to \`$ac_old_val' in the previous run" >&2;} ac_cache_corrupted=: ;; ,set) { $as_echo "$as_me:${as_lineno-$LINENO}: error: \`$ac_var' was not set in the previous run" >&5 $as_echo "$as_me: error: \`$ac_var' was not set in the previous run" >&2;} ac_cache_corrupted=: ;; ,);; *) if test "x$ac_old_val" != "x$ac_new_val"; then # differences in whitespace do not lead to failure. ac_old_val_w=`echo x $ac_old_val` ac_new_val_w=`echo x $ac_new_val` if test "$ac_old_val_w" != "$ac_new_val_w"; then { $as_echo "$as_me:${as_lineno-$LINENO}: error: \`$ac_var' has changed since the previous run:" >&5 $as_echo "$as_me: error: \`$ac_var' has changed since the previous run:" >&2;} ac_cache_corrupted=: else { $as_echo "$as_me:${as_lineno-$LINENO}: warning: ignoring whitespace changes in \`$ac_var' since the previous run:" >&5 $as_echo "$as_me: warning: ignoring whitespace changes in \`$ac_var' since the previous run:" >&2;} eval $ac_var=\$ac_old_val fi { $as_echo "$as_me:${as_lineno-$LINENO}: former value: \`$ac_old_val'" >&5 $as_echo "$as_me: former value: \`$ac_old_val'" >&2;} { $as_echo "$as_me:${as_lineno-$LINENO}: current value: \`$ac_new_val'" >&5 $as_echo "$as_me: current value: \`$ac_new_val'" >&2;} fi;; esac # Pass precious variables to config.status. if test "$ac_new_set" = set; then case $ac_new_val in *\'*) ac_arg=$ac_var=`$as_echo "$ac_new_val" | sed "s/'/'\\\\\\\\''/g"` ;; *) ac_arg=$ac_var=$ac_new_val ;; esac case " $ac_configure_args " in *" '$ac_arg' "*) ;; # Avoid dups. Use of quotes ensures accuracy. *) as_fn_append ac_configure_args " '$ac_arg'" ;; esac fi done if $ac_cache_corrupted; then { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 $as_echo "$as_me: error: in \`$ac_pwd':" >&2;} { $as_echo "$as_me:${as_lineno-$LINENO}: error: changes in the environment can compromise the build" >&5 $as_echo "$as_me: error: changes in the environment can compromise the build" >&2;} as_fn_error $? "run \`make distclean' and/or \`rm $cache_file' and start over" "$LINENO" 5 fi ## -------------------- ## ## Main body of script. ## ## -------------------- ## ac_ext=c ac_cpp='$CPP $CPPFLAGS' ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5' ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5' ac_compiler_gnu=$ac_cv_c_compiler_gnu ac_aux_dir= for ac_dir in config "$srcdir"/config; do if test -f "$ac_dir/install-sh"; then ac_aux_dir=$ac_dir ac_install_sh="$ac_aux_dir/install-sh -c" break elif test -f "$ac_dir/install.sh"; then ac_aux_dir=$ac_dir ac_install_sh="$ac_aux_dir/install.sh -c" break elif test -f "$ac_dir/shtool"; then ac_aux_dir=$ac_dir ac_install_sh="$ac_aux_dir/shtool install -c" break fi done if test -z "$ac_aux_dir"; then as_fn_error $? "cannot find install-sh, install.sh, or shtool in config \"$srcdir\"/config" "$LINENO" 5 fi # These three variables are undocumented and unsupported, # and are intended to be withdrawn in a future Autoconf release. # They can cause serious problems if a builder's source tree is in a directory # whose full name contains unusual characters. ac_config_guess="$SHELL $ac_aux_dir/config.guess" # Please don't use this var. ac_config_sub="$SHELL $ac_aux_dir/config.sub" # Please don't use this var. ac_configure="$SHELL $ac_aux_dir/configure" # Please don't use this var. am__api_version='1.15' # Find a good install program. We prefer a C program (faster), # so one script is as good as another. But avoid the broken or # incompatible versions: # SysV /etc/install, /usr/sbin/install # SunOS /usr/etc/install # IRIX /sbin/install # AIX /bin/install # AmigaOS /C/install, which installs bootblocks on floppy discs # AIX 4 /usr/bin/installbsd, which doesn't work without a -g flag # AFS /usr/afsws/bin/install, which mishandles nonexistent args # SVR4 /usr/ucb/install, which tries to use the nonexistent group "staff" # OS/2's system install, which has a completely different semantic # ./install, which can be erroneously created by make from ./install.sh. # Reject install programs that cannot install multiple files. { $as_echo "$as_me:${as_lineno-$LINENO}: checking for a BSD-compatible install" >&5 $as_echo_n "checking for a BSD-compatible install... " >&6; } if test -z "$INSTALL"; then if ${ac_cv_path_install+:} false; then : $as_echo_n "(cached) " >&6 else as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. # Account for people who put trailing slashes in PATH elements. case $as_dir/ in #(( ./ | .// | /[cC]/* | \ /etc/* | /usr/sbin/* | /usr/etc/* | /sbin/* | /usr/afsws/bin/* | \ ?:[\\/]os2[\\/]install[\\/]* | ?:[\\/]OS2[\\/]INSTALL[\\/]* | \ /usr/ucb/* ) ;; *) # OSF1 and SCO ODT 3.0 have their own names for install. # Don't use installbsd from OSF since it installs stuff as root # by default. for ac_prog in ginstall scoinst install; do for ac_exec_ext in '' $ac_executable_extensions; do if as_fn_executable_p "$as_dir/$ac_prog$ac_exec_ext"; then if test $ac_prog = install && grep dspmsg "$as_dir/$ac_prog$ac_exec_ext" >/dev/null 2>&1; then # AIX install. It has an incompatible calling convention. : elif test $ac_prog = install && grep pwplus "$as_dir/$ac_prog$ac_exec_ext" >/dev/null 2>&1; then # program-specific install script used by HP pwplus--don't use. : else rm -rf conftest.one conftest.two conftest.dir echo one > conftest.one echo two > conftest.two mkdir conftest.dir if "$as_dir/$ac_prog$ac_exec_ext" -c conftest.one conftest.two "`pwd`/conftest.dir" && test -s conftest.one && test -s conftest.two && test -s conftest.dir/conftest.one && test -s conftest.dir/conftest.two then ac_cv_path_install="$as_dir/$ac_prog$ac_exec_ext -c" break 3 fi fi fi done done ;; esac done IFS=$as_save_IFS rm -rf conftest.one conftest.two conftest.dir fi if test "${ac_cv_path_install+set}" = set; then INSTALL=$ac_cv_path_install else # As a last resort, use the slow shell script. Don't cache a # value for INSTALL within a source directory, because that will # break other packages using the cache if that directory is # removed, or if the value is a relative name. INSTALL=$ac_install_sh fi fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $INSTALL" >&5 $as_echo "$INSTALL" >&6; } # Use test -z because SunOS4 sh mishandles braces in ${var-val}. # It thinks the first close brace ends the variable substitution. test -z "$INSTALL_PROGRAM" && INSTALL_PROGRAM='${INSTALL}' test -z "$INSTALL_SCRIPT" && INSTALL_SCRIPT='${INSTALL}' test -z "$INSTALL_DATA" && INSTALL_DATA='${INSTALL} -m 644' { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether build environment is sane" >&5 $as_echo_n "checking whether build environment is sane... " >&6; } # Reject unsafe characters in $srcdir or the absolute working directory # name. Accept space and tab only in the latter. am_lf=' ' case `pwd` in *[\\\"\#\$\&\'\`$am_lf]*) as_fn_error $? "unsafe absolute working directory name" "$LINENO" 5;; esac case $srcdir in *[\\\"\#\$\&\'\`$am_lf\ \ ]*) as_fn_error $? "unsafe srcdir value: '$srcdir'" "$LINENO" 5;; esac # Do 'set' in a subshell so we don't clobber the current shell's # arguments. Must try -L first in case configure is actually a # symlink; some systems play weird games with the mod time of symlinks # (eg FreeBSD returns the mod time of the symlink's containing # directory). if ( am_has_slept=no for am_try in 1 2; do echo "timestamp, slept: $am_has_slept" > conftest.file set X `ls -Lt "$srcdir/configure" conftest.file 2> /dev/null` if test "$*" = "X"; then # -L didn't work. set X `ls -t "$srcdir/configure" conftest.file` fi if test "$*" != "X $srcdir/configure conftest.file" \ && test "$*" != "X conftest.file $srcdir/configure"; then # If neither matched, then we have a broken ls. This can happen # if, for instance, CONFIG_SHELL is bash and it inherits a # broken ls alias from the environment. This has actually # happened. Such a system could not be considered "sane". as_fn_error $? "ls -t appears to fail. Make sure there is not a broken alias in your environment" "$LINENO" 5 fi if test "$2" = conftest.file || test $am_try -eq 2; then break fi # Just in case. sleep 1 am_has_slept=yes done test "$2" = conftest.file ) then # Ok. : else as_fn_error $? "newly created file is older than distributed files! Check your system clock" "$LINENO" 5 fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 $as_echo "yes" >&6; } # If we didn't sleep, we still need to ensure time stamps of config.status and # generated files are strictly newer. am_sleep_pid= if grep 'slept: no' conftest.file >/dev/null 2>&1; then ( sleep 1 ) & am_sleep_pid=$! fi rm -f conftest.file test "$program_prefix" != NONE && program_transform_name="s&^&$program_prefix&;$program_transform_name" # Use a double $ so make ignores it. test "$program_suffix" != NONE && program_transform_name="s&\$&$program_suffix&;$program_transform_name" # Double any \ or $. # By default was `s,x,x', remove it if useless. ac_script='s/[\\$]/&&/g;s/;s,x,x,$//' program_transform_name=`$as_echo "$program_transform_name" | sed "$ac_script"` # Expand $ac_aux_dir to an absolute path. am_aux_dir=`cd "$ac_aux_dir" && pwd` if test x"${MISSING+set}" != xset; then case $am_aux_dir in *\ * | *\ *) MISSING="\${SHELL} \"$am_aux_dir/missing\"" ;; *) MISSING="\${SHELL} $am_aux_dir/missing" ;; esac fi # Use eval to expand $SHELL if eval "$MISSING --is-lightweight"; then am_missing_run="$MISSING " else am_missing_run= { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: 'missing' script is too old or missing" >&5 $as_echo "$as_me: WARNING: 'missing' script is too old or missing" >&2;} fi if test x"${install_sh+set}" != xset; then case $am_aux_dir in *\ * | *\ *) install_sh="\${SHELL} '$am_aux_dir/install-sh'" ;; *) install_sh="\${SHELL} $am_aux_dir/install-sh" esac fi # Installed binaries are usually stripped using 'strip' when the user # run "make install-strip". However 'strip' might not be the right # tool to use in cross-compilation environments, therefore Automake # will honor the 'STRIP' environment variable to overrule this program. if test "$cross_compiling" != no; then if test -n "$ac_tool_prefix"; then # Extract the first word of "${ac_tool_prefix}strip", so it can be a program name with args. set dummy ${ac_tool_prefix}strip; ac_word=$2 { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 $as_echo_n "checking for $ac_word... " >&6; } if ${ac_cv_prog_STRIP+:} false; then : $as_echo_n "(cached) " >&6 else if test -n "$STRIP"; then ac_cv_prog_STRIP="$STRIP" # Let the user override the test. else as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. for ac_exec_ext in '' $ac_executable_extensions; do if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then ac_cv_prog_STRIP="${ac_tool_prefix}strip" $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 break 2 fi done done IFS=$as_save_IFS fi fi STRIP=$ac_cv_prog_STRIP if test -n "$STRIP"; then { $as_echo "$as_me:${as_lineno-$LINENO}: result: $STRIP" >&5 $as_echo "$STRIP" >&6; } else { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 $as_echo "no" >&6; } fi fi if test -z "$ac_cv_prog_STRIP"; then ac_ct_STRIP=$STRIP # Extract the first word of "strip", so it can be a program name with args. set dummy strip; ac_word=$2 { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 $as_echo_n "checking for $ac_word... " >&6; } if ${ac_cv_prog_ac_ct_STRIP+:} false; then : $as_echo_n "(cached) " >&6 else if test -n "$ac_ct_STRIP"; then ac_cv_prog_ac_ct_STRIP="$ac_ct_STRIP" # Let the user override the test. else as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. for ac_exec_ext in '' $ac_executable_extensions; do if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then ac_cv_prog_ac_ct_STRIP="strip" $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 break 2 fi done done IFS=$as_save_IFS fi fi ac_ct_STRIP=$ac_cv_prog_ac_ct_STRIP if test -n "$ac_ct_STRIP"; then { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_ct_STRIP" >&5 $as_echo "$ac_ct_STRIP" >&6; } else { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 $as_echo "no" >&6; } fi if test "x$ac_ct_STRIP" = x; then STRIP=":" else case $cross_compiling:$ac_tool_warned in yes:) { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: using cross tools not prefixed with host triplet" >&5 $as_echo "$as_me: WARNING: using cross tools not prefixed with host triplet" >&2;} ac_tool_warned=yes ;; esac STRIP=$ac_ct_STRIP fi else STRIP="$ac_cv_prog_STRIP" fi fi INSTALL_STRIP_PROGRAM="\$(install_sh) -c -s" { $as_echo "$as_me:${as_lineno-$LINENO}: checking for a thread-safe mkdir -p" >&5 $as_echo_n "checking for a thread-safe mkdir -p... " >&6; } if test -z "$MKDIR_P"; then if ${ac_cv_path_mkdir+:} false; then : $as_echo_n "(cached) " >&6 else as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH$PATH_SEPARATOR/opt/sfw/bin do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. for ac_prog in mkdir gmkdir; do for ac_exec_ext in '' $ac_executable_extensions; do as_fn_executable_p "$as_dir/$ac_prog$ac_exec_ext" || continue case `"$as_dir/$ac_prog$ac_exec_ext" --version 2>&1` in #( 'mkdir (GNU coreutils) '* | \ 'mkdir (coreutils) '* | \ 'mkdir (fileutils) '4.1*) ac_cv_path_mkdir=$as_dir/$ac_prog$ac_exec_ext break 3;; esac done done done IFS=$as_save_IFS fi test -d ./--version && rmdir ./--version if test "${ac_cv_path_mkdir+set}" = set; then MKDIR_P="$ac_cv_path_mkdir -p" else # As a last resort, use the slow shell script. Don't cache a # value for MKDIR_P within a source directory, because that will # break other packages using the cache if that directory is # removed, or if the value is a relative name. MKDIR_P="$ac_install_sh -d" fi fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $MKDIR_P" >&5 $as_echo "$MKDIR_P" >&6; } for ac_prog in gawk mawk nawk awk do # Extract the first word of "$ac_prog", so it can be a program name with args. set dummy $ac_prog; ac_word=$2 { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 $as_echo_n "checking for $ac_word... " >&6; } if ${ac_cv_prog_AWK+:} false; then : $as_echo_n "(cached) " >&6 else if test -n "$AWK"; then ac_cv_prog_AWK="$AWK" # Let the user override the test. else as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. for ac_exec_ext in '' $ac_executable_extensions; do if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then ac_cv_prog_AWK="$ac_prog" $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 break 2 fi done done IFS=$as_save_IFS fi fi AWK=$ac_cv_prog_AWK if test -n "$AWK"; then { $as_echo "$as_me:${as_lineno-$LINENO}: result: $AWK" >&5 $as_echo "$AWK" >&6; } else { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 $as_echo "no" >&6; } fi test -n "$AWK" && break done { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether ${MAKE-make} sets \$(MAKE)" >&5 $as_echo_n "checking whether ${MAKE-make} sets \$(MAKE)... " >&6; } set x ${MAKE-make} ac_make=`$as_echo "$2" | sed 's/+/p/g; s/[^a-zA-Z0-9_]/_/g'` if eval \${ac_cv_prog_make_${ac_make}_set+:} false; then : $as_echo_n "(cached) " >&6 else cat >conftest.make <<\_ACEOF SHELL = /bin/sh all: @echo '@@@%%%=$(MAKE)=@@@%%%' _ACEOF # GNU make sometimes prints "make[1]: Entering ...", which would confuse us. case `${MAKE-make} -f conftest.make 2>/dev/null` in *@@@%%%=?*=@@@%%%*) eval ac_cv_prog_make_${ac_make}_set=yes;; *) eval ac_cv_prog_make_${ac_make}_set=no;; esac rm -f conftest.make fi if eval test \$ac_cv_prog_make_${ac_make}_set = yes; then { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 $as_echo "yes" >&6; } SET_MAKE= else { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 $as_echo "no" >&6; } SET_MAKE="MAKE=${MAKE-make}" fi rm -rf .tst 2>/dev/null mkdir .tst 2>/dev/null if test -d .tst; then am__leading_dot=. else am__leading_dot=_ fi rmdir .tst 2>/dev/null # Check whether --enable-silent-rules was given. if test "${enable_silent_rules+set}" = set; then : enableval=$enable_silent_rules; fi case $enable_silent_rules in # ((( yes) AM_DEFAULT_VERBOSITY=0;; no) AM_DEFAULT_VERBOSITY=1;; *) AM_DEFAULT_VERBOSITY=1;; esac am_make=${MAKE-make} { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether $am_make supports nested variables" >&5 $as_echo_n "checking whether $am_make supports nested variables... " >&6; } if ${am_cv_make_support_nested_variables+:} false; then : $as_echo_n "(cached) " >&6 else if $as_echo 'TRUE=$(BAR$(V)) BAR0=false BAR1=true V=1 am__doit: @$(TRUE) .PHONY: am__doit' | $am_make -f - >/dev/null 2>&1; then am_cv_make_support_nested_variables=yes else am_cv_make_support_nested_variables=no fi fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $am_cv_make_support_nested_variables" >&5 $as_echo "$am_cv_make_support_nested_variables" >&6; } if test $am_cv_make_support_nested_variables = yes; then AM_V='$(V)' AM_DEFAULT_V='$(AM_DEFAULT_VERBOSITY)' else AM_V=$AM_DEFAULT_VERBOSITY AM_DEFAULT_V=$AM_DEFAULT_VERBOSITY fi AM_BACKSLASH='\' if test "`cd $srcdir && pwd`" != "`pwd`"; then # Use -I$(srcdir) only when $(srcdir) != ., so that make's output # is not polluted with repeated "-I." am__isrc=' -I$(srcdir)' # test to see if srcdir already configured if test -f $srcdir/config.status; then as_fn_error $? "source directory already configured; run \"make distclean\" there first" "$LINENO" 5 fi fi # test whether we have cygpath if test -z "$CYGPATH_W"; then if (cygpath --version) >/dev/null 2>/dev/null; then CYGPATH_W='cygpath -w' else CYGPATH_W=echo fi fi # Define the identity of the package. PACKAGE='stacks' VERSION='1.44' cat >>confdefs.h <<_ACEOF #define PACKAGE "$PACKAGE" _ACEOF cat >>confdefs.h <<_ACEOF #define VERSION "$VERSION" _ACEOF # Some tools Automake needs. ACLOCAL=${ACLOCAL-"${am_missing_run}aclocal-${am__api_version}"} AUTOCONF=${AUTOCONF-"${am_missing_run}autoconf"} AUTOMAKE=${AUTOMAKE-"${am_missing_run}automake-${am__api_version}"} AUTOHEADER=${AUTOHEADER-"${am_missing_run}autoheader"} MAKEINFO=${MAKEINFO-"${am_missing_run}makeinfo"} # For better backward compatibility. To be removed once Automake 1.9.x # dies out for good. For more background, see: # # mkdir_p='$(MKDIR_P)' # We need awk for the "check" target (and possibly the TAP driver). The # system "awk" is bad on some platforms. # Always define AMTAR for backward compatibility. Yes, it's still used # in the wild :-( We should find a proper way to deprecate it ... AMTAR='$${TAR-tar}' # We'll loop over all known methods to create a tar archive until one works. _am_tools='gnutar pax cpio none' am__tar='$${TAR-tar} chof - "$$tardir"' am__untar='$${TAR-tar} xf -' # POSIX will say in a future version that running "rm -f" with no argument # is OK; and we want to be able to make that assumption in our Makefile # recipes. So use an aggressive probe to check that the usage we want is # actually supported "in the wild" to an acceptable degree. # See automake bug#10828. # To make any issue more visible, cause the running configure to be aborted # by default if the 'rm' program in use doesn't match our expectations; the # user can still override this though. if rm -f && rm -fr && rm -rf; then : OK; else cat >&2 <<'END' Oops! Your 'rm' program seems unable to run without file operands specified on the command line, even when the '-f' option is present. This is contrary to the behaviour of most rm programs out there, and not conforming with the upcoming POSIX standard: Please tell bug-automake@gnu.org about your system, including the value of your $PATH and any error possibly output before this message. This can help us improve future automake versions. END if test x"$ACCEPT_INFERIOR_RM_PROGRAM" = x"yes"; then echo 'Configuration will proceed anyway, since you have set the' >&2 echo 'ACCEPT_INFERIOR_RM_PROGRAM variable to "yes"' >&2 echo >&2 else cat >&2 <<'END' Aborting the configuration process, to ensure you take notice of the issue. You can download and install GNU coreutils to get an 'rm' implementation that behaves properly: . If you want to complete the configuration process using your problematic 'rm' anyway, export the environment variable ACCEPT_INFERIOR_RM_PROGRAM to "yes", and re-run configure. END as_fn_error $? "Your 'rm' program is bad, sorry." "$LINENO" 5 fi fi ac_config_headers="$ac_config_headers config.h" # # Get BAM, aka HTSLib library and include locations if requested # # Check whether --enable-bam was given. if test "${enable_bam+set}" = set; then : enableval=$enable_bam; fi if test "x$enable_bam" != "xno"; then : $as_echo "#define HAVE_BAM 1" >>confdefs.h BAM_CFLAGS='-I./htslib/htslib' BAM_LIBS='./htslib/libhts.a' fi # # Enable use of Google Sparsehash and get include location if requested. # # Check whether --enable-sparsehash was given. if test "${enable_sparsehash+set}" = set; then : enableval=$enable_sparsehash; fi if test "x$enable_sparsehash" = "xyes"; then : $as_echo "#define HAVE_SPARSEHASH 1" >>confdefs.h # Check whether --with-sparsehash-include-path was given. if test "${with_sparsehash_include_path+set}" = set; then : withval=$with_sparsehash_include_path; SPARSEHASH_CFLAGS="-I$withval" else SPARSEHASH_CFLAGS="" fi fi # Checks for programs. ac_ext=cpp ac_cpp='$CXXCPP $CPPFLAGS' ac_compile='$CXX -c $CXXFLAGS $CPPFLAGS conftest.$ac_ext >&5' ac_link='$CXX -o conftest$ac_exeext $CXXFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5' ac_compiler_gnu=$ac_cv_cxx_compiler_gnu if test -z "$CXX"; then if test -n "$CCC"; then CXX=$CCC else if test -n "$ac_tool_prefix"; then for ac_prog in g++ c++ gpp aCC CC cxx cc++ cl.exe FCC KCC RCC xlC_r xlC do # Extract the first word of "$ac_tool_prefix$ac_prog", so it can be a program name with args. set dummy $ac_tool_prefix$ac_prog; ac_word=$2 { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 $as_echo_n "checking for $ac_word... " >&6; } if ${ac_cv_prog_CXX+:} false; then : $as_echo_n "(cached) " >&6 else if test -n "$CXX"; then ac_cv_prog_CXX="$CXX" # Let the user override the test. else as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. for ac_exec_ext in '' $ac_executable_extensions; do if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then ac_cv_prog_CXX="$ac_tool_prefix$ac_prog" $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 break 2 fi done done IFS=$as_save_IFS fi fi CXX=$ac_cv_prog_CXX if test -n "$CXX"; then { $as_echo "$as_me:${as_lineno-$LINENO}: result: $CXX" >&5 $as_echo "$CXX" >&6; } else { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 $as_echo "no" >&6; } fi test -n "$CXX" && break done fi if test -z "$CXX"; then ac_ct_CXX=$CXX for ac_prog in g++ c++ gpp aCC CC cxx cc++ cl.exe FCC KCC RCC xlC_r xlC do # Extract the first word of "$ac_prog", so it can be a program name with args. set dummy $ac_prog; ac_word=$2 { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 $as_echo_n "checking for $ac_word... " >&6; } if ${ac_cv_prog_ac_ct_CXX+:} false; then : $as_echo_n "(cached) " >&6 else if test -n "$ac_ct_CXX"; then ac_cv_prog_ac_ct_CXX="$ac_ct_CXX" # Let the user override the test. else as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. for ac_exec_ext in '' $ac_executable_extensions; do if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then ac_cv_prog_ac_ct_CXX="$ac_prog" $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 break 2 fi done done IFS=$as_save_IFS fi fi ac_ct_CXX=$ac_cv_prog_ac_ct_CXX if test -n "$ac_ct_CXX"; then { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_ct_CXX" >&5 $as_echo "$ac_ct_CXX" >&6; } else { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 $as_echo "no" >&6; } fi test -n "$ac_ct_CXX" && break done if test "x$ac_ct_CXX" = x; then CXX="g++" else case $cross_compiling:$ac_tool_warned in yes:) { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: using cross tools not prefixed with host triplet" >&5 $as_echo "$as_me: WARNING: using cross tools not prefixed with host triplet" >&2;} ac_tool_warned=yes ;; esac CXX=$ac_ct_CXX fi fi fi fi # Provide some information about the compiler. $as_echo "$as_me:${as_lineno-$LINENO}: checking for C++ compiler version" >&5 set X $ac_compile ac_compiler=$2 for ac_option in --version -v -V -qversion; do { { ac_try="$ac_compiler $ac_option >&5" case "(($ac_try" in *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; *) ac_try_echo=$ac_try;; esac eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" $as_echo "$ac_try_echo"; } >&5 (eval "$ac_compiler $ac_option >&5") 2>conftest.err ac_status=$? if test -s conftest.err; then sed '10a\ ... rest of stderr output deleted ... 10q' conftest.err >conftest.er1 cat conftest.er1 >&5 fi rm -f conftest.er1 conftest.err $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 test $ac_status = 0; } done cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ int main () { ; return 0; } _ACEOF ac_clean_files_save=$ac_clean_files ac_clean_files="$ac_clean_files a.out a.out.dSYM a.exe b.out" # Try to create an executable without -o first, disregard a.out. # It will help us diagnose broken compilers, and finding out an intuition # of exeext. { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether the C++ compiler works" >&5 $as_echo_n "checking whether the C++ compiler works... " >&6; } ac_link_default=`$as_echo "$ac_link" | sed 's/ -o *conftest[^ ]*//'` # The possible output files: ac_files="a.out conftest.exe conftest a.exe a_out.exe b.out conftest.*" ac_rmfiles= for ac_file in $ac_files do case $ac_file in *.$ac_ext | *.xcoff | *.tds | *.d | *.pdb | *.xSYM | *.bb | *.bbg | *.map | *.inf | *.dSYM | *.o | *.obj ) ;; * ) ac_rmfiles="$ac_rmfiles $ac_file";; esac done rm -f $ac_rmfiles if { { ac_try="$ac_link_default" case "(($ac_try" in *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; *) ac_try_echo=$ac_try;; esac eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" $as_echo "$ac_try_echo"; } >&5 (eval "$ac_link_default") 2>&5 ac_status=$? $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 test $ac_status = 0; }; then : # Autoconf-2.13 could set the ac_cv_exeext variable to `no'. # So ignore a value of `no', otherwise this would lead to `EXEEXT = no' # in a Makefile. We should not override ac_cv_exeext if it was cached, # so that the user can short-circuit this test for compilers unknown to # Autoconf. for ac_file in $ac_files '' do test -f "$ac_file" || continue case $ac_file in *.$ac_ext | *.xcoff | *.tds | *.d | *.pdb | *.xSYM | *.bb | *.bbg | *.map | *.inf | *.dSYM | *.o | *.obj ) ;; [ab].out ) # We found the default executable, but exeext='' is most # certainly right. break;; *.* ) if test "${ac_cv_exeext+set}" = set && test "$ac_cv_exeext" != no; then :; else ac_cv_exeext=`expr "$ac_file" : '[^.]*\(\..*\)'` fi # We set ac_cv_exeext here because the later test for it is not # safe: cross compilers may not add the suffix if given an `-o' # argument, so we may need to know it at that point already. # Even if this section looks crufty: it has the advantage of # actually working. break;; * ) break;; esac done test "$ac_cv_exeext" = no && ac_cv_exeext= else ac_file='' fi if test -z "$ac_file"; then : { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 $as_echo "no" >&6; } $as_echo "$as_me: failed program was:" >&5 sed 's/^/| /' conftest.$ac_ext >&5 { { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 $as_echo "$as_me: error: in \`$ac_pwd':" >&2;} as_fn_error 77 "C++ compiler cannot create executables See \`config.log' for more details" "$LINENO" 5; } else { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 $as_echo "yes" >&6; } fi { $as_echo "$as_me:${as_lineno-$LINENO}: checking for C++ compiler default output file name" >&5 $as_echo_n "checking for C++ compiler default output file name... " >&6; } { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_file" >&5 $as_echo "$ac_file" >&6; } ac_exeext=$ac_cv_exeext rm -f -r a.out a.out.dSYM a.exe conftest$ac_cv_exeext b.out ac_clean_files=$ac_clean_files_save { $as_echo "$as_me:${as_lineno-$LINENO}: checking for suffix of executables" >&5 $as_echo_n "checking for suffix of executables... " >&6; } if { { ac_try="$ac_link" case "(($ac_try" in *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; *) ac_try_echo=$ac_try;; esac eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" $as_echo "$ac_try_echo"; } >&5 (eval "$ac_link") 2>&5 ac_status=$? $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 test $ac_status = 0; }; then : # If both `conftest.exe' and `conftest' are `present' (well, observable) # catch `conftest.exe'. For instance with Cygwin, `ls conftest' will # work properly (i.e., refer to `conftest.exe'), while it won't with # `rm'. for ac_file in conftest.exe conftest conftest.*; do test -f "$ac_file" || continue case $ac_file in *.$ac_ext | *.xcoff | *.tds | *.d | *.pdb | *.xSYM | *.bb | *.bbg | *.map | *.inf | *.dSYM | *.o | *.obj ) ;; *.* ) ac_cv_exeext=`expr "$ac_file" : '[^.]*\(\..*\)'` break;; * ) break;; esac done else { { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 $as_echo "$as_me: error: in \`$ac_pwd':" >&2;} as_fn_error $? "cannot compute suffix of executables: cannot compile and link See \`config.log' for more details" "$LINENO" 5; } fi rm -f conftest conftest$ac_cv_exeext { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_exeext" >&5 $as_echo "$ac_cv_exeext" >&6; } rm -f conftest.$ac_ext EXEEXT=$ac_cv_exeext ac_exeext=$EXEEXT cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ #include int main () { FILE *f = fopen ("conftest.out", "w"); return ferror (f) || fclose (f) != 0; ; return 0; } _ACEOF ac_clean_files="$ac_clean_files conftest.out" # Check that the compiler produces executables we can run. If not, either # the compiler is broken, or we cross compile. { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether we are cross compiling" >&5 $as_echo_n "checking whether we are cross compiling... " >&6; } if test "$cross_compiling" != yes; then { { ac_try="$ac_link" case "(($ac_try" in *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; *) ac_try_echo=$ac_try;; esac eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" $as_echo "$ac_try_echo"; } >&5 (eval "$ac_link") 2>&5 ac_status=$? $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 test $ac_status = 0; } if { ac_try='./conftest$ac_cv_exeext' { { case "(($ac_try" in *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; *) ac_try_echo=$ac_try;; esac eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" $as_echo "$ac_try_echo"; } >&5 (eval "$ac_try") 2>&5 ac_status=$? $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 test $ac_status = 0; }; }; then cross_compiling=no else if test "$cross_compiling" = maybe; then cross_compiling=yes else { { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 $as_echo "$as_me: error: in \`$ac_pwd':" >&2;} as_fn_error $? "cannot run C++ compiled programs. If you meant to cross compile, use \`--host'. See \`config.log' for more details" "$LINENO" 5; } fi fi fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $cross_compiling" >&5 $as_echo "$cross_compiling" >&6; } rm -f conftest.$ac_ext conftest$ac_cv_exeext conftest.out ac_clean_files=$ac_clean_files_save { $as_echo "$as_me:${as_lineno-$LINENO}: checking for suffix of object files" >&5 $as_echo_n "checking for suffix of object files... " >&6; } if ${ac_cv_objext+:} false; then : $as_echo_n "(cached) " >&6 else cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ int main () { ; return 0; } _ACEOF rm -f conftest.o conftest.obj if { { ac_try="$ac_compile" case "(($ac_try" in *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; *) ac_try_echo=$ac_try;; esac eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" $as_echo "$ac_try_echo"; } >&5 (eval "$ac_compile") 2>&5 ac_status=$? $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 test $ac_status = 0; }; then : for ac_file in conftest.o conftest.obj conftest.*; do test -f "$ac_file" || continue; case $ac_file in *.$ac_ext | *.xcoff | *.tds | *.d | *.pdb | *.xSYM | *.bb | *.bbg | *.map | *.inf | *.dSYM ) ;; *) ac_cv_objext=`expr "$ac_file" : '.*\.\(.*\)'` break;; esac done else $as_echo "$as_me: failed program was:" >&5 sed 's/^/| /' conftest.$ac_ext >&5 { { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 $as_echo "$as_me: error: in \`$ac_pwd':" >&2;} as_fn_error $? "cannot compute suffix of object files: cannot compile See \`config.log' for more details" "$LINENO" 5; } fi rm -f conftest.$ac_cv_objext conftest.$ac_ext fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_objext" >&5 $as_echo "$ac_cv_objext" >&6; } OBJEXT=$ac_cv_objext ac_objext=$OBJEXT { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether we are using the GNU C++ compiler" >&5 $as_echo_n "checking whether we are using the GNU C++ compiler... " >&6; } if ${ac_cv_cxx_compiler_gnu+:} false; then : $as_echo_n "(cached) " >&6 else cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ int main () { #ifndef __GNUC__ choke me #endif ; return 0; } _ACEOF if ac_fn_cxx_try_compile "$LINENO"; then : ac_compiler_gnu=yes else ac_compiler_gnu=no fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext ac_cv_cxx_compiler_gnu=$ac_compiler_gnu fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_cxx_compiler_gnu" >&5 $as_echo "$ac_cv_cxx_compiler_gnu" >&6; } if test $ac_compiler_gnu = yes; then GXX=yes else GXX= fi ac_test_CXXFLAGS=${CXXFLAGS+set} ac_save_CXXFLAGS=$CXXFLAGS { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether $CXX accepts -g" >&5 $as_echo_n "checking whether $CXX accepts -g... " >&6; } if ${ac_cv_prog_cxx_g+:} false; then : $as_echo_n "(cached) " >&6 else ac_save_cxx_werror_flag=$ac_cxx_werror_flag ac_cxx_werror_flag=yes ac_cv_prog_cxx_g=no CXXFLAGS="-g" cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ int main () { ; return 0; } _ACEOF if ac_fn_cxx_try_compile "$LINENO"; then : ac_cv_prog_cxx_g=yes else CXXFLAGS="" cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ int main () { ; return 0; } _ACEOF if ac_fn_cxx_try_compile "$LINENO"; then : else ac_cxx_werror_flag=$ac_save_cxx_werror_flag CXXFLAGS="-g" cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ int main () { ; return 0; } _ACEOF if ac_fn_cxx_try_compile "$LINENO"; then : ac_cv_prog_cxx_g=yes fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext ac_cxx_werror_flag=$ac_save_cxx_werror_flag fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_prog_cxx_g" >&5 $as_echo "$ac_cv_prog_cxx_g" >&6; } if test "$ac_test_CXXFLAGS" = set; then CXXFLAGS=$ac_save_CXXFLAGS elif test $ac_cv_prog_cxx_g = yes; then if test "$GXX" = yes; then CXXFLAGS="-g -O2" else CXXFLAGS="-g" fi else if test "$GXX" = yes; then CXXFLAGS="-O2" else CXXFLAGS= fi fi ac_ext=c ac_cpp='$CPP $CPPFLAGS' ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5' ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5' ac_compiler_gnu=$ac_cv_c_compiler_gnu DEPDIR="${am__leading_dot}deps" ac_config_commands="$ac_config_commands depfiles" am_make=${MAKE-make} cat > confinc << 'END' am__doit: @echo this is the am__doit target .PHONY: am__doit END # If we don't find an include directive, just comment out the code. { $as_echo "$as_me:${as_lineno-$LINENO}: checking for style of include used by $am_make" >&5 $as_echo_n "checking for style of include used by $am_make... " >&6; } am__include="#" am__quote= _am_result=none # First try GNU make style include. echo "include confinc" > confmf # Ignore all kinds of additional output from 'make'. case `$am_make -s -f confmf 2> /dev/null` in #( *the\ am__doit\ target*) am__include=include am__quote= _am_result=GNU ;; esac # Now try BSD make style include. if test "$am__include" = "#"; then echo '.include "confinc"' > confmf case `$am_make -s -f confmf 2> /dev/null` in #( *the\ am__doit\ target*) am__include=.include am__quote="\"" _am_result=BSD ;; esac fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $_am_result" >&5 $as_echo "$_am_result" >&6; } rm -f confinc confmf # Check whether --enable-dependency-tracking was given. if test "${enable_dependency_tracking+set}" = set; then : enableval=$enable_dependency_tracking; fi if test "x$enable_dependency_tracking" != xno; then am_depcomp="$ac_aux_dir/depcomp" AMDEPBACKSLASH='\' am__nodep='_no' fi if test "x$enable_dependency_tracking" != xno; then AMDEP_TRUE= AMDEP_FALSE='#' else AMDEP_TRUE='#' AMDEP_FALSE= fi depcc="$CXX" am_compiler_list= { $as_echo "$as_me:${as_lineno-$LINENO}: checking dependency style of $depcc" >&5 $as_echo_n "checking dependency style of $depcc... " >&6; } if ${am_cv_CXX_dependencies_compiler_type+:} false; then : $as_echo_n "(cached) " >&6 else if test -z "$AMDEP_TRUE" && test -f "$am_depcomp"; then # We make a subdir and do the tests there. Otherwise we can end up # making bogus files that we don't know about and never remove. For # instance it was reported that on HP-UX the gcc test will end up # making a dummy file named 'D' -- because '-MD' means "put the output # in D". rm -rf conftest.dir mkdir conftest.dir # Copy depcomp to subdir because otherwise we won't find it if we're # using a relative directory. cp "$am_depcomp" conftest.dir cd conftest.dir # We will build objects and dependencies in a subdirectory because # it helps to detect inapplicable dependency modes. For instance # both Tru64's cc and ICC support -MD to output dependencies as a # side effect of compilation, but ICC will put the dependencies in # the current directory while Tru64 will put them in the object # directory. mkdir sub am_cv_CXX_dependencies_compiler_type=none if test "$am_compiler_list" = ""; then am_compiler_list=`sed -n 's/^#*\([a-zA-Z0-9]*\))$/\1/p' < ./depcomp` fi am__universal=false case " $depcc " in #( *\ -arch\ *\ -arch\ *) am__universal=true ;; esac for depmode in $am_compiler_list; do # Setup a source with many dependencies, because some compilers # like to wrap large dependency lists on column 80 (with \), and # we should not choose a depcomp mode which is confused by this. # # We need to recreate these files for each test, as the compiler may # overwrite some of them when testing with obscure command lines. # This happens at least with the AIX C compiler. : > sub/conftest.c for i in 1 2 3 4 5 6; do echo '#include "conftst'$i'.h"' >> sub/conftest.c # Using ": > sub/conftst$i.h" creates only sub/conftst1.h with # Solaris 10 /bin/sh. echo '/* dummy */' > sub/conftst$i.h done echo "${am__include} ${am__quote}sub/conftest.Po${am__quote}" > confmf # We check with '-c' and '-o' for the sake of the "dashmstdout" # mode. It turns out that the SunPro C++ compiler does not properly # handle '-M -o', and we need to detect this. Also, some Intel # versions had trouble with output in subdirs. am__obj=sub/conftest.${OBJEXT-o} am__minus_obj="-o $am__obj" case $depmode in gcc) # This depmode causes a compiler race in universal mode. test "$am__universal" = false || continue ;; nosideeffect) # After this tag, mechanisms are not by side-effect, so they'll # only be used when explicitly requested. if test "x$enable_dependency_tracking" = xyes; then continue else break fi ;; msvc7 | msvc7msys | msvisualcpp | msvcmsys) # This compiler won't grok '-c -o', but also, the minuso test has # not run yet. These depmodes are late enough in the game, and # so weak that their functioning should not be impacted. am__obj=conftest.${OBJEXT-o} am__minus_obj= ;; none) break ;; esac if depmode=$depmode \ source=sub/conftest.c object=$am__obj \ depfile=sub/conftest.Po tmpdepfile=sub/conftest.TPo \ $SHELL ./depcomp $depcc -c $am__minus_obj sub/conftest.c \ >/dev/null 2>conftest.err && grep sub/conftst1.h sub/conftest.Po > /dev/null 2>&1 && grep sub/conftst6.h sub/conftest.Po > /dev/null 2>&1 && grep $am__obj sub/conftest.Po > /dev/null 2>&1 && ${MAKE-make} -s -f confmf > /dev/null 2>&1; then # icc doesn't choke on unknown options, it will just issue warnings # or remarks (even with -Werror). So we grep stderr for any message # that says an option was ignored or not supported. # When given -MP, icc 7.0 and 7.1 complain thusly: # icc: Command line warning: ignoring option '-M'; no argument required # The diagnosis changed in icc 8.0: # icc: Command line remark: option '-MP' not supported if (grep 'ignoring option' conftest.err || grep 'not supported' conftest.err) >/dev/null 2>&1; then :; else am_cv_CXX_dependencies_compiler_type=$depmode break fi fi done cd .. rm -rf conftest.dir else am_cv_CXX_dependencies_compiler_type=none fi fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $am_cv_CXX_dependencies_compiler_type" >&5 $as_echo "$am_cv_CXX_dependencies_compiler_type" >&6; } CXXDEPMODE=depmode=$am_cv_CXX_dependencies_compiler_type if test "x$enable_dependency_tracking" != xno \ && test "$am_cv_CXX_dependencies_compiler_type" = gcc3; then am__fastdepCXX_TRUE= am__fastdepCXX_FALSE='#' else am__fastdepCXX_TRUE='#' am__fastdepCXX_FALSE= fi ac_ext=c ac_cpp='$CPP $CPPFLAGS' ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5' ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5' ac_compiler_gnu=$ac_cv_c_compiler_gnu if test -n "$ac_tool_prefix"; then # Extract the first word of "${ac_tool_prefix}gcc", so it can be a program name with args. set dummy ${ac_tool_prefix}gcc; ac_word=$2 { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 $as_echo_n "checking for $ac_word... " >&6; } if ${ac_cv_prog_CC+:} false; then : $as_echo_n "(cached) " >&6 else if test -n "$CC"; then ac_cv_prog_CC="$CC" # Let the user override the test. else as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. for ac_exec_ext in '' $ac_executable_extensions; do if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then ac_cv_prog_CC="${ac_tool_prefix}gcc" $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 break 2 fi done done IFS=$as_save_IFS fi fi CC=$ac_cv_prog_CC if test -n "$CC"; then { $as_echo "$as_me:${as_lineno-$LINENO}: result: $CC" >&5 $as_echo "$CC" >&6; } else { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 $as_echo "no" >&6; } fi fi if test -z "$ac_cv_prog_CC"; then ac_ct_CC=$CC # Extract the first word of "gcc", so it can be a program name with args. set dummy gcc; ac_word=$2 { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 $as_echo_n "checking for $ac_word... " >&6; } if ${ac_cv_prog_ac_ct_CC+:} false; then : $as_echo_n "(cached) " >&6 else if test -n "$ac_ct_CC"; then ac_cv_prog_ac_ct_CC="$ac_ct_CC" # Let the user override the test. else as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. for ac_exec_ext in '' $ac_executable_extensions; do if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then ac_cv_prog_ac_ct_CC="gcc" $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 break 2 fi done done IFS=$as_save_IFS fi fi ac_ct_CC=$ac_cv_prog_ac_ct_CC if test -n "$ac_ct_CC"; then { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_ct_CC" >&5 $as_echo "$ac_ct_CC" >&6; } else { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 $as_echo "no" >&6; } fi if test "x$ac_ct_CC" = x; then CC="" else case $cross_compiling:$ac_tool_warned in yes:) { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: using cross tools not prefixed with host triplet" >&5 $as_echo "$as_me: WARNING: using cross tools not prefixed with host triplet" >&2;} ac_tool_warned=yes ;; esac CC=$ac_ct_CC fi else CC="$ac_cv_prog_CC" fi if test -z "$CC"; then if test -n "$ac_tool_prefix"; then # Extract the first word of "${ac_tool_prefix}cc", so it can be a program name with args. set dummy ${ac_tool_prefix}cc; ac_word=$2 { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 $as_echo_n "checking for $ac_word... " >&6; } if ${ac_cv_prog_CC+:} false; then : $as_echo_n "(cached) " >&6 else if test -n "$CC"; then ac_cv_prog_CC="$CC" # Let the user override the test. else as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. for ac_exec_ext in '' $ac_executable_extensions; do if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then ac_cv_prog_CC="${ac_tool_prefix}cc" $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 break 2 fi done done IFS=$as_save_IFS fi fi CC=$ac_cv_prog_CC if test -n "$CC"; then { $as_echo "$as_me:${as_lineno-$LINENO}: result: $CC" >&5 $as_echo "$CC" >&6; } else { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 $as_echo "no" >&6; } fi fi fi if test -z "$CC"; then # Extract the first word of "cc", so it can be a program name with args. set dummy cc; ac_word=$2 { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 $as_echo_n "checking for $ac_word... " >&6; } if ${ac_cv_prog_CC+:} false; then : $as_echo_n "(cached) " >&6 else if test -n "$CC"; then ac_cv_prog_CC="$CC" # Let the user override the test. else ac_prog_rejected=no as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. for ac_exec_ext in '' $ac_executable_extensions; do if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then if test "$as_dir/$ac_word$ac_exec_ext" = "/usr/ucb/cc"; then ac_prog_rejected=yes continue fi ac_cv_prog_CC="cc" $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 break 2 fi done done IFS=$as_save_IFS if test $ac_prog_rejected = yes; then # We found a bogon in the path, so make sure we never use it. set dummy $ac_cv_prog_CC shift if test $# != 0; then # We chose a different compiler from the bogus one. # However, it has the same basename, so the bogon will be chosen # first if we set CC to just the basename; use the full file name. shift ac_cv_prog_CC="$as_dir/$ac_word${1+' '}$@" fi fi fi fi CC=$ac_cv_prog_CC if test -n "$CC"; then { $as_echo "$as_me:${as_lineno-$LINENO}: result: $CC" >&5 $as_echo "$CC" >&6; } else { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 $as_echo "no" >&6; } fi fi if test -z "$CC"; then if test -n "$ac_tool_prefix"; then for ac_prog in cl.exe do # Extract the first word of "$ac_tool_prefix$ac_prog", so it can be a program name with args. set dummy $ac_tool_prefix$ac_prog; ac_word=$2 { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 $as_echo_n "checking for $ac_word... " >&6; } if ${ac_cv_prog_CC+:} false; then : $as_echo_n "(cached) " >&6 else if test -n "$CC"; then ac_cv_prog_CC="$CC" # Let the user override the test. else as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. for ac_exec_ext in '' $ac_executable_extensions; do if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then ac_cv_prog_CC="$ac_tool_prefix$ac_prog" $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 break 2 fi done done IFS=$as_save_IFS fi fi CC=$ac_cv_prog_CC if test -n "$CC"; then { $as_echo "$as_me:${as_lineno-$LINENO}: result: $CC" >&5 $as_echo "$CC" >&6; } else { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 $as_echo "no" >&6; } fi test -n "$CC" && break done fi if test -z "$CC"; then ac_ct_CC=$CC for ac_prog in cl.exe do # Extract the first word of "$ac_prog", so it can be a program name with args. set dummy $ac_prog; ac_word=$2 { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 $as_echo_n "checking for $ac_word... " >&6; } if ${ac_cv_prog_ac_ct_CC+:} false; then : $as_echo_n "(cached) " >&6 else if test -n "$ac_ct_CC"; then ac_cv_prog_ac_ct_CC="$ac_ct_CC" # Let the user override the test. else as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. for ac_exec_ext in '' $ac_executable_extensions; do if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then ac_cv_prog_ac_ct_CC="$ac_prog" $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 break 2 fi done done IFS=$as_save_IFS fi fi ac_ct_CC=$ac_cv_prog_ac_ct_CC if test -n "$ac_ct_CC"; then { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_ct_CC" >&5 $as_echo "$ac_ct_CC" >&6; } else { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 $as_echo "no" >&6; } fi test -n "$ac_ct_CC" && break done if test "x$ac_ct_CC" = x; then CC="" else case $cross_compiling:$ac_tool_warned in yes:) { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: using cross tools not prefixed with host triplet" >&5 $as_echo "$as_me: WARNING: using cross tools not prefixed with host triplet" >&2;} ac_tool_warned=yes ;; esac CC=$ac_ct_CC fi fi fi test -z "$CC" && { { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 $as_echo "$as_me: error: in \`$ac_pwd':" >&2;} as_fn_error $? "no acceptable C compiler found in \$PATH See \`config.log' for more details" "$LINENO" 5; } # Provide some information about the compiler. $as_echo "$as_me:${as_lineno-$LINENO}: checking for C compiler version" >&5 set X $ac_compile ac_compiler=$2 for ac_option in --version -v -V -qversion; do { { ac_try="$ac_compiler $ac_option >&5" case "(($ac_try" in *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; *) ac_try_echo=$ac_try;; esac eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" $as_echo "$ac_try_echo"; } >&5 (eval "$ac_compiler $ac_option >&5") 2>conftest.err ac_status=$? if test -s conftest.err; then sed '10a\ ... rest of stderr output deleted ... 10q' conftest.err >conftest.er1 cat conftest.er1 >&5 fi rm -f conftest.er1 conftest.err $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 test $ac_status = 0; } done { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether we are using the GNU C compiler" >&5 $as_echo_n "checking whether we are using the GNU C compiler... " >&6; } if ${ac_cv_c_compiler_gnu+:} false; then : $as_echo_n "(cached) " >&6 else cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ int main () { #ifndef __GNUC__ choke me #endif ; return 0; } _ACEOF if ac_fn_c_try_compile "$LINENO"; then : ac_compiler_gnu=yes else ac_compiler_gnu=no fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext ac_cv_c_compiler_gnu=$ac_compiler_gnu fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_c_compiler_gnu" >&5 $as_echo "$ac_cv_c_compiler_gnu" >&6; } if test $ac_compiler_gnu = yes; then GCC=yes else GCC= fi ac_test_CFLAGS=${CFLAGS+set} ac_save_CFLAGS=$CFLAGS { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether $CC accepts -g" >&5 $as_echo_n "checking whether $CC accepts -g... " >&6; } if ${ac_cv_prog_cc_g+:} false; then : $as_echo_n "(cached) " >&6 else ac_save_c_werror_flag=$ac_c_werror_flag ac_c_werror_flag=yes ac_cv_prog_cc_g=no CFLAGS="-g" cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ int main () { ; return 0; } _ACEOF if ac_fn_c_try_compile "$LINENO"; then : ac_cv_prog_cc_g=yes else CFLAGS="" cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ int main () { ; return 0; } _ACEOF if ac_fn_c_try_compile "$LINENO"; then : else ac_c_werror_flag=$ac_save_c_werror_flag CFLAGS="-g" cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ int main () { ; return 0; } _ACEOF if ac_fn_c_try_compile "$LINENO"; then : ac_cv_prog_cc_g=yes fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext ac_c_werror_flag=$ac_save_c_werror_flag fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_prog_cc_g" >&5 $as_echo "$ac_cv_prog_cc_g" >&6; } if test "$ac_test_CFLAGS" = set; then CFLAGS=$ac_save_CFLAGS elif test $ac_cv_prog_cc_g = yes; then if test "$GCC" = yes; then CFLAGS="-g -O2" else CFLAGS="-g" fi else if test "$GCC" = yes; then CFLAGS="-O2" else CFLAGS= fi fi { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $CC option to accept ISO C89" >&5 $as_echo_n "checking for $CC option to accept ISO C89... " >&6; } if ${ac_cv_prog_cc_c89+:} false; then : $as_echo_n "(cached) " >&6 else ac_cv_prog_cc_c89=no ac_save_CC=$CC cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ #include #include struct stat; /* Most of the following tests are stolen from RCS 5.7's src/conf.sh. */ struct buf { int x; }; FILE * (*rcsopen) (struct buf *, struct stat *, int); static char *e (p, i) char **p; int i; { return p[i]; } static char *f (char * (*g) (char **, int), char **p, ...) { char *s; va_list v; va_start (v,p); s = g (p, va_arg (v,int)); va_end (v); return s; } /* OSF 4.0 Compaq cc is some sort of almost-ANSI by default. It has function prototypes and stuff, but not '\xHH' hex character constants. These don't provoke an error unfortunately, instead are silently treated as 'x'. The following induces an error, until -std is added to get proper ANSI mode. Curiously '\x00'!='x' always comes out true, for an array size at least. It's necessary to write '\x00'==0 to get something that's true only with -std. */ int osf4_cc_array ['\x00' == 0 ? 1 : -1]; /* IBM C 6 for AIX is almost-ANSI by default, but it replaces macro parameters inside strings and character constants. */ #define FOO(x) 'x' int xlc6_cc_array[FOO(a) == 'x' ? 1 : -1]; int test (int i, double x); struct s1 {int (*f) (int a);}; struct s2 {int (*f) (double a);}; int pairnames (int, char **, FILE *(*)(struct buf *, struct stat *, int), int, int); int argc; char **argv; int main () { return f (e, argv, 0) != argv[0] || f (e, argv, 1) != argv[1]; ; return 0; } _ACEOF for ac_arg in '' -qlanglvl=extc89 -qlanglvl=ansi -std \ -Ae "-Aa -D_HPUX_SOURCE" "-Xc -D__EXTENSIONS__" do CC="$ac_save_CC $ac_arg" if ac_fn_c_try_compile "$LINENO"; then : ac_cv_prog_cc_c89=$ac_arg fi rm -f core conftest.err conftest.$ac_objext test "x$ac_cv_prog_cc_c89" != "xno" && break done rm -f conftest.$ac_ext CC=$ac_save_CC fi # AC_CACHE_VAL case "x$ac_cv_prog_cc_c89" in x) { $as_echo "$as_me:${as_lineno-$LINENO}: result: none needed" >&5 $as_echo "none needed" >&6; } ;; xno) { $as_echo "$as_me:${as_lineno-$LINENO}: result: unsupported" >&5 $as_echo "unsupported" >&6; } ;; *) CC="$CC $ac_cv_prog_cc_c89" { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_prog_cc_c89" >&5 $as_echo "$ac_cv_prog_cc_c89" >&6; } ;; esac if test "x$ac_cv_prog_cc_c89" != xno; then : fi ac_ext=c ac_cpp='$CPP $CPPFLAGS' ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5' ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5' ac_compiler_gnu=$ac_cv_c_compiler_gnu ac_ext=c ac_cpp='$CPP $CPPFLAGS' ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5' ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5' ac_compiler_gnu=$ac_cv_c_compiler_gnu { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether $CC understands -c and -o together" >&5 $as_echo_n "checking whether $CC understands -c and -o together... " >&6; } if ${am_cv_prog_cc_c_o+:} false; then : $as_echo_n "(cached) " >&6 else cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ int main () { ; return 0; } _ACEOF # Make sure it works both with $CC and with simple cc. # Following AC_PROG_CC_C_O, we do the test twice because some # compilers refuse to overwrite an existing .o file with -o, # though they will create one. am_cv_prog_cc_c_o=yes for am_i in 1 2; do if { echo "$as_me:$LINENO: $CC -c conftest.$ac_ext -o conftest2.$ac_objext" >&5 ($CC -c conftest.$ac_ext -o conftest2.$ac_objext) >&5 2>&5 ac_status=$? echo "$as_me:$LINENO: \$? = $ac_status" >&5 (exit $ac_status); } \ && test -f conftest2.$ac_objext; then : OK else am_cv_prog_cc_c_o=no break fi done rm -f core conftest* unset am_i fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $am_cv_prog_cc_c_o" >&5 $as_echo "$am_cv_prog_cc_c_o" >&6; } if test "$am_cv_prog_cc_c_o" != yes; then # Losing compiler, so override with the script. # FIXME: It is wrong to rewrite CC. # But if we don't then we get into trouble of one sort or another. # A longer-term fix would be to have automake use am__CC in this case, # and then we could set am__CC="\$(top_srcdir)/compile \$(CC)" CC="$am_aux_dir/compile $CC" fi ac_ext=c ac_cpp='$CPP $CPPFLAGS' ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5' ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5' ac_compiler_gnu=$ac_cv_c_compiler_gnu depcc="$CC" am_compiler_list= { $as_echo "$as_me:${as_lineno-$LINENO}: checking dependency style of $depcc" >&5 $as_echo_n "checking dependency style of $depcc... " >&6; } if ${am_cv_CC_dependencies_compiler_type+:} false; then : $as_echo_n "(cached) " >&6 else if test -z "$AMDEP_TRUE" && test -f "$am_depcomp"; then # We make a subdir and do the tests there. Otherwise we can end up # making bogus files that we don't know about and never remove. For # instance it was reported that on HP-UX the gcc test will end up # making a dummy file named 'D' -- because '-MD' means "put the output # in D". rm -rf conftest.dir mkdir conftest.dir # Copy depcomp to subdir because otherwise we won't find it if we're # using a relative directory. cp "$am_depcomp" conftest.dir cd conftest.dir # We will build objects and dependencies in a subdirectory because # it helps to detect inapplicable dependency modes. For instance # both Tru64's cc and ICC support -MD to output dependencies as a # side effect of compilation, but ICC will put the dependencies in # the current directory while Tru64 will put them in the object # directory. mkdir sub am_cv_CC_dependencies_compiler_type=none if test "$am_compiler_list" = ""; then am_compiler_list=`sed -n 's/^#*\([a-zA-Z0-9]*\))$/\1/p' < ./depcomp` fi am__universal=false case " $depcc " in #( *\ -arch\ *\ -arch\ *) am__universal=true ;; esac for depmode in $am_compiler_list; do # Setup a source with many dependencies, because some compilers # like to wrap large dependency lists on column 80 (with \), and # we should not choose a depcomp mode which is confused by this. # # We need to recreate these files for each test, as the compiler may # overwrite some of them when testing with obscure command lines. # This happens at least with the AIX C compiler. : > sub/conftest.c for i in 1 2 3 4 5 6; do echo '#include "conftst'$i'.h"' >> sub/conftest.c # Using ": > sub/conftst$i.h" creates only sub/conftst1.h with # Solaris 10 /bin/sh. echo '/* dummy */' > sub/conftst$i.h done echo "${am__include} ${am__quote}sub/conftest.Po${am__quote}" > confmf # We check with '-c' and '-o' for the sake of the "dashmstdout" # mode. It turns out that the SunPro C++ compiler does not properly # handle '-M -o', and we need to detect this. Also, some Intel # versions had trouble with output in subdirs. am__obj=sub/conftest.${OBJEXT-o} am__minus_obj="-o $am__obj" case $depmode in gcc) # This depmode causes a compiler race in universal mode. test "$am__universal" = false || continue ;; nosideeffect) # After this tag, mechanisms are not by side-effect, so they'll # only be used when explicitly requested. if test "x$enable_dependency_tracking" = xyes; then continue else break fi ;; msvc7 | msvc7msys | msvisualcpp | msvcmsys) # This compiler won't grok '-c -o', but also, the minuso test has # not run yet. These depmodes are late enough in the game, and # so weak that their functioning should not be impacted. am__obj=conftest.${OBJEXT-o} am__minus_obj= ;; none) break ;; esac if depmode=$depmode \ source=sub/conftest.c object=$am__obj \ depfile=sub/conftest.Po tmpdepfile=sub/conftest.TPo \ $SHELL ./depcomp $depcc -c $am__minus_obj sub/conftest.c \ >/dev/null 2>conftest.err && grep sub/conftst1.h sub/conftest.Po > /dev/null 2>&1 && grep sub/conftst6.h sub/conftest.Po > /dev/null 2>&1 && grep $am__obj sub/conftest.Po > /dev/null 2>&1 && ${MAKE-make} -s -f confmf > /dev/null 2>&1; then # icc doesn't choke on unknown options, it will just issue warnings # or remarks (even with -Werror). So we grep stderr for any message # that says an option was ignored or not supported. # When given -MP, icc 7.0 and 7.1 complain thusly: # icc: Command line warning: ignoring option '-M'; no argument required # The diagnosis changed in icc 8.0: # icc: Command line remark: option '-MP' not supported if (grep 'ignoring option' conftest.err || grep 'not supported' conftest.err) >/dev/null 2>&1; then :; else am_cv_CC_dependencies_compiler_type=$depmode break fi fi done cd .. rm -rf conftest.dir else am_cv_CC_dependencies_compiler_type=none fi fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $am_cv_CC_dependencies_compiler_type" >&5 $as_echo "$am_cv_CC_dependencies_compiler_type" >&6; } CCDEPMODE=depmode=$am_cv_CC_dependencies_compiler_type if test "x$enable_dependency_tracking" != xno \ && test "$am_cv_CC_dependencies_compiler_type" = gcc3; then am__fastdepCC_TRUE= am__fastdepCC_FALSE='#' else am__fastdepCC_TRUE='#' am__fastdepCC_FALSE= fi ax_cxx_compile_cxx11_required=true ac_ext=cpp ac_cpp='$CXXCPP $CPPFLAGS' ac_compile='$CXX -c $CXXFLAGS $CPPFLAGS conftest.$ac_ext >&5' ac_link='$CXX -o conftest$ac_exeext $CXXFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5' ac_compiler_gnu=$ac_cv_cxx_compiler_gnu ac_success=no { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether $CXX supports C++11 features by default" >&5 $as_echo_n "checking whether $CXX supports C++11 features by default... " >&6; } if ${ax_cv_cxx_compile_cxx11+:} false; then : $as_echo_n "(cached) " >&6 else cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ // If the compiler admits that it is not ready for C++11, why torture it? // Hopefully, this will speed up the test. #ifndef __cplusplus #error "This is not a C++ compiler" #elif __cplusplus < 201103L #error "This is not a C++11 compiler" #else namespace cxx11 { namespace test_static_assert { template struct check { static_assert(sizeof(int) <= sizeof(T), "not big enough"); }; } namespace test_final_override { struct Base { virtual void f() {} }; struct Derived : public Base { virtual void f() override {} }; } namespace test_double_right_angle_brackets { template < typename T > struct check {}; typedef check single_type; typedef check> double_type; typedef check>> triple_type; typedef check>>> quadruple_type; } namespace test_decltype { int f() { int a = 1; decltype(a) b = 2; return a + b; } } namespace test_type_deduction { template < typename T1, typename T2 > struct is_same { static const bool value = false; }; template < typename T > struct is_same { static const bool value = true; }; template < typename T1, typename T2 > auto add(T1 a1, T2 a2) -> decltype(a1 + a2) { return a1 + a2; } int test(const int c, volatile int v) { static_assert(is_same::value == true, ""); static_assert(is_same::value == false, ""); static_assert(is_same::value == false, ""); auto ac = c; auto av = v; auto sumi = ac + av + 'x'; auto sumf = ac + av + 1.0; static_assert(is_same::value == true, ""); static_assert(is_same::value == true, ""); static_assert(is_same::value == true, ""); static_assert(is_same::value == false, ""); static_assert(is_same::value == true, ""); return (sumf > 0.0) ? sumi : add(c, v); } } namespace test_noexcept { int f() { return 0; } int g() noexcept { return 0; } static_assert(noexcept(f()) == false, ""); static_assert(noexcept(g()) == true, ""); } namespace test_constexpr { template < typename CharT > unsigned long constexpr strlen_c_r(const CharT *const s, const unsigned long acc) noexcept { return *s ? strlen_c_r(s + 1, acc + 1) : acc; } template < typename CharT > unsigned long constexpr strlen_c(const CharT *const s) noexcept { return strlen_c_r(s, 0UL); } static_assert(strlen_c("") == 0UL, ""); static_assert(strlen_c("1") == 1UL, ""); static_assert(strlen_c("example") == 7UL, ""); static_assert(strlen_c("another\0example") == 7UL, ""); } namespace test_rvalue_references { template < int N > struct answer { static constexpr int value = N; }; answer<1> f(int&) { return answer<1>(); } answer<2> f(const int&) { return answer<2>(); } answer<3> f(int&&) { return answer<3>(); } void test() { int i = 0; const int c = 0; static_assert(decltype(f(i))::value == 1, ""); static_assert(decltype(f(c))::value == 2, ""); static_assert(decltype(f(0))::value == 3, ""); } } namespace test_uniform_initialization { struct test { static const int zero {}; static const int one {1}; }; static_assert(test::zero == 0, ""); static_assert(test::one == 1, ""); } namespace test_lambdas { void test1() { auto lambda1 = [](){}; auto lambda2 = lambda1; lambda1(); lambda2(); } int test2() { auto a = [](int i, int j){ return i + j; }(1, 2); auto b = []() -> int { return '0'; }(); auto c = [=](){ return a + b; }(); auto d = [&](){ return c; }(); auto e = [a, &b](int x) mutable { const auto identity = [](int y){ return y; }; for (auto i = 0; i < a; ++i) a += b--; return x + identity(a + b); }(0); return a + b + c + d + e; } int test3() { const auto nullary = [](){ return 0; }; const auto unary = [](int x){ return x; }; using nullary_t = decltype(nullary); using unary_t = decltype(unary); const auto higher1st = [](nullary_t f){ return f(); }; const auto higher2nd = [unary](nullary_t f1){ return [unary, f1](unary_t f2){ return f2(unary(f1())); }; }; return higher1st(nullary) + higher2nd(nullary)(unary); } } namespace test_variadic_templates { template struct sum; template struct sum { static constexpr auto value = N0 + sum::value; }; template <> struct sum<> { static constexpr auto value = 0; }; static_assert(sum<>::value == 0, ""); static_assert(sum<1>::value == 1, ""); static_assert(sum<23>::value == 23, ""); static_assert(sum<1, 2>::value == 3, ""); static_assert(sum<5, 5, 11>::value == 21, ""); static_assert(sum<2, 3, 5, 7, 11, 13>::value == 41, ""); } // http://stackoverflow.com/questions/13728184/template-aliases-and-sfinae // Clang 3.1 fails with headers of libstd++ 4.8.3 when using std::function // because of this. namespace test_template_alias_sfinae { struct foo {}; template using member = typename T::member_type; template void func(...) {} template void func(member*) {} void test(); void test() { func(0); } } } // namespace cxx11 #endif // __cplusplus >= 201103L _ACEOF if ac_fn_cxx_try_compile "$LINENO"; then : ax_cv_cxx_compile_cxx11=yes else ax_cv_cxx_compile_cxx11=no fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ax_cv_cxx_compile_cxx11" >&5 $as_echo "$ax_cv_cxx_compile_cxx11" >&6; } if test x$ax_cv_cxx_compile_cxx11 = xyes; then ac_success=yes fi if test x$ac_success = xno; then for switch in -std=gnu++11 -std=gnu++0x; do cachevar=`$as_echo "ax_cv_cxx_compile_cxx11_$switch" | $as_tr_sh` { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether $CXX supports C++11 features with $switch" >&5 $as_echo_n "checking whether $CXX supports C++11 features with $switch... " >&6; } if eval \${$cachevar+:} false; then : $as_echo_n "(cached) " >&6 else ac_save_CXX="$CXX" CXX="$CXX $switch" cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ // If the compiler admits that it is not ready for C++11, why torture it? // Hopefully, this will speed up the test. #ifndef __cplusplus #error "This is not a C++ compiler" #elif __cplusplus < 201103L #error "This is not a C++11 compiler" #else namespace cxx11 { namespace test_static_assert { template struct check { static_assert(sizeof(int) <= sizeof(T), "not big enough"); }; } namespace test_final_override { struct Base { virtual void f() {} }; struct Derived : public Base { virtual void f() override {} }; } namespace test_double_right_angle_brackets { template < typename T > struct check {}; typedef check single_type; typedef check> double_type; typedef check>> triple_type; typedef check>>> quadruple_type; } namespace test_decltype { int f() { int a = 1; decltype(a) b = 2; return a + b; } } namespace test_type_deduction { template < typename T1, typename T2 > struct is_same { static const bool value = false; }; template < typename T > struct is_same { static const bool value = true; }; template < typename T1, typename T2 > auto add(T1 a1, T2 a2) -> decltype(a1 + a2) { return a1 + a2; } int test(const int c, volatile int v) { static_assert(is_same::value == true, ""); static_assert(is_same::value == false, ""); static_assert(is_same::value == false, ""); auto ac = c; auto av = v; auto sumi = ac + av + 'x'; auto sumf = ac + av + 1.0; static_assert(is_same::value == true, ""); static_assert(is_same::value == true, ""); static_assert(is_same::value == true, ""); static_assert(is_same::value == false, ""); static_assert(is_same::value == true, ""); return (sumf > 0.0) ? sumi : add(c, v); } } namespace test_noexcept { int f() { return 0; } int g() noexcept { return 0; } static_assert(noexcept(f()) == false, ""); static_assert(noexcept(g()) == true, ""); } namespace test_constexpr { template < typename CharT > unsigned long constexpr strlen_c_r(const CharT *const s, const unsigned long acc) noexcept { return *s ? strlen_c_r(s + 1, acc + 1) : acc; } template < typename CharT > unsigned long constexpr strlen_c(const CharT *const s) noexcept { return strlen_c_r(s, 0UL); } static_assert(strlen_c("") == 0UL, ""); static_assert(strlen_c("1") == 1UL, ""); static_assert(strlen_c("example") == 7UL, ""); static_assert(strlen_c("another\0example") == 7UL, ""); } namespace test_rvalue_references { template < int N > struct answer { static constexpr int value = N; }; answer<1> f(int&) { return answer<1>(); } answer<2> f(const int&) { return answer<2>(); } answer<3> f(int&&) { return answer<3>(); } void test() { int i = 0; const int c = 0; static_assert(decltype(f(i))::value == 1, ""); static_assert(decltype(f(c))::value == 2, ""); static_assert(decltype(f(0))::value == 3, ""); } } namespace test_uniform_initialization { struct test { static const int zero {}; static const int one {1}; }; static_assert(test::zero == 0, ""); static_assert(test::one == 1, ""); } namespace test_lambdas { void test1() { auto lambda1 = [](){}; auto lambda2 = lambda1; lambda1(); lambda2(); } int test2() { auto a = [](int i, int j){ return i + j; }(1, 2); auto b = []() -> int { return '0'; }(); auto c = [=](){ return a + b; }(); auto d = [&](){ return c; }(); auto e = [a, &b](int x) mutable { const auto identity = [](int y){ return y; }; for (auto i = 0; i < a; ++i) a += b--; return x + identity(a + b); }(0); return a + b + c + d + e; } int test3() { const auto nullary = [](){ return 0; }; const auto unary = [](int x){ return x; }; using nullary_t = decltype(nullary); using unary_t = decltype(unary); const auto higher1st = [](nullary_t f){ return f(); }; const auto higher2nd = [unary](nullary_t f1){ return [unary, f1](unary_t f2){ return f2(unary(f1())); }; }; return higher1st(nullary) + higher2nd(nullary)(unary); } } namespace test_variadic_templates { template struct sum; template struct sum { static constexpr auto value = N0 + sum::value; }; template <> struct sum<> { static constexpr auto value = 0; }; static_assert(sum<>::value == 0, ""); static_assert(sum<1>::value == 1, ""); static_assert(sum<23>::value == 23, ""); static_assert(sum<1, 2>::value == 3, ""); static_assert(sum<5, 5, 11>::value == 21, ""); static_assert(sum<2, 3, 5, 7, 11, 13>::value == 41, ""); } // http://stackoverflow.com/questions/13728184/template-aliases-and-sfinae // Clang 3.1 fails with headers of libstd++ 4.8.3 when using std::function // because of this. namespace test_template_alias_sfinae { struct foo {}; template using member = typename T::member_type; template void func(...) {} template void func(member*) {} void test(); void test() { func(0); } } } // namespace cxx11 #endif // __cplusplus >= 201103L _ACEOF if ac_fn_cxx_try_compile "$LINENO"; then : eval $cachevar=yes else eval $cachevar=no fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext CXX="$ac_save_CXX" fi eval ac_res=\$$cachevar { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_res" >&5 $as_echo "$ac_res" >&6; } if eval test x\$$cachevar = xyes; then CXX="$CXX $switch" if test -n "$CXXCPP" ; then CXXCPP="$CXXCPP $switch" fi ac_success=yes break fi done fi if test x$ac_success = xno; then for switch in -std=c++11 -std=c++0x +std=c++11 "-h std=c++11"; do cachevar=`$as_echo "ax_cv_cxx_compile_cxx11_$switch" | $as_tr_sh` { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether $CXX supports C++11 features with $switch" >&5 $as_echo_n "checking whether $CXX supports C++11 features with $switch... " >&6; } if eval \${$cachevar+:} false; then : $as_echo_n "(cached) " >&6 else ac_save_CXX="$CXX" CXX="$CXX $switch" cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ // If the compiler admits that it is not ready for C++11, why torture it? // Hopefully, this will speed up the test. #ifndef __cplusplus #error "This is not a C++ compiler" #elif __cplusplus < 201103L #error "This is not a C++11 compiler" #else namespace cxx11 { namespace test_static_assert { template struct check { static_assert(sizeof(int) <= sizeof(T), "not big enough"); }; } namespace test_final_override { struct Base { virtual void f() {} }; struct Derived : public Base { virtual void f() override {} }; } namespace test_double_right_angle_brackets { template < typename T > struct check {}; typedef check single_type; typedef check> double_type; typedef check>> triple_type; typedef check>>> quadruple_type; } namespace test_decltype { int f() { int a = 1; decltype(a) b = 2; return a + b; } } namespace test_type_deduction { template < typename T1, typename T2 > struct is_same { static const bool value = false; }; template < typename T > struct is_same { static const bool value = true; }; template < typename T1, typename T2 > auto add(T1 a1, T2 a2) -> decltype(a1 + a2) { return a1 + a2; } int test(const int c, volatile int v) { static_assert(is_same::value == true, ""); static_assert(is_same::value == false, ""); static_assert(is_same::value == false, ""); auto ac = c; auto av = v; auto sumi = ac + av + 'x'; auto sumf = ac + av + 1.0; static_assert(is_same::value == true, ""); static_assert(is_same::value == true, ""); static_assert(is_same::value == true, ""); static_assert(is_same::value == false, ""); static_assert(is_same::value == true, ""); return (sumf > 0.0) ? sumi : add(c, v); } } namespace test_noexcept { int f() { return 0; } int g() noexcept { return 0; } static_assert(noexcept(f()) == false, ""); static_assert(noexcept(g()) == true, ""); } namespace test_constexpr { template < typename CharT > unsigned long constexpr strlen_c_r(const CharT *const s, const unsigned long acc) noexcept { return *s ? strlen_c_r(s + 1, acc + 1) : acc; } template < typename CharT > unsigned long constexpr strlen_c(const CharT *const s) noexcept { return strlen_c_r(s, 0UL); } static_assert(strlen_c("") == 0UL, ""); static_assert(strlen_c("1") == 1UL, ""); static_assert(strlen_c("example") == 7UL, ""); static_assert(strlen_c("another\0example") == 7UL, ""); } namespace test_rvalue_references { template < int N > struct answer { static constexpr int value = N; }; answer<1> f(int&) { return answer<1>(); } answer<2> f(const int&) { return answer<2>(); } answer<3> f(int&&) { return answer<3>(); } void test() { int i = 0; const int c = 0; static_assert(decltype(f(i))::value == 1, ""); static_assert(decltype(f(c))::value == 2, ""); static_assert(decltype(f(0))::value == 3, ""); } } namespace test_uniform_initialization { struct test { static const int zero {}; static const int one {1}; }; static_assert(test::zero == 0, ""); static_assert(test::one == 1, ""); } namespace test_lambdas { void test1() { auto lambda1 = [](){}; auto lambda2 = lambda1; lambda1(); lambda2(); } int test2() { auto a = [](int i, int j){ return i + j; }(1, 2); auto b = []() -> int { return '0'; }(); auto c = [=](){ return a + b; }(); auto d = [&](){ return c; }(); auto e = [a, &b](int x) mutable { const auto identity = [](int y){ return y; }; for (auto i = 0; i < a; ++i) a += b--; return x + identity(a + b); }(0); return a + b + c + d + e; } int test3() { const auto nullary = [](){ return 0; }; const auto unary = [](int x){ return x; }; using nullary_t = decltype(nullary); using unary_t = decltype(unary); const auto higher1st = [](nullary_t f){ return f(); }; const auto higher2nd = [unary](nullary_t f1){ return [unary, f1](unary_t f2){ return f2(unary(f1())); }; }; return higher1st(nullary) + higher2nd(nullary)(unary); } } namespace test_variadic_templates { template struct sum; template struct sum { static constexpr auto value = N0 + sum::value; }; template <> struct sum<> { static constexpr auto value = 0; }; static_assert(sum<>::value == 0, ""); static_assert(sum<1>::value == 1, ""); static_assert(sum<23>::value == 23, ""); static_assert(sum<1, 2>::value == 3, ""); static_assert(sum<5, 5, 11>::value == 21, ""); static_assert(sum<2, 3, 5, 7, 11, 13>::value == 41, ""); } // http://stackoverflow.com/questions/13728184/template-aliases-and-sfinae // Clang 3.1 fails with headers of libstd++ 4.8.3 when using std::function // because of this. namespace test_template_alias_sfinae { struct foo {}; template using member = typename T::member_type; template void func(...) {} template void func(member*) {} void test(); void test() { func(0); } } } // namespace cxx11 #endif // __cplusplus >= 201103L _ACEOF if ac_fn_cxx_try_compile "$LINENO"; then : eval $cachevar=yes else eval $cachevar=no fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext CXX="$ac_save_CXX" fi eval ac_res=\$$cachevar { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_res" >&5 $as_echo "$ac_res" >&6; } if eval test x\$$cachevar = xyes; then CXX="$CXX $switch" if test -n "$CXXCPP" ; then CXXCPP="$CXXCPP $switch" fi ac_success=yes break fi done fi ac_ext=c ac_cpp='$CPP $CPPFLAGS' ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5' ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5' ac_compiler_gnu=$ac_cv_c_compiler_gnu if test x$ax_cxx_compile_cxx11_required = xtrue; then if test x$ac_success = xno; then as_fn_error $? "*** A compiler with support for C++11 language features is required." "$LINENO" 5 fi fi if test x$ac_success = xno; then HAVE_CXX11=0 { $as_echo "$as_me:${as_lineno-$LINENO}: No compiler with C++11 support was found" >&5 $as_echo "$as_me: No compiler with C++11 support was found" >&6;} else HAVE_CXX11=1 $as_echo "#define HAVE_CXX11 1" >>confdefs.h fi # Checks for libraries. { $as_echo "$as_me:${as_lineno-$LINENO}: checking for omp_set_num_threads in -lgomp" >&5 $as_echo_n "checking for omp_set_num_threads in -lgomp... " >&6; } if ${ac_cv_lib_gomp_omp_set_num_threads+:} false; then : $as_echo_n "(cached) " >&6 else ac_check_lib_save_LIBS=$LIBS LIBS="-lgomp $LIBS" cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ /* Override any GCC internal prototype to avoid an error. Use char because int might match the return type of a GCC builtin and then its argument prototype would still apply. */ #ifdef __cplusplus extern "C" #endif char omp_set_num_threads (); int main () { return omp_set_num_threads (); ; return 0; } _ACEOF if ac_fn_c_try_link "$LINENO"; then : ac_cv_lib_gomp_omp_set_num_threads=yes else ac_cv_lib_gomp_omp_set_num_threads=no fi rm -f core conftest.err conftest.$ac_objext \ conftest$ac_exeext conftest.$ac_ext LIBS=$ac_check_lib_save_LIBS fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_lib_gomp_omp_set_num_threads" >&5 $as_echo "$ac_cv_lib_gomp_omp_set_num_threads" >&6; } if test "x$ac_cv_lib_gomp_omp_set_num_threads" = xyes; then : cat >>confdefs.h <<_ACEOF #define HAVE_LIBGOMP 1 _ACEOF LIBS="-lgomp $LIBS" else { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: Unable to locate OpenMP library, you should probably specify '--disable-openmp'." >&5 $as_echo "$as_me: WARNING: Unable to locate OpenMP library, you should probably specify '--disable-openmp'." >&2;} fi { $as_echo "$as_me:${as_lineno-$LINENO}: checking for gzread in -lz" >&5 $as_echo_n "checking for gzread in -lz... " >&6; } if ${ac_cv_lib_z_gzread+:} false; then : $as_echo_n "(cached) " >&6 else ac_check_lib_save_LIBS=$LIBS LIBS="-lz $LIBS" cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ /* Override any GCC internal prototype to avoid an error. Use char because int might match the return type of a GCC builtin and then its argument prototype would still apply. */ #ifdef __cplusplus extern "C" #endif char gzread (); int main () { return gzread (); ; return 0; } _ACEOF if ac_fn_c_try_link "$LINENO"; then : ac_cv_lib_z_gzread=yes else ac_cv_lib_z_gzread=no fi rm -f core conftest.err conftest.$ac_objext \ conftest$ac_exeext conftest.$ac_ext LIBS=$ac_check_lib_save_LIBS fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_lib_z_gzread" >&5 $as_echo "$ac_cv_lib_z_gzread" >&6; } if test "x$ac_cv_lib_z_gzread" = xyes; then : cat >>confdefs.h <<_ACEOF #define HAVE_LIBZ 1 _ACEOF LIBS="-lz $LIBS" else as_fn_error $? "Zlib not found, reading gzipped files will not be possible." "$LINENO" 5 fi # Checks for header files. ac_header_dirent=no for ac_hdr in dirent.h sys/ndir.h sys/dir.h ndir.h; do as_ac_Header=`$as_echo "ac_cv_header_dirent_$ac_hdr" | $as_tr_sh` { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_hdr that defines DIR" >&5 $as_echo_n "checking for $ac_hdr that defines DIR... " >&6; } if eval \${$as_ac_Header+:} false; then : $as_echo_n "(cached) " >&6 else cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ #include #include <$ac_hdr> int main () { if ((DIR *) 0) return 0; ; return 0; } _ACEOF if ac_fn_c_try_compile "$LINENO"; then : eval "$as_ac_Header=yes" else eval "$as_ac_Header=no" fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext fi eval ac_res=\$$as_ac_Header { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_res" >&5 $as_echo "$ac_res" >&6; } if eval test \"x\$"$as_ac_Header"\" = x"yes"; then : cat >>confdefs.h <<_ACEOF #define `$as_echo "HAVE_$ac_hdr" | $as_tr_cpp` 1 _ACEOF ac_header_dirent=$ac_hdr; break fi done # Two versions of opendir et al. are in -ldir and -lx on SCO Xenix. if test $ac_header_dirent = dirent.h; then { $as_echo "$as_me:${as_lineno-$LINENO}: checking for library containing opendir" >&5 $as_echo_n "checking for library containing opendir... " >&6; } if ${ac_cv_search_opendir+:} false; then : $as_echo_n "(cached) " >&6 else ac_func_search_save_LIBS=$LIBS cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ /* Override any GCC internal prototype to avoid an error. Use char because int might match the return type of a GCC builtin and then its argument prototype would still apply. */ #ifdef __cplusplus extern "C" #endif char opendir (); int main () { return opendir (); ; return 0; } _ACEOF for ac_lib in '' dir; do if test -z "$ac_lib"; then ac_res="none required" else ac_res=-l$ac_lib LIBS="-l$ac_lib $ac_func_search_save_LIBS" fi if ac_fn_c_try_link "$LINENO"; then : ac_cv_search_opendir=$ac_res fi rm -f core conftest.err conftest.$ac_objext \ conftest$ac_exeext if ${ac_cv_search_opendir+:} false; then : break fi done if ${ac_cv_search_opendir+:} false; then : else ac_cv_search_opendir=no fi rm conftest.$ac_ext LIBS=$ac_func_search_save_LIBS fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_search_opendir" >&5 $as_echo "$ac_cv_search_opendir" >&6; } ac_res=$ac_cv_search_opendir if test "$ac_res" != no; then : test "$ac_res" = "none required" || LIBS="$ac_res $LIBS" fi else { $as_echo "$as_me:${as_lineno-$LINENO}: checking for library containing opendir" >&5 $as_echo_n "checking for library containing opendir... " >&6; } if ${ac_cv_search_opendir+:} false; then : $as_echo_n "(cached) " >&6 else ac_func_search_save_LIBS=$LIBS cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ /* Override any GCC internal prototype to avoid an error. Use char because int might match the return type of a GCC builtin and then its argument prototype would still apply. */ #ifdef __cplusplus extern "C" #endif char opendir (); int main () { return opendir (); ; return 0; } _ACEOF for ac_lib in '' x; do if test -z "$ac_lib"; then ac_res="none required" else ac_res=-l$ac_lib LIBS="-l$ac_lib $ac_func_search_save_LIBS" fi if ac_fn_c_try_link "$LINENO"; then : ac_cv_search_opendir=$ac_res fi rm -f core conftest.err conftest.$ac_objext \ conftest$ac_exeext if ${ac_cv_search_opendir+:} false; then : break fi done if ${ac_cv_search_opendir+:} false; then : else ac_cv_search_opendir=no fi rm conftest.$ac_ext LIBS=$ac_func_search_save_LIBS fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_search_opendir" >&5 $as_echo "$ac_cv_search_opendir" >&6; } ac_res=$ac_cv_search_opendir if test "$ac_res" != no; then : test "$ac_res" = "none required" || LIBS="$ac_res $LIBS" fi fi ac_ext=c ac_cpp='$CPP $CPPFLAGS' ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5' ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5' ac_compiler_gnu=$ac_cv_c_compiler_gnu { $as_echo "$as_me:${as_lineno-$LINENO}: checking how to run the C preprocessor" >&5 $as_echo_n "checking how to run the C preprocessor... " >&6; } # On Suns, sometimes $CPP names a directory. if test -n "$CPP" && test -d "$CPP"; then CPP= fi if test -z "$CPP"; then if ${ac_cv_prog_CPP+:} false; then : $as_echo_n "(cached) " >&6 else # Double quotes because CPP needs to be expanded for CPP in "$CC -E" "$CC -E -traditional-cpp" "/lib/cpp" do ac_preproc_ok=false for ac_c_preproc_warn_flag in '' yes do # Use a header file that comes with gcc, so configuring glibc # with a fresh cross-compiler works. # Prefer to if __STDC__ is defined, since # exists even on freestanding compilers. # On the NeXT, cc -E runs the code through the compiler's parser, # not just through cpp. "Syntax error" is here to catch this case. cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ #ifdef __STDC__ # include #else # include #endif Syntax error _ACEOF if ac_fn_c_try_cpp "$LINENO"; then : else # Broken: fails on valid input. continue fi rm -f conftest.err conftest.i conftest.$ac_ext # OK, works on sane cases. Now check whether nonexistent headers # can be detected and how. cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ #include _ACEOF if ac_fn_c_try_cpp "$LINENO"; then : # Broken: success on invalid input. continue else # Passes both tests. ac_preproc_ok=: break fi rm -f conftest.err conftest.i conftest.$ac_ext done # Because of `break', _AC_PREPROC_IFELSE's cleaning code was skipped. rm -f conftest.i conftest.err conftest.$ac_ext if $ac_preproc_ok; then : break fi done ac_cv_prog_CPP=$CPP fi CPP=$ac_cv_prog_CPP else ac_cv_prog_CPP=$CPP fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $CPP" >&5 $as_echo "$CPP" >&6; } ac_preproc_ok=false for ac_c_preproc_warn_flag in '' yes do # Use a header file that comes with gcc, so configuring glibc # with a fresh cross-compiler works. # Prefer to if __STDC__ is defined, since # exists even on freestanding compilers. # On the NeXT, cc -E runs the code through the compiler's parser, # not just through cpp. "Syntax error" is here to catch this case. cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ #ifdef __STDC__ # include #else # include #endif Syntax error _ACEOF if ac_fn_c_try_cpp "$LINENO"; then : else # Broken: fails on valid input. continue fi rm -f conftest.err conftest.i conftest.$ac_ext # OK, works on sane cases. Now check whether nonexistent headers # can be detected and how. cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ #include _ACEOF if ac_fn_c_try_cpp "$LINENO"; then : # Broken: success on invalid input. continue else # Passes both tests. ac_preproc_ok=: break fi rm -f conftest.err conftest.i conftest.$ac_ext done # Because of `break', _AC_PREPROC_IFELSE's cleaning code was skipped. rm -f conftest.i conftest.err conftest.$ac_ext if $ac_preproc_ok; then : else { { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 $as_echo "$as_me: error: in \`$ac_pwd':" >&2;} as_fn_error $? "C preprocessor \"$CPP\" fails sanity check See \`config.log' for more details" "$LINENO" 5; } fi ac_ext=c ac_cpp='$CPP $CPPFLAGS' ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5' ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5' ac_compiler_gnu=$ac_cv_c_compiler_gnu { $as_echo "$as_me:${as_lineno-$LINENO}: checking for grep that handles long lines and -e" >&5 $as_echo_n "checking for grep that handles long lines and -e... " >&6; } if ${ac_cv_path_GREP+:} false; then : $as_echo_n "(cached) " >&6 else if test -z "$GREP"; then ac_path_GREP_found=false # Loop through the user's path and test for each of PROGNAME-LIST as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH$PATH_SEPARATOR/usr/xpg4/bin do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. for ac_prog in grep ggrep; do for ac_exec_ext in '' $ac_executable_extensions; do ac_path_GREP="$as_dir/$ac_prog$ac_exec_ext" as_fn_executable_p "$ac_path_GREP" || continue # Check for GNU ac_path_GREP and select it if it is found. # Check for GNU $ac_path_GREP case `"$ac_path_GREP" --version 2>&1` in *GNU*) ac_cv_path_GREP="$ac_path_GREP" ac_path_GREP_found=:;; *) ac_count=0 $as_echo_n 0123456789 >"conftest.in" while : do cat "conftest.in" "conftest.in" >"conftest.tmp" mv "conftest.tmp" "conftest.in" cp "conftest.in" "conftest.nl" $as_echo 'GREP' >> "conftest.nl" "$ac_path_GREP" -e 'GREP$' -e '-(cannot match)-' < "conftest.nl" >"conftest.out" 2>/dev/null || break diff "conftest.out" "conftest.nl" >/dev/null 2>&1 || break as_fn_arith $ac_count + 1 && ac_count=$as_val if test $ac_count -gt ${ac_path_GREP_max-0}; then # Best one so far, save it but keep looking for a better one ac_cv_path_GREP="$ac_path_GREP" ac_path_GREP_max=$ac_count fi # 10*(2^10) chars as input seems more than enough test $ac_count -gt 10 && break done rm -f conftest.in conftest.tmp conftest.nl conftest.out;; esac $ac_path_GREP_found && break 3 done done done IFS=$as_save_IFS if test -z "$ac_cv_path_GREP"; then as_fn_error $? "no acceptable grep could be found in $PATH$PATH_SEPARATOR/usr/xpg4/bin" "$LINENO" 5 fi else ac_cv_path_GREP=$GREP fi fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_path_GREP" >&5 $as_echo "$ac_cv_path_GREP" >&6; } GREP="$ac_cv_path_GREP" { $as_echo "$as_me:${as_lineno-$LINENO}: checking for egrep" >&5 $as_echo_n "checking for egrep... " >&6; } if ${ac_cv_path_EGREP+:} false; then : $as_echo_n "(cached) " >&6 else if echo a | $GREP -E '(a|b)' >/dev/null 2>&1 then ac_cv_path_EGREP="$GREP -E" else if test -z "$EGREP"; then ac_path_EGREP_found=false # Loop through the user's path and test for each of PROGNAME-LIST as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH$PATH_SEPARATOR/usr/xpg4/bin do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. for ac_prog in egrep; do for ac_exec_ext in '' $ac_executable_extensions; do ac_path_EGREP="$as_dir/$ac_prog$ac_exec_ext" as_fn_executable_p "$ac_path_EGREP" || continue # Check for GNU ac_path_EGREP and select it if it is found. # Check for GNU $ac_path_EGREP case `"$ac_path_EGREP" --version 2>&1` in *GNU*) ac_cv_path_EGREP="$ac_path_EGREP" ac_path_EGREP_found=:;; *) ac_count=0 $as_echo_n 0123456789 >"conftest.in" while : do cat "conftest.in" "conftest.in" >"conftest.tmp" mv "conftest.tmp" "conftest.in" cp "conftest.in" "conftest.nl" $as_echo 'EGREP' >> "conftest.nl" "$ac_path_EGREP" 'EGREP$' < "conftest.nl" >"conftest.out" 2>/dev/null || break diff "conftest.out" "conftest.nl" >/dev/null 2>&1 || break as_fn_arith $ac_count + 1 && ac_count=$as_val if test $ac_count -gt ${ac_path_EGREP_max-0}; then # Best one so far, save it but keep looking for a better one ac_cv_path_EGREP="$ac_path_EGREP" ac_path_EGREP_max=$ac_count fi # 10*(2^10) chars as input seems more than enough test $ac_count -gt 10 && break done rm -f conftest.in conftest.tmp conftest.nl conftest.out;; esac $ac_path_EGREP_found && break 3 done done done IFS=$as_save_IFS if test -z "$ac_cv_path_EGREP"; then as_fn_error $? "no acceptable egrep could be found in $PATH$PATH_SEPARATOR/usr/xpg4/bin" "$LINENO" 5 fi else ac_cv_path_EGREP=$EGREP fi fi fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_path_EGREP" >&5 $as_echo "$ac_cv_path_EGREP" >&6; } EGREP="$ac_cv_path_EGREP" { $as_echo "$as_me:${as_lineno-$LINENO}: checking for ANSI C header files" >&5 $as_echo_n "checking for ANSI C header files... " >&6; } if ${ac_cv_header_stdc+:} false; then : $as_echo_n "(cached) " >&6 else cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ #include #include #include #include int main () { ; return 0; } _ACEOF if ac_fn_c_try_compile "$LINENO"; then : ac_cv_header_stdc=yes else ac_cv_header_stdc=no fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext if test $ac_cv_header_stdc = yes; then # SunOS 4.x string.h does not declare mem*, contrary to ANSI. cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ #include _ACEOF if (eval "$ac_cpp conftest.$ac_ext") 2>&5 | $EGREP "memchr" >/dev/null 2>&1; then : else ac_cv_header_stdc=no fi rm -f conftest* fi if test $ac_cv_header_stdc = yes; then # ISC 2.0.2 stdlib.h does not declare free, contrary to ANSI. cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ #include _ACEOF if (eval "$ac_cpp conftest.$ac_ext") 2>&5 | $EGREP "free" >/dev/null 2>&1; then : else ac_cv_header_stdc=no fi rm -f conftest* fi if test $ac_cv_header_stdc = yes; then # /bin/cc in Irix-4.0.5 gets non-ANSI ctype macros unless using -ansi. if test "$cross_compiling" = yes; then : : else cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ #include #include #if ((' ' & 0x0FF) == 0x020) # define ISLOWER(c) ('a' <= (c) && (c) <= 'z') # define TOUPPER(c) (ISLOWER(c) ? 'A' + ((c) - 'a') : (c)) #else # define ISLOWER(c) \ (('a' <= (c) && (c) <= 'i') \ || ('j' <= (c) && (c) <= 'r') \ || ('s' <= (c) && (c) <= 'z')) # define TOUPPER(c) (ISLOWER(c) ? ((c) | 0x40) : (c)) #endif #define XOR(e, f) (((e) && !(f)) || (!(e) && (f))) int main () { int i; for (i = 0; i < 256; i++) if (XOR (islower (i), ISLOWER (i)) || toupper (i) != TOUPPER (i)) return 2; return 0; } _ACEOF if ac_fn_c_try_run "$LINENO"; then : else ac_cv_header_stdc=no fi rm -f core *.core core.conftest.* gmon.out bb.out conftest$ac_exeext \ conftest.$ac_objext conftest.beam conftest.$ac_ext fi fi fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_header_stdc" >&5 $as_echo "$ac_cv_header_stdc" >&6; } if test $ac_cv_header_stdc = yes; then $as_echo "#define STDC_HEADERS 1" >>confdefs.h fi # On IRIX 5.3, sys/types and inttypes.h are conflicting. for ac_header in sys/types.h sys/stat.h stdlib.h string.h memory.h strings.h \ inttypes.h stdint.h unistd.h do : as_ac_Header=`$as_echo "ac_cv_header_$ac_header" | $as_tr_sh` ac_fn_c_check_header_compile "$LINENO" "$ac_header" "$as_ac_Header" "$ac_includes_default " if eval test \"x\$"$as_ac_Header"\" = x"yes"; then : cat >>confdefs.h <<_ACEOF #define `$as_echo "HAVE_$ac_header" | $as_tr_cpp` 1 _ACEOF fi done for ac_header in float.h limits.h stdlib.h string.h do : as_ac_Header=`$as_echo "ac_cv_header_$ac_header" | $as_tr_sh` ac_fn_c_check_header_mongrel "$LINENO" "$ac_header" "$as_ac_Header" "$ac_includes_default" if eval test \"x\$"$as_ac_Header"\" = x"yes"; then : cat >>confdefs.h <<_ACEOF #define `$as_echo "HAVE_$ac_header" | $as_tr_cpp` 1 _ACEOF fi done # Check for OpenMP parallel execution support OPENMP_CFLAGS= # Check whether --enable-openmp was given. if test "${enable_openmp+set}" = set; then : enableval=$enable_openmp; fi if test "$enable_openmp" != no; then { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $CC option to support OpenMP" >&5 $as_echo_n "checking for $CC option to support OpenMP... " >&6; } if ${ac_cv_prog_c_openmp+:} false; then : $as_echo_n "(cached) " >&6 else cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ #ifndef _OPENMP choke me #endif #include int main () { return omp_get_num_threads (); } _ACEOF if ac_fn_c_try_link "$LINENO"; then : ac_cv_prog_c_openmp='none needed' else ac_cv_prog_c_openmp='unsupported' for ac_option in -fopenmp -xopenmp -openmp -mp -omp -qsmp=omp -homp \ -Popenmp --openmp; do ac_save_CFLAGS=$CFLAGS CFLAGS="$CFLAGS $ac_option" cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ #ifndef _OPENMP choke me #endif #include int main () { return omp_get_num_threads (); } _ACEOF if ac_fn_c_try_link "$LINENO"; then : ac_cv_prog_c_openmp=$ac_option fi rm -f core conftest.err conftest.$ac_objext \ conftest$ac_exeext conftest.$ac_ext CFLAGS=$ac_save_CFLAGS if test "$ac_cv_prog_c_openmp" != unsupported; then break fi done fi rm -f core conftest.err conftest.$ac_objext \ conftest$ac_exeext conftest.$ac_ext fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_prog_c_openmp" >&5 $as_echo "$ac_cv_prog_c_openmp" >&6; } case $ac_cv_prog_c_openmp in #( "none needed" | unsupported) ;; #( *) OPENMP_CFLAGS=$ac_cv_prog_c_openmp ;; esac fi # Checks for typedefs, structures, and compiler characteristics. { $as_echo "$as_me:${as_lineno-$LINENO}: checking for stdbool.h that conforms to C99" >&5 $as_echo_n "checking for stdbool.h that conforms to C99... " >&6; } if ${ac_cv_header_stdbool_h+:} false; then : $as_echo_n "(cached) " >&6 else cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ #include #ifndef bool "error: bool is not defined" #endif #ifndef false "error: false is not defined" #endif #if false "error: false is not 0" #endif #ifndef true "error: true is not defined" #endif #if true != 1 "error: true is not 1" #endif #ifndef __bool_true_false_are_defined "error: __bool_true_false_are_defined is not defined" #endif struct s { _Bool s: 1; _Bool t; } s; char a[true == 1 ? 1 : -1]; char b[false == 0 ? 1 : -1]; char c[__bool_true_false_are_defined == 1 ? 1 : -1]; char d[(bool) 0.5 == true ? 1 : -1]; /* See body of main program for 'e'. */ char f[(_Bool) 0.0 == false ? 1 : -1]; char g[true]; char h[sizeof (_Bool)]; char i[sizeof s.t]; enum { j = false, k = true, l = false * true, m = true * 256 }; /* The following fails for HP aC++/ANSI C B3910B A.05.55 [Dec 04 2003]. */ _Bool n[m]; char o[sizeof n == m * sizeof n[0] ? 1 : -1]; char p[-1 - (_Bool) 0 < 0 && -1 - (bool) 0 < 0 ? 1 : -1]; /* Catch a bug in an HP-UX C compiler. See http://gcc.gnu.org/ml/gcc-patches/2003-12/msg02303.html http://lists.gnu.org/archive/html/bug-coreutils/2005-11/msg00161.html */ _Bool q = true; _Bool *pq = &q; int main () { bool e = &s; *pq |= q; *pq |= ! q; /* Refer to every declared value, to avoid compiler optimizations. */ return (!a + !b + !c + !d + !e + !f + !g + !h + !i + !!j + !k + !!l + !m + !n + !o + !p + !q + !pq); ; return 0; } _ACEOF if ac_fn_c_try_compile "$LINENO"; then : ac_cv_header_stdbool_h=yes else ac_cv_header_stdbool_h=no fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_header_stdbool_h" >&5 $as_echo "$ac_cv_header_stdbool_h" >&6; } ac_fn_c_check_type "$LINENO" "_Bool" "ac_cv_type__Bool" "$ac_includes_default" if test "x$ac_cv_type__Bool" = xyes; then : cat >>confdefs.h <<_ACEOF #define HAVE__BOOL 1 _ACEOF fi if test $ac_cv_header_stdbool_h = yes; then $as_echo "#define HAVE_STDBOOL_H 1" >>confdefs.h fi { $as_echo "$as_me:${as_lineno-$LINENO}: checking for an ANSI C-conforming const" >&5 $as_echo_n "checking for an ANSI C-conforming const... " >&6; } if ${ac_cv_c_const+:} false; then : $as_echo_n "(cached) " >&6 else cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ int main () { #ifndef __cplusplus /* Ultrix mips cc rejects this sort of thing. */ typedef int charset[2]; const charset cs = { 0, 0 }; /* SunOS 4.1.1 cc rejects this. */ char const *const *pcpcc; char **ppc; /* NEC SVR4.0.2 mips cc rejects this. */ struct point {int x, y;}; static struct point const zero = {0,0}; /* AIX XL C 1.02.0.0 rejects this. It does not let you subtract one const X* pointer from another in an arm of an if-expression whose if-part is not a constant expression */ const char *g = "string"; pcpcc = &g + (g ? g-g : 0); /* HPUX 7.0 cc rejects these. */ ++pcpcc; ppc = (char**) pcpcc; pcpcc = (char const *const *) ppc; { /* SCO 3.2v4 cc rejects this sort of thing. */ char tx; char *t = &tx; char const *s = 0 ? (char *) 0 : (char const *) 0; *t++ = 0; if (s) return 0; } { /* Someone thinks the Sun supposedly-ANSI compiler will reject this. */ int x[] = {25, 17}; const int *foo = &x[0]; ++foo; } { /* Sun SC1.0 ANSI compiler rejects this -- but not the above. */ typedef const int *iptr; iptr p = 0; ++p; } { /* AIX XL C 1.02.0.0 rejects this sort of thing, saying "k.c", line 2.27: 1506-025 (S) Operand must be a modifiable lvalue. */ struct s { int j; const int *ap[3]; } bx; struct s *b = &bx; b->j = 5; } { /* ULTRIX-32 V3.1 (Rev 9) vcc rejects this */ const int foo = 10; if (!foo) return 0; } return !cs[0] && !zero.x; #endif ; return 0; } _ACEOF if ac_fn_c_try_compile "$LINENO"; then : ac_cv_c_const=yes else ac_cv_c_const=no fi rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_c_const" >&5 $as_echo "$ac_cv_c_const" >&6; } if test $ac_cv_c_const = no; then $as_echo "#define const /**/" >>confdefs.h fi ac_fn_c_check_type "$LINENO" "size_t" "ac_cv_type_size_t" "$ac_includes_default" if test "x$ac_cv_type_size_t" = xyes; then : else cat >>confdefs.h <<_ACEOF #define size_t unsigned int _ACEOF fi # Checks for library functions. { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether closedir returns void" >&5 $as_echo_n "checking whether closedir returns void... " >&6; } if ${ac_cv_func_closedir_void+:} false; then : $as_echo_n "(cached) " >&6 else if test "$cross_compiling" = yes; then : ac_cv_func_closedir_void=yes else cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ $ac_includes_default #include <$ac_header_dirent> #ifndef __cplusplus int closedir (); #endif int main () { return closedir (opendir (".")) != 0; ; return 0; } _ACEOF if ac_fn_c_try_run "$LINENO"; then : ac_cv_func_closedir_void=no else ac_cv_func_closedir_void=yes fi rm -f core *.core core.conftest.* gmon.out bb.out conftest$ac_exeext \ conftest.$ac_objext conftest.beam conftest.$ac_ext fi fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_func_closedir_void" >&5 $as_echo "$ac_cv_func_closedir_void" >&6; } if test $ac_cv_func_closedir_void = yes; then $as_echo "#define CLOSEDIR_VOID 1" >>confdefs.h fi for ac_header in stdlib.h do : ac_fn_c_check_header_mongrel "$LINENO" "stdlib.h" "ac_cv_header_stdlib_h" "$ac_includes_default" if test "x$ac_cv_header_stdlib_h" = xyes; then : cat >>confdefs.h <<_ACEOF #define HAVE_STDLIB_H 1 _ACEOF fi done { $as_echo "$as_me:${as_lineno-$LINENO}: checking for GNU libc compatible malloc" >&5 $as_echo_n "checking for GNU libc compatible malloc... " >&6; } if ${ac_cv_func_malloc_0_nonnull+:} false; then : $as_echo_n "(cached) " >&6 else if test "$cross_compiling" = yes; then : ac_cv_func_malloc_0_nonnull=no else cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ #if defined STDC_HEADERS || defined HAVE_STDLIB_H # include #else char *malloc (); #endif int main () { return ! malloc (0); ; return 0; } _ACEOF if ac_fn_c_try_run "$LINENO"; then : ac_cv_func_malloc_0_nonnull=yes else ac_cv_func_malloc_0_nonnull=no fi rm -f core *.core core.conftest.* gmon.out bb.out conftest$ac_exeext \ conftest.$ac_objext conftest.beam conftest.$ac_ext fi fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_func_malloc_0_nonnull" >&5 $as_echo "$ac_cv_func_malloc_0_nonnull" >&6; } if test $ac_cv_func_malloc_0_nonnull = yes; then : $as_echo "#define HAVE_MALLOC 1" >>confdefs.h else $as_echo "#define HAVE_MALLOC 0" >>confdefs.h case " $LIBOBJS " in *" malloc.$ac_objext "* ) ;; *) LIBOBJS="$LIBOBJS malloc.$ac_objext" ;; esac $as_echo "#define malloc rpl_malloc" >>confdefs.h fi for ac_header in stdlib.h do : ac_fn_c_check_header_mongrel "$LINENO" "stdlib.h" "ac_cv_header_stdlib_h" "$ac_includes_default" if test "x$ac_cv_header_stdlib_h" = xyes; then : cat >>confdefs.h <<_ACEOF #define HAVE_STDLIB_H 1 _ACEOF fi done { $as_echo "$as_me:${as_lineno-$LINENO}: checking for GNU libc compatible realloc" >&5 $as_echo_n "checking for GNU libc compatible realloc... " >&6; } if ${ac_cv_func_realloc_0_nonnull+:} false; then : $as_echo_n "(cached) " >&6 else if test "$cross_compiling" = yes; then : ac_cv_func_realloc_0_nonnull=no else cat confdefs.h - <<_ACEOF >conftest.$ac_ext /* end confdefs.h. */ #if defined STDC_HEADERS || defined HAVE_STDLIB_H # include #else char *realloc (); #endif int main () { return ! realloc (0, 0); ; return 0; } _ACEOF if ac_fn_c_try_run "$LINENO"; then : ac_cv_func_realloc_0_nonnull=yes else ac_cv_func_realloc_0_nonnull=no fi rm -f core *.core core.conftest.* gmon.out bb.out conftest$ac_exeext \ conftest.$ac_objext conftest.beam conftest.$ac_ext fi fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_func_realloc_0_nonnull" >&5 $as_echo "$ac_cv_func_realloc_0_nonnull" >&6; } if test $ac_cv_func_realloc_0_nonnull = yes; then : $as_echo "#define HAVE_REALLOC 1" >>confdefs.h else $as_echo "#define HAVE_REALLOC 0" >>confdefs.h case " $LIBOBJS " in *" realloc.$ac_objext "* ) ;; *) LIBOBJS="$LIBOBJS realloc.$ac_objext" ;; esac $as_echo "#define realloc rpl_realloc" >>confdefs.h fi for ac_func in floor memset pow sqrt do : as_ac_var=`$as_echo "ac_cv_func_$ac_func" | $as_tr_sh` ac_fn_c_check_func "$LINENO" "$ac_func" "$as_ac_var" if eval test \"x\$"$as_ac_var"\" = x"yes"; then : cat >>confdefs.h <<_ACEOF #define `$as_echo "HAVE_$ac_func" | $as_tr_cpp` 1 _ACEOF fi done # For test harness for ac_prog in gawk mawk nawk awk do # Extract the first word of "$ac_prog", so it can be a program name with args. set dummy $ac_prog; ac_word=$2 { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 $as_echo_n "checking for $ac_word... " >&6; } if ${ac_cv_prog_AWK+:} false; then : $as_echo_n "(cached) " >&6 else if test -n "$AWK"; then ac_cv_prog_AWK="$AWK" # Let the user override the test. else as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. for ac_exec_ext in '' $ac_executable_extensions; do if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then ac_cv_prog_AWK="$ac_prog" $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 break 2 fi done done IFS=$as_save_IFS fi fi AWK=$ac_cv_prog_AWK if test -n "$AWK"; then { $as_echo "$as_me:${as_lineno-$LINENO}: result: $AWK" >&5 $as_echo "$AWK" >&6; } else { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 $as_echo "no" >&6; } fi test -n "$AWK" && break done ac_config_files="$ac_config_files Makefile" cat >confcache <<\_ACEOF # This file is a shell script that caches the results of configure # tests run on this system so they can be shared between configure # scripts and configure runs, see configure's option --config-cache. # It is not useful on other systems. If it contains results you don't # want to keep, you may remove or edit it. # # config.status only pays attention to the cache file if you give it # the --recheck option to rerun configure. # # `ac_cv_env_foo' variables (set or unset) will be overridden when # loading this file, other *unset* `ac_cv_foo' will be assigned the # following values. _ACEOF # The following way of writing the cache mishandles newlines in values, # but we know of no workaround that is simple, portable, and efficient. # So, we kill variables containing newlines. # Ultrix sh set writes to stderr and can't be redirected directly, # and sets the high bit in the cache file unless we assign to the vars. ( for ac_var in `(set) 2>&1 | sed -n 's/^\([a-zA-Z_][a-zA-Z0-9_]*\)=.*/\1/p'`; do eval ac_val=\$$ac_var case $ac_val in #( *${as_nl}*) case $ac_var in #( *_cv_*) { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: cache variable $ac_var contains a newline" >&5 $as_echo "$as_me: WARNING: cache variable $ac_var contains a newline" >&2;} ;; esac case $ac_var in #( _ | IFS | as_nl) ;; #( BASH_ARGV | BASH_SOURCE) eval $ac_var= ;; #( *) { eval $ac_var=; unset $ac_var;} ;; esac ;; esac done (set) 2>&1 | case $as_nl`(ac_space=' '; set) 2>&1` in #( *${as_nl}ac_space=\ *) # `set' does not quote correctly, so add quotes: double-quote # substitution turns \\\\ into \\, and sed turns \\ into \. sed -n \ "s/'/'\\\\''/g; s/^\\([_$as_cr_alnum]*_cv_[_$as_cr_alnum]*\\)=\\(.*\\)/\\1='\\2'/p" ;; #( *) # `set' quotes correctly as required by POSIX, so do not add quotes. sed -n "/^[_$as_cr_alnum]*_cv_[_$as_cr_alnum]*=/p" ;; esac | sort ) | sed ' /^ac_cv_env_/b end t clear :clear s/^\([^=]*\)=\(.*[{}].*\)$/test "${\1+set}" = set || &/ t end s/^\([^=]*\)=\(.*\)$/\1=${\1=\2}/ :end' >>confcache if diff "$cache_file" confcache >/dev/null 2>&1; then :; else if test -w "$cache_file"; then if test "x$cache_file" != "x/dev/null"; then { $as_echo "$as_me:${as_lineno-$LINENO}: updating cache $cache_file" >&5 $as_echo "$as_me: updating cache $cache_file" >&6;} if test ! -f "$cache_file" || test -h "$cache_file"; then cat confcache >"$cache_file" else case $cache_file in #( */* | ?:*) mv -f confcache "$cache_file"$$ && mv -f "$cache_file"$$ "$cache_file" ;; #( *) mv -f confcache "$cache_file" ;; esac fi fi else { $as_echo "$as_me:${as_lineno-$LINENO}: not updating unwritable cache $cache_file" >&5 $as_echo "$as_me: not updating unwritable cache $cache_file" >&6;} fi fi rm -f confcache test "x$prefix" = xNONE && prefix=$ac_default_prefix # Let make expand exec_prefix. test "x$exec_prefix" = xNONE && exec_prefix='${prefix}' DEFS=-DHAVE_CONFIG_H ac_libobjs= ac_ltlibobjs= U= for ac_i in : $LIBOBJS; do test "x$ac_i" = x: && continue # 1. Remove the extension, and $U if already installed. ac_script='s/\$U\././;s/\.o$//;s/\.obj$//' ac_i=`$as_echo "$ac_i" | sed "$ac_script"` # 2. Prepend LIBOBJDIR. When used with automake>=1.10 LIBOBJDIR # will be set to the directory where LIBOBJS objects are built. as_fn_append ac_libobjs " \${LIBOBJDIR}$ac_i\$U.$ac_objext" as_fn_append ac_ltlibobjs " \${LIBOBJDIR}$ac_i"'$U.lo' done LIBOBJS=$ac_libobjs LTLIBOBJS=$ac_ltlibobjs { $as_echo "$as_me:${as_lineno-$LINENO}: checking that generated files are newer than configure" >&5 $as_echo_n "checking that generated files are newer than configure... " >&6; } if test -n "$am_sleep_pid"; then # Hide warnings about reused PIDs. wait $am_sleep_pid 2>/dev/null fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: done" >&5 $as_echo "done" >&6; } if test -n "$EXEEXT"; then am__EXEEXT_TRUE= am__EXEEXT_FALSE='#' else am__EXEEXT_TRUE='#' am__EXEEXT_FALSE= fi if test -z "${AMDEP_TRUE}" && test -z "${AMDEP_FALSE}"; then as_fn_error $? "conditional \"AMDEP\" was never defined. Usually this means the macro was only invoked conditionally." "$LINENO" 5 fi if test -z "${am__fastdepCXX_TRUE}" && test -z "${am__fastdepCXX_FALSE}"; then as_fn_error $? "conditional \"am__fastdepCXX\" was never defined. Usually this means the macro was only invoked conditionally." "$LINENO" 5 fi if test -z "${am__fastdepCC_TRUE}" && test -z "${am__fastdepCC_FALSE}"; then as_fn_error $? "conditional \"am__fastdepCC\" was never defined. Usually this means the macro was only invoked conditionally." "$LINENO" 5 fi : "${CONFIG_STATUS=./config.status}" ac_write_fail=0 ac_clean_files_save=$ac_clean_files ac_clean_files="$ac_clean_files $CONFIG_STATUS" { $as_echo "$as_me:${as_lineno-$LINENO}: creating $CONFIG_STATUS" >&5 $as_echo "$as_me: creating $CONFIG_STATUS" >&6;} as_write_fail=0 cat >$CONFIG_STATUS <<_ASEOF || as_write_fail=1 #! $SHELL # Generated by $as_me. # Run this file to recreate the current configuration. # Compiler output produced by configure, useful for debugging # configure, is in config.log if it exists. debug=false ac_cs_recheck=false ac_cs_silent=false SHELL=\${CONFIG_SHELL-$SHELL} export SHELL _ASEOF cat >>$CONFIG_STATUS <<\_ASEOF || as_write_fail=1 ## -------------------- ## ## M4sh Initialization. ## ## -------------------- ## # Be more Bourne compatible DUALCASE=1; export DUALCASE # for MKS sh if test -n "${ZSH_VERSION+set}" && (emulate sh) >/dev/null 2>&1; then : emulate sh NULLCMD=: # Pre-4.2 versions of Zsh do word splitting on ${1+"$@"}, which # is contrary to our usage. Disable this feature. alias -g '${1+"$@"}'='"$@"' setopt NO_GLOB_SUBST else case `(set -o) 2>/dev/null` in #( *posix*) : set -o posix ;; #( *) : ;; esac fi as_nl=' ' export as_nl # Printing a long string crashes Solaris 7 /usr/bin/printf. as_echo='\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\' as_echo=$as_echo$as_echo$as_echo$as_echo$as_echo as_echo=$as_echo$as_echo$as_echo$as_echo$as_echo$as_echo # Prefer a ksh shell builtin over an external printf program on Solaris, # but without wasting forks for bash or zsh. if test -z "$BASH_VERSION$ZSH_VERSION" \ && (test "X`print -r -- $as_echo`" = "X$as_echo") 2>/dev/null; then as_echo='print -r --' as_echo_n='print -rn --' elif (test "X`printf %s $as_echo`" = "X$as_echo") 2>/dev/null; then as_echo='printf %s\n' as_echo_n='printf %s' else if test "X`(/usr/ucb/echo -n -n $as_echo) 2>/dev/null`" = "X-n $as_echo"; then as_echo_body='eval /usr/ucb/echo -n "$1$as_nl"' as_echo_n='/usr/ucb/echo -n' else as_echo_body='eval expr "X$1" : "X\\(.*\\)"' as_echo_n_body='eval arg=$1; case $arg in #( *"$as_nl"*) expr "X$arg" : "X\\(.*\\)$as_nl"; arg=`expr "X$arg" : ".*$as_nl\\(.*\\)"`;; esac; expr "X$arg" : "X\\(.*\\)" | tr -d "$as_nl" ' export as_echo_n_body as_echo_n='sh -c $as_echo_n_body as_echo' fi export as_echo_body as_echo='sh -c $as_echo_body as_echo' fi # The user is always right. if test "${PATH_SEPARATOR+set}" != set; then PATH_SEPARATOR=: (PATH='/bin;/bin'; FPATH=$PATH; sh -c :) >/dev/null 2>&1 && { (PATH='/bin:/bin'; FPATH=$PATH; sh -c :) >/dev/null 2>&1 || PATH_SEPARATOR=';' } fi # IFS # We need space, tab and new line, in precisely that order. Quoting is # there to prevent editors from complaining about space-tab. # (If _AS_PATH_WALK were called with IFS unset, it would disable word # splitting by setting IFS to empty value.) IFS=" "" $as_nl" # Find who we are. Look in the path if we contain no directory separator. as_myself= case $0 in #(( *[\\/]* ) as_myself=$0 ;; *) as_save_IFS=$IFS; IFS=$PATH_SEPARATOR for as_dir in $PATH do IFS=$as_save_IFS test -z "$as_dir" && as_dir=. test -r "$as_dir/$0" && as_myself=$as_dir/$0 && break done IFS=$as_save_IFS ;; esac # We did not find ourselves, most probably we were run as `sh COMMAND' # in which case we are not to be found in the path. if test "x$as_myself" = x; then as_myself=$0 fi if test ! -f "$as_myself"; then $as_echo "$as_myself: error: cannot find myself; rerun with an absolute file name" >&2 exit 1 fi # Unset variables that we do not need and which cause bugs (e.g. in # pre-3.0 UWIN ksh). But do not cause bugs in bash 2.01; the "|| exit 1" # suppresses any "Segmentation fault" message there. '((' could # trigger a bug in pdksh 5.2.14. for as_var in BASH_ENV ENV MAIL MAILPATH do eval test x\${$as_var+set} = xset \ && ( (unset $as_var) || exit 1) >/dev/null 2>&1 && unset $as_var || : done PS1='$ ' PS2='> ' PS4='+ ' # NLS nuisances. LC_ALL=C export LC_ALL LANGUAGE=C export LANGUAGE # CDPATH. (unset CDPATH) >/dev/null 2>&1 && unset CDPATH # as_fn_error STATUS ERROR [LINENO LOG_FD] # ---------------------------------------- # Output "`basename $0`: error: ERROR" to stderr. If LINENO and LOG_FD are # provided, also output the error to LOG_FD, referencing LINENO. Then exit the # script with STATUS, using 1 if that was 0. as_fn_error () { as_status=$1; test $as_status -eq 0 && as_status=1 if test "$4"; then as_lineno=${as_lineno-"$3"} as_lineno_stack=as_lineno_stack=$as_lineno_stack $as_echo "$as_me:${as_lineno-$LINENO}: error: $2" >&$4 fi $as_echo "$as_me: error: $2" >&2 as_fn_exit $as_status } # as_fn_error # as_fn_set_status STATUS # ----------------------- # Set $? to STATUS, without forking. as_fn_set_status () { return $1 } # as_fn_set_status # as_fn_exit STATUS # ----------------- # Exit the shell with STATUS, even in a "trap 0" or "set -e" context. as_fn_exit () { set +e as_fn_set_status $1 exit $1 } # as_fn_exit # as_fn_unset VAR # --------------- # Portably unset VAR. as_fn_unset () { { eval $1=; unset $1;} } as_unset=as_fn_unset # as_fn_append VAR VALUE # ---------------------- # Append the text in VALUE to the end of the definition contained in VAR. Take # advantage of any shell optimizations that allow amortized linear growth over # repeated appends, instead of the typical quadratic growth present in naive # implementations. if (eval "as_var=1; as_var+=2; test x\$as_var = x12") 2>/dev/null; then : eval 'as_fn_append () { eval $1+=\$2 }' else as_fn_append () { eval $1=\$$1\$2 } fi # as_fn_append # as_fn_arith ARG... # ------------------ # Perform arithmetic evaluation on the ARGs, and store the result in the # global $as_val. Take advantage of shells that can avoid forks. The arguments # must be portable across $(()) and expr. if (eval "test \$(( 1 + 1 )) = 2") 2>/dev/null; then : eval 'as_fn_arith () { as_val=$(( $* )) }' else as_fn_arith () { as_val=`expr "$@" || test $? -eq 1` } fi # as_fn_arith if expr a : '\(a\)' >/dev/null 2>&1 && test "X`expr 00001 : '.*\(...\)'`" = X001; then as_expr=expr else as_expr=false fi if (basename -- /) >/dev/null 2>&1 && test "X`basename -- / 2>&1`" = "X/"; then as_basename=basename else as_basename=false fi if (as_dir=`dirname -- /` && test "X$as_dir" = X/) >/dev/null 2>&1; then as_dirname=dirname else as_dirname=false fi as_me=`$as_basename -- "$0" || $as_expr X/"$0" : '.*/\([^/][^/]*\)/*$' \| \ X"$0" : 'X\(//\)$' \| \ X"$0" : 'X\(/\)' \| . 2>/dev/null || $as_echo X/"$0" | sed '/^.*\/\([^/][^/]*\)\/*$/{ s//\1/ q } /^X\/\(\/\/\)$/{ s//\1/ q } /^X\/\(\/\).*/{ s//\1/ q } s/.*/./; q'` # Avoid depending upon Character Ranges. as_cr_letters='abcdefghijklmnopqrstuvwxyz' as_cr_LETTERS='ABCDEFGHIJKLMNOPQRSTUVWXYZ' as_cr_Letters=$as_cr_letters$as_cr_LETTERS as_cr_digits='0123456789' as_cr_alnum=$as_cr_Letters$as_cr_digits ECHO_C= ECHO_N= ECHO_T= case `echo -n x` in #((((( -n*) case `echo 'xy\c'` in *c*) ECHO_T=' ';; # ECHO_T is single tab character. xy) ECHO_C='\c';; *) echo `echo ksh88 bug on AIX 6.1` > /dev/null ECHO_T=' ';; esac;; *) ECHO_N='-n';; esac rm -f conf$$ conf$$.exe conf$$.file if test -d conf$$.dir; then rm -f conf$$.dir/conf$$.file else rm -f conf$$.dir mkdir conf$$.dir 2>/dev/null fi if (echo >conf$$.file) 2>/dev/null; then if ln -s conf$$.file conf$$ 2>/dev/null; then as_ln_s='ln -s' # ... but there are two gotchas: # 1) On MSYS, both `ln -s file dir' and `ln file dir' fail. # 2) DJGPP < 2.04 has no symlinks; `ln -s' creates a wrapper executable. # In both cases, we have to default to `cp -pR'. ln -s conf$$.file conf$$.dir 2>/dev/null && test ! -f conf$$.exe || as_ln_s='cp -pR' elif ln conf$$.file conf$$ 2>/dev/null; then as_ln_s=ln else as_ln_s='cp -pR' fi else as_ln_s='cp -pR' fi rm -f conf$$ conf$$.exe conf$$.dir/conf$$.file conf$$.file rmdir conf$$.dir 2>/dev/null # as_fn_mkdir_p # ------------- # Create "$as_dir" as a directory, including parents if necessary. as_fn_mkdir_p () { case $as_dir in #( -*) as_dir=./$as_dir;; esac test -d "$as_dir" || eval $as_mkdir_p || { as_dirs= while :; do case $as_dir in #( *\'*) as_qdir=`$as_echo "$as_dir" | sed "s/'/'\\\\\\\\''/g"`;; #'( *) as_qdir=$as_dir;; esac as_dirs="'$as_qdir' $as_dirs" as_dir=`$as_dirname -- "$as_dir" || $as_expr X"$as_dir" : 'X\(.*[^/]\)//*[^/][^/]*/*$' \| \ X"$as_dir" : 'X\(//\)[^/]' \| \ X"$as_dir" : 'X\(//\)$' \| \ X"$as_dir" : 'X\(/\)' \| . 2>/dev/null || $as_echo X"$as_dir" | sed '/^X\(.*[^/]\)\/\/*[^/][^/]*\/*$/{ s//\1/ q } /^X\(\/\/\)[^/].*/{ s//\1/ q } /^X\(\/\/\)$/{ s//\1/ q } /^X\(\/\).*/{ s//\1/ q } s/.*/./; q'` test -d "$as_dir" && break done test -z "$as_dirs" || eval "mkdir $as_dirs" } || test -d "$as_dir" || as_fn_error $? "cannot create directory $as_dir" } # as_fn_mkdir_p if mkdir -p . 2>/dev/null; then as_mkdir_p='mkdir -p "$as_dir"' else test -d ./-p && rmdir ./-p as_mkdir_p=false fi # as_fn_executable_p FILE # ----------------------- # Test if FILE is an executable regular file. as_fn_executable_p () { test -f "$1" && test -x "$1" } # as_fn_executable_p as_test_x='test -x' as_executable_p=as_fn_executable_p # Sed expression to map a string onto a valid CPP name. as_tr_cpp="eval sed 'y%*$as_cr_letters%P$as_cr_LETTERS%;s%[^_$as_cr_alnum]%_%g'" # Sed expression to map a string onto a valid variable name. as_tr_sh="eval sed 'y%*+%pp%;s%[^_$as_cr_alnum]%_%g'" exec 6>&1 ## ----------------------------------- ## ## Main body of $CONFIG_STATUS script. ## ## ----------------------------------- ## _ASEOF test $as_write_fail = 0 && chmod +x $CONFIG_STATUS || ac_write_fail=1 cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 # Save the log message, to keep $0 and so on meaningful, and to # report actual input values of CONFIG_FILES etc. instead of their # values after options handling. ac_log=" This file was extended by Stacks $as_me 1.44, which was generated by GNU Autoconf 2.69. Invocation command line was CONFIG_FILES = $CONFIG_FILES CONFIG_HEADERS = $CONFIG_HEADERS CONFIG_LINKS = $CONFIG_LINKS CONFIG_COMMANDS = $CONFIG_COMMANDS $ $0 $@ on `(hostname || uname -n) 2>/dev/null | sed 1q` " _ACEOF case $ac_config_files in *" "*) set x $ac_config_files; shift; ac_config_files=$*;; esac case $ac_config_headers in *" "*) set x $ac_config_headers; shift; ac_config_headers=$*;; esac cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 # Files that config.status was made for. config_files="$ac_config_files" config_headers="$ac_config_headers" config_commands="$ac_config_commands" _ACEOF cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 ac_cs_usage="\ \`$as_me' instantiates files and other configuration actions from templates according to the current configuration. Unless the files and actions are specified as TAGs, all are instantiated by default. Usage: $0 [OPTION]... [TAG]... -h, --help print this help, then exit -V, --version print version number and configuration settings, then exit --config print configuration, then exit -q, --quiet, --silent do not print progress messages -d, --debug don't remove temporary files --recheck update $as_me by reconfiguring in the same conditions --file=FILE[:TEMPLATE] instantiate the configuration file FILE --header=FILE[:TEMPLATE] instantiate the configuration header FILE Configuration files: $config_files Configuration headers: $config_headers Configuration commands: $config_commands Report bugs to the package provider." _ACEOF cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 ac_cs_config="`$as_echo "$ac_configure_args" | sed 's/^ //; s/[\\""\`\$]/\\\\&/g'`" ac_cs_version="\\ Stacks config.status 1.44 configured by $0, generated by GNU Autoconf 2.69, with options \\"\$ac_cs_config\\" Copyright (C) 2012 Free Software Foundation, Inc. This config.status script is free software; the Free Software Foundation gives unlimited permission to copy, distribute and modify it." ac_pwd='$ac_pwd' srcdir='$srcdir' INSTALL='$INSTALL' MKDIR_P='$MKDIR_P' AWK='$AWK' test -n "\$AWK" || AWK=awk _ACEOF cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 # The default lists apply if the user does not specify any file. ac_need_defaults=: while test $# != 0 do case $1 in --*=?*) ac_option=`expr "X$1" : 'X\([^=]*\)='` ac_optarg=`expr "X$1" : 'X[^=]*=\(.*\)'` ac_shift=: ;; --*=) ac_option=`expr "X$1" : 'X\([^=]*\)='` ac_optarg= ac_shift=: ;; *) ac_option=$1 ac_optarg=$2 ac_shift=shift ;; esac case $ac_option in # Handling of the options. -recheck | --recheck | --rechec | --reche | --rech | --rec | --re | --r) ac_cs_recheck=: ;; --version | --versio | --versi | --vers | --ver | --ve | --v | -V ) $as_echo "$ac_cs_version"; exit ;; --config | --confi | --conf | --con | --co | --c ) $as_echo "$ac_cs_config"; exit ;; --debug | --debu | --deb | --de | --d | -d ) debug=: ;; --file | --fil | --fi | --f ) $ac_shift case $ac_optarg in *\'*) ac_optarg=`$as_echo "$ac_optarg" | sed "s/'/'\\\\\\\\''/g"` ;; '') as_fn_error $? "missing file argument" ;; esac as_fn_append CONFIG_FILES " '$ac_optarg'" ac_need_defaults=false;; --header | --heade | --head | --hea ) $ac_shift case $ac_optarg in *\'*) ac_optarg=`$as_echo "$ac_optarg" | sed "s/'/'\\\\\\\\''/g"` ;; esac as_fn_append CONFIG_HEADERS " '$ac_optarg'" ac_need_defaults=false;; --he | --h) # Conflict between --help and --header as_fn_error $? "ambiguous option: \`$1' Try \`$0 --help' for more information.";; --help | --hel | -h ) $as_echo "$ac_cs_usage"; exit ;; -q | -quiet | --quiet | --quie | --qui | --qu | --q \ | -silent | --silent | --silen | --sile | --sil | --si | --s) ac_cs_silent=: ;; # This is an error. -*) as_fn_error $? "unrecognized option: \`$1' Try \`$0 --help' for more information." ;; *) as_fn_append ac_config_targets " $1" ac_need_defaults=false ;; esac shift done ac_configure_extra_args= if $ac_cs_silent; then exec 6>/dev/null ac_configure_extra_args="$ac_configure_extra_args --silent" fi _ACEOF cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 if \$ac_cs_recheck; then set X $SHELL '$0' $ac_configure_args \$ac_configure_extra_args --no-create --no-recursion shift \$as_echo "running CONFIG_SHELL=$SHELL \$*" >&6 CONFIG_SHELL='$SHELL' export CONFIG_SHELL exec "\$@" fi _ACEOF cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 exec 5>>config.log { echo sed 'h;s/./-/g;s/^.../## /;s/...$/ ##/;p;x;p;x' <<_ASBOX ## Running $as_me. ## _ASBOX $as_echo "$ac_log" } >&5 _ACEOF cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 # # INIT-COMMANDS # AMDEP_TRUE="$AMDEP_TRUE" ac_aux_dir="$ac_aux_dir" _ACEOF cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 # Handling of arguments. for ac_config_target in $ac_config_targets do case $ac_config_target in "config.h") CONFIG_HEADERS="$CONFIG_HEADERS config.h" ;; "depfiles") CONFIG_COMMANDS="$CONFIG_COMMANDS depfiles" ;; "Makefile") CONFIG_FILES="$CONFIG_FILES Makefile" ;; *) as_fn_error $? "invalid argument: \`$ac_config_target'" "$LINENO" 5;; esac done # If the user did not use the arguments to specify the items to instantiate, # then the envvar interface is used. Set only those that are not. # We use the long form for the default assignment because of an extremely # bizarre bug on SunOS 4.1.3. if $ac_need_defaults; then test "${CONFIG_FILES+set}" = set || CONFIG_FILES=$config_files test "${CONFIG_HEADERS+set}" = set || CONFIG_HEADERS=$config_headers test "${CONFIG_COMMANDS+set}" = set || CONFIG_COMMANDS=$config_commands fi # Have a temporary directory for convenience. Make it in the build tree # simply because there is no reason against having it here, and in addition, # creating and moving files from /tmp can sometimes cause problems. # Hook for its removal unless debugging. # Note that there is a small window in which the directory will not be cleaned: # after its creation but before its name has been assigned to `$tmp'. $debug || { tmp= ac_tmp= trap 'exit_status=$? : "${ac_tmp:=$tmp}" { test ! -d "$ac_tmp" || rm -fr "$ac_tmp"; } && exit $exit_status ' 0 trap 'as_fn_exit 1' 1 2 13 15 } # Create a (secure) tmp directory for tmp files. { tmp=`(umask 077 && mktemp -d "./confXXXXXX") 2>/dev/null` && test -d "$tmp" } || { tmp=./conf$$-$RANDOM (umask 077 && mkdir "$tmp") } || as_fn_error $? "cannot create a temporary directory in ." "$LINENO" 5 ac_tmp=$tmp # Set up the scripts for CONFIG_FILES section. # No need to generate them if there are no CONFIG_FILES. # This happens for instance with `./config.status config.h'. if test -n "$CONFIG_FILES"; then ac_cr=`echo X | tr X '\015'` # On cygwin, bash can eat \r inside `` if the user requested igncr. # But we know of no other shell where ac_cr would be empty at this # point, so we can use a bashism as a fallback. if test "x$ac_cr" = x; then eval ac_cr=\$\'\\r\' fi ac_cs_awk_cr=`$AWK 'BEGIN { print "a\rb" }' /dev/null` if test "$ac_cs_awk_cr" = "a${ac_cr}b"; then ac_cs_awk_cr='\\r' else ac_cs_awk_cr=$ac_cr fi echo 'BEGIN {' >"$ac_tmp/subs1.awk" && _ACEOF { echo "cat >conf$$subs.awk <<_ACEOF" && echo "$ac_subst_vars" | sed 's/.*/&!$&$ac_delim/' && echo "_ACEOF" } >conf$$subs.sh || as_fn_error $? "could not make $CONFIG_STATUS" "$LINENO" 5 ac_delim_num=`echo "$ac_subst_vars" | grep -c '^'` ac_delim='%!_!# ' for ac_last_try in false false false false false :; do . ./conf$$subs.sh || as_fn_error $? "could not make $CONFIG_STATUS" "$LINENO" 5 ac_delim_n=`sed -n "s/.*$ac_delim\$/X/p" conf$$subs.awk | grep -c X` if test $ac_delim_n = $ac_delim_num; then break elif $ac_last_try; then as_fn_error $? "could not make $CONFIG_STATUS" "$LINENO" 5 else ac_delim="$ac_delim!$ac_delim _$ac_delim!! " fi done rm -f conf$$subs.sh cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 cat >>"\$ac_tmp/subs1.awk" <<\\_ACAWK && _ACEOF sed -n ' h s/^/S["/; s/!.*/"]=/ p g s/^[^!]*!// :repl t repl s/'"$ac_delim"'$// t delim :nl h s/\(.\{148\}\)..*/\1/ t more1 s/["\\]/\\&/g; s/^/"/; s/$/\\n"\\/ p n b repl :more1 s/["\\]/\\&/g; s/^/"/; s/$/"\\/ p g s/.\{148\}// t nl :delim h s/\(.\{148\}\)..*/\1/ t more2 s/["\\]/\\&/g; s/^/"/; s/$/"/ p b :more2 s/["\\]/\\&/g; s/^/"/; s/$/"\\/ p g s/.\{148\}// t delim ' >$CONFIG_STATUS || ac_write_fail=1 rm -f conf$$subs.awk cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 _ACAWK cat >>"\$ac_tmp/subs1.awk" <<_ACAWK && for (key in S) S_is_set[key] = 1 FS = "" } { line = $ 0 nfields = split(line, field, "@") substed = 0 len = length(field[1]) for (i = 2; i < nfields; i++) { key = field[i] keylen = length(key) if (S_is_set[key]) { value = S[key] line = substr(line, 1, len) "" value "" substr(line, len + keylen + 3) len += length(value) + length(field[++i]) substed = 1 } else len += 1 + keylen } print line } _ACAWK _ACEOF cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 if sed "s/$ac_cr//" < /dev/null > /dev/null 2>&1; then sed "s/$ac_cr\$//; s/$ac_cr/$ac_cs_awk_cr/g" else cat fi < "$ac_tmp/subs1.awk" > "$ac_tmp/subs.awk" \ || as_fn_error $? "could not setup config files machinery" "$LINENO" 5 _ACEOF # VPATH may cause trouble with some makes, so we remove sole $(srcdir), # ${srcdir} and @srcdir@ entries from VPATH if srcdir is ".", strip leading and # trailing colons and then remove the whole line if VPATH becomes empty # (actually we leave an empty line to preserve line numbers). if test "x$srcdir" = x.; then ac_vpsub='/^[ ]*VPATH[ ]*=[ ]*/{ h s/// s/^/:/ s/[ ]*$/:/ s/:\$(srcdir):/:/g s/:\${srcdir}:/:/g s/:@srcdir@:/:/g s/^:*// s/:*$// x s/\(=[ ]*\).*/\1/ G s/\n// s/^[^=]*=[ ]*$// }' fi cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 fi # test -n "$CONFIG_FILES" # Set up the scripts for CONFIG_HEADERS section. # No need to generate them if there are no CONFIG_HEADERS. # This happens for instance with `./config.status Makefile'. if test -n "$CONFIG_HEADERS"; then cat >"$ac_tmp/defines.awk" <<\_ACAWK || BEGIN { _ACEOF # Transform confdefs.h into an awk script `defines.awk', embedded as # here-document in config.status, that substitutes the proper values into # config.h.in to produce config.h. # Create a delimiter string that does not exist in confdefs.h, to ease # handling of long lines. ac_delim='%!_!# ' for ac_last_try in false false :; do ac_tt=`sed -n "/$ac_delim/p" confdefs.h` if test -z "$ac_tt"; then break elif $ac_last_try; then as_fn_error $? "could not make $CONFIG_HEADERS" "$LINENO" 5 else ac_delim="$ac_delim!$ac_delim _$ac_delim!! " fi done # For the awk script, D is an array of macro values keyed by name, # likewise P contains macro parameters if any. Preserve backslash # newline sequences. ac_word_re=[_$as_cr_Letters][_$as_cr_alnum]* sed -n ' s/.\{148\}/&'"$ac_delim"'/g t rset :rset s/^[ ]*#[ ]*define[ ][ ]*/ / t def d :def s/\\$// t bsnl s/["\\]/\\&/g s/^ \('"$ac_word_re"'\)\(([^()]*)\)[ ]*\(.*\)/P["\1"]="\2"\ D["\1"]=" \3"/p s/^ \('"$ac_word_re"'\)[ ]*\(.*\)/D["\1"]=" \2"/p d :bsnl s/["\\]/\\&/g s/^ \('"$ac_word_re"'\)\(([^()]*)\)[ ]*\(.*\)/P["\1"]="\2"\ D["\1"]=" \3\\\\\\n"\\/p t cont s/^ \('"$ac_word_re"'\)[ ]*\(.*\)/D["\1"]=" \2\\\\\\n"\\/p t cont d :cont n s/.\{148\}/&'"$ac_delim"'/g t clear :clear s/\\$// t bsnlc s/["\\]/\\&/g; s/^/"/; s/$/"/p d :bsnlc s/["\\]/\\&/g; s/^/"/; s/$/\\\\\\n"\\/p b cont ' >$CONFIG_STATUS || ac_write_fail=1 cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 for (key in D) D_is_set[key] = 1 FS = "" } /^[\t ]*#[\t ]*(define|undef)[\t ]+$ac_word_re([\t (]|\$)/ { line = \$ 0 split(line, arg, " ") if (arg[1] == "#") { defundef = arg[2] mac1 = arg[3] } else { defundef = substr(arg[1], 2) mac1 = arg[2] } split(mac1, mac2, "(") #) macro = mac2[1] prefix = substr(line, 1, index(line, defundef) - 1) if (D_is_set[macro]) { # Preserve the white space surrounding the "#". print prefix "define", macro P[macro] D[macro] next } else { # Replace #undef with comments. This is necessary, for example, # in the case of _POSIX_SOURCE, which is predefined and required # on some systems where configure will not decide to define it. if (defundef == "undef") { print "/*", prefix defundef, macro, "*/" next } } } { print } _ACAWK _ACEOF cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 as_fn_error $? "could not setup config headers machinery" "$LINENO" 5 fi # test -n "$CONFIG_HEADERS" eval set X " :F $CONFIG_FILES :H $CONFIG_HEADERS :C $CONFIG_COMMANDS" shift for ac_tag do case $ac_tag in :[FHLC]) ac_mode=$ac_tag; continue;; esac case $ac_mode$ac_tag in :[FHL]*:*);; :L* | :C*:*) as_fn_error $? "invalid tag \`$ac_tag'" "$LINENO" 5;; :[FH]-) ac_tag=-:-;; :[FH]*) ac_tag=$ac_tag:$ac_tag.in;; esac ac_save_IFS=$IFS IFS=: set x $ac_tag IFS=$ac_save_IFS shift ac_file=$1 shift case $ac_mode in :L) ac_source=$1;; :[FH]) ac_file_inputs= for ac_f do case $ac_f in -) ac_f="$ac_tmp/stdin";; *) # Look for the file first in the build tree, then in the source tree # (if the path is not absolute). The absolute path cannot be DOS-style, # because $ac_f cannot contain `:'. test -f "$ac_f" || case $ac_f in [\\/$]*) false;; *) test -f "$srcdir/$ac_f" && ac_f="$srcdir/$ac_f";; esac || as_fn_error 1 "cannot find input file: \`$ac_f'" "$LINENO" 5;; esac case $ac_f in *\'*) ac_f=`$as_echo "$ac_f" | sed "s/'/'\\\\\\\\''/g"`;; esac as_fn_append ac_file_inputs " '$ac_f'" done # Let's still pretend it is `configure' which instantiates (i.e., don't # use $as_me), people would be surprised to read: # /* config.h. Generated by config.status. */ configure_input='Generated from '` $as_echo "$*" | sed 's|^[^:]*/||;s|:[^:]*/|, |g' `' by configure.' if test x"$ac_file" != x-; then configure_input="$ac_file. $configure_input" { $as_echo "$as_me:${as_lineno-$LINENO}: creating $ac_file" >&5 $as_echo "$as_me: creating $ac_file" >&6;} fi # Neutralize special characters interpreted by sed in replacement strings. case $configure_input in #( *\&* | *\|* | *\\* ) ac_sed_conf_input=`$as_echo "$configure_input" | sed 's/[\\\\&|]/\\\\&/g'`;; #( *) ac_sed_conf_input=$configure_input;; esac case $ac_tag in *:-:* | *:-) cat >"$ac_tmp/stdin" \ || as_fn_error $? "could not create $ac_file" "$LINENO" 5 ;; esac ;; esac ac_dir=`$as_dirname -- "$ac_file" || $as_expr X"$ac_file" : 'X\(.*[^/]\)//*[^/][^/]*/*$' \| \ X"$ac_file" : 'X\(//\)[^/]' \| \ X"$ac_file" : 'X\(//\)$' \| \ X"$ac_file" : 'X\(/\)' \| . 2>/dev/null || $as_echo X"$ac_file" | sed '/^X\(.*[^/]\)\/\/*[^/][^/]*\/*$/{ s//\1/ q } /^X\(\/\/\)[^/].*/{ s//\1/ q } /^X\(\/\/\)$/{ s//\1/ q } /^X\(\/\).*/{ s//\1/ q } s/.*/./; q'` as_dir="$ac_dir"; as_fn_mkdir_p ac_builddir=. case "$ac_dir" in .) ac_dir_suffix= ac_top_builddir_sub=. ac_top_build_prefix= ;; *) ac_dir_suffix=/`$as_echo "$ac_dir" | sed 's|^\.[\\/]||'` # A ".." for each directory in $ac_dir_suffix. ac_top_builddir_sub=`$as_echo "$ac_dir_suffix" | sed 's|/[^\\/]*|/..|g;s|/||'` case $ac_top_builddir_sub in "") ac_top_builddir_sub=. ac_top_build_prefix= ;; *) ac_top_build_prefix=$ac_top_builddir_sub/ ;; esac ;; esac ac_abs_top_builddir=$ac_pwd ac_abs_builddir=$ac_pwd$ac_dir_suffix # for backward compatibility: ac_top_builddir=$ac_top_build_prefix case $srcdir in .) # We are building in place. ac_srcdir=. ac_top_srcdir=$ac_top_builddir_sub ac_abs_top_srcdir=$ac_pwd ;; [\\/]* | ?:[\\/]* ) # Absolute name. ac_srcdir=$srcdir$ac_dir_suffix; ac_top_srcdir=$srcdir ac_abs_top_srcdir=$srcdir ;; *) # Relative name. ac_srcdir=$ac_top_build_prefix$srcdir$ac_dir_suffix ac_top_srcdir=$ac_top_build_prefix$srcdir ac_abs_top_srcdir=$ac_pwd/$srcdir ;; esac ac_abs_srcdir=$ac_abs_top_srcdir$ac_dir_suffix case $ac_mode in :F) # # CONFIG_FILE # case $INSTALL in [\\/$]* | ?:[\\/]* ) ac_INSTALL=$INSTALL ;; *) ac_INSTALL=$ac_top_build_prefix$INSTALL ;; esac ac_MKDIR_P=$MKDIR_P case $MKDIR_P in [\\/$]* | ?:[\\/]* ) ;; */*) ac_MKDIR_P=$ac_top_build_prefix$MKDIR_P ;; esac _ACEOF cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 # If the template does not know about datarootdir, expand it. # FIXME: This hack should be removed a few years after 2.60. ac_datarootdir_hack=; ac_datarootdir_seen= ac_sed_dataroot=' /datarootdir/ { p q } /@datadir@/p /@docdir@/p /@infodir@/p /@localedir@/p /@mandir@/p' case `eval "sed -n \"\$ac_sed_dataroot\" $ac_file_inputs"` in *datarootdir*) ac_datarootdir_seen=yes;; *@datadir@*|*@docdir@*|*@infodir@*|*@localedir@*|*@mandir@*) { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $ac_file_inputs seems to ignore the --datarootdir setting" >&5 $as_echo "$as_me: WARNING: $ac_file_inputs seems to ignore the --datarootdir setting" >&2;} _ACEOF cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 ac_datarootdir_hack=' s&@datadir@&$datadir&g s&@docdir@&$docdir&g s&@infodir@&$infodir&g s&@localedir@&$localedir&g s&@mandir@&$mandir&g s&\\\${datarootdir}&$datarootdir&g' ;; esac _ACEOF # Neutralize VPATH when `$srcdir' = `.'. # Shell code in configure.ac might set extrasub. # FIXME: do we really want to maintain this feature? cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 ac_sed_extra="$ac_vpsub $extrasub _ACEOF cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 :t /@[a-zA-Z_][a-zA-Z_0-9]*@/!b s|@configure_input@|$ac_sed_conf_input|;t t s&@top_builddir@&$ac_top_builddir_sub&;t t s&@top_build_prefix@&$ac_top_build_prefix&;t t s&@srcdir@&$ac_srcdir&;t t s&@abs_srcdir@&$ac_abs_srcdir&;t t s&@top_srcdir@&$ac_top_srcdir&;t t s&@abs_top_srcdir@&$ac_abs_top_srcdir&;t t s&@builddir@&$ac_builddir&;t t s&@abs_builddir@&$ac_abs_builddir&;t t s&@abs_top_builddir@&$ac_abs_top_builddir&;t t s&@INSTALL@&$ac_INSTALL&;t t s&@MKDIR_P@&$ac_MKDIR_P&;t t $ac_datarootdir_hack " eval sed \"\$ac_sed_extra\" "$ac_file_inputs" | $AWK -f "$ac_tmp/subs.awk" \ >$ac_tmp/out || as_fn_error $? "could not create $ac_file" "$LINENO" 5 test -z "$ac_datarootdir_hack$ac_datarootdir_seen" && { ac_out=`sed -n '/\${datarootdir}/p' "$ac_tmp/out"`; test -n "$ac_out"; } && { ac_out=`sed -n '/^[ ]*datarootdir[ ]*:*=/p' \ "$ac_tmp/out"`; test -z "$ac_out"; } && { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $ac_file contains a reference to the variable \`datarootdir' which seems to be undefined. Please make sure it is defined" >&5 $as_echo "$as_me: WARNING: $ac_file contains a reference to the variable \`datarootdir' which seems to be undefined. Please make sure it is defined" >&2;} rm -f "$ac_tmp/stdin" case $ac_file in -) cat "$ac_tmp/out" && rm -f "$ac_tmp/out";; *) rm -f "$ac_file" && mv "$ac_tmp/out" "$ac_file";; esac \ || as_fn_error $? "could not create $ac_file" "$LINENO" 5 ;; :H) # # CONFIG_HEADER # if test x"$ac_file" != x-; then { $as_echo "/* $configure_input */" \ && eval '$AWK -f "$ac_tmp/defines.awk"' "$ac_file_inputs" } >"$ac_tmp/config.h" \ || as_fn_error $? "could not create $ac_file" "$LINENO" 5 if diff "$ac_file" "$ac_tmp/config.h" >/dev/null 2>&1; then { $as_echo "$as_me:${as_lineno-$LINENO}: $ac_file is unchanged" >&5 $as_echo "$as_me: $ac_file is unchanged" >&6;} else rm -f "$ac_file" mv "$ac_tmp/config.h" "$ac_file" \ || as_fn_error $? "could not create $ac_file" "$LINENO" 5 fi else $as_echo "/* $configure_input */" \ && eval '$AWK -f "$ac_tmp/defines.awk"' "$ac_file_inputs" \ || as_fn_error $? "could not create -" "$LINENO" 5 fi # Compute "$ac_file"'s index in $config_headers. _am_arg="$ac_file" _am_stamp_count=1 for _am_header in $config_headers :; do case $_am_header in $_am_arg | $_am_arg:* ) break ;; * ) _am_stamp_count=`expr $_am_stamp_count + 1` ;; esac done echo "timestamp for $_am_arg" >`$as_dirname -- "$_am_arg" || $as_expr X"$_am_arg" : 'X\(.*[^/]\)//*[^/][^/]*/*$' \| \ X"$_am_arg" : 'X\(//\)[^/]' \| \ X"$_am_arg" : 'X\(//\)$' \| \ X"$_am_arg" : 'X\(/\)' \| . 2>/dev/null || $as_echo X"$_am_arg" | sed '/^X\(.*[^/]\)\/\/*[^/][^/]*\/*$/{ s//\1/ q } /^X\(\/\/\)[^/].*/{ s//\1/ q } /^X\(\/\/\)$/{ s//\1/ q } /^X\(\/\).*/{ s//\1/ q } s/.*/./; q'`/stamp-h$_am_stamp_count ;; :C) { $as_echo "$as_me:${as_lineno-$LINENO}: executing $ac_file commands" >&5 $as_echo "$as_me: executing $ac_file commands" >&6;} ;; esac case $ac_file$ac_mode in "depfiles":C) test x"$AMDEP_TRUE" != x"" || { # Older Autoconf quotes --file arguments for eval, but not when files # are listed without --file. Let's play safe and only enable the eval # if we detect the quoting. case $CONFIG_FILES in *\'*) eval set x "$CONFIG_FILES" ;; *) set x $CONFIG_FILES ;; esac shift for mf do # Strip MF so we end up with the name of the file. mf=`echo "$mf" | sed -e 's/:.*$//'` # Check whether this is an Automake generated Makefile or not. # We used to match only the files named 'Makefile.in', but # some people rename them; so instead we look at the file content. # Grep'ing the first line is not enough: some people post-process # each Makefile.in and add a new line on top of each file to say so. # Grep'ing the whole file is not good either: AIX grep has a line # limit of 2048, but all sed's we know have understand at least 4000. if sed -n 's,^#.*generated by automake.*,X,p' "$mf" | grep X >/dev/null 2>&1; then dirpart=`$as_dirname -- "$mf" || $as_expr X"$mf" : 'X\(.*[^/]\)//*[^/][^/]*/*$' \| \ X"$mf" : 'X\(//\)[^/]' \| \ X"$mf" : 'X\(//\)$' \| \ X"$mf" : 'X\(/\)' \| . 2>/dev/null || $as_echo X"$mf" | sed '/^X\(.*[^/]\)\/\/*[^/][^/]*\/*$/{ s//\1/ q } /^X\(\/\/\)[^/].*/{ s//\1/ q } /^X\(\/\/\)$/{ s//\1/ q } /^X\(\/\).*/{ s//\1/ q } s/.*/./; q'` else continue fi # Extract the definition of DEPDIR, am__include, and am__quote # from the Makefile without running 'make'. DEPDIR=`sed -n 's/^DEPDIR = //p' < "$mf"` test -z "$DEPDIR" && continue am__include=`sed -n 's/^am__include = //p' < "$mf"` test -z "$am__include" && continue am__quote=`sed -n 's/^am__quote = //p' < "$mf"` # Find all dependency output files, they are included files with # $(DEPDIR) in their names. We invoke sed twice because it is the # simplest approach to changing $(DEPDIR) to its actual value in the # expansion. for file in `sed -n " s/^$am__include $am__quote\(.*(DEPDIR).*\)$am__quote"'$/\1/p' <"$mf" | \ sed -e 's/\$(DEPDIR)/'"$DEPDIR"'/g'`; do # Make sure the directory exists. test -f "$dirpart/$file" && continue fdir=`$as_dirname -- "$file" || $as_expr X"$file" : 'X\(.*[^/]\)//*[^/][^/]*/*$' \| \ X"$file" : 'X\(//\)[^/]' \| \ X"$file" : 'X\(//\)$' \| \ X"$file" : 'X\(/\)' \| . 2>/dev/null || $as_echo X"$file" | sed '/^X\(.*[^/]\)\/\/*[^/][^/]*\/*$/{ s//\1/ q } /^X\(\/\/\)[^/].*/{ s//\1/ q } /^X\(\/\/\)$/{ s//\1/ q } /^X\(\/\).*/{ s//\1/ q } s/.*/./; q'` as_dir=$dirpart/$fdir; as_fn_mkdir_p # echo "creating $dirpart/$file" echo '# dummy' > "$dirpart/$file" done done } ;; esac done # for ac_tag as_fn_exit 0 _ACEOF ac_clean_files=$ac_clean_files_save test $ac_write_fail = 0 || as_fn_error $? "write failure creating $CONFIG_STATUS" "$LINENO" 5 # configure is writing to config.log, and then calls config.status. # config.status does its own redirection, appending to config.log. # Unfortunately, on DOS this fails, as config.log is still kept open # by configure, so config.status won't be able to write to it; its # output is simply discarded. So we exec the FD to /dev/null, # effectively closing config.log, so it can be properly (re)opened and # appended to by config.status. When coming back to configure, we # need to make the FD available again. if test "$no_create" != yes; then ac_cs_success=: ac_config_status_args= test "$silent" = yes && ac_config_status_args="$ac_config_status_args --quiet" exec 5>/dev/null $SHELL $CONFIG_STATUS $ac_config_status_args || ac_cs_success=false exec 5>>config.log # Use ||, not &&, to avoid exiting from the if with $? = 1, which # would make configure fail if this is the last instruction. $ac_cs_success || as_fn_exit 1 fi if test -n "$ac_unrecognized_opts" && test "$enable_option_checking" != no; then { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: unrecognized options: $ac_unrecognized_opts" >&5 $as_echo "$as_me: WARNING: unrecognized options: $ac_unrecognized_opts" >&2;} fi stacks-1.44/configure.ac000644 000765 000024 00000004131 13001003376 015760 0ustar00catchenstaff000000 000000 # -*- Autoconf -*- # Process this file with autoconf to produce a configure script. AC_PREREQ(2.59) AC_INIT([Stacks], [1.44]) AC_CONFIG_AUX_DIR([config]) AM_INIT_AUTOMAKE([-Wall -Werror foreign parallel-tests subdir-objects]) AC_CONFIG_SRCDIR([src/ustacks.cc]) AC_CONFIG_HEADERS([config.h]) m4_pattern_allow([AC_OPENMP]) # # Get BAM, aka HTSLib library and include locations if requested # AC_ARG_ENABLE([bam], AS_HELP_STRING([--disable-bam], [Disable use of BAM files through HTSLib.])) AS_IF([test "x$enable_bam" != "xno"], [ AC_DEFINE([HAVE_BAM], [1], [Enable compilation with Samtools BAM library]) BAM_CFLAGS='-I./htslib/htslib' AC_SUBST([BAM_CFLAGS]) BAM_LIBS='./htslib/libhts.a' AC_SUBST([BAM_LIBS]) ]) # # Enable use of Google Sparsehash and get include location if requested. # AC_ARG_ENABLE([sparsehash], AS_HELP_STRING([--enable-sparsehash], [Enable the use of Google Sparsehash (must be installed).])) AS_IF([test "x$enable_sparsehash" = "xyes"], [ AC_DEFINE([HAVE_SPARSEHASH], [1], [Enable compilation with Google Sparsehash]) AC_ARG_WITH([sparsehash-include-path], [AS_HELP_STRING([--with-sparsehash-include-path], [location of Google Sparsehash headers])], [SPARSEHASH_CFLAGS="-I$withval"], [SPARSEHASH_CFLAGS=""]) AC_SUBST([SPARSEHASH_CFLAGS]) ]) # Checks for programs. AC_PROG_CXX AM_PROG_CC_C_O AX_CXX_COMPILE_STDCXX(11,, [mandatory]) # Checks for libraries. AC_CHECK_LIB([gomp], [omp_set_num_threads],, [AC_MSG_WARN([Unable to locate OpenMP library, you should probably specify '--disable-openmp'.])]) AC_CHECK_LIB([z], [gzread],, [AC_MSG_ERROR([Zlib not found, reading gzipped files will not be possible.])]) # Checks for header files. AC_HEADER_DIRENT AC_HEADER_STDC AC_CHECK_HEADERS([float.h limits.h stdlib.h string.h]) # Check for OpenMP parallel execution support AC_OPENMP # Checks for typedefs, structures, and compiler characteristics. AC_HEADER_STDBOOL AC_C_CONST AC_TYPE_SIZE_T # Checks for library functions. AC_FUNC_CLOSEDIR_VOID AC_FUNC_MALLOC AC_FUNC_REALLOC AC_CHECK_FUNCS([floor memset pow sqrt]) # For test harness AC_PROG_AWK AC_CONFIG_FILES([Makefile]) AC_OUTPUT stacks-1.44/htslib/000755 000765 000024 00000000000 13001012452 014753 5ustar00catchenstaff000000 000000 stacks-1.44/INSTALL000644 000765 000024 00000022432 12335173442 014542 0ustar00catchenstaff000000 000000 Installation Instructions ************************* Copyright (C) 1994, 1995, 1996, 1999, 2000, 2001, 2002, 2004, 2005 Free Software Foundation, Inc. This file is free documentation; the Free Software Foundation gives unlimited permission to copy, distribute and modify it. Basic Installation ================== These are generic installation instructions. The `configure' shell script attempts to guess correct values for various system-dependent variables used during compilation. It uses those values to create a `Makefile' in each directory of the package. It may also create one or more `.h' files containing system-dependent definitions. Finally, it creates a shell script `config.status' that you can run in the future to recreate the current configuration, and a file `config.log' containing compiler output (useful mainly for debugging `configure'). It can also use an optional file (typically called `config.cache' and enabled with `--cache-file=config.cache' or simply `-C') that saves the results of its tests to speed up reconfiguring. (Caching is disabled by default to prevent problems with accidental use of stale cache files.) If you need to do unusual things to compile the package, please try to figure out how `configure' could check whether to do them, and mail diffs or instructions to the address given in the `README' so they can be considered for the next release. If you are using the cache, and at some point `config.cache' contains results you don't want to keep, you may remove or edit it. The file `configure.ac' (or `configure.in') is used to create `configure' by a program called `autoconf'. You only need `configure.ac' if you want to change it or regenerate `configure' using a newer version of `autoconf'. The simplest way to compile this package is: 1. `cd' to the directory containing the package's source code and type `./configure' to configure the package for your system. If you're using `csh' on an old version of System V, you might need to type `sh ./configure' instead to prevent `csh' from trying to execute `configure' itself. Running `configure' takes awhile. While running, it prints some messages telling which features it is checking for. 2. Type `make' to compile the package. 3. Optionally, type `make check' to run any self-tests that come with the package. 4. Type `make install' to install the programs and any data files and documentation. 5. You can remove the program binaries and object files from the source code directory by typing `make clean'. To also remove the files that `configure' created (so you can compile the package for a different kind of computer), type `make distclean'. There is also a `make maintainer-clean' target, but that is intended mainly for the package's developers. If you use it, you may have to get all sorts of other programs in order to regenerate files that came with the distribution. Compilers and Options ===================== Some systems require unusual options for compilation or linking that the `configure' script does not know about. Run `./configure --help' for details on some of the pertinent environment variables. You can give `configure' initial values for configuration parameters by setting variables in the command line or in the environment. Here is an example: ./configure CC=c89 CFLAGS=-O2 LIBS=-lposix *Note Defining Variables::, for more details. Compiling For Multiple Architectures ==================================== You can compile the package for more than one kind of computer at the same time, by placing the object files for each architecture in their own directory. To do this, you must use a version of `make' that supports the `VPATH' variable, such as GNU `make'. `cd' to the directory where you want the object files and executables to go and run the `configure' script. `configure' automatically checks for the source code in the directory that `configure' is in and in `..'. If you have to use a `make' that does not support the `VPATH' variable, you have to compile the package for one architecture at a time in the source code directory. After you have installed the package for one architecture, use `make distclean' before reconfiguring for another architecture. Installation Names ================== By default, `make install' installs the package's commands under `/usr/local/bin', include files under `/usr/local/include', etc. You can specify an installation prefix other than `/usr/local' by giving `configure' the option `--prefix=PREFIX'. You can specify separate installation prefixes for architecture-specific files and architecture-independent files. If you pass the option `--exec-prefix=PREFIX' to `configure', the package uses PREFIX as the prefix for installing programs and libraries. Documentation and other data files still use the regular prefix. In addition, if you use an unusual directory layout you can give options like `--bindir=DIR' to specify different values for particular kinds of files. Run `configure --help' for a list of the directories you can set and what kinds of files go in them. If the package supports it, you can cause programs to be installed with an extra prefix or suffix on their names by giving `configure' the option `--program-prefix=PREFIX' or `--program-suffix=SUFFIX'. Optional Features ================= Some packages pay attention to `--enable-FEATURE' options to `configure', where FEATURE indicates an optional part of the package. They may also pay attention to `--with-PACKAGE' options, where PACKAGE is something like `gnu-as' or `x' (for the X Window System). The `README' should mention any `--enable-' and `--with-' options that the package recognizes. For packages that use the X Window System, `configure' can usually find the X include and library files automatically, but if it doesn't, you can use the `configure' options `--x-includes=DIR' and `--x-libraries=DIR' to specify their locations. Specifying the System Type ========================== There may be some features `configure' cannot figure out automatically, but needs to determine by the type of machine the package will run on. Usually, assuming the package is built to be run on the _same_ architectures, `configure' can figure that out, but if it prints a message saying it cannot guess the machine type, give it the `--build=TYPE' option. TYPE can either be a short name for the system type, such as `sun4', or a canonical name which has the form: CPU-COMPANY-SYSTEM where SYSTEM can have one of these forms: OS KERNEL-OS See the file `config.sub' for the possible values of each field. If `config.sub' isn't included in this package, then this package doesn't need to know the machine type. If you are _building_ compiler tools for cross-compiling, you should use the option `--target=TYPE' to select the type of system they will produce code for. If you want to _use_ a cross compiler, that generates code for a platform different from the build platform, you should specify the "host" platform (i.e., that on which the generated programs will eventually be run) with `--host=TYPE'. Sharing Defaults ================ If you want to set default values for `configure' scripts to share, you can create a site shell script called `config.site' that gives default values for variables like `CC', `cache_file', and `prefix'. `configure' looks for `PREFIX/share/config.site' if it exists, then `PREFIX/etc/config.site' if it exists. Or, you can set the `CONFIG_SITE' environment variable to the location of the site script. A warning: not all `configure' scripts look for a site script. Defining Variables ================== Variables not defined in a site shell script can be set in the environment passed to `configure'. However, some packages may run configure again during the build, and the customized values of these variables may be lost. In order to avoid this problem, you should set them in the `configure' command line, using `VAR=value'. For example: ./configure CC=/usr/local2/bin/gcc causes the specified `gcc' to be used as the C compiler (unless it is overridden in the site shell script). Here is a another example: /bin/bash ./configure CONFIG_SHELL=/bin/bash Here the `CONFIG_SHELL=/bin/bash' operand causes subsequent configuration-related scripts to be executed by `/bin/bash'. `configure' Invocation ====================== `configure' recognizes the following options to control how it operates. `--help' `-h' Print a summary of the options to `configure', and exit. `--version' `-V' Print the version of Autoconf used to generate the `configure' script, and exit. `--cache-file=FILE' Enable the cache: use and save the results of the tests in FILE, traditionally `config.cache'. FILE defaults to `/dev/null' to disable caching. `--config-cache' `-C' Alias for `--cache-file=config.cache'. `--quiet' `--silent' `-q' Do not print messages saying which checks are being made. To suppress all normal output, redirect it to `/dev/null' (any error messages will still be shown). `--srcdir=DIR' Look for the package's source code in directory DIR. Usually `configure' can determine that directory automatically. `configure' also accepts some other, not widely useful, options. Run `configure --help' for more details. stacks-1.44/LICENSE000644 000765 000024 00000104513 12335173442 014517 0ustar00catchenstaff000000 000000 GNU GENERAL PUBLIC LICENSE Version 3, 29 June 2007 Copyright (C) 2007 Free Software Foundation, Inc. Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed. Preamble The GNU General Public License is a free, copyleft license for software and other kinds of works. The licenses for most software and other practical works are designed to take away your freedom to share and change the works. By contrast, the GNU General Public License is intended to guarantee your freedom to share and change all versions of a program--to make sure it remains free software for all its users. We, the Free Software Foundation, use the GNU General Public License for most of our software; it applies also to any other work released this way by its authors. You can apply it to your programs, too. When we speak of free software, we are referring to freedom, not price. Our General Public Licenses are designed to make sure that you have the freedom to distribute copies of free software (and charge for them if you wish), that you receive source code or can get it if you want it, that you can change the software or use pieces of it in new free programs, and that you know you can do these things. To protect your rights, we need to prevent others from denying you these rights or asking you to surrender the rights. Therefore, you have certain responsibilities if you distribute copies of the software, or if you modify it: responsibilities to respect the freedom of others. For example, if you distribute copies of such a program, whether gratis or for a fee, you must pass on to the recipients the same freedoms that you received. You must make sure that they, too, receive or can get the source code. And you must show them these terms so they know their rights. Developers that use the GNU GPL protect your rights with two steps: (1) assert copyright on the software, and (2) offer you this License giving you legal permission to copy, distribute and/or modify it. For the developers' and authors' protection, the GPL clearly explains that there is no warranty for this free software. For both users' and authors' sake, the GPL requires that modified versions be marked as changed, so that their problems will not be attributed erroneously to authors of previous versions. Some devices are designed to deny users access to install or run modified versions of the software inside them, although the manufacturer can do so. This is fundamentally incompatible with the aim of protecting users' freedom to change the software. The systematic pattern of such abuse occurs in the area of products for individuals to use, which is precisely where it is most unacceptable. Therefore, we have designed this version of the GPL to prohibit the practice for those products. If such problems arise substantially in other domains, we stand ready to extend this provision to those domains in future versions of the GPL, as needed to protect the freedom of users. Finally, every program is threatened constantly by software patents. States should not allow patents to restrict development and use of software on general-purpose computers, but in those that do, we wish to avoid the special danger that patents applied to a free program could make it effectively proprietary. To prevent this, the GPL assures that patents cannot be used to render the program non-free. The precise terms and conditions for copying, distribution and modification follow. TERMS AND CONDITIONS 0. Definitions. "This License" refers to version 3 of the GNU General Public License. "Copyright" also means copyright-like laws that apply to other kinds of works, such as semiconductor masks. "The Program" refers to any copyrightable work licensed under this License. Each licensee is addressed as "you". "Licensees" and "recipients" may be individuals or organizations. To "modify" a work means to copy from or adapt all or part of the work in a fashion requiring copyright permission, other than the making of an exact copy. The resulting work is called a "modified version" of the earlier work or a work "based on" the earlier work. A "covered work" means either the unmodified Program or a work based on the Program. To "propagate" a work means to do anything with it that, without permission, would make you directly or secondarily liable for infringement under applicable copyright law, except executing it on a computer or modifying a private copy. Propagation includes copying, distribution (with or without modification), making available to the public, and in some countries other activities as well. To "convey" a work means any kind of propagation that enables other parties to make or receive copies. Mere interaction with a user through a computer network, with no transfer of a copy, is not conveying. An interactive user interface displays "Appropriate Legal Notices" to the extent that it includes a convenient and prominently visible feature that (1) displays an appropriate copyright notice, and (2) tells the user that there is no warranty for the work (except to the extent that warranties are provided), that licensees may convey the work under this License, and how to view a copy of this License. If the interface presents a list of user commands or options, such as a menu, a prominent item in the list meets this criterion. 1. Source Code. The "source code" for a work means the preferred form of the work for making modifications to it. "Object code" means any non-source form of a work. A "Standard Interface" means an interface that either is an official standard defined by a recognized standards body, or, in the case of interfaces specified for a particular programming language, one that is widely used among developers working in that language. The "System Libraries" of an executable work include anything, other than the work as a whole, that (a) is included in the normal form of packaging a Major Component, but which is not part of that Major Component, and (b) serves only to enable use of the work with that Major Component, or to implement a Standard Interface for which an implementation is available to the public in source code form. A "Major Component", in this context, means a major essential component (kernel, window system, and so on) of the specific operating system (if any) on which the executable work runs, or a compiler used to produce the work, or an object code interpreter used to run it. The "Corresponding Source" for a work in object code form means all the source code needed to generate, install, and (for an executable work) run the object code and to modify the work, including scripts to control those activities. However, it does not include the work's System Libraries, or general-purpose tools or generally available free programs which are used unmodified in performing those activities but which are not part of the work. For example, Corresponding Source includes interface definition files associated with source files for the work, and the source code for shared libraries and dynamically linked subprograms that the work is specifically designed to require, such as by intimate data communication or control flow between those subprograms and other parts of the work. The Corresponding Source need not include anything that users can regenerate automatically from other parts of the Corresponding Source. The Corresponding Source for a work in source code form is that same work. 2. Basic Permissions. All rights granted under this License are granted for the term of copyright on the Program, and are irrevocable provided the stated conditions are met. This License explicitly affirms your unlimited permission to run the unmodified Program. The output from running a covered work is covered by this License only if the output, given its content, constitutes a covered work. This License acknowledges your rights of fair use or other equivalent, as provided by copyright law. You may make, run and propagate covered works that you do not convey, without conditions so long as your license otherwise remains in force. You may convey covered works to others for the sole purpose of having them make modifications exclusively for you, or provide you with facilities for running those works, provided that you comply with the terms of this License in conveying all material for which you do not control copyright. Those thus making or running the covered works for you must do so exclusively on your behalf, under your direction and control, on terms that prohibit them from making any copies of your copyrighted material outside their relationship with you. Conveying under any other circumstances is permitted solely under the conditions stated below. Sublicensing is not allowed; section 10 makes it unnecessary. 3. Protecting Users' Legal Rights From Anti-Circumvention Law. No covered work shall be deemed part of an effective technological measure under any applicable law fulfilling obligations under article 11 of the WIPO copyright treaty adopted on 20 December 1996, or similar laws prohibiting or restricting circumvention of such measures. When you convey a covered work, you waive any legal power to forbid circumvention of technological measures to the extent such circumvention is effected by exercising rights under this License with respect to the covered work, and you disclaim any intention to limit operation or modification of the work as a means of enforcing, against the work's users, your or third parties' legal rights to forbid circumvention of technological measures. 4. Conveying Verbatim Copies. You may convey verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice; keep intact all notices stating that this License and any non-permissive terms added in accord with section 7 apply to the code; keep intact all notices of the absence of any warranty; and give all recipients a copy of this License along with the Program. You may charge any price or no price for each copy that you convey, and you may offer support or warranty protection for a fee. 5. Conveying Modified Source Versions. You may convey a work based on the Program, or the modifications to produce it from the Program, in the form of source code under the terms of section 4, provided that you also meet all of these conditions: a) The work must carry prominent notices stating that you modified it, and giving a relevant date. b) The work must carry prominent notices stating that it is released under this License and any conditions added under section 7. This requirement modifies the requirement in section 4 to "keep intact all notices". c) You must license the entire work, as a whole, under this License to anyone who comes into possession of a copy. This License will therefore apply, along with any applicable section 7 additional terms, to the whole of the work, and all its parts, regardless of how they are packaged. This License gives no permission to license the work in any other way, but it does not invalidate such permission if you have separately received it. d) If the work has interactive user interfaces, each must display Appropriate Legal Notices; however, if the Program has interactive interfaces that do not display Appropriate Legal Notices, your work need not make them do so. A compilation of a covered work with other separate and independent works, which are not by their nature extensions of the covered work, and which are not combined with it such as to form a larger program, in or on a volume of a storage or distribution medium, is called an "aggregate" if the compilation and its resulting copyright are not used to limit the access or legal rights of the compilation's users beyond what the individual works permit. Inclusion of a covered work in an aggregate does not cause this License to apply to the other parts of the aggregate. 6. Conveying Non-Source Forms. You may convey a covered work in object code form under the terms of sections 4 and 5, provided that you also convey the machine-readable Corresponding Source under the terms of this License, in one of these ways: a) Convey the object code in, or embodied in, a physical product (including a physical distribution medium), accompanied by the Corresponding Source fixed on a durable physical medium customarily used for software interchange. b) Convey the object code in, or embodied in, a physical product (including a physical distribution medium), accompanied by a written offer, valid for at least three years and valid for as long as you offer spare parts or customer support for that product model, to give anyone who possesses the object code either (1) a copy of the Corresponding Source for all the software in the product that is covered by this License, on a durable physical medium customarily used for software interchange, for a price no more than your reasonable cost of physically performing this conveying of source, or (2) access to copy the Corresponding Source from a network server at no charge. c) Convey individual copies of the object code with a copy of the written offer to provide the Corresponding Source. This alternative is allowed only occasionally and noncommercially, and only if you received the object code with such an offer, in accord with subsection 6b. d) Convey the object code by offering access from a designated place (gratis or for a charge), and offer equivalent access to the Corresponding Source in the same way through the same place at no further charge. You need not require recipients to copy the Corresponding Source along with the object code. If the place to copy the object code is a network server, the Corresponding Source may be on a different server (operated by you or a third party) that supports equivalent copying facilities, provided you maintain clear directions next to the object code saying where to find the Corresponding Source. Regardless of what server hosts the Corresponding Source, you remain obligated to ensure that it is available for as long as needed to satisfy these requirements. e) Convey the object code using peer-to-peer transmission, provided you inform other peers where the object code and Corresponding Source of the work are being offered to the general public at no charge under subsection 6d. A separable portion of the object code, whose source code is excluded from the Corresponding Source as a System Library, need not be included in conveying the object code work. A "User Product" is either (1) a "consumer product", which means any tangible personal property which is normally used for personal, family, or household purposes, or (2) anything designed or sold for incorporation into a dwelling. In determining whether a product is a consumer product, doubtful cases shall be resolved in favor of coverage. For a particular product received by a particular user, "normally used" refers to a typical or common use of that class of product, regardless of the status of the particular user or of the way in which the particular user actually uses, or expects or is expected to use, the product. A product is a consumer product regardless of whether the product has substantial commercial, industrial or non-consumer uses, unless such uses represent the only significant mode of use of the product. "Installation Information" for a User Product means any methods, procedures, authorization keys, or other information required to install and execute modified versions of a covered work in that User Product from a modified version of its Corresponding Source. The information must suffice to ensure that the continued functioning of the modified object code is in no case prevented or interfered with solely because modification has been made. If you convey an object code work under this section in, or with, or specifically for use in, a User Product, and the conveying occurs as part of a transaction in which the right of possession and use of the User Product is transferred to the recipient in perpetuity or for a fixed term (regardless of how the transaction is characterized), the Corresponding Source conveyed under this section must be accompanied by the Installation Information. But this requirement does not apply if neither you nor any third party retains the ability to install modified object code on the User Product (for example, the work has been installed in ROM). The requirement to provide Installation Information does not include a requirement to continue to provide support service, warranty, or updates for a work that has been modified or installed by the recipient, or for the User Product in which it has been modified or installed. Access to a network may be denied when the modification itself materially and adversely affects the operation of the network or violates the rules and protocols for communication across the network. Corresponding Source conveyed, and Installation Information provided, in accord with this section must be in a format that is publicly documented (and with an implementation available to the public in source code form), and must require no special password or key for unpacking, reading or copying. 7. Additional Terms. "Additional permissions" are terms that supplement the terms of this License by making exceptions from one or more of its conditions. Additional permissions that are applicable to the entire Program shall be treated as though they were included in this License, to the extent that they are valid under applicable law. If additional permissions apply only to part of the Program, that part may be used separately under those permissions, but the entire Program remains governed by this License without regard to the additional permissions. When you convey a copy of a covered work, you may at your option remove any additional permissions from that copy, or from any part of it. (Additional permissions may be written to require their own removal in certain cases when you modify the work.) You may place additional permissions on material, added by you to a covered work, for which you have or can give appropriate copyright permission. Notwithstanding any other provision of this License, for material you add to a covered work, you may (if authorized by the copyright holders of that material) supplement the terms of this License with terms: a) Disclaiming warranty or limiting liability differently from the terms of sections 15 and 16 of this License; or b) Requiring preservation of specified reasonable legal notices or author attributions in that material or in the Appropriate Legal Notices displayed by works containing it; or c) Prohibiting misrepresentation of the origin of that material, or requiring that modified versions of such material be marked in reasonable ways as different from the original version; or d) Limiting the use for publicity purposes of names of licensors or authors of the material; or e) Declining to grant rights under trademark law for use of some trade names, trademarks, or service marks; or f) Requiring indemnification of licensors and authors of that material by anyone who conveys the material (or modified versions of it) with contractual assumptions of liability to the recipient, for any liability that these contractual assumptions directly impose on those licensors and authors. All other non-permissive additional terms are considered "further restrictions" within the meaning of section 10. If the Program as you received it, or any part of it, contains a notice stating that it is governed by this License along with a term that is a further restriction, you may remove that term. If a license document contains a further restriction but permits relicensing or conveying under this License, you may add to a covered work material governed by the terms of that license document, provided that the further restriction does not survive such relicensing or conveying. If you add terms to a covered work in accord with this section, you must place, in the relevant source files, a statement of the additional terms that apply to those files, or a notice indicating where to find the applicable terms. Additional terms, permissive or non-permissive, may be stated in the form of a separately written license, or stated as exceptions; the above requirements apply either way. 8. Termination. You may not propagate or modify a covered work except as expressly provided under this License. Any attempt otherwise to propagate or modify it is void, and will automatically terminate your rights under this License (including any patent licenses granted under the third paragraph of section 11). However, if you cease all violation of this License, then your license from a particular copyright holder is reinstated (a) provisionally, unless and until the copyright holder explicitly and finally terminates your license, and (b) permanently, if the copyright holder fails to notify you of the violation by some reasonable means prior to 60 days after the cessation. Moreover, your license from a particular copyright holder is reinstated permanently if the copyright holder notifies you of the violation by some reasonable means, this is the first time you have received notice of violation of this License (for any work) from that copyright holder, and you cure the violation prior to 30 days after your receipt of the notice. Termination of your rights under this section does not terminate the licenses of parties who have received copies or rights from you under this License. If your rights have been terminated and not permanently reinstated, you do not qualify to receive new licenses for the same material under section 10. 9. Acceptance Not Required for Having Copies. You are not required to accept this License in order to receive or run a copy of the Program. Ancillary propagation of a covered work occurring solely as a consequence of using peer-to-peer transmission to receive a copy likewise does not require acceptance. However, nothing other than this License grants you permission to propagate or modify any covered work. These actions infringe copyright if you do not accept this License. Therefore, by modifying or propagating a covered work, you indicate your acceptance of this License to do so. 10. Automatic Licensing of Downstream Recipients. Each time you convey a covered work, the recipient automatically receives a license from the original licensors, to run, modify and propagate that work, subject to this License. You are not responsible for enforcing compliance by third parties with this License. An "entity transaction" is a transaction transferring control of an organization, or substantially all assets of one, or subdividing an organization, or merging organizations. If propagation of a covered work results from an entity transaction, each party to that transaction who receives a copy of the work also receives whatever licenses to the work the party's predecessor in interest had or could give under the previous paragraph, plus a right to possession of the Corresponding Source of the work from the predecessor in interest, if the predecessor has it or can get it with reasonable efforts. You may not impose any further restrictions on the exercise of the rights granted or affirmed under this License. For example, you may not impose a license fee, royalty, or other charge for exercise of rights granted under this License, and you may not initiate litigation (including a cross-claim or counterclaim in a lawsuit) alleging that any patent claim is infringed by making, using, selling, offering for sale, or importing the Program or any portion of it. 11. Patents. A "contributor" is a copyright holder who authorizes use under this License of the Program or a work on which the Program is based. The work thus licensed is called the contributor's "contributor version". A contributor's "essential patent claims" are all patent claims owned or controlled by the contributor, whether already acquired or hereafter acquired, that would be infringed by some manner, permitted by this License, of making, using, or selling its contributor version, but do not include claims that would be infringed only as a consequence of further modification of the contributor version. For purposes of this definition, "control" includes the right to grant patent sublicenses in a manner consistent with the requirements of this License. Each contributor grants you a non-exclusive, worldwide, royalty-free patent license under the contributor's essential patent claims, to make, use, sell, offer for sale, import and otherwise run, modify and propagate the contents of its contributor version. In the following three paragraphs, a "patent license" is any express agreement or commitment, however denominated, not to enforce a patent (such as an express permission to practice a patent or covenant not to sue for patent infringement). To "grant" such a patent license to a party means to make such an agreement or commitment not to enforce a patent against the party. If you convey a covered work, knowingly relying on a patent license, and the Corresponding Source of the work is not available for anyone to copy, free of charge and under the terms of this License, through a publicly available network server or other readily accessible means, then you must either (1) cause the Corresponding Source to be so available, or (2) arrange to deprive yourself of the benefit of the patent license for this particular work, or (3) arrange, in a manner consistent with the requirements of this License, to extend the patent license to downstream recipients. "Knowingly relying" means you have actual knowledge that, but for the patent license, your conveying the covered work in a country, or your recipient's use of the covered work in a country, would infringe one or more identifiable patents in that country that you have reason to believe are valid. If, pursuant to or in connection with a single transaction or arrangement, you convey, or propagate by procuring conveyance of, a covered work, and grant a patent license to some of the parties receiving the covered work authorizing them to use, propagate, modify or convey a specific copy of the covered work, then the patent license you grant is automatically extended to all recipients of the covered work and works based on it. A patent license is "discriminatory" if it does not include within the scope of its coverage, prohibits the exercise of, or is conditioned on the non-exercise of one or more of the rights that are specifically granted under this License. You may not convey a covered work if you are a party to an arrangement with a third party that is in the business of distributing software, under which you make payment to the third party based on the extent of your activity of conveying the work, and under which the third party grants, to any of the parties who would receive the covered work from you, a discriminatory patent license (a) in connection with copies of the covered work conveyed by you (or copies made from those copies), or (b) primarily for and in connection with specific products or compilations that contain the covered work, unless you entered into that arrangement, or that patent license was granted, prior to 28 March 2007. Nothing in this License shall be construed as excluding or limiting any implied license or other defenses to infringement that may otherwise be available to you under applicable patent law. 12. No Surrender of Others' Freedom. If conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot convey a covered work so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not convey it at all. For example, if you agree to terms that obligate you to collect a royalty for further conveying from those to whom you convey the Program, the only way you could satisfy both those terms and this License would be to refrain entirely from conveying the Program. 13. Use with the GNU Affero General Public License. Notwithstanding any other provision of this License, you have permission to link or combine any covered work with a work licensed under version 3 of the GNU Affero General Public License into a single combined work, and to convey the resulting work. The terms of this License will continue to apply to the part which is the covered work, but the special requirements of the GNU Affero General Public License, section 13, concerning interaction through a network will apply to the combination as such. 14. Revised Versions of this License. The Free Software Foundation may publish revised and/or new versions of the GNU General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns. Each version is given a distinguishing version number. If the Program specifies that a certain numbered version of the GNU General Public License "or any later version" applies to it, you have the option of following the terms and conditions either of that numbered version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of the GNU General Public License, you may choose any version ever published by the Free Software Foundation. If the Program specifies that a proxy can decide which future versions of the GNU General Public License can be used, that proxy's public statement of acceptance of a version permanently authorizes you to choose that version for the Program. Later license versions may give you additional or different permissions. However, no additional obligations are imposed on any author or copyright holder as a result of your choosing to follow a later version. 15. Disclaimer of Warranty. THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION. 16. Limitation of Liability. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. 17. Interpretation of Sections 15 and 16. If the disclaimer of warranty and limitation of liability provided above cannot be given local legal effect according to their terms, reviewing courts shall apply local law that most closely approximates an absolute waiver of all civil liability in connection with the Program, unless a warranty or assumption of liability accompanies a copy of the Program in return for a fee. END OF TERMS AND CONDITIONS How to Apply These Terms to Your New Programs If you develop a new program, and you want it to be of the greatest possible use to the public, the best way to achieve this is to make it free software which everyone can redistribute and change under these terms. To do so, attach the following notices to the program. It is safest to attach them to the start of each source file to most effectively state the exclusion of warranty; and each file should have at least the "copyright" line and a pointer to where the full notice is found. Copyright (C) This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program. If not, see . Also add information on how to contact you by electronic and paper mail. If the program does terminal interaction, make it output a short notice like this when it starts in an interactive mode: Copyright (C) This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'. This is free software, and you are welcome to redistribute it under certain conditions; type `show c' for details. The hypothetical commands `show w' and `show c' should show the appropriate parts of the General Public License. Of course, your program's commands might be different; for a GUI interface, you would use an "about box". You should also get your employer (if you work as a programmer) or school, if any, to sign a "copyright disclaimer" for the program, if necessary. For more information on this, and how to apply and follow the GNU GPL, see . The GNU General Public License does not permit incorporating your program into proprietary programs. If your program is a subroutine library, you may consider it more useful to permit linking proprietary applications with the library. If this is what you want to do, use the GNU Lesser General Public License instead of this License. But first, please read . stacks-1.44/Makefile.am000644 000765 000024 00000027174 12750756352 015565 0ustar00catchenstaff000000 000000 bin_PROGRAMS = ustacks pstacks estacks cstacks sstacks rxstacks hstacks process_radtags process_shortreads \ kmer_filter clone_filter genotypes populations phasedstacks ustacks_SOURCES = src/ustacks.h src/ustacks.cc src/stacks.h src/stacks.cc src/constants.h \ src/mstack.h src/mstack.cc src/mst.h src/mst.cc src/cmb.h src/cmb.cc \ src/DNASeq.h src/DNASeq.cc src/DNANSeq.h src/DNANSeq.cc \ src/models.h src/models.cc src/utils.h src/utils.cc \ src/kmers.h src/kmers.cc src/GappedAln.h src/aln_utils.h src/aln_utils.cc \ src/input.h src/input.cc src/Tsv.h src/BowtieI.h src/FastaI.h src/FastqI.h src/SamI.h \ src/gzFasta.h src/gzFastq.h pstacks_SOURCES = src/pstacks.h src/pstacks.cc src/stacks.h src/stacks.cc src/constants.h \ src/mstack.h src/mstack.cc \ src/DNANSeq.h src/DNANSeq.cc src/DNASeq.h src/DNASeq.cc \ src/models.h src/models.cc src/utils.h src/utils.cc \ src/input.h src/input.cc src/Tsv.h src/BowtieI.h src/FastaI.h src/FastqI.h src/SamI.h \ src/BamI.h estacks_SOURCES = src/estacks.h src/estacks.cc src/stacks.h src/stacks.cc src/constants.h \ src/mstack.h src/mstack.cc \ src/models.h src/models.cc src/utils.h src/utils.cc \ src/DNASeq.h src/DNASeq.cc src/DNANSeq.h src/DNANSeq.cc \ src/input.h src/input.cc src/Tsv.h src/BowtieI.h src/FastaI.h src/FastqI.h src/SamI.h cstacks_SOURCES = src/cstacks.h src/cstacks.cc src/constants.h \ src/stacks.h src/stacks.cc src/locus.h src/locus.cc \ src/kmers.h src/kmers.cc src/utils.h src/utils.cc \ src/GappedAln.h src/aln_utils.h src/aln_utils.cc \ src/DNASeq.h src/DNASeq.cc src/DNANSeq.h src/DNANSeq.cc \ src/input.h src/input.cc src/sql_utilities.h src/sql_utilities.cc hstacks_SOURCES = src/hstacks.h src/hstacks.cc src/constants.h \ src/stacks.h src/stacks.cc src/locus.h src/locus.cc \ src/kmers.h src/kmers.cc \ src/DNASeq.h src/DNASeq.cc src/DNANSeq.h src/DNANSeq.cc \ src/utils.h src/utils.cc \ src/input.h src/input.cc src/sql_utilities.h src/sql_utilities.cc sstacks_SOURCES = src/sstacks.h src/sstacks.cc src/constants.h \ src/stacks.h src/stacks.cc src/locus.h src/locus.cc \ src/kmers.h src/kmers.cc \ src/GappedAln.h src/aln_utils.h src/aln_utils.cc \ src/DNASeq.h src/DNASeq.cc src/DNANSeq.h src/DNANSeq.cc src/utils.h src/utils.cc \ src/input.h src/input.cc src/sql_utilities.h src/sql_utilities.cc rxstacks_SOURCES = src/rxstacks.h src/rxstacks.cc src/constants.h \ src/stacks.h src/stacks.cc src/locus.h src/locus.cc \ src/DNANSeq.h src/DNANSeq.cc src/DNASeq.h src/DNASeq.cc \ src/mst.h src/mst.cc \ src/models.h src/models.cc \ src/aln_utils.h src/aln_utils.cc src/utils.h src/utils.cc \ src/input.h src/input.cc src/sql_utilities.h src/sql_utilities.cc \ src/MetaPopInfo.h src/MetaPopInfo.cc process_radtags_SOURCES = src/process_radtags.h src/process_radtags.cc src/constants.h \ src/utils.h src/utils.cc src/log_utils.h src/log_utils.cc \ src/write.h src/write.cc \ src/clean.h src/clean.cc \ src/file_io.h src/file_io.cc \ src/input.h src/input.cc src/BustardI.h src/BamUnalignedI.h src/FastqI.h src/gzFastq.h \ src/renz.h src/renz.cc process_shortreads_SOURCES = src/process_shortreads.h src/process_shortreads.cc src/constants.h \ src/clean.h src/clean.cc \ src/file_io.h src/file_io.cc \ src/utils.h src/utils.cc src/log_utils.h src/log_utils.cc \ src/write.h src/write.cc \ src/kmers.h src/kmers.cc \ src/input.h src/input.cc src/BustardI.h src/BamUnalignedI.h src/FastqI.h src/gzFastq.h kmer_filter_SOURCES = src/kmer_filter.h src/kmer_filter.cc src/constants.h \ src/utils.h src/utils.cc \ src/write.h src/write.cc \ src/kmers.h src/kmers.cc \ src/input.h src/input.cc src/BustardI.h src/FastqI.h src/FastaI.h \ src/gzFastq.h src/gzFasta.h clone_filter_SOURCES = src/clone_filter.h src/clone_filter.cc src/constants.h \ src/kmers.h src/kmers.cc \ src/utils.h src/utils.cc \ src/write.h src/write.cc \ src/clean.h src/clean.cc src/file_io.h src/file_io.cc \ src/input.h src/input.cc src/BustardI.h src/FastqI.h src/FastaI.h \ src/gzFastq.h src/gzFasta.h genotypes_SOURCES = src/genotypes.h src/genotypes.cc src/constants.h \ src/utils.h src/utils.cc src/catalog_utils.h src/catalog_utils.cc \ src/log_utils.h src/log_utils.cc \ src/stacks.h src/stacks.cc src/locus.h src/locus.cc \ src/DNASeq.h src/DNASeq.cc src/DNANSeq.h src/DNANSeq.cc \ src/aln_utils.h src/aln_utils.cc \ src/PopMap.h src/genotype_dictionaries.h src/genotype_dictionaries.cc\ src/input.h src/input.cc src/sql_utilities.h src/sql_utilities.cc src/renz.h src/renz.cc \ src/MetaPopInfo.h src/MetaPopInfo.cc populations_SOURCES = src/populations.h src/populations.cc src/constants.h \ src/utils.h src/utils.cc src/catalog_utils.h src/catalog_utils.cc \ src/log_utils.h src/log_utils.cc \ src/stacks.h src/stacks.cc src/locus.h src/locus.cc \ src/DNASeq.h src/DNASeq.cc src/DNANSeq.h src/DNANSeq.cc \ src/PopMap.h src/PopSum.h src/aln_utils.h src/aln_utils.cc src/genotype_dictionaries.h src/genotype_dictionaries.cc \ src/input.h src/input.cc src/sql_utilities.h src/sql_utilities.cc src/renz.h src/renz.cc \ src/bootstrap.h src/ordered.h src/smoothing.h src/smoothing_utils.h \ src/export_formats.h src/export_formats.cc \ src/MetaPopInfo.h src/MetaPopInfo.cc src/Vcf.h src/Vcf.cc phasedstacks_SOURCES = src/phasedstacks.h src/phasedstacks.cc src/constants.h \ src/locus.h src/locus.cc \ src/input.h src/input.cc src/sql_utilities.h src/sql_utilities.cc \ src/log_utils.h src/log_utils.cc \ src/utils.h src/utils.cc src/catalog_utils.h src/catalog_utils.cc ustacks_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) ustacks_LDFLAGS = $(OPENMP_CFLAGS) pstacks_CXXFLAGS = $(OPENMP_CFLAGS) $(BAM_CFLAGS) $(SPARSEHASH_CFLAGS) pstacks_LDFLAGS = $(OPENMP_CFLAGS) pstacks_LDADD = $(BAM_LIBS) estacks_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) estacks_LDFLAGS = $(OPENMP_CFLAGS) cstacks_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) cstacks_LDFLAGS = $(OPENMP_CFLAGS) hstacks_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) hstacks_LDFLAGS = $(OPENMP_CFLAGS) sstacks_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) sstacks_LDFLAGS = $(OPENMP_CFLAGS) rxstacks_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) rxstacks_LDFLAGS = $(OPENMP_CFLAGS) process_radtags_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) $(BAM_CFLAGS) process_radtags_LDFLAGS = $(OPENMP_CFLAGS) process_radtags_LDADD = $(BAM_LIBS) process_shortreads_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) $(BAM_CFLAGS) process_shortreads_LDFLAGS = $(OPENMP_CFLAGS) process_shortreads_LDADD = $(BAM_LIBS) kmer_filter_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) kmer_filter_LDFLAGS = $(OPENMP_CFLAGS) genotypes_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) genotypes_LDFLAGS = $(OPENMP_CFLAGS) populations_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) populations_LDFLAGS = $(OPENMP_CFLAGS) phasedstacks_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) phasedstacks_LDFLAGS = $(OPENMP_CFLAGS) clone_filter_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) $(BAM_CFLAGS) clone_filter_LDFLAGS = $(OPENMP_CFLAGS) clone_filter_LDADD = $(BAM_LIBS) dist_bin_SCRIPTS = scripts/denovo_map.pl scripts/ref_map.pl scripts/export_sql.pl \ scripts/sort_read_pairs.pl scripts/exec_velvet.pl scripts/load_sequences.pl \ scripts/index_radtags.pl scripts/load_radtags.pl scripts/stacks_export_notify.pl \ scripts/integrate_alignments.py dist_noinst_SCRIPTS = autogen.sh scripts/extract_interpop_chars.pl scripts/convert_stacks.pl nobase_pkgdata_DATA = sql/mysql.cnf.dist sql/catalog_index.sql sql/stacks.sql sql/tag_index.sql sql/chr_index.sql \ php/CatalogClass.php php/annotate_marker.php php/constants.php.dist php/index.php php/tags.php php/tag.php \ php/Locus.php php/catalog.php php/correct_genotypes.php php/correct_genotype.php php/export_batch.php php/last_modified.php \ php/version.php php/catalog_genotypes.php php/db_functions.php php/header.php php/samples.php \ php/stacks_functions.php php/view_sequence.php php/sequence_blast.php \ php/pop_view.php php/sumstat_view.php php/hapstat_view.php php/fst_view.php php/phist_view.php php/stack_view.php \ php/population_view.js php/ajax.js php/annotate.js php/stacks.js php/export.js php/stacks.css \ php/images/caret-d.png php/images/caret-u.png php/images/excel_icon.png php/images/l-arrow-disabled.png \ php/images/l-arrow.png php/images/r-arrow-disabled.png php/images/r-arrow.png php/images/stacks_bg.png \ php/images/stacks_logo_rev_small.png LOG_DRIVER = env AM_TAP_AWK='$(AWK)' $(SHELL) $(abs_top_srcdir)/tests/tap-driver.sh TESTS = tests/process_radtags.t tests/kmer_filter.t tests/ustacks.t tests/pstacks.t EXTRA_DIST = $(nobase_pkgdata_DATA) LICENSE INSTALL README ChangeLog $(TESTS) SUBDIRS = htslib . pkglocalstatedir = $(localstatedir)/$(PACKAGE) debug: $(MAKE) all "CXXFLAGS=-g -Wall -DDEBUG -std=gnu++0x" install-data-hook: sed -e 's,_VERSION_,$(VERSION),' -e 's,_BINDIR_,$(bindir)/,g' -e 's,_PKGDATADIR_,$(pkgdatadir)/,g' $(DESTDIR)$(bindir)/denovo_map.pl > $(DESTDIR)$(bindir)/denovo_map.pl.subst mv $(DESTDIR)$(bindir)/denovo_map.pl.subst $(DESTDIR)$(bindir)/denovo_map.pl chmod +x $(DESTDIR)$(bindir)/denovo_map.pl sed -e 's,_VERSION_,$(VERSION),' -e 's,_BINDIR_,$(bindir)/,g' -e 's,_PKGDATADIR_,$(pkgdatadir)/,g' $(DESTDIR)$(bindir)/ref_map.pl > $(DESTDIR)$(bindir)/ref_map.pl.subst mv $(DESTDIR)$(bindir)/ref_map.pl.subst $(DESTDIR)$(bindir)/ref_map.pl chmod +x $(DESTDIR)$(bindir)/ref_map.pl sed -e 's,_VERSION_,$(VERSION),' -e 's,_PKGDATADIR_,$(pkgdatadir)/,g' $(DESTDIR)$(bindir)/export_sql.pl > $(DESTDIR)$(bindir)/export_sql.pl.subst mv $(DESTDIR)$(bindir)/export_sql.pl.subst $(DESTDIR)$(bindir)/export_sql.pl chmod +x $(DESTDIR)$(bindir)/export_sql.pl sed -e 's,_VERSION_,$(VERSION),' -e 's,_PKGDATADIR_,$(pkgdatadir)/,g' $(DESTDIR)$(bindir)/index_radtags.pl > $(DESTDIR)$(bindir)/index_radtags.pl.subst mv $(DESTDIR)$(bindir)/index_radtags.pl.subst $(DESTDIR)$(bindir)/index_radtags.pl chmod +x $(DESTDIR)$(bindir)/index_radtags.pl sed -e 's,_VERSION_,$(VERSION),' -e 's,_PKGDATADIR_,$(pkgdatadir)/,g' $(DESTDIR)$(bindir)/load_radtags.pl > $(DESTDIR)$(bindir)/load_radtags.pl.subst mv $(DESTDIR)$(bindir)/load_radtags.pl.subst $(DESTDIR)$(bindir)/load_radtags.pl chmod +x $(DESTDIR)$(bindir)/load_radtags.pl sed -e 's,_VERSION_,$(VERSION),' $(DESTDIR)$(bindir)/sort_read_pairs.pl > $(DESTDIR)$(bindir)/sort_read_pairs.pl.subst mv $(DESTDIR)$(bindir)/sort_read_pairs.pl.subst $(DESTDIR)$(bindir)/sort_read_pairs.pl chmod +x $(DESTDIR)$(bindir)/sort_read_pairs.pl sed -e 's,_VERSION_,$(VERSION),' $(DESTDIR)$(bindir)/exec_velvet.pl > $(DESTDIR)$(bindir)/exec_velvet.pl.subst mv $(DESTDIR)$(bindir)/exec_velvet.pl.subst $(DESTDIR)$(bindir)/exec_velvet.pl chmod +x $(DESTDIR)$(bindir)/exec_velvet.pl sed -e 's,_VERSION_,$(VERSION),' -e 's,_PKGDATADIR_,$(pkgdatadir)/,g' $(DESTDIR)$(bindir)/load_sequences.pl > $(DESTDIR)$(bindir)/load_sequences.pl.subst mv $(DESTDIR)$(bindir)/load_sequences.pl.subst $(DESTDIR)$(bindir)/load_sequences.pl chmod +x $(DESTDIR)$(bindir)/load_sequences.pl sed -e 's,_VERSION_,$(VERSION),' -e 's,_PKGDATADIR_,$(pkgdatadir)/,g' -e 's,_BINDIR_,$(bindir)/,g' \ $(DESTDIR)$(bindir)/stacks_export_notify.pl > $(DESTDIR)$(bindir)/stacks_export_notify.pl.subst mv $(DESTDIR)$(bindir)/stacks_export_notify.pl.subst $(DESTDIR)$(bindir)/stacks_export_notify.pl chmod +x $(DESTDIR)$(bindir)/stacks_export_notify.pl sed -e 's,_PKGDATADIR_,$(pkgdatadir)/,g' -e 's,_BINDIR_,$(bindir)/,g' \ $(DESTDIR)$(pkgdatadir)/php/constants.php.dist > $(DESTDIR)$(pkgdatadir)/php/constants.php.dist.subst mv $(DESTDIR)$(pkgdatadir)/php/constants.php.dist.subst $(DESTDIR)$(pkgdatadir)/php/constants.php.dist echo $(VERSION) > $(DESTDIR)$(pkgdatadir)/php/version.php $(install_sh) -d -m 755 $(DESTDIR)$(pkgdatadir)/php/export uninstall-hook: rm -rf $(DESTDIR)$(pkglocalstatedir) stacks-1.44/Makefile.in000644 000765 000024 00001301353 13001003445 015543 0ustar00catchenstaff000000 000000 # Makefile.in generated by automake 1.15 from Makefile.am. # @configure_input@ # Copyright (C) 1994-2014 Free Software Foundation, Inc. # This Makefile.in is free software; the Free Software Foundation # gives unlimited permission to copy and/or distribute it, # with or without modifications, as long as this notice is preserved. # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY, to the extent permitted by law; without # even the implied warranty of MERCHANTABILITY or FITNESS FOR A # PARTICULAR PURPOSE. @SET_MAKE@ VPATH = @srcdir@ am__is_gnu_make = { \ if test -z '$(MAKELEVEL)'; then \ false; \ elif test -n '$(MAKE_HOST)'; then \ true; \ elif test -n '$(MAKE_VERSION)' && test -n '$(CURDIR)'; then \ true; \ else \ false; \ fi; \ } am__make_running_with_option = \ case $${target_option-} in \ ?) ;; \ *) echo "am__make_running_with_option: internal error: invalid" \ "target option '$${target_option-}' specified" >&2; \ exit 1;; \ esac; \ has_opt=no; \ sane_makeflags=$$MAKEFLAGS; \ if $(am__is_gnu_make); then \ sane_makeflags=$$MFLAGS; \ else \ case $$MAKEFLAGS in \ *\\[\ \ ]*) \ bs=\\; \ sane_makeflags=`printf '%s\n' "$$MAKEFLAGS" \ | sed "s/$$bs$$bs[$$bs $$bs ]*//g"`;; \ esac; \ fi; \ skip_next=no; \ strip_trailopt () \ { \ flg=`printf '%s\n' "$$flg" | sed "s/$$1.*$$//"`; \ }; \ for flg in $$sane_makeflags; do \ test $$skip_next = yes && { skip_next=no; continue; }; \ case $$flg in \ *=*|--*) continue;; \ -*I) strip_trailopt 'I'; skip_next=yes;; \ -*I?*) strip_trailopt 'I';; \ -*O) strip_trailopt 'O'; skip_next=yes;; \ -*O?*) strip_trailopt 'O';; \ -*l) strip_trailopt 'l'; skip_next=yes;; \ -*l?*) strip_trailopt 'l';; \ -[dEDm]) skip_next=yes;; \ -[JT]) skip_next=yes;; \ esac; \ case $$flg in \ *$$target_option*) has_opt=yes; break;; \ esac; \ done; \ test $$has_opt = yes am__make_dryrun = (target_option=n; $(am__make_running_with_option)) am__make_keepgoing = (target_option=k; $(am__make_running_with_option)) pkgdatadir = $(datadir)/@PACKAGE@ pkgincludedir = $(includedir)/@PACKAGE@ pkglibdir = $(libdir)/@PACKAGE@ pkglibexecdir = $(libexecdir)/@PACKAGE@ am__cd = CDPATH="$${ZSH_VERSION+.}$(PATH_SEPARATOR)" && cd install_sh_DATA = $(install_sh) -c -m 644 install_sh_PROGRAM = $(install_sh) -c install_sh_SCRIPT = $(install_sh) -c INSTALL_HEADER = $(INSTALL_DATA) transform = $(program_transform_name) NORMAL_INSTALL = : PRE_INSTALL = : POST_INSTALL = : NORMAL_UNINSTALL = : PRE_UNINSTALL = : POST_UNINSTALL = : bin_PROGRAMS = ustacks$(EXEEXT) pstacks$(EXEEXT) estacks$(EXEEXT) \ cstacks$(EXEEXT) sstacks$(EXEEXT) rxstacks$(EXEEXT) \ hstacks$(EXEEXT) process_radtags$(EXEEXT) \ process_shortreads$(EXEEXT) kmer_filter$(EXEEXT) \ clone_filter$(EXEEXT) genotypes$(EXEEXT) populations$(EXEEXT) \ phasedstacks$(EXEEXT) subdir = . ACLOCAL_M4 = $(top_srcdir)/aclocal.m4 am__aclocal_m4_deps = $(top_srcdir)/acinclude.m4 \ $(top_srcdir)/configure.ac am__configure_deps = $(am__aclocal_m4_deps) $(CONFIGURE_DEPENDENCIES) \ $(ACLOCAL_M4) DIST_COMMON = $(srcdir)/Makefile.am $(top_srcdir)/configure \ $(am__configure_deps) $(dist_bin_SCRIPTS) \ $(dist_noinst_SCRIPTS) $(am__DIST_COMMON) am__CONFIG_DISTCLEAN_FILES = config.status config.cache config.log \ configure.lineno config.status.lineno mkinstalldirs = $(install_sh) -d CONFIG_HEADER = config.h CONFIG_CLEAN_FILES = CONFIG_CLEAN_VPATH_FILES = am__installdirs = "$(DESTDIR)$(bindir)" "$(DESTDIR)$(bindir)" \ "$(DESTDIR)$(pkgdatadir)" PROGRAMS = $(bin_PROGRAMS) am__dirstamp = $(am__leading_dot)dirstamp am_clone_filter_OBJECTS = src/clone_filter-clone_filter.$(OBJEXT) \ src/clone_filter-kmers.$(OBJEXT) \ src/clone_filter-utils.$(OBJEXT) \ src/clone_filter-write.$(OBJEXT) \ src/clone_filter-clean.$(OBJEXT) \ src/clone_filter-file_io.$(OBJEXT) \ src/clone_filter-input.$(OBJEXT) clone_filter_OBJECTS = $(am_clone_filter_OBJECTS) am__DEPENDENCIES_1 = clone_filter_DEPENDENCIES = $(am__DEPENDENCIES_1) clone_filter_LINK = $(CXXLD) $(clone_filter_CXXFLAGS) $(CXXFLAGS) \ $(clone_filter_LDFLAGS) $(LDFLAGS) -o $@ am_cstacks_OBJECTS = src/cstacks-cstacks.$(OBJEXT) \ src/cstacks-stacks.$(OBJEXT) src/cstacks-locus.$(OBJEXT) \ src/cstacks-kmers.$(OBJEXT) src/cstacks-utils.$(OBJEXT) \ src/cstacks-aln_utils.$(OBJEXT) src/cstacks-DNASeq.$(OBJEXT) \ src/cstacks-DNANSeq.$(OBJEXT) src/cstacks-input.$(OBJEXT) \ src/cstacks-sql_utilities.$(OBJEXT) cstacks_OBJECTS = $(am_cstacks_OBJECTS) cstacks_LDADD = $(LDADD) cstacks_LINK = $(CXXLD) $(cstacks_CXXFLAGS) $(CXXFLAGS) \ $(cstacks_LDFLAGS) $(LDFLAGS) -o $@ am_estacks_OBJECTS = src/estacks-estacks.$(OBJEXT) \ src/estacks-stacks.$(OBJEXT) src/estacks-mstack.$(OBJEXT) \ src/estacks-models.$(OBJEXT) src/estacks-utils.$(OBJEXT) \ src/estacks-DNASeq.$(OBJEXT) src/estacks-DNANSeq.$(OBJEXT) \ src/estacks-input.$(OBJEXT) estacks_OBJECTS = $(am_estacks_OBJECTS) estacks_LDADD = $(LDADD) estacks_LINK = $(CXXLD) $(estacks_CXXFLAGS) $(CXXFLAGS) \ $(estacks_LDFLAGS) $(LDFLAGS) -o $@ am_genotypes_OBJECTS = src/genotypes-genotypes.$(OBJEXT) \ src/genotypes-utils.$(OBJEXT) \ src/genotypes-catalog_utils.$(OBJEXT) \ src/genotypes-log_utils.$(OBJEXT) \ src/genotypes-stacks.$(OBJEXT) src/genotypes-locus.$(OBJEXT) \ src/genotypes-DNASeq.$(OBJEXT) src/genotypes-DNANSeq.$(OBJEXT) \ src/genotypes-aln_utils.$(OBJEXT) \ src/genotypes-genotype_dictionaries.$(OBJEXT) \ src/genotypes-input.$(OBJEXT) \ src/genotypes-sql_utilities.$(OBJEXT) \ src/genotypes-renz.$(OBJEXT) \ src/genotypes-MetaPopInfo.$(OBJEXT) genotypes_OBJECTS = $(am_genotypes_OBJECTS) genotypes_LDADD = $(LDADD) genotypes_LINK = $(CXXLD) $(genotypes_CXXFLAGS) $(CXXFLAGS) \ $(genotypes_LDFLAGS) $(LDFLAGS) -o $@ am_hstacks_OBJECTS = src/hstacks-hstacks.$(OBJEXT) \ src/hstacks-stacks.$(OBJEXT) src/hstacks-locus.$(OBJEXT) \ src/hstacks-kmers.$(OBJEXT) src/hstacks-DNASeq.$(OBJEXT) \ src/hstacks-DNANSeq.$(OBJEXT) src/hstacks-utils.$(OBJEXT) \ src/hstacks-input.$(OBJEXT) \ src/hstacks-sql_utilities.$(OBJEXT) hstacks_OBJECTS = $(am_hstacks_OBJECTS) hstacks_LDADD = $(LDADD) hstacks_LINK = $(CXXLD) $(hstacks_CXXFLAGS) $(CXXFLAGS) \ $(hstacks_LDFLAGS) $(LDFLAGS) -o $@ am_kmer_filter_OBJECTS = src/kmer_filter-kmer_filter.$(OBJEXT) \ src/kmer_filter-utils.$(OBJEXT) \ src/kmer_filter-write.$(OBJEXT) \ src/kmer_filter-kmers.$(OBJEXT) \ src/kmer_filter-input.$(OBJEXT) kmer_filter_OBJECTS = $(am_kmer_filter_OBJECTS) kmer_filter_LDADD = $(LDADD) kmer_filter_LINK = $(CXXLD) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) \ $(kmer_filter_LDFLAGS) $(LDFLAGS) -o $@ am_phasedstacks_OBJECTS = src/phasedstacks-phasedstacks.$(OBJEXT) \ src/phasedstacks-locus.$(OBJEXT) \ src/phasedstacks-input.$(OBJEXT) \ src/phasedstacks-sql_utilities.$(OBJEXT) \ src/phasedstacks-log_utils.$(OBJEXT) \ src/phasedstacks-utils.$(OBJEXT) \ src/phasedstacks-catalog_utils.$(OBJEXT) phasedstacks_OBJECTS = $(am_phasedstacks_OBJECTS) phasedstacks_LDADD = $(LDADD) phasedstacks_LINK = $(CXXLD) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) \ $(phasedstacks_LDFLAGS) $(LDFLAGS) -o $@ am_populations_OBJECTS = src/populations-populations.$(OBJEXT) \ src/populations-utils.$(OBJEXT) \ src/populations-catalog_utils.$(OBJEXT) \ src/populations-log_utils.$(OBJEXT) \ src/populations-stacks.$(OBJEXT) \ src/populations-locus.$(OBJEXT) \ src/populations-DNASeq.$(OBJEXT) \ src/populations-DNANSeq.$(OBJEXT) \ src/populations-aln_utils.$(OBJEXT) \ src/populations-genotype_dictionaries.$(OBJEXT) \ src/populations-input.$(OBJEXT) \ src/populations-sql_utilities.$(OBJEXT) \ src/populations-renz.$(OBJEXT) \ src/populations-export_formats.$(OBJEXT) \ src/populations-MetaPopInfo.$(OBJEXT) \ src/populations-Vcf.$(OBJEXT) populations_OBJECTS = $(am_populations_OBJECTS) populations_LDADD = $(LDADD) populations_LINK = $(CXXLD) $(populations_CXXFLAGS) $(CXXFLAGS) \ $(populations_LDFLAGS) $(LDFLAGS) -o $@ am_process_radtags_OBJECTS = \ src/process_radtags-process_radtags.$(OBJEXT) \ src/process_radtags-utils.$(OBJEXT) \ src/process_radtags-log_utils.$(OBJEXT) \ src/process_radtags-write.$(OBJEXT) \ src/process_radtags-clean.$(OBJEXT) \ src/process_radtags-file_io.$(OBJEXT) \ src/process_radtags-input.$(OBJEXT) \ src/process_radtags-renz.$(OBJEXT) process_radtags_OBJECTS = $(am_process_radtags_OBJECTS) process_radtags_DEPENDENCIES = $(am__DEPENDENCIES_1) process_radtags_LINK = $(CXXLD) $(process_radtags_CXXFLAGS) \ $(CXXFLAGS) $(process_radtags_LDFLAGS) $(LDFLAGS) -o $@ am_process_shortreads_OBJECTS = \ src/process_shortreads-process_shortreads.$(OBJEXT) \ src/process_shortreads-clean.$(OBJEXT) \ src/process_shortreads-file_io.$(OBJEXT) \ src/process_shortreads-utils.$(OBJEXT) \ src/process_shortreads-log_utils.$(OBJEXT) \ src/process_shortreads-write.$(OBJEXT) \ src/process_shortreads-kmers.$(OBJEXT) \ src/process_shortreads-input.$(OBJEXT) process_shortreads_OBJECTS = $(am_process_shortreads_OBJECTS) process_shortreads_DEPENDENCIES = $(am__DEPENDENCIES_1) process_shortreads_LINK = $(CXXLD) $(process_shortreads_CXXFLAGS) \ $(CXXFLAGS) $(process_shortreads_LDFLAGS) $(LDFLAGS) -o $@ am_pstacks_OBJECTS = src/pstacks-pstacks.$(OBJEXT) \ src/pstacks-stacks.$(OBJEXT) src/pstacks-mstack.$(OBJEXT) \ src/pstacks-DNANSeq.$(OBJEXT) src/pstacks-DNASeq.$(OBJEXT) \ src/pstacks-models.$(OBJEXT) src/pstacks-utils.$(OBJEXT) \ src/pstacks-input.$(OBJEXT) pstacks_OBJECTS = $(am_pstacks_OBJECTS) pstacks_DEPENDENCIES = $(am__DEPENDENCIES_1) pstacks_LINK = $(CXXLD) $(pstacks_CXXFLAGS) $(CXXFLAGS) \ $(pstacks_LDFLAGS) $(LDFLAGS) -o $@ am_rxstacks_OBJECTS = src/rxstacks-rxstacks.$(OBJEXT) \ src/rxstacks-stacks.$(OBJEXT) src/rxstacks-locus.$(OBJEXT) \ src/rxstacks-DNANSeq.$(OBJEXT) src/rxstacks-DNASeq.$(OBJEXT) \ src/rxstacks-mst.$(OBJEXT) src/rxstacks-models.$(OBJEXT) \ src/rxstacks-aln_utils.$(OBJEXT) src/rxstacks-utils.$(OBJEXT) \ src/rxstacks-input.$(OBJEXT) \ src/rxstacks-sql_utilities.$(OBJEXT) \ src/rxstacks-MetaPopInfo.$(OBJEXT) rxstacks_OBJECTS = $(am_rxstacks_OBJECTS) rxstacks_LDADD = $(LDADD) rxstacks_LINK = $(CXXLD) $(rxstacks_CXXFLAGS) $(CXXFLAGS) \ $(rxstacks_LDFLAGS) $(LDFLAGS) -o $@ am_sstacks_OBJECTS = src/sstacks-sstacks.$(OBJEXT) \ src/sstacks-stacks.$(OBJEXT) src/sstacks-locus.$(OBJEXT) \ src/sstacks-kmers.$(OBJEXT) src/sstacks-aln_utils.$(OBJEXT) \ src/sstacks-DNASeq.$(OBJEXT) src/sstacks-DNANSeq.$(OBJEXT) \ src/sstacks-utils.$(OBJEXT) src/sstacks-input.$(OBJEXT) \ src/sstacks-sql_utilities.$(OBJEXT) sstacks_OBJECTS = $(am_sstacks_OBJECTS) sstacks_LDADD = $(LDADD) sstacks_LINK = $(CXXLD) $(sstacks_CXXFLAGS) $(CXXFLAGS) \ $(sstacks_LDFLAGS) $(LDFLAGS) -o $@ am_ustacks_OBJECTS = src/ustacks-ustacks.$(OBJEXT) \ src/ustacks-stacks.$(OBJEXT) src/ustacks-mstack.$(OBJEXT) \ src/ustacks-mst.$(OBJEXT) src/ustacks-cmb.$(OBJEXT) \ src/ustacks-DNASeq.$(OBJEXT) src/ustacks-DNANSeq.$(OBJEXT) \ src/ustacks-models.$(OBJEXT) src/ustacks-utils.$(OBJEXT) \ src/ustacks-kmers.$(OBJEXT) src/ustacks-aln_utils.$(OBJEXT) \ src/ustacks-input.$(OBJEXT) ustacks_OBJECTS = $(am_ustacks_OBJECTS) ustacks_LDADD = $(LDADD) ustacks_LINK = $(CXXLD) $(ustacks_CXXFLAGS) $(CXXFLAGS) \ $(ustacks_LDFLAGS) $(LDFLAGS) -o $@ am__vpath_adj_setup = srcdirstrip=`echo "$(srcdir)" | sed 's|.|.|g'`; am__vpath_adj = case $$p in \ $(srcdir)/*) f=`echo "$$p" | sed "s|^$$srcdirstrip/||"`;; \ *) f=$$p;; \ esac; am__strip_dir = f=`echo $$p | sed -e 's|^.*/||'`; am__install_max = 40 am__nobase_strip_setup = \ srcdirstrip=`echo "$(srcdir)" | sed 's/[].[^$$\\*|]/\\\\&/g'` am__nobase_strip = \ for p in $$list; do echo "$$p"; done | sed -e "s|$$srcdirstrip/||" am__nobase_list = $(am__nobase_strip_setup); \ for p in $$list; do echo "$$p $$p"; done | \ sed "s| $$srcdirstrip/| |;"' / .*\//!s/ .*/ ./; s,\( .*\)/[^/]*$$,\1,' | \ $(AWK) 'BEGIN { files["."] = "" } { files[$$2] = files[$$2] " " $$1; \ if (++n[$$2] == $(am__install_max)) \ { print $$2, files[$$2]; n[$$2] = 0; files[$$2] = "" } } \ END { for (dir in files) print dir, files[dir] }' am__base_list = \ sed '$$!N;$$!N;$$!N;$$!N;$$!N;$$!N;$$!N;s/\n/ /g' | \ sed '$$!N;$$!N;$$!N;$$!N;s/\n/ /g' am__uninstall_files_from_dir = { \ test -z "$$files" \ || { test ! -d "$$dir" && test ! -f "$$dir" && test ! -r "$$dir"; } \ || { echo " ( cd '$$dir' && rm -f" $$files ")"; \ $(am__cd) "$$dir" && rm -f $$files; }; \ } SCRIPTS = $(dist_bin_SCRIPTS) $(dist_noinst_SCRIPTS) AM_V_P = $(am__v_P_@AM_V@) am__v_P_ = $(am__v_P_@AM_DEFAULT_V@) am__v_P_0 = false am__v_P_1 = : AM_V_GEN = $(am__v_GEN_@AM_V@) am__v_GEN_ = $(am__v_GEN_@AM_DEFAULT_V@) am__v_GEN_0 = @echo " GEN " $@; am__v_GEN_1 = AM_V_at = $(am__v_at_@AM_V@) am__v_at_ = $(am__v_at_@AM_DEFAULT_V@) am__v_at_0 = @ am__v_at_1 = DEFAULT_INCLUDES = -I.@am__isrc@ depcomp = $(SHELL) $(top_srcdir)/config/depcomp am__depfiles_maybe = depfiles am__mv = mv -f AM_V_lt = $(am__v_lt_@AM_V@) am__v_lt_ = $(am__v_lt_@AM_DEFAULT_V@) am__v_lt_0 = --silent am__v_lt_1 = CXXCOMPILE = $(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) \ $(AM_CPPFLAGS) $(CPPFLAGS) $(AM_CXXFLAGS) $(CXXFLAGS) AM_V_CXX = $(am__v_CXX_@AM_V@) am__v_CXX_ = $(am__v_CXX_@AM_DEFAULT_V@) am__v_CXX_0 = @echo " CXX " $@; am__v_CXX_1 = CXXLD = $(CXX) CXXLINK = $(CXXLD) $(AM_CXXFLAGS) $(CXXFLAGS) $(AM_LDFLAGS) $(LDFLAGS) \ -o $@ AM_V_CXXLD = $(am__v_CXXLD_@AM_V@) am__v_CXXLD_ = $(am__v_CXXLD_@AM_DEFAULT_V@) am__v_CXXLD_0 = @echo " CXXLD " $@; am__v_CXXLD_1 = COMPILE = $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) \ $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) AM_V_CC = $(am__v_CC_@AM_V@) am__v_CC_ = $(am__v_CC_@AM_DEFAULT_V@) am__v_CC_0 = @echo " CC " $@; am__v_CC_1 = CCLD = $(CC) LINK = $(CCLD) $(AM_CFLAGS) $(CFLAGS) $(AM_LDFLAGS) $(LDFLAGS) -o $@ AM_V_CCLD = $(am__v_CCLD_@AM_V@) am__v_CCLD_ = $(am__v_CCLD_@AM_DEFAULT_V@) am__v_CCLD_0 = @echo " CCLD " $@; am__v_CCLD_1 = SOURCES = $(clone_filter_SOURCES) $(cstacks_SOURCES) \ $(estacks_SOURCES) $(genotypes_SOURCES) $(hstacks_SOURCES) \ $(kmer_filter_SOURCES) $(phasedstacks_SOURCES) \ $(populations_SOURCES) $(process_radtags_SOURCES) \ $(process_shortreads_SOURCES) $(pstacks_SOURCES) \ $(rxstacks_SOURCES) $(sstacks_SOURCES) $(ustacks_SOURCES) DIST_SOURCES = $(clone_filter_SOURCES) $(cstacks_SOURCES) \ $(estacks_SOURCES) $(genotypes_SOURCES) $(hstacks_SOURCES) \ $(kmer_filter_SOURCES) $(phasedstacks_SOURCES) \ $(populations_SOURCES) $(process_radtags_SOURCES) \ $(process_shortreads_SOURCES) $(pstacks_SOURCES) \ $(rxstacks_SOURCES) $(sstacks_SOURCES) $(ustacks_SOURCES) RECURSIVE_TARGETS = all-recursive check-recursive cscopelist-recursive \ ctags-recursive dvi-recursive html-recursive info-recursive \ install-data-recursive install-dvi-recursive \ install-exec-recursive install-html-recursive \ install-info-recursive install-pdf-recursive \ install-ps-recursive install-recursive installcheck-recursive \ installdirs-recursive pdf-recursive ps-recursive \ tags-recursive uninstall-recursive am__can_run_installinfo = \ case $$AM_UPDATE_INFO_DIR in \ n|no|NO) false;; \ *) (install-info --version) >/dev/null 2>&1;; \ esac DATA = $(nobase_pkgdata_DATA) RECURSIVE_CLEAN_TARGETS = mostlyclean-recursive clean-recursive \ distclean-recursive maintainer-clean-recursive am__recursive_targets = \ $(RECURSIVE_TARGETS) \ $(RECURSIVE_CLEAN_TARGETS) \ $(am__extra_recursive_targets) AM_RECURSIVE_TARGETS = $(am__recursive_targets:-recursive=) TAGS CTAGS \ cscope check recheck distdir dist dist-all distcheck am__tagged_files = $(HEADERS) $(SOURCES) $(TAGS_FILES) \ $(LISP)config.h.in # Read a list of newline-separated strings from the standard input, # and print each of them once, without duplicates. Input order is # *not* preserved. am__uniquify_input = $(AWK) '\ BEGIN { nonempty = 0; } \ { items[$$0] = 1; nonempty = 1; } \ END { if (nonempty) { for (i in items) print i; }; } \ ' # Make sure the list of sources is unique. This is necessary because, # e.g., the same source file might be shared among _SOURCES variables # for different programs/libraries. am__define_uniq_tagged_files = \ list='$(am__tagged_files)'; \ unique=`for i in $$list; do \ if test -f "$$i"; then echo $$i; else echo $(srcdir)/$$i; fi; \ done | $(am__uniquify_input)` ETAGS = etags CTAGS = ctags CSCOPE = cscope am__tty_colors_dummy = \ mgn= red= grn= lgn= blu= brg= std=; \ am__color_tests=no am__tty_colors = { \ $(am__tty_colors_dummy); \ if test "X$(AM_COLOR_TESTS)" = Xno; then \ am__color_tests=no; \ elif test "X$(AM_COLOR_TESTS)" = Xalways; then \ am__color_tests=yes; \ elif test "X$$TERM" != Xdumb && { test -t 1; } 2>/dev/null; then \ am__color_tests=yes; \ fi; \ if test $$am__color_tests = yes; then \ red=''; \ grn=''; \ lgn=''; \ blu=''; \ mgn=''; \ brg=''; \ std=''; \ fi; \ } am__recheck_rx = ^[ ]*:recheck:[ ]* am__global_test_result_rx = ^[ ]*:global-test-result:[ ]* am__copy_in_global_log_rx = ^[ ]*:copy-in-global-log:[ ]* # A command that, given a newline-separated list of test names on the # standard input, print the name of the tests that are to be re-run # upon "make recheck". am__list_recheck_tests = $(AWK) '{ \ recheck = 1; \ while ((rc = (getline line < ($$0 ".trs"))) != 0) \ { \ if (rc < 0) \ { \ if ((getline line2 < ($$0 ".log")) < 0) \ recheck = 0; \ break; \ } \ else if (line ~ /$(am__recheck_rx)[nN][Oo]/) \ { \ recheck = 0; \ break; \ } \ else if (line ~ /$(am__recheck_rx)[yY][eE][sS]/) \ { \ break; \ } \ }; \ if (recheck) \ print $$0; \ close ($$0 ".trs"); \ close ($$0 ".log"); \ }' # A command that, given a newline-separated list of test names on the # standard input, create the global log from their .trs and .log files. am__create_global_log = $(AWK) ' \ function fatal(msg) \ { \ print "fatal: making $@: " msg | "cat >&2"; \ exit 1; \ } \ function rst_section(header) \ { \ print header; \ len = length(header); \ for (i = 1; i <= len; i = i + 1) \ printf "="; \ printf "\n\n"; \ } \ { \ copy_in_global_log = 1; \ global_test_result = "RUN"; \ while ((rc = (getline line < ($$0 ".trs"))) != 0) \ { \ if (rc < 0) \ fatal("failed to read from " $$0 ".trs"); \ if (line ~ /$(am__global_test_result_rx)/) \ { \ sub("$(am__global_test_result_rx)", "", line); \ sub("[ ]*$$", "", line); \ global_test_result = line; \ } \ else if (line ~ /$(am__copy_in_global_log_rx)[nN][oO]/) \ copy_in_global_log = 0; \ }; \ if (copy_in_global_log) \ { \ rst_section(global_test_result ": " $$0); \ while ((rc = (getline line < ($$0 ".log"))) != 0) \ { \ if (rc < 0) \ fatal("failed to read from " $$0 ".log"); \ print line; \ }; \ printf "\n"; \ }; \ close ($$0 ".trs"); \ close ($$0 ".log"); \ }' # Restructured Text title. am__rst_title = { sed 's/.*/ & /;h;s/./=/g;p;x;s/ *$$//;p;g' && echo; } # Solaris 10 'make', and several other traditional 'make' implementations, # pass "-e" to $(SHELL), and POSIX 2008 even requires this. Work around it # by disabling -e (using the XSI extension "set +e") if it's set. am__sh_e_setup = case $$- in *e*) set +e;; esac # Default flags passed to test drivers. am__common_driver_flags = \ --color-tests "$$am__color_tests" \ --enable-hard-errors "$$am__enable_hard_errors" \ --expect-failure "$$am__expect_failure" # To be inserted before the command running the test. Creates the # directory for the log if needed. Stores in $dir the directory # containing $f, in $tst the test, in $log the log. Executes the # developer- defined test setup AM_TESTS_ENVIRONMENT (if any), and # passes TESTS_ENVIRONMENT. Set up options for the wrapper that # will run the test scripts (or their associated LOG_COMPILER, if # thy have one). am__check_pre = \ $(am__sh_e_setup); \ $(am__vpath_adj_setup) $(am__vpath_adj) \ $(am__tty_colors); \ srcdir=$(srcdir); export srcdir; \ case "$@" in \ */*) am__odir=`echo "./$@" | sed 's|/[^/]*$$||'`;; \ *) am__odir=.;; \ esac; \ test "x$$am__odir" = x"." || test -d "$$am__odir" \ || $(MKDIR_P) "$$am__odir" || exit $$?; \ if test -f "./$$f"; then dir=./; \ elif test -f "$$f"; then dir=; \ else dir="$(srcdir)/"; fi; \ tst=$$dir$$f; log='$@'; \ if test -n '$(DISABLE_HARD_ERRORS)'; then \ am__enable_hard_errors=no; \ else \ am__enable_hard_errors=yes; \ fi; \ case " $(XFAIL_TESTS) " in \ *[\ \ ]$$f[\ \ ]* | *[\ \ ]$$dir$$f[\ \ ]*) \ am__expect_failure=yes;; \ *) \ am__expect_failure=no;; \ esac; \ $(AM_TESTS_ENVIRONMENT) $(TESTS_ENVIRONMENT) # A shell command to get the names of the tests scripts with any registered # extension removed (i.e., equivalently, the names of the test logs, with # the '.log' extension removed). The result is saved in the shell variable # '$bases'. This honors runtime overriding of TESTS and TEST_LOGS. Sadly, # we cannot use something simpler, involving e.g., "$(TEST_LOGS:.log=)", # since that might cause problem with VPATH rewrites for suffix-less tests. # See also 'test-harness-vpath-rewrite.sh' and 'test-trs-basic.sh'. am__set_TESTS_bases = \ bases='$(TEST_LOGS)'; \ bases=`for i in $$bases; do echo $$i; done | sed 's/\.log$$//'`; \ bases=`echo $$bases` RECHECK_LOGS = $(TEST_LOGS) TEST_SUITE_LOG = test-suite.log TEST_EXTENSIONS = @EXEEXT@ .test LOG_COMPILE = $(LOG_COMPILER) $(AM_LOG_FLAGS) $(LOG_FLAGS) am__set_b = \ case '$@' in \ */*) \ case '$*' in \ */*) b='$*';; \ *) b=`echo '$@' | sed 's/\.log$$//'`; \ esac;; \ *) \ b='$*';; \ esac am__test_logs1 = $(TESTS:=.log) am__test_logs2 = $(am__test_logs1:@EXEEXT@.log=.log) TEST_LOGS = $(am__test_logs2:.test.log=.log) TEST_LOG_DRIVER = $(SHELL) $(top_srcdir)/config/test-driver TEST_LOG_COMPILE = $(TEST_LOG_COMPILER) $(AM_TEST_LOG_FLAGS) \ $(TEST_LOG_FLAGS) DIST_SUBDIRS = $(SUBDIRS) am__DIST_COMMON = $(srcdir)/Makefile.in $(srcdir)/config.h.in \ $(top_srcdir)/config/compile $(top_srcdir)/config/depcomp \ $(top_srcdir)/config/install-sh $(top_srcdir)/config/missing \ $(top_srcdir)/config/test-driver ChangeLog INSTALL README \ config/compile config/depcomp config/install-sh config/missing DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST) distdir = $(PACKAGE)-$(VERSION) top_distdir = $(distdir) am__remove_distdir = \ if test -d "$(distdir)"; then \ find "$(distdir)" -type d ! -perm -200 -exec chmod u+w {} ';' \ && rm -rf "$(distdir)" \ || { sleep 5 && rm -rf "$(distdir)"; }; \ else :; fi am__post_remove_distdir = $(am__remove_distdir) am__relativize = \ dir0=`pwd`; \ sed_first='s,^\([^/]*\)/.*$$,\1,'; \ sed_rest='s,^[^/]*/*,,'; \ sed_last='s,^.*/\([^/]*\)$$,\1,'; \ sed_butlast='s,/*[^/]*$$,,'; \ while test -n "$$dir1"; do \ first=`echo "$$dir1" | sed -e "$$sed_first"`; \ if test "$$first" != "."; then \ if test "$$first" = ".."; then \ dir2=`echo "$$dir0" | sed -e "$$sed_last"`/"$$dir2"; \ dir0=`echo "$$dir0" | sed -e "$$sed_butlast"`; \ else \ first2=`echo "$$dir2" | sed -e "$$sed_first"`; \ if test "$$first2" = "$$first"; then \ dir2=`echo "$$dir2" | sed -e "$$sed_rest"`; \ else \ dir2="../$$dir2"; \ fi; \ dir0="$$dir0"/"$$first"; \ fi; \ fi; \ dir1=`echo "$$dir1" | sed -e "$$sed_rest"`; \ done; \ reldir="$$dir2" DIST_ARCHIVES = $(distdir).tar.gz GZIP_ENV = --best DIST_TARGETS = dist-gzip distuninstallcheck_listfiles = find . -type f -print am__distuninstallcheck_listfiles = $(distuninstallcheck_listfiles) \ | sed 's|^\./|$(prefix)/|' | grep -v '$(infodir)/dir$$' distcleancheck_listfiles = find . -type f -print ACLOCAL = @ACLOCAL@ AMTAR = @AMTAR@ AM_DEFAULT_VERBOSITY = @AM_DEFAULT_VERBOSITY@ AUTOCONF = @AUTOCONF@ AUTOHEADER = @AUTOHEADER@ AUTOMAKE = @AUTOMAKE@ AWK = @AWK@ BAM_CFLAGS = @BAM_CFLAGS@ BAM_LIBS = @BAM_LIBS@ CC = @CC@ CCDEPMODE = @CCDEPMODE@ CFLAGS = @CFLAGS@ CPP = @CPP@ CPPFLAGS = @CPPFLAGS@ CXX = @CXX@ CXXDEPMODE = @CXXDEPMODE@ CXXFLAGS = @CXXFLAGS@ CYGPATH_W = @CYGPATH_W@ DEFS = @DEFS@ DEPDIR = @DEPDIR@ ECHO_C = @ECHO_C@ ECHO_N = @ECHO_N@ ECHO_T = @ECHO_T@ EGREP = @EGREP@ EXEEXT = @EXEEXT@ GREP = @GREP@ HAVE_CXX11 = @HAVE_CXX11@ INSTALL = @INSTALL@ INSTALL_DATA = @INSTALL_DATA@ INSTALL_PROGRAM = @INSTALL_PROGRAM@ INSTALL_SCRIPT = @INSTALL_SCRIPT@ INSTALL_STRIP_PROGRAM = @INSTALL_STRIP_PROGRAM@ LDFLAGS = @LDFLAGS@ LIBOBJS = @LIBOBJS@ LIBS = @LIBS@ LTLIBOBJS = @LTLIBOBJS@ MAKEINFO = @MAKEINFO@ MKDIR_P = @MKDIR_P@ OBJEXT = @OBJEXT@ OPENMP_CFLAGS = @OPENMP_CFLAGS@ PACKAGE = @PACKAGE@ PACKAGE_BUGREPORT = @PACKAGE_BUGREPORT@ PACKAGE_NAME = @PACKAGE_NAME@ PACKAGE_STRING = @PACKAGE_STRING@ PACKAGE_TARNAME = @PACKAGE_TARNAME@ PACKAGE_URL = @PACKAGE_URL@ PACKAGE_VERSION = @PACKAGE_VERSION@ PATH_SEPARATOR = @PATH_SEPARATOR@ SET_MAKE = @SET_MAKE@ SHELL = @SHELL@ SPARSEHASH_CFLAGS = @SPARSEHASH_CFLAGS@ STRIP = @STRIP@ VERSION = @VERSION@ abs_builddir = @abs_builddir@ abs_srcdir = @abs_srcdir@ abs_top_builddir = @abs_top_builddir@ abs_top_srcdir = @abs_top_srcdir@ ac_ct_CC = @ac_ct_CC@ ac_ct_CXX = @ac_ct_CXX@ am__include = @am__include@ am__leading_dot = @am__leading_dot@ am__quote = @am__quote@ am__tar = @am__tar@ am__untar = @am__untar@ bindir = @bindir@ build_alias = @build_alias@ builddir = @builddir@ datadir = @datadir@ datarootdir = @datarootdir@ docdir = @docdir@ dvidir = @dvidir@ exec_prefix = @exec_prefix@ host_alias = @host_alias@ htmldir = @htmldir@ includedir = @includedir@ infodir = @infodir@ install_sh = @install_sh@ libdir = @libdir@ libexecdir = @libexecdir@ localedir = @localedir@ localstatedir = @localstatedir@ mandir = @mandir@ mkdir_p = @mkdir_p@ oldincludedir = @oldincludedir@ pdfdir = @pdfdir@ prefix = @prefix@ program_transform_name = @program_transform_name@ psdir = @psdir@ sbindir = @sbindir@ sharedstatedir = @sharedstatedir@ srcdir = @srcdir@ sysconfdir = @sysconfdir@ target_alias = @target_alias@ top_build_prefix = @top_build_prefix@ top_builddir = @top_builddir@ top_srcdir = @top_srcdir@ ustacks_SOURCES = src/ustacks.h src/ustacks.cc src/stacks.h src/stacks.cc src/constants.h \ src/mstack.h src/mstack.cc src/mst.h src/mst.cc src/cmb.h src/cmb.cc \ src/DNASeq.h src/DNASeq.cc src/DNANSeq.h src/DNANSeq.cc \ src/models.h src/models.cc src/utils.h src/utils.cc \ src/kmers.h src/kmers.cc src/GappedAln.h src/aln_utils.h src/aln_utils.cc \ src/input.h src/input.cc src/Tsv.h src/BowtieI.h src/FastaI.h src/FastqI.h src/SamI.h \ src/gzFasta.h src/gzFastq.h pstacks_SOURCES = src/pstacks.h src/pstacks.cc src/stacks.h src/stacks.cc src/constants.h \ src/mstack.h src/mstack.cc \ src/DNANSeq.h src/DNANSeq.cc src/DNASeq.h src/DNASeq.cc \ src/models.h src/models.cc src/utils.h src/utils.cc \ src/input.h src/input.cc src/Tsv.h src/BowtieI.h src/FastaI.h src/FastqI.h src/SamI.h \ src/BamI.h estacks_SOURCES = src/estacks.h src/estacks.cc src/stacks.h src/stacks.cc src/constants.h \ src/mstack.h src/mstack.cc \ src/models.h src/models.cc src/utils.h src/utils.cc \ src/DNASeq.h src/DNASeq.cc src/DNANSeq.h src/DNANSeq.cc \ src/input.h src/input.cc src/Tsv.h src/BowtieI.h src/FastaI.h src/FastqI.h src/SamI.h cstacks_SOURCES = src/cstacks.h src/cstacks.cc src/constants.h \ src/stacks.h src/stacks.cc src/locus.h src/locus.cc \ src/kmers.h src/kmers.cc src/utils.h src/utils.cc \ src/GappedAln.h src/aln_utils.h src/aln_utils.cc \ src/DNASeq.h src/DNASeq.cc src/DNANSeq.h src/DNANSeq.cc \ src/input.h src/input.cc src/sql_utilities.h src/sql_utilities.cc hstacks_SOURCES = src/hstacks.h src/hstacks.cc src/constants.h \ src/stacks.h src/stacks.cc src/locus.h src/locus.cc \ src/kmers.h src/kmers.cc \ src/DNASeq.h src/DNASeq.cc src/DNANSeq.h src/DNANSeq.cc \ src/utils.h src/utils.cc \ src/input.h src/input.cc src/sql_utilities.h src/sql_utilities.cc sstacks_SOURCES = src/sstacks.h src/sstacks.cc src/constants.h \ src/stacks.h src/stacks.cc src/locus.h src/locus.cc \ src/kmers.h src/kmers.cc \ src/GappedAln.h src/aln_utils.h src/aln_utils.cc \ src/DNASeq.h src/DNASeq.cc src/DNANSeq.h src/DNANSeq.cc src/utils.h src/utils.cc \ src/input.h src/input.cc src/sql_utilities.h src/sql_utilities.cc rxstacks_SOURCES = src/rxstacks.h src/rxstacks.cc src/constants.h \ src/stacks.h src/stacks.cc src/locus.h src/locus.cc \ src/DNANSeq.h src/DNANSeq.cc src/DNASeq.h src/DNASeq.cc \ src/mst.h src/mst.cc \ src/models.h src/models.cc \ src/aln_utils.h src/aln_utils.cc src/utils.h src/utils.cc \ src/input.h src/input.cc src/sql_utilities.h src/sql_utilities.cc \ src/MetaPopInfo.h src/MetaPopInfo.cc process_radtags_SOURCES = src/process_radtags.h src/process_radtags.cc src/constants.h \ src/utils.h src/utils.cc src/log_utils.h src/log_utils.cc \ src/write.h src/write.cc \ src/clean.h src/clean.cc \ src/file_io.h src/file_io.cc \ src/input.h src/input.cc src/BustardI.h src/BamUnalignedI.h src/FastqI.h src/gzFastq.h \ src/renz.h src/renz.cc process_shortreads_SOURCES = src/process_shortreads.h src/process_shortreads.cc src/constants.h \ src/clean.h src/clean.cc \ src/file_io.h src/file_io.cc \ src/utils.h src/utils.cc src/log_utils.h src/log_utils.cc \ src/write.h src/write.cc \ src/kmers.h src/kmers.cc \ src/input.h src/input.cc src/BustardI.h src/BamUnalignedI.h src/FastqI.h src/gzFastq.h kmer_filter_SOURCES = src/kmer_filter.h src/kmer_filter.cc src/constants.h \ src/utils.h src/utils.cc \ src/write.h src/write.cc \ src/kmers.h src/kmers.cc \ src/input.h src/input.cc src/BustardI.h src/FastqI.h src/FastaI.h \ src/gzFastq.h src/gzFasta.h clone_filter_SOURCES = src/clone_filter.h src/clone_filter.cc src/constants.h \ src/kmers.h src/kmers.cc \ src/utils.h src/utils.cc \ src/write.h src/write.cc \ src/clean.h src/clean.cc src/file_io.h src/file_io.cc \ src/input.h src/input.cc src/BustardI.h src/FastqI.h src/FastaI.h \ src/gzFastq.h src/gzFasta.h genotypes_SOURCES = src/genotypes.h src/genotypes.cc src/constants.h \ src/utils.h src/utils.cc src/catalog_utils.h src/catalog_utils.cc \ src/log_utils.h src/log_utils.cc \ src/stacks.h src/stacks.cc src/locus.h src/locus.cc \ src/DNASeq.h src/DNASeq.cc src/DNANSeq.h src/DNANSeq.cc \ src/aln_utils.h src/aln_utils.cc \ src/PopMap.h src/genotype_dictionaries.h src/genotype_dictionaries.cc\ src/input.h src/input.cc src/sql_utilities.h src/sql_utilities.cc src/renz.h src/renz.cc \ src/MetaPopInfo.h src/MetaPopInfo.cc populations_SOURCES = src/populations.h src/populations.cc src/constants.h \ src/utils.h src/utils.cc src/catalog_utils.h src/catalog_utils.cc \ src/log_utils.h src/log_utils.cc \ src/stacks.h src/stacks.cc src/locus.h src/locus.cc \ src/DNASeq.h src/DNASeq.cc src/DNANSeq.h src/DNANSeq.cc \ src/PopMap.h src/PopSum.h src/aln_utils.h src/aln_utils.cc src/genotype_dictionaries.h src/genotype_dictionaries.cc \ src/input.h src/input.cc src/sql_utilities.h src/sql_utilities.cc src/renz.h src/renz.cc \ src/bootstrap.h src/ordered.h src/smoothing.h src/smoothing_utils.h \ src/export_formats.h src/export_formats.cc \ src/MetaPopInfo.h src/MetaPopInfo.cc src/Vcf.h src/Vcf.cc phasedstacks_SOURCES = src/phasedstacks.h src/phasedstacks.cc src/constants.h \ src/locus.h src/locus.cc \ src/input.h src/input.cc src/sql_utilities.h src/sql_utilities.cc \ src/log_utils.h src/log_utils.cc \ src/utils.h src/utils.cc src/catalog_utils.h src/catalog_utils.cc ustacks_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) ustacks_LDFLAGS = $(OPENMP_CFLAGS) pstacks_CXXFLAGS = $(OPENMP_CFLAGS) $(BAM_CFLAGS) $(SPARSEHASH_CFLAGS) pstacks_LDFLAGS = $(OPENMP_CFLAGS) pstacks_LDADD = $(BAM_LIBS) estacks_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) estacks_LDFLAGS = $(OPENMP_CFLAGS) cstacks_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) cstacks_LDFLAGS = $(OPENMP_CFLAGS) hstacks_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) hstacks_LDFLAGS = $(OPENMP_CFLAGS) sstacks_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) sstacks_LDFLAGS = $(OPENMP_CFLAGS) rxstacks_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) rxstacks_LDFLAGS = $(OPENMP_CFLAGS) process_radtags_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) $(BAM_CFLAGS) process_radtags_LDFLAGS = $(OPENMP_CFLAGS) process_radtags_LDADD = $(BAM_LIBS) process_shortreads_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) $(BAM_CFLAGS) process_shortreads_LDFLAGS = $(OPENMP_CFLAGS) process_shortreads_LDADD = $(BAM_LIBS) kmer_filter_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) kmer_filter_LDFLAGS = $(OPENMP_CFLAGS) genotypes_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) genotypes_LDFLAGS = $(OPENMP_CFLAGS) populations_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) populations_LDFLAGS = $(OPENMP_CFLAGS) phasedstacks_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) phasedstacks_LDFLAGS = $(OPENMP_CFLAGS) clone_filter_CXXFLAGS = $(OPENMP_CFLAGS) $(SPARSEHASH_CFLAGS) $(BAM_CFLAGS) clone_filter_LDFLAGS = $(OPENMP_CFLAGS) clone_filter_LDADD = $(BAM_LIBS) dist_bin_SCRIPTS = scripts/denovo_map.pl scripts/ref_map.pl scripts/export_sql.pl \ scripts/sort_read_pairs.pl scripts/exec_velvet.pl scripts/load_sequences.pl \ scripts/index_radtags.pl scripts/load_radtags.pl scripts/stacks_export_notify.pl \ scripts/integrate_alignments.py dist_noinst_SCRIPTS = autogen.sh scripts/extract_interpop_chars.pl scripts/convert_stacks.pl nobase_pkgdata_DATA = sql/mysql.cnf.dist sql/catalog_index.sql sql/stacks.sql sql/tag_index.sql sql/chr_index.sql \ php/CatalogClass.php php/annotate_marker.php php/constants.php.dist php/index.php php/tags.php php/tag.php \ php/Locus.php php/catalog.php php/correct_genotypes.php php/correct_genotype.php php/export_batch.php php/last_modified.php \ php/version.php php/catalog_genotypes.php php/db_functions.php php/header.php php/samples.php \ php/stacks_functions.php php/view_sequence.php php/sequence_blast.php \ php/pop_view.php php/sumstat_view.php php/hapstat_view.php php/fst_view.php php/phist_view.php php/stack_view.php \ php/population_view.js php/ajax.js php/annotate.js php/stacks.js php/export.js php/stacks.css \ php/images/caret-d.png php/images/caret-u.png php/images/excel_icon.png php/images/l-arrow-disabled.png \ php/images/l-arrow.png php/images/r-arrow-disabled.png php/images/r-arrow.png php/images/stacks_bg.png \ php/images/stacks_logo_rev_small.png LOG_DRIVER = env AM_TAP_AWK='$(AWK)' $(SHELL) $(abs_top_srcdir)/tests/tap-driver.sh TESTS = tests/process_radtags.t tests/kmer_filter.t tests/ustacks.t tests/pstacks.t EXTRA_DIST = $(nobase_pkgdata_DATA) LICENSE INSTALL README ChangeLog $(TESTS) SUBDIRS = htslib . pkglocalstatedir = $(localstatedir)/$(PACKAGE) all: config.h $(MAKE) $(AM_MAKEFLAGS) all-recursive .SUFFIXES: .SUFFIXES: .cc .log .o .obj .test .test$(EXEEXT) .trs am--refresh: Makefile @: $(srcdir)/Makefile.in: $(srcdir)/Makefile.am $(am__configure_deps) @for dep in $?; do \ case '$(am__configure_deps)' in \ *$$dep*) \ echo ' cd $(srcdir) && $(AUTOMAKE) --foreign'; \ $(am__cd) $(srcdir) && $(AUTOMAKE) --foreign \ && exit 0; \ exit 1;; \ esac; \ done; \ echo ' cd $(top_srcdir) && $(AUTOMAKE) --foreign Makefile'; \ $(am__cd) $(top_srcdir) && \ $(AUTOMAKE) --foreign Makefile Makefile: $(srcdir)/Makefile.in $(top_builddir)/config.status @case '$?' in \ *config.status*) \ echo ' $(SHELL) ./config.status'; \ $(SHELL) ./config.status;; \ *) \ echo ' cd $(top_builddir) && $(SHELL) ./config.status $@ $(am__depfiles_maybe)'; \ cd $(top_builddir) && $(SHELL) ./config.status $@ $(am__depfiles_maybe);; \ esac; $(top_builddir)/config.status: $(top_srcdir)/configure $(CONFIG_STATUS_DEPENDENCIES) $(SHELL) ./config.status --recheck $(top_srcdir)/configure: $(am__configure_deps) $(am__cd) $(srcdir) && $(AUTOCONF) $(ACLOCAL_M4): $(am__aclocal_m4_deps) $(am__cd) $(srcdir) && $(ACLOCAL) $(ACLOCAL_AMFLAGS) $(am__aclocal_m4_deps): config.h: stamp-h1 @test -f $@ || rm -f stamp-h1 @test -f $@ || $(MAKE) $(AM_MAKEFLAGS) stamp-h1 stamp-h1: $(srcdir)/config.h.in $(top_builddir)/config.status @rm -f stamp-h1 cd $(top_builddir) && $(SHELL) ./config.status config.h $(srcdir)/config.h.in: $(am__configure_deps) ($(am__cd) $(top_srcdir) && $(AUTOHEADER)) rm -f stamp-h1 touch $@ distclean-hdr: -rm -f config.h stamp-h1 install-binPROGRAMS: $(bin_PROGRAMS) @$(NORMAL_INSTALL) @list='$(bin_PROGRAMS)'; test -n "$(bindir)" || list=; \ if test -n "$$list"; then \ echo " $(MKDIR_P) '$(DESTDIR)$(bindir)'"; \ $(MKDIR_P) "$(DESTDIR)$(bindir)" || exit 1; \ fi; \ for p in $$list; do echo "$$p $$p"; done | \ sed 's/$(EXEEXT)$$//' | \ while read p p1; do if test -f $$p \ ; then echo "$$p"; echo "$$p"; else :; fi; \ done | \ sed -e 'p;s,.*/,,;n;h' \ -e 's|.*|.|' \ -e 'p;x;s,.*/,,;s/$(EXEEXT)$$//;$(transform);s/$$/$(EXEEXT)/' | \ sed 'N;N;N;s,\n, ,g' | \ $(AWK) 'BEGIN { files["."] = ""; dirs["."] = 1 } \ { d=$$3; if (dirs[d] != 1) { print "d", d; dirs[d] = 1 } \ if ($$2 == $$4) files[d] = files[d] " " $$1; \ else { print "f", $$3 "/" $$4, $$1; } } \ END { for (d in files) print "f", d, files[d] }' | \ while read type dir files; do \ if test "$$dir" = .; then dir=; else dir=/$$dir; fi; \ test -z "$$files" || { \ echo " $(INSTALL_PROGRAM_ENV) $(INSTALL_PROGRAM) $$files '$(DESTDIR)$(bindir)$$dir'"; \ $(INSTALL_PROGRAM_ENV) $(INSTALL_PROGRAM) $$files "$(DESTDIR)$(bindir)$$dir" || exit $$?; \ } \ ; done uninstall-binPROGRAMS: @$(NORMAL_UNINSTALL) @list='$(bin_PROGRAMS)'; test -n "$(bindir)" || list=; \ files=`for p in $$list; do echo "$$p"; done | \ sed -e 'h;s,^.*/,,;s/$(EXEEXT)$$//;$(transform)' \ -e 's/$$/$(EXEEXT)/' \ `; \ test -n "$$list" || exit 0; \ echo " ( cd '$(DESTDIR)$(bindir)' && rm -f" $$files ")"; \ cd "$(DESTDIR)$(bindir)" && rm -f $$files clean-binPROGRAMS: -test -z "$(bin_PROGRAMS)" || rm -f $(bin_PROGRAMS) src/$(am__dirstamp): @$(MKDIR_P) src @: > src/$(am__dirstamp) src/$(DEPDIR)/$(am__dirstamp): @$(MKDIR_P) src/$(DEPDIR) @: > src/$(DEPDIR)/$(am__dirstamp) src/clone_filter-clone_filter.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/clone_filter-kmers.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/clone_filter-utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/clone_filter-write.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/clone_filter-clean.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/clone_filter-file_io.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/clone_filter-input.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) clone_filter$(EXEEXT): $(clone_filter_OBJECTS) $(clone_filter_DEPENDENCIES) $(EXTRA_clone_filter_DEPENDENCIES) @rm -f clone_filter$(EXEEXT) $(AM_V_CXXLD)$(clone_filter_LINK) $(clone_filter_OBJECTS) $(clone_filter_LDADD) $(LIBS) src/cstacks-cstacks.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/cstacks-stacks.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/cstacks-locus.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/cstacks-kmers.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/cstacks-utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/cstacks-aln_utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/cstacks-DNASeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/cstacks-DNANSeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/cstacks-input.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/cstacks-sql_utilities.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) cstacks$(EXEEXT): $(cstacks_OBJECTS) $(cstacks_DEPENDENCIES) $(EXTRA_cstacks_DEPENDENCIES) @rm -f cstacks$(EXEEXT) $(AM_V_CXXLD)$(cstacks_LINK) $(cstacks_OBJECTS) $(cstacks_LDADD) $(LIBS) src/estacks-estacks.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/estacks-stacks.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/estacks-mstack.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/estacks-models.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/estacks-utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/estacks-DNASeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/estacks-DNANSeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/estacks-input.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) estacks$(EXEEXT): $(estacks_OBJECTS) $(estacks_DEPENDENCIES) $(EXTRA_estacks_DEPENDENCIES) @rm -f estacks$(EXEEXT) $(AM_V_CXXLD)$(estacks_LINK) $(estacks_OBJECTS) $(estacks_LDADD) $(LIBS) src/genotypes-genotypes.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/genotypes-utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/genotypes-catalog_utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/genotypes-log_utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/genotypes-stacks.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/genotypes-locus.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/genotypes-DNASeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/genotypes-DNANSeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/genotypes-aln_utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/genotypes-genotype_dictionaries.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/genotypes-input.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/genotypes-sql_utilities.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/genotypes-renz.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/genotypes-MetaPopInfo.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) genotypes$(EXEEXT): $(genotypes_OBJECTS) $(genotypes_DEPENDENCIES) $(EXTRA_genotypes_DEPENDENCIES) @rm -f genotypes$(EXEEXT) $(AM_V_CXXLD)$(genotypes_LINK) $(genotypes_OBJECTS) $(genotypes_LDADD) $(LIBS) src/hstacks-hstacks.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/hstacks-stacks.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/hstacks-locus.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/hstacks-kmers.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/hstacks-DNASeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/hstacks-DNANSeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/hstacks-utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/hstacks-input.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/hstacks-sql_utilities.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) hstacks$(EXEEXT): $(hstacks_OBJECTS) $(hstacks_DEPENDENCIES) $(EXTRA_hstacks_DEPENDENCIES) @rm -f hstacks$(EXEEXT) $(AM_V_CXXLD)$(hstacks_LINK) $(hstacks_OBJECTS) $(hstacks_LDADD) $(LIBS) src/kmer_filter-kmer_filter.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/kmer_filter-utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/kmer_filter-write.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/kmer_filter-kmers.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/kmer_filter-input.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) kmer_filter$(EXEEXT): $(kmer_filter_OBJECTS) $(kmer_filter_DEPENDENCIES) $(EXTRA_kmer_filter_DEPENDENCIES) @rm -f kmer_filter$(EXEEXT) $(AM_V_CXXLD)$(kmer_filter_LINK) $(kmer_filter_OBJECTS) $(kmer_filter_LDADD) $(LIBS) src/phasedstacks-phasedstacks.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/phasedstacks-locus.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/phasedstacks-input.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/phasedstacks-sql_utilities.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/phasedstacks-log_utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/phasedstacks-utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/phasedstacks-catalog_utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) phasedstacks$(EXEEXT): $(phasedstacks_OBJECTS) $(phasedstacks_DEPENDENCIES) $(EXTRA_phasedstacks_DEPENDENCIES) @rm -f phasedstacks$(EXEEXT) $(AM_V_CXXLD)$(phasedstacks_LINK) $(phasedstacks_OBJECTS) $(phasedstacks_LDADD) $(LIBS) src/populations-populations.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/populations-utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/populations-catalog_utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/populations-log_utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/populations-stacks.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/populations-locus.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/populations-DNASeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/populations-DNANSeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/populations-aln_utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/populations-genotype_dictionaries.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/populations-input.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/populations-sql_utilities.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/populations-renz.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/populations-export_formats.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/populations-MetaPopInfo.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/populations-Vcf.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) populations$(EXEEXT): $(populations_OBJECTS) $(populations_DEPENDENCIES) $(EXTRA_populations_DEPENDENCIES) @rm -f populations$(EXEEXT) $(AM_V_CXXLD)$(populations_LINK) $(populations_OBJECTS) $(populations_LDADD) $(LIBS) src/process_radtags-process_radtags.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/process_radtags-utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/process_radtags-log_utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/process_radtags-write.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/process_radtags-clean.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/process_radtags-file_io.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/process_radtags-input.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/process_radtags-renz.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) process_radtags$(EXEEXT): $(process_radtags_OBJECTS) $(process_radtags_DEPENDENCIES) $(EXTRA_process_radtags_DEPENDENCIES) @rm -f process_radtags$(EXEEXT) $(AM_V_CXXLD)$(process_radtags_LINK) $(process_radtags_OBJECTS) $(process_radtags_LDADD) $(LIBS) src/process_shortreads-process_shortreads.$(OBJEXT): \ src/$(am__dirstamp) src/$(DEPDIR)/$(am__dirstamp) src/process_shortreads-clean.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/process_shortreads-file_io.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/process_shortreads-utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/process_shortreads-log_utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/process_shortreads-write.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/process_shortreads-kmers.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/process_shortreads-input.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) process_shortreads$(EXEEXT): $(process_shortreads_OBJECTS) $(process_shortreads_DEPENDENCIES) $(EXTRA_process_shortreads_DEPENDENCIES) @rm -f process_shortreads$(EXEEXT) $(AM_V_CXXLD)$(process_shortreads_LINK) $(process_shortreads_OBJECTS) $(process_shortreads_LDADD) $(LIBS) src/pstacks-pstacks.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/pstacks-stacks.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/pstacks-mstack.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/pstacks-DNANSeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/pstacks-DNASeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/pstacks-models.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/pstacks-utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/pstacks-input.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) pstacks$(EXEEXT): $(pstacks_OBJECTS) $(pstacks_DEPENDENCIES) $(EXTRA_pstacks_DEPENDENCIES) @rm -f pstacks$(EXEEXT) $(AM_V_CXXLD)$(pstacks_LINK) $(pstacks_OBJECTS) $(pstacks_LDADD) $(LIBS) src/rxstacks-rxstacks.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/rxstacks-stacks.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/rxstacks-locus.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/rxstacks-DNANSeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/rxstacks-DNASeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/rxstacks-mst.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/rxstacks-models.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/rxstacks-aln_utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/rxstacks-utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/rxstacks-input.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/rxstacks-sql_utilities.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/rxstacks-MetaPopInfo.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) rxstacks$(EXEEXT): $(rxstacks_OBJECTS) $(rxstacks_DEPENDENCIES) $(EXTRA_rxstacks_DEPENDENCIES) @rm -f rxstacks$(EXEEXT) $(AM_V_CXXLD)$(rxstacks_LINK) $(rxstacks_OBJECTS) $(rxstacks_LDADD) $(LIBS) src/sstacks-sstacks.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/sstacks-stacks.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/sstacks-locus.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/sstacks-kmers.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/sstacks-aln_utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/sstacks-DNASeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/sstacks-DNANSeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/sstacks-utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/sstacks-input.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/sstacks-sql_utilities.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) sstacks$(EXEEXT): $(sstacks_OBJECTS) $(sstacks_DEPENDENCIES) $(EXTRA_sstacks_DEPENDENCIES) @rm -f sstacks$(EXEEXT) $(AM_V_CXXLD)$(sstacks_LINK) $(sstacks_OBJECTS) $(sstacks_LDADD) $(LIBS) src/ustacks-ustacks.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/ustacks-stacks.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/ustacks-mstack.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/ustacks-mst.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/ustacks-cmb.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/ustacks-DNASeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/ustacks-DNANSeq.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/ustacks-models.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/ustacks-utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/ustacks-kmers.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/ustacks-aln_utils.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) src/ustacks-input.$(OBJEXT): src/$(am__dirstamp) \ src/$(DEPDIR)/$(am__dirstamp) ustacks$(EXEEXT): $(ustacks_OBJECTS) $(ustacks_DEPENDENCIES) $(EXTRA_ustacks_DEPENDENCIES) @rm -f ustacks$(EXEEXT) $(AM_V_CXXLD)$(ustacks_LINK) $(ustacks_OBJECTS) $(ustacks_LDADD) $(LIBS) install-dist_binSCRIPTS: $(dist_bin_SCRIPTS) @$(NORMAL_INSTALL) @list='$(dist_bin_SCRIPTS)'; test -n "$(bindir)" || list=; \ if test -n "$$list"; then \ echo " $(MKDIR_P) '$(DESTDIR)$(bindir)'"; \ $(MKDIR_P) "$(DESTDIR)$(bindir)" || exit 1; \ fi; \ for p in $$list; do \ if test -f "$$p"; then d=; else d="$(srcdir)/"; fi; \ if test -f "$$d$$p"; then echo "$$d$$p"; echo "$$p"; else :; fi; \ done | \ sed -e 'p;s,.*/,,;n' \ -e 'h;s|.*|.|' \ -e 'p;x;s,.*/,,;$(transform)' | sed 'N;N;N;s,\n, ,g' | \ $(AWK) 'BEGIN { files["."] = ""; dirs["."] = 1; } \ { d=$$3; if (dirs[d] != 1) { print "d", d; dirs[d] = 1 } \ if ($$2 == $$4) { files[d] = files[d] " " $$1; \ if (++n[d] == $(am__install_max)) { \ print "f", d, files[d]; n[d] = 0; files[d] = "" } } \ else { print "f", d "/" $$4, $$1 } } \ END { for (d in files) print "f", d, files[d] }' | \ while read type dir files; do \ if test "$$dir" = .; then dir=; else dir=/$$dir; fi; \ test -z "$$files" || { \ echo " $(INSTALL_SCRIPT) $$files '$(DESTDIR)$(bindir)$$dir'"; \ $(INSTALL_SCRIPT) $$files "$(DESTDIR)$(bindir)$$dir" || exit $$?; \ } \ ; done uninstall-dist_binSCRIPTS: @$(NORMAL_UNINSTALL) @list='$(dist_bin_SCRIPTS)'; test -n "$(bindir)" || exit 0; \ files=`for p in $$list; do echo "$$p"; done | \ sed -e 's,.*/,,;$(transform)'`; \ dir='$(DESTDIR)$(bindir)'; $(am__uninstall_files_from_dir) mostlyclean-compile: -rm -f *.$(OBJEXT) -rm -f src/*.$(OBJEXT) distclean-compile: -rm -f *.tab.c @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/clone_filter-clean.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/clone_filter-clone_filter.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/clone_filter-file_io.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/clone_filter-input.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/clone_filter-kmers.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/clone_filter-utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/clone_filter-write.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/cstacks-DNANSeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/cstacks-DNASeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/cstacks-aln_utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/cstacks-cstacks.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/cstacks-input.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/cstacks-kmers.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/cstacks-locus.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/cstacks-sql_utilities.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/cstacks-stacks.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/cstacks-utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/estacks-DNANSeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/estacks-DNASeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/estacks-estacks.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/estacks-input.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/estacks-models.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/estacks-mstack.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/estacks-stacks.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/estacks-utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/genotypes-DNANSeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/genotypes-DNASeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/genotypes-MetaPopInfo.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/genotypes-aln_utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/genotypes-catalog_utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/genotypes-genotype_dictionaries.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/genotypes-genotypes.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/genotypes-input.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/genotypes-locus.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/genotypes-log_utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/genotypes-renz.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/genotypes-sql_utilities.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/genotypes-stacks.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/genotypes-utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/hstacks-DNANSeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/hstacks-DNASeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/hstacks-hstacks.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/hstacks-input.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/hstacks-kmers.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/hstacks-locus.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/hstacks-sql_utilities.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/hstacks-stacks.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/hstacks-utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/kmer_filter-input.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/kmer_filter-kmer_filter.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/kmer_filter-kmers.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/kmer_filter-utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/kmer_filter-write.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/phasedstacks-catalog_utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/phasedstacks-input.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/phasedstacks-locus.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/phasedstacks-log_utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/phasedstacks-phasedstacks.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/phasedstacks-sql_utilities.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/phasedstacks-utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/populations-DNANSeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/populations-DNASeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/populations-MetaPopInfo.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/populations-Vcf.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/populations-aln_utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/populations-catalog_utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/populations-export_formats.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/populations-genotype_dictionaries.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/populations-input.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/populations-locus.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/populations-log_utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/populations-populations.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/populations-renz.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/populations-sql_utilities.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/populations-stacks.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/populations-utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/process_radtags-clean.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/process_radtags-file_io.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/process_radtags-input.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/process_radtags-log_utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/process_radtags-process_radtags.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/process_radtags-renz.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/process_radtags-utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/process_radtags-write.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/process_shortreads-clean.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/process_shortreads-file_io.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/process_shortreads-input.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/process_shortreads-kmers.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/process_shortreads-log_utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/process_shortreads-process_shortreads.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/process_shortreads-utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/process_shortreads-write.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/pstacks-DNANSeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/pstacks-DNASeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/pstacks-input.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/pstacks-models.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/pstacks-mstack.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/pstacks-pstacks.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/pstacks-stacks.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/pstacks-utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/rxstacks-DNANSeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/rxstacks-DNASeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/rxstacks-MetaPopInfo.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/rxstacks-aln_utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/rxstacks-input.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/rxstacks-locus.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/rxstacks-models.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/rxstacks-mst.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/rxstacks-rxstacks.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/rxstacks-sql_utilities.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/rxstacks-stacks.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/rxstacks-utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/sstacks-DNANSeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/sstacks-DNASeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/sstacks-aln_utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/sstacks-input.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/sstacks-kmers.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/sstacks-locus.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/sstacks-sql_utilities.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/sstacks-sstacks.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/sstacks-stacks.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/sstacks-utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/ustacks-DNANSeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/ustacks-DNASeq.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/ustacks-aln_utils.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/ustacks-cmb.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/ustacks-input.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/ustacks-kmers.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/ustacks-models.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/ustacks-mst.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/ustacks-mstack.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/ustacks-stacks.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/ustacks-ustacks.Po@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/ustacks-utils.Po@am__quote@ .cc.o: @am__fastdepCXX_TRUE@ $(AM_V_CXX)depbase=`echo $@ | sed 's|[^/]*$$|$(DEPDIR)/&|;s|\.o$$||'`;\ @am__fastdepCXX_TRUE@ $(CXXCOMPILE) -MT $@ -MD -MP -MF $$depbase.Tpo -c -o $@ $< &&\ @am__fastdepCXX_TRUE@ $(am__mv) $$depbase.Tpo $$depbase.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='$<' object='$@' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXXCOMPILE) -c -o $@ $< .cc.obj: @am__fastdepCXX_TRUE@ $(AM_V_CXX)depbase=`echo $@ | sed 's|[^/]*$$|$(DEPDIR)/&|;s|\.obj$$||'`;\ @am__fastdepCXX_TRUE@ $(CXXCOMPILE) -MT $@ -MD -MP -MF $$depbase.Tpo -c -o $@ `$(CYGPATH_W) '$<'` &&\ @am__fastdepCXX_TRUE@ $(am__mv) $$depbase.Tpo $$depbase.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='$<' object='$@' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXXCOMPILE) -c -o $@ `$(CYGPATH_W) '$<'` src/clone_filter-clone_filter.o: src/clone_filter.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -MT src/clone_filter-clone_filter.o -MD -MP -MF src/$(DEPDIR)/clone_filter-clone_filter.Tpo -c -o src/clone_filter-clone_filter.o `test -f 'src/clone_filter.cc' || echo '$(srcdir)/'`src/clone_filter.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/clone_filter-clone_filter.Tpo src/$(DEPDIR)/clone_filter-clone_filter.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/clone_filter.cc' object='src/clone_filter-clone_filter.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/clone_filter-clone_filter.o `test -f 'src/clone_filter.cc' || echo '$(srcdir)/'`src/clone_filter.cc src/clone_filter-clone_filter.obj: src/clone_filter.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -MT src/clone_filter-clone_filter.obj -MD -MP -MF src/$(DEPDIR)/clone_filter-clone_filter.Tpo -c -o src/clone_filter-clone_filter.obj `if test -f 'src/clone_filter.cc'; then $(CYGPATH_W) 'src/clone_filter.cc'; else $(CYGPATH_W) '$(srcdir)/src/clone_filter.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/clone_filter-clone_filter.Tpo src/$(DEPDIR)/clone_filter-clone_filter.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/clone_filter.cc' object='src/clone_filter-clone_filter.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/clone_filter-clone_filter.obj `if test -f 'src/clone_filter.cc'; then $(CYGPATH_W) 'src/clone_filter.cc'; else $(CYGPATH_W) '$(srcdir)/src/clone_filter.cc'; fi` src/clone_filter-kmers.o: src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -MT src/clone_filter-kmers.o -MD -MP -MF src/$(DEPDIR)/clone_filter-kmers.Tpo -c -o src/clone_filter-kmers.o `test -f 'src/kmers.cc' || echo '$(srcdir)/'`src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/clone_filter-kmers.Tpo src/$(DEPDIR)/clone_filter-kmers.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/kmers.cc' object='src/clone_filter-kmers.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/clone_filter-kmers.o `test -f 'src/kmers.cc' || echo '$(srcdir)/'`src/kmers.cc src/clone_filter-kmers.obj: src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -MT src/clone_filter-kmers.obj -MD -MP -MF src/$(DEPDIR)/clone_filter-kmers.Tpo -c -o src/clone_filter-kmers.obj `if test -f 'src/kmers.cc'; then $(CYGPATH_W) 'src/kmers.cc'; else $(CYGPATH_W) '$(srcdir)/src/kmers.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/clone_filter-kmers.Tpo src/$(DEPDIR)/clone_filter-kmers.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/kmers.cc' object='src/clone_filter-kmers.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/clone_filter-kmers.obj `if test -f 'src/kmers.cc'; then $(CYGPATH_W) 'src/kmers.cc'; else $(CYGPATH_W) '$(srcdir)/src/kmers.cc'; fi` src/clone_filter-utils.o: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -MT src/clone_filter-utils.o -MD -MP -MF src/$(DEPDIR)/clone_filter-utils.Tpo -c -o src/clone_filter-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/clone_filter-utils.Tpo src/$(DEPDIR)/clone_filter-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/clone_filter-utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/clone_filter-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc src/clone_filter-utils.obj: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -MT src/clone_filter-utils.obj -MD -MP -MF src/$(DEPDIR)/clone_filter-utils.Tpo -c -o src/clone_filter-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/clone_filter-utils.Tpo src/$(DEPDIR)/clone_filter-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/clone_filter-utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/clone_filter-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` src/clone_filter-write.o: src/write.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -MT src/clone_filter-write.o -MD -MP -MF src/$(DEPDIR)/clone_filter-write.Tpo -c -o src/clone_filter-write.o `test -f 'src/write.cc' || echo '$(srcdir)/'`src/write.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/clone_filter-write.Tpo src/$(DEPDIR)/clone_filter-write.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/write.cc' object='src/clone_filter-write.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/clone_filter-write.o `test -f 'src/write.cc' || echo '$(srcdir)/'`src/write.cc src/clone_filter-write.obj: src/write.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -MT src/clone_filter-write.obj -MD -MP -MF src/$(DEPDIR)/clone_filter-write.Tpo -c -o src/clone_filter-write.obj `if test -f 'src/write.cc'; then $(CYGPATH_W) 'src/write.cc'; else $(CYGPATH_W) '$(srcdir)/src/write.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/clone_filter-write.Tpo src/$(DEPDIR)/clone_filter-write.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/write.cc' object='src/clone_filter-write.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/clone_filter-write.obj `if test -f 'src/write.cc'; then $(CYGPATH_W) 'src/write.cc'; else $(CYGPATH_W) '$(srcdir)/src/write.cc'; fi` src/clone_filter-clean.o: src/clean.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -MT src/clone_filter-clean.o -MD -MP -MF src/$(DEPDIR)/clone_filter-clean.Tpo -c -o src/clone_filter-clean.o `test -f 'src/clean.cc' || echo '$(srcdir)/'`src/clean.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/clone_filter-clean.Tpo src/$(DEPDIR)/clone_filter-clean.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/clean.cc' object='src/clone_filter-clean.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/clone_filter-clean.o `test -f 'src/clean.cc' || echo '$(srcdir)/'`src/clean.cc src/clone_filter-clean.obj: src/clean.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -MT src/clone_filter-clean.obj -MD -MP -MF src/$(DEPDIR)/clone_filter-clean.Tpo -c -o src/clone_filter-clean.obj `if test -f 'src/clean.cc'; then $(CYGPATH_W) 'src/clean.cc'; else $(CYGPATH_W) '$(srcdir)/src/clean.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/clone_filter-clean.Tpo src/$(DEPDIR)/clone_filter-clean.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/clean.cc' object='src/clone_filter-clean.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/clone_filter-clean.obj `if test -f 'src/clean.cc'; then $(CYGPATH_W) 'src/clean.cc'; else $(CYGPATH_W) '$(srcdir)/src/clean.cc'; fi` src/clone_filter-file_io.o: src/file_io.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -MT src/clone_filter-file_io.o -MD -MP -MF src/$(DEPDIR)/clone_filter-file_io.Tpo -c -o src/clone_filter-file_io.o `test -f 'src/file_io.cc' || echo '$(srcdir)/'`src/file_io.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/clone_filter-file_io.Tpo src/$(DEPDIR)/clone_filter-file_io.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/file_io.cc' object='src/clone_filter-file_io.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/clone_filter-file_io.o `test -f 'src/file_io.cc' || echo '$(srcdir)/'`src/file_io.cc src/clone_filter-file_io.obj: src/file_io.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -MT src/clone_filter-file_io.obj -MD -MP -MF src/$(DEPDIR)/clone_filter-file_io.Tpo -c -o src/clone_filter-file_io.obj `if test -f 'src/file_io.cc'; then $(CYGPATH_W) 'src/file_io.cc'; else $(CYGPATH_W) '$(srcdir)/src/file_io.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/clone_filter-file_io.Tpo src/$(DEPDIR)/clone_filter-file_io.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/file_io.cc' object='src/clone_filter-file_io.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/clone_filter-file_io.obj `if test -f 'src/file_io.cc'; then $(CYGPATH_W) 'src/file_io.cc'; else $(CYGPATH_W) '$(srcdir)/src/file_io.cc'; fi` src/clone_filter-input.o: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -MT src/clone_filter-input.o -MD -MP -MF src/$(DEPDIR)/clone_filter-input.Tpo -c -o src/clone_filter-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/clone_filter-input.Tpo src/$(DEPDIR)/clone_filter-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/clone_filter-input.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/clone_filter-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc src/clone_filter-input.obj: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -MT src/clone_filter-input.obj -MD -MP -MF src/$(DEPDIR)/clone_filter-input.Tpo -c -o src/clone_filter-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/clone_filter-input.Tpo src/$(DEPDIR)/clone_filter-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/clone_filter-input.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(clone_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/clone_filter-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` src/cstacks-cstacks.o: src/cstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-cstacks.o -MD -MP -MF src/$(DEPDIR)/cstacks-cstacks.Tpo -c -o src/cstacks-cstacks.o `test -f 'src/cstacks.cc' || echo '$(srcdir)/'`src/cstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-cstacks.Tpo src/$(DEPDIR)/cstacks-cstacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/cstacks.cc' object='src/cstacks-cstacks.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-cstacks.o `test -f 'src/cstacks.cc' || echo '$(srcdir)/'`src/cstacks.cc src/cstacks-cstacks.obj: src/cstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-cstacks.obj -MD -MP -MF src/$(DEPDIR)/cstacks-cstacks.Tpo -c -o src/cstacks-cstacks.obj `if test -f 'src/cstacks.cc'; then $(CYGPATH_W) 'src/cstacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/cstacks.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-cstacks.Tpo src/$(DEPDIR)/cstacks-cstacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/cstacks.cc' object='src/cstacks-cstacks.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-cstacks.obj `if test -f 'src/cstacks.cc'; then $(CYGPATH_W) 'src/cstacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/cstacks.cc'; fi` src/cstacks-stacks.o: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-stacks.o -MD -MP -MF src/$(DEPDIR)/cstacks-stacks.Tpo -c -o src/cstacks-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-stacks.Tpo src/$(DEPDIR)/cstacks-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/cstacks-stacks.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc src/cstacks-stacks.obj: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-stacks.obj -MD -MP -MF src/$(DEPDIR)/cstacks-stacks.Tpo -c -o src/cstacks-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-stacks.Tpo src/$(DEPDIR)/cstacks-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/cstacks-stacks.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` src/cstacks-locus.o: src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-locus.o -MD -MP -MF src/$(DEPDIR)/cstacks-locus.Tpo -c -o src/cstacks-locus.o `test -f 'src/locus.cc' || echo '$(srcdir)/'`src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-locus.Tpo src/$(DEPDIR)/cstacks-locus.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/locus.cc' object='src/cstacks-locus.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-locus.o `test -f 'src/locus.cc' || echo '$(srcdir)/'`src/locus.cc src/cstacks-locus.obj: src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-locus.obj -MD -MP -MF src/$(DEPDIR)/cstacks-locus.Tpo -c -o src/cstacks-locus.obj `if test -f 'src/locus.cc'; then $(CYGPATH_W) 'src/locus.cc'; else $(CYGPATH_W) '$(srcdir)/src/locus.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-locus.Tpo src/$(DEPDIR)/cstacks-locus.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/locus.cc' object='src/cstacks-locus.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-locus.obj `if test -f 'src/locus.cc'; then $(CYGPATH_W) 'src/locus.cc'; else $(CYGPATH_W) '$(srcdir)/src/locus.cc'; fi` src/cstacks-kmers.o: src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-kmers.o -MD -MP -MF src/$(DEPDIR)/cstacks-kmers.Tpo -c -o src/cstacks-kmers.o `test -f 'src/kmers.cc' || echo '$(srcdir)/'`src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-kmers.Tpo src/$(DEPDIR)/cstacks-kmers.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/kmers.cc' object='src/cstacks-kmers.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-kmers.o `test -f 'src/kmers.cc' || echo '$(srcdir)/'`src/kmers.cc src/cstacks-kmers.obj: src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-kmers.obj -MD -MP -MF src/$(DEPDIR)/cstacks-kmers.Tpo -c -o src/cstacks-kmers.obj `if test -f 'src/kmers.cc'; then $(CYGPATH_W) 'src/kmers.cc'; else $(CYGPATH_W) '$(srcdir)/src/kmers.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-kmers.Tpo src/$(DEPDIR)/cstacks-kmers.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/kmers.cc' object='src/cstacks-kmers.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-kmers.obj `if test -f 'src/kmers.cc'; then $(CYGPATH_W) 'src/kmers.cc'; else $(CYGPATH_W) '$(srcdir)/src/kmers.cc'; fi` src/cstacks-utils.o: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-utils.o -MD -MP -MF src/$(DEPDIR)/cstacks-utils.Tpo -c -o src/cstacks-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-utils.Tpo src/$(DEPDIR)/cstacks-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/cstacks-utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc src/cstacks-utils.obj: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-utils.obj -MD -MP -MF src/$(DEPDIR)/cstacks-utils.Tpo -c -o src/cstacks-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-utils.Tpo src/$(DEPDIR)/cstacks-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/cstacks-utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` src/cstacks-aln_utils.o: src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-aln_utils.o -MD -MP -MF src/$(DEPDIR)/cstacks-aln_utils.Tpo -c -o src/cstacks-aln_utils.o `test -f 'src/aln_utils.cc' || echo '$(srcdir)/'`src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-aln_utils.Tpo src/$(DEPDIR)/cstacks-aln_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/aln_utils.cc' object='src/cstacks-aln_utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-aln_utils.o `test -f 'src/aln_utils.cc' || echo '$(srcdir)/'`src/aln_utils.cc src/cstacks-aln_utils.obj: src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-aln_utils.obj -MD -MP -MF src/$(DEPDIR)/cstacks-aln_utils.Tpo -c -o src/cstacks-aln_utils.obj `if test -f 'src/aln_utils.cc'; then $(CYGPATH_W) 'src/aln_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/aln_utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-aln_utils.Tpo src/$(DEPDIR)/cstacks-aln_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/aln_utils.cc' object='src/cstacks-aln_utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-aln_utils.obj `if test -f 'src/aln_utils.cc'; then $(CYGPATH_W) 'src/aln_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/aln_utils.cc'; fi` src/cstacks-DNASeq.o: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-DNASeq.o -MD -MP -MF src/$(DEPDIR)/cstacks-DNASeq.Tpo -c -o src/cstacks-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-DNASeq.Tpo src/$(DEPDIR)/cstacks-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/cstacks-DNASeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc src/cstacks-DNASeq.obj: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-DNASeq.obj -MD -MP -MF src/$(DEPDIR)/cstacks-DNASeq.Tpo -c -o src/cstacks-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-DNASeq.Tpo src/$(DEPDIR)/cstacks-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/cstacks-DNASeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` src/cstacks-DNANSeq.o: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-DNANSeq.o -MD -MP -MF src/$(DEPDIR)/cstacks-DNANSeq.Tpo -c -o src/cstacks-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-DNANSeq.Tpo src/$(DEPDIR)/cstacks-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/cstacks-DNANSeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc src/cstacks-DNANSeq.obj: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-DNANSeq.obj -MD -MP -MF src/$(DEPDIR)/cstacks-DNANSeq.Tpo -c -o src/cstacks-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-DNANSeq.Tpo src/$(DEPDIR)/cstacks-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/cstacks-DNANSeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` src/cstacks-input.o: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-input.o -MD -MP -MF src/$(DEPDIR)/cstacks-input.Tpo -c -o src/cstacks-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-input.Tpo src/$(DEPDIR)/cstacks-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/cstacks-input.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc src/cstacks-input.obj: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-input.obj -MD -MP -MF src/$(DEPDIR)/cstacks-input.Tpo -c -o src/cstacks-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-input.Tpo src/$(DEPDIR)/cstacks-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/cstacks-input.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` src/cstacks-sql_utilities.o: src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-sql_utilities.o -MD -MP -MF src/$(DEPDIR)/cstacks-sql_utilities.Tpo -c -o src/cstacks-sql_utilities.o `test -f 'src/sql_utilities.cc' || echo '$(srcdir)/'`src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-sql_utilities.Tpo src/$(DEPDIR)/cstacks-sql_utilities.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/sql_utilities.cc' object='src/cstacks-sql_utilities.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-sql_utilities.o `test -f 'src/sql_utilities.cc' || echo '$(srcdir)/'`src/sql_utilities.cc src/cstacks-sql_utilities.obj: src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -MT src/cstacks-sql_utilities.obj -MD -MP -MF src/$(DEPDIR)/cstacks-sql_utilities.Tpo -c -o src/cstacks-sql_utilities.obj `if test -f 'src/sql_utilities.cc'; then $(CYGPATH_W) 'src/sql_utilities.cc'; else $(CYGPATH_W) '$(srcdir)/src/sql_utilities.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/cstacks-sql_utilities.Tpo src/$(DEPDIR)/cstacks-sql_utilities.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/sql_utilities.cc' object='src/cstacks-sql_utilities.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(cstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/cstacks-sql_utilities.obj `if test -f 'src/sql_utilities.cc'; then $(CYGPATH_W) 'src/sql_utilities.cc'; else $(CYGPATH_W) '$(srcdir)/src/sql_utilities.cc'; fi` src/estacks-estacks.o: src/estacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -MT src/estacks-estacks.o -MD -MP -MF src/$(DEPDIR)/estacks-estacks.Tpo -c -o src/estacks-estacks.o `test -f 'src/estacks.cc' || echo '$(srcdir)/'`src/estacks.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/estacks-estacks.Tpo src/$(DEPDIR)/estacks-estacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/estacks.cc' object='src/estacks-estacks.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -c -o src/estacks-estacks.o `test -f 'src/estacks.cc' || echo '$(srcdir)/'`src/estacks.cc src/estacks-estacks.obj: src/estacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -MT src/estacks-estacks.obj -MD -MP -MF src/$(DEPDIR)/estacks-estacks.Tpo -c -o src/estacks-estacks.obj `if test -f 'src/estacks.cc'; then $(CYGPATH_W) 'src/estacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/estacks.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/estacks-estacks.Tpo src/$(DEPDIR)/estacks-estacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/estacks.cc' object='src/estacks-estacks.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -c -o src/estacks-estacks.obj `if test -f 'src/estacks.cc'; then $(CYGPATH_W) 'src/estacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/estacks.cc'; fi` src/estacks-stacks.o: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -MT src/estacks-stacks.o -MD -MP -MF src/$(DEPDIR)/estacks-stacks.Tpo -c -o src/estacks-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/estacks-stacks.Tpo src/$(DEPDIR)/estacks-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/estacks-stacks.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -c -o src/estacks-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc src/estacks-stacks.obj: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -MT src/estacks-stacks.obj -MD -MP -MF src/$(DEPDIR)/estacks-stacks.Tpo -c -o src/estacks-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/estacks-stacks.Tpo src/$(DEPDIR)/estacks-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/estacks-stacks.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -c -o src/estacks-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` src/estacks-mstack.o: src/mstack.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -MT src/estacks-mstack.o -MD -MP -MF src/$(DEPDIR)/estacks-mstack.Tpo -c -o src/estacks-mstack.o `test -f 'src/mstack.cc' || echo '$(srcdir)/'`src/mstack.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/estacks-mstack.Tpo src/$(DEPDIR)/estacks-mstack.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/mstack.cc' object='src/estacks-mstack.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -c -o src/estacks-mstack.o `test -f 'src/mstack.cc' || echo '$(srcdir)/'`src/mstack.cc src/estacks-mstack.obj: src/mstack.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -MT src/estacks-mstack.obj -MD -MP -MF src/$(DEPDIR)/estacks-mstack.Tpo -c -o src/estacks-mstack.obj `if test -f 'src/mstack.cc'; then $(CYGPATH_W) 'src/mstack.cc'; else $(CYGPATH_W) '$(srcdir)/src/mstack.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/estacks-mstack.Tpo src/$(DEPDIR)/estacks-mstack.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/mstack.cc' object='src/estacks-mstack.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -c -o src/estacks-mstack.obj `if test -f 'src/mstack.cc'; then $(CYGPATH_W) 'src/mstack.cc'; else $(CYGPATH_W) '$(srcdir)/src/mstack.cc'; fi` src/estacks-models.o: src/models.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -MT src/estacks-models.o -MD -MP -MF src/$(DEPDIR)/estacks-models.Tpo -c -o src/estacks-models.o `test -f 'src/models.cc' || echo '$(srcdir)/'`src/models.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/estacks-models.Tpo src/$(DEPDIR)/estacks-models.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/models.cc' object='src/estacks-models.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -c -o src/estacks-models.o `test -f 'src/models.cc' || echo '$(srcdir)/'`src/models.cc src/estacks-models.obj: src/models.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -MT src/estacks-models.obj -MD -MP -MF src/$(DEPDIR)/estacks-models.Tpo -c -o src/estacks-models.obj `if test -f 'src/models.cc'; then $(CYGPATH_W) 'src/models.cc'; else $(CYGPATH_W) '$(srcdir)/src/models.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/estacks-models.Tpo src/$(DEPDIR)/estacks-models.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/models.cc' object='src/estacks-models.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -c -o src/estacks-models.obj `if test -f 'src/models.cc'; then $(CYGPATH_W) 'src/models.cc'; else $(CYGPATH_W) '$(srcdir)/src/models.cc'; fi` src/estacks-utils.o: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -MT src/estacks-utils.o -MD -MP -MF src/$(DEPDIR)/estacks-utils.Tpo -c -o src/estacks-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/estacks-utils.Tpo src/$(DEPDIR)/estacks-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/estacks-utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -c -o src/estacks-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc src/estacks-utils.obj: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -MT src/estacks-utils.obj -MD -MP -MF src/$(DEPDIR)/estacks-utils.Tpo -c -o src/estacks-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/estacks-utils.Tpo src/$(DEPDIR)/estacks-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/estacks-utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -c -o src/estacks-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` src/estacks-DNASeq.o: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -MT src/estacks-DNASeq.o -MD -MP -MF src/$(DEPDIR)/estacks-DNASeq.Tpo -c -o src/estacks-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/estacks-DNASeq.Tpo src/$(DEPDIR)/estacks-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/estacks-DNASeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -c -o src/estacks-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc src/estacks-DNASeq.obj: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -MT src/estacks-DNASeq.obj -MD -MP -MF src/$(DEPDIR)/estacks-DNASeq.Tpo -c -o src/estacks-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/estacks-DNASeq.Tpo src/$(DEPDIR)/estacks-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/estacks-DNASeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -c -o src/estacks-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` src/estacks-DNANSeq.o: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -MT src/estacks-DNANSeq.o -MD -MP -MF src/$(DEPDIR)/estacks-DNANSeq.Tpo -c -o src/estacks-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/estacks-DNANSeq.Tpo src/$(DEPDIR)/estacks-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/estacks-DNANSeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -c -o src/estacks-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc src/estacks-DNANSeq.obj: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -MT src/estacks-DNANSeq.obj -MD -MP -MF src/$(DEPDIR)/estacks-DNANSeq.Tpo -c -o src/estacks-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/estacks-DNANSeq.Tpo src/$(DEPDIR)/estacks-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/estacks-DNANSeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -c -o src/estacks-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` src/estacks-input.o: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -MT src/estacks-input.o -MD -MP -MF src/$(DEPDIR)/estacks-input.Tpo -c -o src/estacks-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/estacks-input.Tpo src/$(DEPDIR)/estacks-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/estacks-input.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -c -o src/estacks-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc src/estacks-input.obj: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -MT src/estacks-input.obj -MD -MP -MF src/$(DEPDIR)/estacks-input.Tpo -c -o src/estacks-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/estacks-input.Tpo src/$(DEPDIR)/estacks-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/estacks-input.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(estacks_CXXFLAGS) $(CXXFLAGS) -c -o src/estacks-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` src/genotypes-genotypes.o: src/genotypes.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-genotypes.o -MD -MP -MF src/$(DEPDIR)/genotypes-genotypes.Tpo -c -o src/genotypes-genotypes.o `test -f 'src/genotypes.cc' || echo '$(srcdir)/'`src/genotypes.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-genotypes.Tpo src/$(DEPDIR)/genotypes-genotypes.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/genotypes.cc' object='src/genotypes-genotypes.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-genotypes.o `test -f 'src/genotypes.cc' || echo '$(srcdir)/'`src/genotypes.cc src/genotypes-genotypes.obj: src/genotypes.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-genotypes.obj -MD -MP -MF src/$(DEPDIR)/genotypes-genotypes.Tpo -c -o src/genotypes-genotypes.obj `if test -f 'src/genotypes.cc'; then $(CYGPATH_W) 'src/genotypes.cc'; else $(CYGPATH_W) '$(srcdir)/src/genotypes.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-genotypes.Tpo src/$(DEPDIR)/genotypes-genotypes.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/genotypes.cc' object='src/genotypes-genotypes.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-genotypes.obj `if test -f 'src/genotypes.cc'; then $(CYGPATH_W) 'src/genotypes.cc'; else $(CYGPATH_W) '$(srcdir)/src/genotypes.cc'; fi` src/genotypes-utils.o: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-utils.o -MD -MP -MF src/$(DEPDIR)/genotypes-utils.Tpo -c -o src/genotypes-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-utils.Tpo src/$(DEPDIR)/genotypes-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/genotypes-utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc src/genotypes-utils.obj: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-utils.obj -MD -MP -MF src/$(DEPDIR)/genotypes-utils.Tpo -c -o src/genotypes-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-utils.Tpo src/$(DEPDIR)/genotypes-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/genotypes-utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` src/genotypes-catalog_utils.o: src/catalog_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-catalog_utils.o -MD -MP -MF src/$(DEPDIR)/genotypes-catalog_utils.Tpo -c -o src/genotypes-catalog_utils.o `test -f 'src/catalog_utils.cc' || echo '$(srcdir)/'`src/catalog_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-catalog_utils.Tpo src/$(DEPDIR)/genotypes-catalog_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/catalog_utils.cc' object='src/genotypes-catalog_utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-catalog_utils.o `test -f 'src/catalog_utils.cc' || echo '$(srcdir)/'`src/catalog_utils.cc src/genotypes-catalog_utils.obj: src/catalog_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-catalog_utils.obj -MD -MP -MF src/$(DEPDIR)/genotypes-catalog_utils.Tpo -c -o src/genotypes-catalog_utils.obj `if test -f 'src/catalog_utils.cc'; then $(CYGPATH_W) 'src/catalog_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/catalog_utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-catalog_utils.Tpo src/$(DEPDIR)/genotypes-catalog_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/catalog_utils.cc' object='src/genotypes-catalog_utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-catalog_utils.obj `if test -f 'src/catalog_utils.cc'; then $(CYGPATH_W) 'src/catalog_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/catalog_utils.cc'; fi` src/genotypes-log_utils.o: src/log_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-log_utils.o -MD -MP -MF src/$(DEPDIR)/genotypes-log_utils.Tpo -c -o src/genotypes-log_utils.o `test -f 'src/log_utils.cc' || echo '$(srcdir)/'`src/log_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-log_utils.Tpo src/$(DEPDIR)/genotypes-log_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/log_utils.cc' object='src/genotypes-log_utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-log_utils.o `test -f 'src/log_utils.cc' || echo '$(srcdir)/'`src/log_utils.cc src/genotypes-log_utils.obj: src/log_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-log_utils.obj -MD -MP -MF src/$(DEPDIR)/genotypes-log_utils.Tpo -c -o src/genotypes-log_utils.obj `if test -f 'src/log_utils.cc'; then $(CYGPATH_W) 'src/log_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/log_utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-log_utils.Tpo src/$(DEPDIR)/genotypes-log_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/log_utils.cc' object='src/genotypes-log_utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-log_utils.obj `if test -f 'src/log_utils.cc'; then $(CYGPATH_W) 'src/log_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/log_utils.cc'; fi` src/genotypes-stacks.o: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-stacks.o -MD -MP -MF src/$(DEPDIR)/genotypes-stacks.Tpo -c -o src/genotypes-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-stacks.Tpo src/$(DEPDIR)/genotypes-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/genotypes-stacks.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc src/genotypes-stacks.obj: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-stacks.obj -MD -MP -MF src/$(DEPDIR)/genotypes-stacks.Tpo -c -o src/genotypes-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-stacks.Tpo src/$(DEPDIR)/genotypes-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/genotypes-stacks.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` src/genotypes-locus.o: src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-locus.o -MD -MP -MF src/$(DEPDIR)/genotypes-locus.Tpo -c -o src/genotypes-locus.o `test -f 'src/locus.cc' || echo '$(srcdir)/'`src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-locus.Tpo src/$(DEPDIR)/genotypes-locus.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/locus.cc' object='src/genotypes-locus.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-locus.o `test -f 'src/locus.cc' || echo '$(srcdir)/'`src/locus.cc src/genotypes-locus.obj: src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-locus.obj -MD -MP -MF src/$(DEPDIR)/genotypes-locus.Tpo -c -o src/genotypes-locus.obj `if test -f 'src/locus.cc'; then $(CYGPATH_W) 'src/locus.cc'; else $(CYGPATH_W) '$(srcdir)/src/locus.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-locus.Tpo src/$(DEPDIR)/genotypes-locus.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/locus.cc' object='src/genotypes-locus.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-locus.obj `if test -f 'src/locus.cc'; then $(CYGPATH_W) 'src/locus.cc'; else $(CYGPATH_W) '$(srcdir)/src/locus.cc'; fi` src/genotypes-DNASeq.o: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-DNASeq.o -MD -MP -MF src/$(DEPDIR)/genotypes-DNASeq.Tpo -c -o src/genotypes-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-DNASeq.Tpo src/$(DEPDIR)/genotypes-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/genotypes-DNASeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc src/genotypes-DNASeq.obj: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-DNASeq.obj -MD -MP -MF src/$(DEPDIR)/genotypes-DNASeq.Tpo -c -o src/genotypes-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-DNASeq.Tpo src/$(DEPDIR)/genotypes-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/genotypes-DNASeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` src/genotypes-DNANSeq.o: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-DNANSeq.o -MD -MP -MF src/$(DEPDIR)/genotypes-DNANSeq.Tpo -c -o src/genotypes-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-DNANSeq.Tpo src/$(DEPDIR)/genotypes-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/genotypes-DNANSeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc src/genotypes-DNANSeq.obj: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-DNANSeq.obj -MD -MP -MF src/$(DEPDIR)/genotypes-DNANSeq.Tpo -c -o src/genotypes-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-DNANSeq.Tpo src/$(DEPDIR)/genotypes-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/genotypes-DNANSeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` src/genotypes-aln_utils.o: src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-aln_utils.o -MD -MP -MF src/$(DEPDIR)/genotypes-aln_utils.Tpo -c -o src/genotypes-aln_utils.o `test -f 'src/aln_utils.cc' || echo '$(srcdir)/'`src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-aln_utils.Tpo src/$(DEPDIR)/genotypes-aln_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/aln_utils.cc' object='src/genotypes-aln_utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-aln_utils.o `test -f 'src/aln_utils.cc' || echo '$(srcdir)/'`src/aln_utils.cc src/genotypes-aln_utils.obj: src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-aln_utils.obj -MD -MP -MF src/$(DEPDIR)/genotypes-aln_utils.Tpo -c -o src/genotypes-aln_utils.obj `if test -f 'src/aln_utils.cc'; then $(CYGPATH_W) 'src/aln_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/aln_utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-aln_utils.Tpo src/$(DEPDIR)/genotypes-aln_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/aln_utils.cc' object='src/genotypes-aln_utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-aln_utils.obj `if test -f 'src/aln_utils.cc'; then $(CYGPATH_W) 'src/aln_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/aln_utils.cc'; fi` src/genotypes-genotype_dictionaries.o: src/genotype_dictionaries.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-genotype_dictionaries.o -MD -MP -MF src/$(DEPDIR)/genotypes-genotype_dictionaries.Tpo -c -o src/genotypes-genotype_dictionaries.o `test -f 'src/genotype_dictionaries.cc' || echo '$(srcdir)/'`src/genotype_dictionaries.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-genotype_dictionaries.Tpo src/$(DEPDIR)/genotypes-genotype_dictionaries.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/genotype_dictionaries.cc' object='src/genotypes-genotype_dictionaries.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-genotype_dictionaries.o `test -f 'src/genotype_dictionaries.cc' || echo '$(srcdir)/'`src/genotype_dictionaries.cc src/genotypes-genotype_dictionaries.obj: src/genotype_dictionaries.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-genotype_dictionaries.obj -MD -MP -MF src/$(DEPDIR)/genotypes-genotype_dictionaries.Tpo -c -o src/genotypes-genotype_dictionaries.obj `if test -f 'src/genotype_dictionaries.cc'; then $(CYGPATH_W) 'src/genotype_dictionaries.cc'; else $(CYGPATH_W) '$(srcdir)/src/genotype_dictionaries.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-genotype_dictionaries.Tpo src/$(DEPDIR)/genotypes-genotype_dictionaries.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/genotype_dictionaries.cc' object='src/genotypes-genotype_dictionaries.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-genotype_dictionaries.obj `if test -f 'src/genotype_dictionaries.cc'; then $(CYGPATH_W) 'src/genotype_dictionaries.cc'; else $(CYGPATH_W) '$(srcdir)/src/genotype_dictionaries.cc'; fi` src/genotypes-input.o: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-input.o -MD -MP -MF src/$(DEPDIR)/genotypes-input.Tpo -c -o src/genotypes-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-input.Tpo src/$(DEPDIR)/genotypes-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/genotypes-input.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc src/genotypes-input.obj: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-input.obj -MD -MP -MF src/$(DEPDIR)/genotypes-input.Tpo -c -o src/genotypes-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-input.Tpo src/$(DEPDIR)/genotypes-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/genotypes-input.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` src/genotypes-sql_utilities.o: src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-sql_utilities.o -MD -MP -MF src/$(DEPDIR)/genotypes-sql_utilities.Tpo -c -o src/genotypes-sql_utilities.o `test -f 'src/sql_utilities.cc' || echo '$(srcdir)/'`src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-sql_utilities.Tpo src/$(DEPDIR)/genotypes-sql_utilities.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/sql_utilities.cc' object='src/genotypes-sql_utilities.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-sql_utilities.o `test -f 'src/sql_utilities.cc' || echo '$(srcdir)/'`src/sql_utilities.cc src/genotypes-sql_utilities.obj: src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-sql_utilities.obj -MD -MP -MF src/$(DEPDIR)/genotypes-sql_utilities.Tpo -c -o src/genotypes-sql_utilities.obj `if test -f 'src/sql_utilities.cc'; then $(CYGPATH_W) 'src/sql_utilities.cc'; else $(CYGPATH_W) '$(srcdir)/src/sql_utilities.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-sql_utilities.Tpo src/$(DEPDIR)/genotypes-sql_utilities.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/sql_utilities.cc' object='src/genotypes-sql_utilities.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-sql_utilities.obj `if test -f 'src/sql_utilities.cc'; then $(CYGPATH_W) 'src/sql_utilities.cc'; else $(CYGPATH_W) '$(srcdir)/src/sql_utilities.cc'; fi` src/genotypes-renz.o: src/renz.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-renz.o -MD -MP -MF src/$(DEPDIR)/genotypes-renz.Tpo -c -o src/genotypes-renz.o `test -f 'src/renz.cc' || echo '$(srcdir)/'`src/renz.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-renz.Tpo src/$(DEPDIR)/genotypes-renz.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/renz.cc' object='src/genotypes-renz.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-renz.o `test -f 'src/renz.cc' || echo '$(srcdir)/'`src/renz.cc src/genotypes-renz.obj: src/renz.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-renz.obj -MD -MP -MF src/$(DEPDIR)/genotypes-renz.Tpo -c -o src/genotypes-renz.obj `if test -f 'src/renz.cc'; then $(CYGPATH_W) 'src/renz.cc'; else $(CYGPATH_W) '$(srcdir)/src/renz.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-renz.Tpo src/$(DEPDIR)/genotypes-renz.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/renz.cc' object='src/genotypes-renz.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-renz.obj `if test -f 'src/renz.cc'; then $(CYGPATH_W) 'src/renz.cc'; else $(CYGPATH_W) '$(srcdir)/src/renz.cc'; fi` src/genotypes-MetaPopInfo.o: src/MetaPopInfo.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-MetaPopInfo.o -MD -MP -MF src/$(DEPDIR)/genotypes-MetaPopInfo.Tpo -c -o src/genotypes-MetaPopInfo.o `test -f 'src/MetaPopInfo.cc' || echo '$(srcdir)/'`src/MetaPopInfo.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-MetaPopInfo.Tpo src/$(DEPDIR)/genotypes-MetaPopInfo.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/MetaPopInfo.cc' object='src/genotypes-MetaPopInfo.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-MetaPopInfo.o `test -f 'src/MetaPopInfo.cc' || echo '$(srcdir)/'`src/MetaPopInfo.cc src/genotypes-MetaPopInfo.obj: src/MetaPopInfo.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -MT src/genotypes-MetaPopInfo.obj -MD -MP -MF src/$(DEPDIR)/genotypes-MetaPopInfo.Tpo -c -o src/genotypes-MetaPopInfo.obj `if test -f 'src/MetaPopInfo.cc'; then $(CYGPATH_W) 'src/MetaPopInfo.cc'; else $(CYGPATH_W) '$(srcdir)/src/MetaPopInfo.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/genotypes-MetaPopInfo.Tpo src/$(DEPDIR)/genotypes-MetaPopInfo.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/MetaPopInfo.cc' object='src/genotypes-MetaPopInfo.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(genotypes_CXXFLAGS) $(CXXFLAGS) -c -o src/genotypes-MetaPopInfo.obj `if test -f 'src/MetaPopInfo.cc'; then $(CYGPATH_W) 'src/MetaPopInfo.cc'; else $(CYGPATH_W) '$(srcdir)/src/MetaPopInfo.cc'; fi` src/hstacks-hstacks.o: src/hstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-hstacks.o -MD -MP -MF src/$(DEPDIR)/hstacks-hstacks.Tpo -c -o src/hstacks-hstacks.o `test -f 'src/hstacks.cc' || echo '$(srcdir)/'`src/hstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-hstacks.Tpo src/$(DEPDIR)/hstacks-hstacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/hstacks.cc' object='src/hstacks-hstacks.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-hstacks.o `test -f 'src/hstacks.cc' || echo '$(srcdir)/'`src/hstacks.cc src/hstacks-hstacks.obj: src/hstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-hstacks.obj -MD -MP -MF src/$(DEPDIR)/hstacks-hstacks.Tpo -c -o src/hstacks-hstacks.obj `if test -f 'src/hstacks.cc'; then $(CYGPATH_W) 'src/hstacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/hstacks.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-hstacks.Tpo src/$(DEPDIR)/hstacks-hstacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/hstacks.cc' object='src/hstacks-hstacks.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-hstacks.obj `if test -f 'src/hstacks.cc'; then $(CYGPATH_W) 'src/hstacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/hstacks.cc'; fi` src/hstacks-stacks.o: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-stacks.o -MD -MP -MF src/$(DEPDIR)/hstacks-stacks.Tpo -c -o src/hstacks-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-stacks.Tpo src/$(DEPDIR)/hstacks-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/hstacks-stacks.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc src/hstacks-stacks.obj: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-stacks.obj -MD -MP -MF src/$(DEPDIR)/hstacks-stacks.Tpo -c -o src/hstacks-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-stacks.Tpo src/$(DEPDIR)/hstacks-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/hstacks-stacks.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` src/hstacks-locus.o: src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-locus.o -MD -MP -MF src/$(DEPDIR)/hstacks-locus.Tpo -c -o src/hstacks-locus.o `test -f 'src/locus.cc' || echo '$(srcdir)/'`src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-locus.Tpo src/$(DEPDIR)/hstacks-locus.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/locus.cc' object='src/hstacks-locus.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-locus.o `test -f 'src/locus.cc' || echo '$(srcdir)/'`src/locus.cc src/hstacks-locus.obj: src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-locus.obj -MD -MP -MF src/$(DEPDIR)/hstacks-locus.Tpo -c -o src/hstacks-locus.obj `if test -f 'src/locus.cc'; then $(CYGPATH_W) 'src/locus.cc'; else $(CYGPATH_W) '$(srcdir)/src/locus.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-locus.Tpo src/$(DEPDIR)/hstacks-locus.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/locus.cc' object='src/hstacks-locus.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-locus.obj `if test -f 'src/locus.cc'; then $(CYGPATH_W) 'src/locus.cc'; else $(CYGPATH_W) '$(srcdir)/src/locus.cc'; fi` src/hstacks-kmers.o: src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-kmers.o -MD -MP -MF src/$(DEPDIR)/hstacks-kmers.Tpo -c -o src/hstacks-kmers.o `test -f 'src/kmers.cc' || echo '$(srcdir)/'`src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-kmers.Tpo src/$(DEPDIR)/hstacks-kmers.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/kmers.cc' object='src/hstacks-kmers.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-kmers.o `test -f 'src/kmers.cc' || echo '$(srcdir)/'`src/kmers.cc src/hstacks-kmers.obj: src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-kmers.obj -MD -MP -MF src/$(DEPDIR)/hstacks-kmers.Tpo -c -o src/hstacks-kmers.obj `if test -f 'src/kmers.cc'; then $(CYGPATH_W) 'src/kmers.cc'; else $(CYGPATH_W) '$(srcdir)/src/kmers.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-kmers.Tpo src/$(DEPDIR)/hstacks-kmers.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/kmers.cc' object='src/hstacks-kmers.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-kmers.obj `if test -f 'src/kmers.cc'; then $(CYGPATH_W) 'src/kmers.cc'; else $(CYGPATH_W) '$(srcdir)/src/kmers.cc'; fi` src/hstacks-DNASeq.o: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-DNASeq.o -MD -MP -MF src/$(DEPDIR)/hstacks-DNASeq.Tpo -c -o src/hstacks-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-DNASeq.Tpo src/$(DEPDIR)/hstacks-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/hstacks-DNASeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc src/hstacks-DNASeq.obj: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-DNASeq.obj -MD -MP -MF src/$(DEPDIR)/hstacks-DNASeq.Tpo -c -o src/hstacks-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-DNASeq.Tpo src/$(DEPDIR)/hstacks-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/hstacks-DNASeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` src/hstacks-DNANSeq.o: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-DNANSeq.o -MD -MP -MF src/$(DEPDIR)/hstacks-DNANSeq.Tpo -c -o src/hstacks-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-DNANSeq.Tpo src/$(DEPDIR)/hstacks-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/hstacks-DNANSeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc src/hstacks-DNANSeq.obj: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-DNANSeq.obj -MD -MP -MF src/$(DEPDIR)/hstacks-DNANSeq.Tpo -c -o src/hstacks-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-DNANSeq.Tpo src/$(DEPDIR)/hstacks-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/hstacks-DNANSeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` src/hstacks-utils.o: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-utils.o -MD -MP -MF src/$(DEPDIR)/hstacks-utils.Tpo -c -o src/hstacks-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-utils.Tpo src/$(DEPDIR)/hstacks-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/hstacks-utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc src/hstacks-utils.obj: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-utils.obj -MD -MP -MF src/$(DEPDIR)/hstacks-utils.Tpo -c -o src/hstacks-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-utils.Tpo src/$(DEPDIR)/hstacks-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/hstacks-utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` src/hstacks-input.o: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-input.o -MD -MP -MF src/$(DEPDIR)/hstacks-input.Tpo -c -o src/hstacks-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-input.Tpo src/$(DEPDIR)/hstacks-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/hstacks-input.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc src/hstacks-input.obj: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-input.obj -MD -MP -MF src/$(DEPDIR)/hstacks-input.Tpo -c -o src/hstacks-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-input.Tpo src/$(DEPDIR)/hstacks-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/hstacks-input.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` src/hstacks-sql_utilities.o: src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-sql_utilities.o -MD -MP -MF src/$(DEPDIR)/hstacks-sql_utilities.Tpo -c -o src/hstacks-sql_utilities.o `test -f 'src/sql_utilities.cc' || echo '$(srcdir)/'`src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-sql_utilities.Tpo src/$(DEPDIR)/hstacks-sql_utilities.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/sql_utilities.cc' object='src/hstacks-sql_utilities.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-sql_utilities.o `test -f 'src/sql_utilities.cc' || echo '$(srcdir)/'`src/sql_utilities.cc src/hstacks-sql_utilities.obj: src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -MT src/hstacks-sql_utilities.obj -MD -MP -MF src/$(DEPDIR)/hstacks-sql_utilities.Tpo -c -o src/hstacks-sql_utilities.obj `if test -f 'src/sql_utilities.cc'; then $(CYGPATH_W) 'src/sql_utilities.cc'; else $(CYGPATH_W) '$(srcdir)/src/sql_utilities.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/hstacks-sql_utilities.Tpo src/$(DEPDIR)/hstacks-sql_utilities.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/sql_utilities.cc' object='src/hstacks-sql_utilities.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(hstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/hstacks-sql_utilities.obj `if test -f 'src/sql_utilities.cc'; then $(CYGPATH_W) 'src/sql_utilities.cc'; else $(CYGPATH_W) '$(srcdir)/src/sql_utilities.cc'; fi` src/kmer_filter-kmer_filter.o: src/kmer_filter.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -MT src/kmer_filter-kmer_filter.o -MD -MP -MF src/$(DEPDIR)/kmer_filter-kmer_filter.Tpo -c -o src/kmer_filter-kmer_filter.o `test -f 'src/kmer_filter.cc' || echo '$(srcdir)/'`src/kmer_filter.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/kmer_filter-kmer_filter.Tpo src/$(DEPDIR)/kmer_filter-kmer_filter.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/kmer_filter.cc' object='src/kmer_filter-kmer_filter.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/kmer_filter-kmer_filter.o `test -f 'src/kmer_filter.cc' || echo '$(srcdir)/'`src/kmer_filter.cc src/kmer_filter-kmer_filter.obj: src/kmer_filter.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -MT src/kmer_filter-kmer_filter.obj -MD -MP -MF src/$(DEPDIR)/kmer_filter-kmer_filter.Tpo -c -o src/kmer_filter-kmer_filter.obj `if test -f 'src/kmer_filter.cc'; then $(CYGPATH_W) 'src/kmer_filter.cc'; else $(CYGPATH_W) '$(srcdir)/src/kmer_filter.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/kmer_filter-kmer_filter.Tpo src/$(DEPDIR)/kmer_filter-kmer_filter.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/kmer_filter.cc' object='src/kmer_filter-kmer_filter.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/kmer_filter-kmer_filter.obj `if test -f 'src/kmer_filter.cc'; then $(CYGPATH_W) 'src/kmer_filter.cc'; else $(CYGPATH_W) '$(srcdir)/src/kmer_filter.cc'; fi` src/kmer_filter-utils.o: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -MT src/kmer_filter-utils.o -MD -MP -MF src/$(DEPDIR)/kmer_filter-utils.Tpo -c -o src/kmer_filter-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/kmer_filter-utils.Tpo src/$(DEPDIR)/kmer_filter-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/kmer_filter-utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/kmer_filter-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc src/kmer_filter-utils.obj: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -MT src/kmer_filter-utils.obj -MD -MP -MF src/$(DEPDIR)/kmer_filter-utils.Tpo -c -o src/kmer_filter-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/kmer_filter-utils.Tpo src/$(DEPDIR)/kmer_filter-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/kmer_filter-utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/kmer_filter-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` src/kmer_filter-write.o: src/write.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -MT src/kmer_filter-write.o -MD -MP -MF src/$(DEPDIR)/kmer_filter-write.Tpo -c -o src/kmer_filter-write.o `test -f 'src/write.cc' || echo '$(srcdir)/'`src/write.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/kmer_filter-write.Tpo src/$(DEPDIR)/kmer_filter-write.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/write.cc' object='src/kmer_filter-write.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/kmer_filter-write.o `test -f 'src/write.cc' || echo '$(srcdir)/'`src/write.cc src/kmer_filter-write.obj: src/write.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -MT src/kmer_filter-write.obj -MD -MP -MF src/$(DEPDIR)/kmer_filter-write.Tpo -c -o src/kmer_filter-write.obj `if test -f 'src/write.cc'; then $(CYGPATH_W) 'src/write.cc'; else $(CYGPATH_W) '$(srcdir)/src/write.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/kmer_filter-write.Tpo src/$(DEPDIR)/kmer_filter-write.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/write.cc' object='src/kmer_filter-write.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/kmer_filter-write.obj `if test -f 'src/write.cc'; then $(CYGPATH_W) 'src/write.cc'; else $(CYGPATH_W) '$(srcdir)/src/write.cc'; fi` src/kmer_filter-kmers.o: src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -MT src/kmer_filter-kmers.o -MD -MP -MF src/$(DEPDIR)/kmer_filter-kmers.Tpo -c -o src/kmer_filter-kmers.o `test -f 'src/kmers.cc' || echo '$(srcdir)/'`src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/kmer_filter-kmers.Tpo src/$(DEPDIR)/kmer_filter-kmers.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/kmers.cc' object='src/kmer_filter-kmers.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/kmer_filter-kmers.o `test -f 'src/kmers.cc' || echo '$(srcdir)/'`src/kmers.cc src/kmer_filter-kmers.obj: src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -MT src/kmer_filter-kmers.obj -MD -MP -MF src/$(DEPDIR)/kmer_filter-kmers.Tpo -c -o src/kmer_filter-kmers.obj `if test -f 'src/kmers.cc'; then $(CYGPATH_W) 'src/kmers.cc'; else $(CYGPATH_W) '$(srcdir)/src/kmers.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/kmer_filter-kmers.Tpo src/$(DEPDIR)/kmer_filter-kmers.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/kmers.cc' object='src/kmer_filter-kmers.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/kmer_filter-kmers.obj `if test -f 'src/kmers.cc'; then $(CYGPATH_W) 'src/kmers.cc'; else $(CYGPATH_W) '$(srcdir)/src/kmers.cc'; fi` src/kmer_filter-input.o: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -MT src/kmer_filter-input.o -MD -MP -MF src/$(DEPDIR)/kmer_filter-input.Tpo -c -o src/kmer_filter-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/kmer_filter-input.Tpo src/$(DEPDIR)/kmer_filter-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/kmer_filter-input.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/kmer_filter-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc src/kmer_filter-input.obj: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -MT src/kmer_filter-input.obj -MD -MP -MF src/$(DEPDIR)/kmer_filter-input.Tpo -c -o src/kmer_filter-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/kmer_filter-input.Tpo src/$(DEPDIR)/kmer_filter-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/kmer_filter-input.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(kmer_filter_CXXFLAGS) $(CXXFLAGS) -c -o src/kmer_filter-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` src/phasedstacks-phasedstacks.o: src/phasedstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -MT src/phasedstacks-phasedstacks.o -MD -MP -MF src/$(DEPDIR)/phasedstacks-phasedstacks.Tpo -c -o src/phasedstacks-phasedstacks.o `test -f 'src/phasedstacks.cc' || echo '$(srcdir)/'`src/phasedstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/phasedstacks-phasedstacks.Tpo src/$(DEPDIR)/phasedstacks-phasedstacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/phasedstacks.cc' object='src/phasedstacks-phasedstacks.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/phasedstacks-phasedstacks.o `test -f 'src/phasedstacks.cc' || echo '$(srcdir)/'`src/phasedstacks.cc src/phasedstacks-phasedstacks.obj: src/phasedstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -MT src/phasedstacks-phasedstacks.obj -MD -MP -MF src/$(DEPDIR)/phasedstacks-phasedstacks.Tpo -c -o src/phasedstacks-phasedstacks.obj `if test -f 'src/phasedstacks.cc'; then $(CYGPATH_W) 'src/phasedstacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/phasedstacks.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/phasedstacks-phasedstacks.Tpo src/$(DEPDIR)/phasedstacks-phasedstacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/phasedstacks.cc' object='src/phasedstacks-phasedstacks.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/phasedstacks-phasedstacks.obj `if test -f 'src/phasedstacks.cc'; then $(CYGPATH_W) 'src/phasedstacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/phasedstacks.cc'; fi` src/phasedstacks-locus.o: src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -MT src/phasedstacks-locus.o -MD -MP -MF src/$(DEPDIR)/phasedstacks-locus.Tpo -c -o src/phasedstacks-locus.o `test -f 'src/locus.cc' || echo '$(srcdir)/'`src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/phasedstacks-locus.Tpo src/$(DEPDIR)/phasedstacks-locus.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/locus.cc' object='src/phasedstacks-locus.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/phasedstacks-locus.o `test -f 'src/locus.cc' || echo '$(srcdir)/'`src/locus.cc src/phasedstacks-locus.obj: src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -MT src/phasedstacks-locus.obj -MD -MP -MF src/$(DEPDIR)/phasedstacks-locus.Tpo -c -o src/phasedstacks-locus.obj `if test -f 'src/locus.cc'; then $(CYGPATH_W) 'src/locus.cc'; else $(CYGPATH_W) '$(srcdir)/src/locus.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/phasedstacks-locus.Tpo src/$(DEPDIR)/phasedstacks-locus.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/locus.cc' object='src/phasedstacks-locus.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/phasedstacks-locus.obj `if test -f 'src/locus.cc'; then $(CYGPATH_W) 'src/locus.cc'; else $(CYGPATH_W) '$(srcdir)/src/locus.cc'; fi` src/phasedstacks-input.o: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -MT src/phasedstacks-input.o -MD -MP -MF src/$(DEPDIR)/phasedstacks-input.Tpo -c -o src/phasedstacks-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/phasedstacks-input.Tpo src/$(DEPDIR)/phasedstacks-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/phasedstacks-input.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/phasedstacks-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc src/phasedstacks-input.obj: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -MT src/phasedstacks-input.obj -MD -MP -MF src/$(DEPDIR)/phasedstacks-input.Tpo -c -o src/phasedstacks-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/phasedstacks-input.Tpo src/$(DEPDIR)/phasedstacks-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/phasedstacks-input.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/phasedstacks-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` src/phasedstacks-sql_utilities.o: src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -MT src/phasedstacks-sql_utilities.o -MD -MP -MF src/$(DEPDIR)/phasedstacks-sql_utilities.Tpo -c -o src/phasedstacks-sql_utilities.o `test -f 'src/sql_utilities.cc' || echo '$(srcdir)/'`src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/phasedstacks-sql_utilities.Tpo src/$(DEPDIR)/phasedstacks-sql_utilities.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/sql_utilities.cc' object='src/phasedstacks-sql_utilities.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/phasedstacks-sql_utilities.o `test -f 'src/sql_utilities.cc' || echo '$(srcdir)/'`src/sql_utilities.cc src/phasedstacks-sql_utilities.obj: src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -MT src/phasedstacks-sql_utilities.obj -MD -MP -MF src/$(DEPDIR)/phasedstacks-sql_utilities.Tpo -c -o src/phasedstacks-sql_utilities.obj `if test -f 'src/sql_utilities.cc'; then $(CYGPATH_W) 'src/sql_utilities.cc'; else $(CYGPATH_W) '$(srcdir)/src/sql_utilities.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/phasedstacks-sql_utilities.Tpo src/$(DEPDIR)/phasedstacks-sql_utilities.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/sql_utilities.cc' object='src/phasedstacks-sql_utilities.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/phasedstacks-sql_utilities.obj `if test -f 'src/sql_utilities.cc'; then $(CYGPATH_W) 'src/sql_utilities.cc'; else $(CYGPATH_W) '$(srcdir)/src/sql_utilities.cc'; fi` src/phasedstacks-log_utils.o: src/log_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -MT src/phasedstacks-log_utils.o -MD -MP -MF src/$(DEPDIR)/phasedstacks-log_utils.Tpo -c -o src/phasedstacks-log_utils.o `test -f 'src/log_utils.cc' || echo '$(srcdir)/'`src/log_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/phasedstacks-log_utils.Tpo src/$(DEPDIR)/phasedstacks-log_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/log_utils.cc' object='src/phasedstacks-log_utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/phasedstacks-log_utils.o `test -f 'src/log_utils.cc' || echo '$(srcdir)/'`src/log_utils.cc src/phasedstacks-log_utils.obj: src/log_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -MT src/phasedstacks-log_utils.obj -MD -MP -MF src/$(DEPDIR)/phasedstacks-log_utils.Tpo -c -o src/phasedstacks-log_utils.obj `if test -f 'src/log_utils.cc'; then $(CYGPATH_W) 'src/log_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/log_utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/phasedstacks-log_utils.Tpo src/$(DEPDIR)/phasedstacks-log_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/log_utils.cc' object='src/phasedstacks-log_utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/phasedstacks-log_utils.obj `if test -f 'src/log_utils.cc'; then $(CYGPATH_W) 'src/log_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/log_utils.cc'; fi` src/phasedstacks-utils.o: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -MT src/phasedstacks-utils.o -MD -MP -MF src/$(DEPDIR)/phasedstacks-utils.Tpo -c -o src/phasedstacks-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/phasedstacks-utils.Tpo src/$(DEPDIR)/phasedstacks-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/phasedstacks-utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/phasedstacks-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc src/phasedstacks-utils.obj: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -MT src/phasedstacks-utils.obj -MD -MP -MF src/$(DEPDIR)/phasedstacks-utils.Tpo -c -o src/phasedstacks-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/phasedstacks-utils.Tpo src/$(DEPDIR)/phasedstacks-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/phasedstacks-utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/phasedstacks-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` src/phasedstacks-catalog_utils.o: src/catalog_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -MT src/phasedstacks-catalog_utils.o -MD -MP -MF src/$(DEPDIR)/phasedstacks-catalog_utils.Tpo -c -o src/phasedstacks-catalog_utils.o `test -f 'src/catalog_utils.cc' || echo '$(srcdir)/'`src/catalog_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/phasedstacks-catalog_utils.Tpo src/$(DEPDIR)/phasedstacks-catalog_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/catalog_utils.cc' object='src/phasedstacks-catalog_utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/phasedstacks-catalog_utils.o `test -f 'src/catalog_utils.cc' || echo '$(srcdir)/'`src/catalog_utils.cc src/phasedstacks-catalog_utils.obj: src/catalog_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -MT src/phasedstacks-catalog_utils.obj -MD -MP -MF src/$(DEPDIR)/phasedstacks-catalog_utils.Tpo -c -o src/phasedstacks-catalog_utils.obj `if test -f 'src/catalog_utils.cc'; then $(CYGPATH_W) 'src/catalog_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/catalog_utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/phasedstacks-catalog_utils.Tpo src/$(DEPDIR)/phasedstacks-catalog_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/catalog_utils.cc' object='src/phasedstacks-catalog_utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(phasedstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/phasedstacks-catalog_utils.obj `if test -f 'src/catalog_utils.cc'; then $(CYGPATH_W) 'src/catalog_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/catalog_utils.cc'; fi` src/populations-populations.o: src/populations.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-populations.o -MD -MP -MF src/$(DEPDIR)/populations-populations.Tpo -c -o src/populations-populations.o `test -f 'src/populations.cc' || echo '$(srcdir)/'`src/populations.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-populations.Tpo src/$(DEPDIR)/populations-populations.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/populations.cc' object='src/populations-populations.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-populations.o `test -f 'src/populations.cc' || echo '$(srcdir)/'`src/populations.cc src/populations-populations.obj: src/populations.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-populations.obj -MD -MP -MF src/$(DEPDIR)/populations-populations.Tpo -c -o src/populations-populations.obj `if test -f 'src/populations.cc'; then $(CYGPATH_W) 'src/populations.cc'; else $(CYGPATH_W) '$(srcdir)/src/populations.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-populations.Tpo src/$(DEPDIR)/populations-populations.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/populations.cc' object='src/populations-populations.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-populations.obj `if test -f 'src/populations.cc'; then $(CYGPATH_W) 'src/populations.cc'; else $(CYGPATH_W) '$(srcdir)/src/populations.cc'; fi` src/populations-utils.o: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-utils.o -MD -MP -MF src/$(DEPDIR)/populations-utils.Tpo -c -o src/populations-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-utils.Tpo src/$(DEPDIR)/populations-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/populations-utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc src/populations-utils.obj: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-utils.obj -MD -MP -MF src/$(DEPDIR)/populations-utils.Tpo -c -o src/populations-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-utils.Tpo src/$(DEPDIR)/populations-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/populations-utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` src/populations-catalog_utils.o: src/catalog_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-catalog_utils.o -MD -MP -MF src/$(DEPDIR)/populations-catalog_utils.Tpo -c -o src/populations-catalog_utils.o `test -f 'src/catalog_utils.cc' || echo '$(srcdir)/'`src/catalog_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-catalog_utils.Tpo src/$(DEPDIR)/populations-catalog_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/catalog_utils.cc' object='src/populations-catalog_utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-catalog_utils.o `test -f 'src/catalog_utils.cc' || echo '$(srcdir)/'`src/catalog_utils.cc src/populations-catalog_utils.obj: src/catalog_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-catalog_utils.obj -MD -MP -MF src/$(DEPDIR)/populations-catalog_utils.Tpo -c -o src/populations-catalog_utils.obj `if test -f 'src/catalog_utils.cc'; then $(CYGPATH_W) 'src/catalog_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/catalog_utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-catalog_utils.Tpo src/$(DEPDIR)/populations-catalog_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/catalog_utils.cc' object='src/populations-catalog_utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-catalog_utils.obj `if test -f 'src/catalog_utils.cc'; then $(CYGPATH_W) 'src/catalog_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/catalog_utils.cc'; fi` src/populations-log_utils.o: src/log_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-log_utils.o -MD -MP -MF src/$(DEPDIR)/populations-log_utils.Tpo -c -o src/populations-log_utils.o `test -f 'src/log_utils.cc' || echo '$(srcdir)/'`src/log_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-log_utils.Tpo src/$(DEPDIR)/populations-log_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/log_utils.cc' object='src/populations-log_utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-log_utils.o `test -f 'src/log_utils.cc' || echo '$(srcdir)/'`src/log_utils.cc src/populations-log_utils.obj: src/log_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-log_utils.obj -MD -MP -MF src/$(DEPDIR)/populations-log_utils.Tpo -c -o src/populations-log_utils.obj `if test -f 'src/log_utils.cc'; then $(CYGPATH_W) 'src/log_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/log_utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-log_utils.Tpo src/$(DEPDIR)/populations-log_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/log_utils.cc' object='src/populations-log_utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-log_utils.obj `if test -f 'src/log_utils.cc'; then $(CYGPATH_W) 'src/log_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/log_utils.cc'; fi` src/populations-stacks.o: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-stacks.o -MD -MP -MF src/$(DEPDIR)/populations-stacks.Tpo -c -o src/populations-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-stacks.Tpo src/$(DEPDIR)/populations-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/populations-stacks.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc src/populations-stacks.obj: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-stacks.obj -MD -MP -MF src/$(DEPDIR)/populations-stacks.Tpo -c -o src/populations-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-stacks.Tpo src/$(DEPDIR)/populations-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/populations-stacks.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` src/populations-locus.o: src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-locus.o -MD -MP -MF src/$(DEPDIR)/populations-locus.Tpo -c -o src/populations-locus.o `test -f 'src/locus.cc' || echo '$(srcdir)/'`src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-locus.Tpo src/$(DEPDIR)/populations-locus.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/locus.cc' object='src/populations-locus.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-locus.o `test -f 'src/locus.cc' || echo '$(srcdir)/'`src/locus.cc src/populations-locus.obj: src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-locus.obj -MD -MP -MF src/$(DEPDIR)/populations-locus.Tpo -c -o src/populations-locus.obj `if test -f 'src/locus.cc'; then $(CYGPATH_W) 'src/locus.cc'; else $(CYGPATH_W) '$(srcdir)/src/locus.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-locus.Tpo src/$(DEPDIR)/populations-locus.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/locus.cc' object='src/populations-locus.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-locus.obj `if test -f 'src/locus.cc'; then $(CYGPATH_W) 'src/locus.cc'; else $(CYGPATH_W) '$(srcdir)/src/locus.cc'; fi` src/populations-DNASeq.o: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-DNASeq.o -MD -MP -MF src/$(DEPDIR)/populations-DNASeq.Tpo -c -o src/populations-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-DNASeq.Tpo src/$(DEPDIR)/populations-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/populations-DNASeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc src/populations-DNASeq.obj: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-DNASeq.obj -MD -MP -MF src/$(DEPDIR)/populations-DNASeq.Tpo -c -o src/populations-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-DNASeq.Tpo src/$(DEPDIR)/populations-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/populations-DNASeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` src/populations-DNANSeq.o: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-DNANSeq.o -MD -MP -MF src/$(DEPDIR)/populations-DNANSeq.Tpo -c -o src/populations-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-DNANSeq.Tpo src/$(DEPDIR)/populations-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/populations-DNANSeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc src/populations-DNANSeq.obj: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-DNANSeq.obj -MD -MP -MF src/$(DEPDIR)/populations-DNANSeq.Tpo -c -o src/populations-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-DNANSeq.Tpo src/$(DEPDIR)/populations-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/populations-DNANSeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` src/populations-aln_utils.o: src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-aln_utils.o -MD -MP -MF src/$(DEPDIR)/populations-aln_utils.Tpo -c -o src/populations-aln_utils.o `test -f 'src/aln_utils.cc' || echo '$(srcdir)/'`src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-aln_utils.Tpo src/$(DEPDIR)/populations-aln_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/aln_utils.cc' object='src/populations-aln_utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-aln_utils.o `test -f 'src/aln_utils.cc' || echo '$(srcdir)/'`src/aln_utils.cc src/populations-aln_utils.obj: src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-aln_utils.obj -MD -MP -MF src/$(DEPDIR)/populations-aln_utils.Tpo -c -o src/populations-aln_utils.obj `if test -f 'src/aln_utils.cc'; then $(CYGPATH_W) 'src/aln_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/aln_utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-aln_utils.Tpo src/$(DEPDIR)/populations-aln_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/aln_utils.cc' object='src/populations-aln_utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-aln_utils.obj `if test -f 'src/aln_utils.cc'; then $(CYGPATH_W) 'src/aln_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/aln_utils.cc'; fi` src/populations-genotype_dictionaries.o: src/genotype_dictionaries.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-genotype_dictionaries.o -MD -MP -MF src/$(DEPDIR)/populations-genotype_dictionaries.Tpo -c -o src/populations-genotype_dictionaries.o `test -f 'src/genotype_dictionaries.cc' || echo '$(srcdir)/'`src/genotype_dictionaries.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-genotype_dictionaries.Tpo src/$(DEPDIR)/populations-genotype_dictionaries.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/genotype_dictionaries.cc' object='src/populations-genotype_dictionaries.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-genotype_dictionaries.o `test -f 'src/genotype_dictionaries.cc' || echo '$(srcdir)/'`src/genotype_dictionaries.cc src/populations-genotype_dictionaries.obj: src/genotype_dictionaries.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-genotype_dictionaries.obj -MD -MP -MF src/$(DEPDIR)/populations-genotype_dictionaries.Tpo -c -o src/populations-genotype_dictionaries.obj `if test -f 'src/genotype_dictionaries.cc'; then $(CYGPATH_W) 'src/genotype_dictionaries.cc'; else $(CYGPATH_W) '$(srcdir)/src/genotype_dictionaries.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-genotype_dictionaries.Tpo src/$(DEPDIR)/populations-genotype_dictionaries.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/genotype_dictionaries.cc' object='src/populations-genotype_dictionaries.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-genotype_dictionaries.obj `if test -f 'src/genotype_dictionaries.cc'; then $(CYGPATH_W) 'src/genotype_dictionaries.cc'; else $(CYGPATH_W) '$(srcdir)/src/genotype_dictionaries.cc'; fi` src/populations-input.o: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-input.o -MD -MP -MF src/$(DEPDIR)/populations-input.Tpo -c -o src/populations-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-input.Tpo src/$(DEPDIR)/populations-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/populations-input.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc src/populations-input.obj: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-input.obj -MD -MP -MF src/$(DEPDIR)/populations-input.Tpo -c -o src/populations-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-input.Tpo src/$(DEPDIR)/populations-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/populations-input.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` src/populations-sql_utilities.o: src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-sql_utilities.o -MD -MP -MF src/$(DEPDIR)/populations-sql_utilities.Tpo -c -o src/populations-sql_utilities.o `test -f 'src/sql_utilities.cc' || echo '$(srcdir)/'`src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-sql_utilities.Tpo src/$(DEPDIR)/populations-sql_utilities.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/sql_utilities.cc' object='src/populations-sql_utilities.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-sql_utilities.o `test -f 'src/sql_utilities.cc' || echo '$(srcdir)/'`src/sql_utilities.cc src/populations-sql_utilities.obj: src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-sql_utilities.obj -MD -MP -MF src/$(DEPDIR)/populations-sql_utilities.Tpo -c -o src/populations-sql_utilities.obj `if test -f 'src/sql_utilities.cc'; then $(CYGPATH_W) 'src/sql_utilities.cc'; else $(CYGPATH_W) '$(srcdir)/src/sql_utilities.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-sql_utilities.Tpo src/$(DEPDIR)/populations-sql_utilities.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/sql_utilities.cc' object='src/populations-sql_utilities.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-sql_utilities.obj `if test -f 'src/sql_utilities.cc'; then $(CYGPATH_W) 'src/sql_utilities.cc'; else $(CYGPATH_W) '$(srcdir)/src/sql_utilities.cc'; fi` src/populations-renz.o: src/renz.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-renz.o -MD -MP -MF src/$(DEPDIR)/populations-renz.Tpo -c -o src/populations-renz.o `test -f 'src/renz.cc' || echo '$(srcdir)/'`src/renz.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-renz.Tpo src/$(DEPDIR)/populations-renz.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/renz.cc' object='src/populations-renz.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-renz.o `test -f 'src/renz.cc' || echo '$(srcdir)/'`src/renz.cc src/populations-renz.obj: src/renz.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-renz.obj -MD -MP -MF src/$(DEPDIR)/populations-renz.Tpo -c -o src/populations-renz.obj `if test -f 'src/renz.cc'; then $(CYGPATH_W) 'src/renz.cc'; else $(CYGPATH_W) '$(srcdir)/src/renz.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-renz.Tpo src/$(DEPDIR)/populations-renz.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/renz.cc' object='src/populations-renz.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-renz.obj `if test -f 'src/renz.cc'; then $(CYGPATH_W) 'src/renz.cc'; else $(CYGPATH_W) '$(srcdir)/src/renz.cc'; fi` src/populations-export_formats.o: src/export_formats.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-export_formats.o -MD -MP -MF src/$(DEPDIR)/populations-export_formats.Tpo -c -o src/populations-export_formats.o `test -f 'src/export_formats.cc' || echo '$(srcdir)/'`src/export_formats.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-export_formats.Tpo src/$(DEPDIR)/populations-export_formats.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/export_formats.cc' object='src/populations-export_formats.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-export_formats.o `test -f 'src/export_formats.cc' || echo '$(srcdir)/'`src/export_formats.cc src/populations-export_formats.obj: src/export_formats.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-export_formats.obj -MD -MP -MF src/$(DEPDIR)/populations-export_formats.Tpo -c -o src/populations-export_formats.obj `if test -f 'src/export_formats.cc'; then $(CYGPATH_W) 'src/export_formats.cc'; else $(CYGPATH_W) '$(srcdir)/src/export_formats.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-export_formats.Tpo src/$(DEPDIR)/populations-export_formats.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/export_formats.cc' object='src/populations-export_formats.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-export_formats.obj `if test -f 'src/export_formats.cc'; then $(CYGPATH_W) 'src/export_formats.cc'; else $(CYGPATH_W) '$(srcdir)/src/export_formats.cc'; fi` src/populations-MetaPopInfo.o: src/MetaPopInfo.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-MetaPopInfo.o -MD -MP -MF src/$(DEPDIR)/populations-MetaPopInfo.Tpo -c -o src/populations-MetaPopInfo.o `test -f 'src/MetaPopInfo.cc' || echo '$(srcdir)/'`src/MetaPopInfo.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-MetaPopInfo.Tpo src/$(DEPDIR)/populations-MetaPopInfo.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/MetaPopInfo.cc' object='src/populations-MetaPopInfo.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-MetaPopInfo.o `test -f 'src/MetaPopInfo.cc' || echo '$(srcdir)/'`src/MetaPopInfo.cc src/populations-MetaPopInfo.obj: src/MetaPopInfo.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-MetaPopInfo.obj -MD -MP -MF src/$(DEPDIR)/populations-MetaPopInfo.Tpo -c -o src/populations-MetaPopInfo.obj `if test -f 'src/MetaPopInfo.cc'; then $(CYGPATH_W) 'src/MetaPopInfo.cc'; else $(CYGPATH_W) '$(srcdir)/src/MetaPopInfo.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-MetaPopInfo.Tpo src/$(DEPDIR)/populations-MetaPopInfo.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/MetaPopInfo.cc' object='src/populations-MetaPopInfo.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-MetaPopInfo.obj `if test -f 'src/MetaPopInfo.cc'; then $(CYGPATH_W) 'src/MetaPopInfo.cc'; else $(CYGPATH_W) '$(srcdir)/src/MetaPopInfo.cc'; fi` src/populations-Vcf.o: src/Vcf.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-Vcf.o -MD -MP -MF src/$(DEPDIR)/populations-Vcf.Tpo -c -o src/populations-Vcf.o `test -f 'src/Vcf.cc' || echo '$(srcdir)/'`src/Vcf.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-Vcf.Tpo src/$(DEPDIR)/populations-Vcf.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/Vcf.cc' object='src/populations-Vcf.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-Vcf.o `test -f 'src/Vcf.cc' || echo '$(srcdir)/'`src/Vcf.cc src/populations-Vcf.obj: src/Vcf.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -MT src/populations-Vcf.obj -MD -MP -MF src/$(DEPDIR)/populations-Vcf.Tpo -c -o src/populations-Vcf.obj `if test -f 'src/Vcf.cc'; then $(CYGPATH_W) 'src/Vcf.cc'; else $(CYGPATH_W) '$(srcdir)/src/Vcf.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/populations-Vcf.Tpo src/$(DEPDIR)/populations-Vcf.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/Vcf.cc' object='src/populations-Vcf.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(populations_CXXFLAGS) $(CXXFLAGS) -c -o src/populations-Vcf.obj `if test -f 'src/Vcf.cc'; then $(CYGPATH_W) 'src/Vcf.cc'; else $(CYGPATH_W) '$(srcdir)/src/Vcf.cc'; fi` src/process_radtags-process_radtags.o: src/process_radtags.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -MT src/process_radtags-process_radtags.o -MD -MP -MF src/$(DEPDIR)/process_radtags-process_radtags.Tpo -c -o src/process_radtags-process_radtags.o `test -f 'src/process_radtags.cc' || echo '$(srcdir)/'`src/process_radtags.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_radtags-process_radtags.Tpo src/$(DEPDIR)/process_radtags-process_radtags.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/process_radtags.cc' object='src/process_radtags-process_radtags.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -c -o src/process_radtags-process_radtags.o `test -f 'src/process_radtags.cc' || echo '$(srcdir)/'`src/process_radtags.cc src/process_radtags-process_radtags.obj: src/process_radtags.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -MT src/process_radtags-process_radtags.obj -MD -MP -MF src/$(DEPDIR)/process_radtags-process_radtags.Tpo -c -o src/process_radtags-process_radtags.obj `if test -f 'src/process_radtags.cc'; then $(CYGPATH_W) 'src/process_radtags.cc'; else $(CYGPATH_W) '$(srcdir)/src/process_radtags.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_radtags-process_radtags.Tpo src/$(DEPDIR)/process_radtags-process_radtags.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/process_radtags.cc' object='src/process_radtags-process_radtags.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -c -o src/process_radtags-process_radtags.obj `if test -f 'src/process_radtags.cc'; then $(CYGPATH_W) 'src/process_radtags.cc'; else $(CYGPATH_W) '$(srcdir)/src/process_radtags.cc'; fi` src/process_radtags-utils.o: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -MT src/process_radtags-utils.o -MD -MP -MF src/$(DEPDIR)/process_radtags-utils.Tpo -c -o src/process_radtags-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_radtags-utils.Tpo src/$(DEPDIR)/process_radtags-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/process_radtags-utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -c -o src/process_radtags-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc src/process_radtags-utils.obj: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -MT src/process_radtags-utils.obj -MD -MP -MF src/$(DEPDIR)/process_radtags-utils.Tpo -c -o src/process_radtags-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_radtags-utils.Tpo src/$(DEPDIR)/process_radtags-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/process_radtags-utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -c -o src/process_radtags-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` src/process_radtags-log_utils.o: src/log_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -MT src/process_radtags-log_utils.o -MD -MP -MF src/$(DEPDIR)/process_radtags-log_utils.Tpo -c -o src/process_radtags-log_utils.o `test -f 'src/log_utils.cc' || echo '$(srcdir)/'`src/log_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_radtags-log_utils.Tpo src/$(DEPDIR)/process_radtags-log_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/log_utils.cc' object='src/process_radtags-log_utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -c -o src/process_radtags-log_utils.o `test -f 'src/log_utils.cc' || echo '$(srcdir)/'`src/log_utils.cc src/process_radtags-log_utils.obj: src/log_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -MT src/process_radtags-log_utils.obj -MD -MP -MF src/$(DEPDIR)/process_radtags-log_utils.Tpo -c -o src/process_radtags-log_utils.obj `if test -f 'src/log_utils.cc'; then $(CYGPATH_W) 'src/log_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/log_utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_radtags-log_utils.Tpo src/$(DEPDIR)/process_radtags-log_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/log_utils.cc' object='src/process_radtags-log_utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -c -o src/process_radtags-log_utils.obj `if test -f 'src/log_utils.cc'; then $(CYGPATH_W) 'src/log_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/log_utils.cc'; fi` src/process_radtags-write.o: src/write.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -MT src/process_radtags-write.o -MD -MP -MF src/$(DEPDIR)/process_radtags-write.Tpo -c -o src/process_radtags-write.o `test -f 'src/write.cc' || echo '$(srcdir)/'`src/write.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_radtags-write.Tpo src/$(DEPDIR)/process_radtags-write.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/write.cc' object='src/process_radtags-write.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -c -o src/process_radtags-write.o `test -f 'src/write.cc' || echo '$(srcdir)/'`src/write.cc src/process_radtags-write.obj: src/write.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -MT src/process_radtags-write.obj -MD -MP -MF src/$(DEPDIR)/process_radtags-write.Tpo -c -o src/process_radtags-write.obj `if test -f 'src/write.cc'; then $(CYGPATH_W) 'src/write.cc'; else $(CYGPATH_W) '$(srcdir)/src/write.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_radtags-write.Tpo src/$(DEPDIR)/process_radtags-write.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/write.cc' object='src/process_radtags-write.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -c -o src/process_radtags-write.obj `if test -f 'src/write.cc'; then $(CYGPATH_W) 'src/write.cc'; else $(CYGPATH_W) '$(srcdir)/src/write.cc'; fi` src/process_radtags-clean.o: src/clean.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -MT src/process_radtags-clean.o -MD -MP -MF src/$(DEPDIR)/process_radtags-clean.Tpo -c -o src/process_radtags-clean.o `test -f 'src/clean.cc' || echo '$(srcdir)/'`src/clean.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_radtags-clean.Tpo src/$(DEPDIR)/process_radtags-clean.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/clean.cc' object='src/process_radtags-clean.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -c -o src/process_radtags-clean.o `test -f 'src/clean.cc' || echo '$(srcdir)/'`src/clean.cc src/process_radtags-clean.obj: src/clean.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -MT src/process_radtags-clean.obj -MD -MP -MF src/$(DEPDIR)/process_radtags-clean.Tpo -c -o src/process_radtags-clean.obj `if test -f 'src/clean.cc'; then $(CYGPATH_W) 'src/clean.cc'; else $(CYGPATH_W) '$(srcdir)/src/clean.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_radtags-clean.Tpo src/$(DEPDIR)/process_radtags-clean.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/clean.cc' object='src/process_radtags-clean.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -c -o src/process_radtags-clean.obj `if test -f 'src/clean.cc'; then $(CYGPATH_W) 'src/clean.cc'; else $(CYGPATH_W) '$(srcdir)/src/clean.cc'; fi` src/process_radtags-file_io.o: src/file_io.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -MT src/process_radtags-file_io.o -MD -MP -MF src/$(DEPDIR)/process_radtags-file_io.Tpo -c -o src/process_radtags-file_io.o `test -f 'src/file_io.cc' || echo '$(srcdir)/'`src/file_io.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_radtags-file_io.Tpo src/$(DEPDIR)/process_radtags-file_io.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/file_io.cc' object='src/process_radtags-file_io.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -c -o src/process_radtags-file_io.o `test -f 'src/file_io.cc' || echo '$(srcdir)/'`src/file_io.cc src/process_radtags-file_io.obj: src/file_io.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -MT src/process_radtags-file_io.obj -MD -MP -MF src/$(DEPDIR)/process_radtags-file_io.Tpo -c -o src/process_radtags-file_io.obj `if test -f 'src/file_io.cc'; then $(CYGPATH_W) 'src/file_io.cc'; else $(CYGPATH_W) '$(srcdir)/src/file_io.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_radtags-file_io.Tpo src/$(DEPDIR)/process_radtags-file_io.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/file_io.cc' object='src/process_radtags-file_io.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -c -o src/process_radtags-file_io.obj `if test -f 'src/file_io.cc'; then $(CYGPATH_W) 'src/file_io.cc'; else $(CYGPATH_W) '$(srcdir)/src/file_io.cc'; fi` src/process_radtags-input.o: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -MT src/process_radtags-input.o -MD -MP -MF src/$(DEPDIR)/process_radtags-input.Tpo -c -o src/process_radtags-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_radtags-input.Tpo src/$(DEPDIR)/process_radtags-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/process_radtags-input.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -c -o src/process_radtags-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc src/process_radtags-input.obj: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -MT src/process_radtags-input.obj -MD -MP -MF src/$(DEPDIR)/process_radtags-input.Tpo -c -o src/process_radtags-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_radtags-input.Tpo src/$(DEPDIR)/process_radtags-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/process_radtags-input.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -c -o src/process_radtags-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` src/process_radtags-renz.o: src/renz.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -MT src/process_radtags-renz.o -MD -MP -MF src/$(DEPDIR)/process_radtags-renz.Tpo -c -o src/process_radtags-renz.o `test -f 'src/renz.cc' || echo '$(srcdir)/'`src/renz.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_radtags-renz.Tpo src/$(DEPDIR)/process_radtags-renz.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/renz.cc' object='src/process_radtags-renz.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -c -o src/process_radtags-renz.o `test -f 'src/renz.cc' || echo '$(srcdir)/'`src/renz.cc src/process_radtags-renz.obj: src/renz.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -MT src/process_radtags-renz.obj -MD -MP -MF src/$(DEPDIR)/process_radtags-renz.Tpo -c -o src/process_radtags-renz.obj `if test -f 'src/renz.cc'; then $(CYGPATH_W) 'src/renz.cc'; else $(CYGPATH_W) '$(srcdir)/src/renz.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_radtags-renz.Tpo src/$(DEPDIR)/process_radtags-renz.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/renz.cc' object='src/process_radtags-renz.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_radtags_CXXFLAGS) $(CXXFLAGS) -c -o src/process_radtags-renz.obj `if test -f 'src/renz.cc'; then $(CYGPATH_W) 'src/renz.cc'; else $(CYGPATH_W) '$(srcdir)/src/renz.cc'; fi` src/process_shortreads-process_shortreads.o: src/process_shortreads.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -MT src/process_shortreads-process_shortreads.o -MD -MP -MF src/$(DEPDIR)/process_shortreads-process_shortreads.Tpo -c -o src/process_shortreads-process_shortreads.o `test -f 'src/process_shortreads.cc' || echo '$(srcdir)/'`src/process_shortreads.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_shortreads-process_shortreads.Tpo src/$(DEPDIR)/process_shortreads-process_shortreads.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/process_shortreads.cc' object='src/process_shortreads-process_shortreads.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -c -o src/process_shortreads-process_shortreads.o `test -f 'src/process_shortreads.cc' || echo '$(srcdir)/'`src/process_shortreads.cc src/process_shortreads-process_shortreads.obj: src/process_shortreads.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -MT src/process_shortreads-process_shortreads.obj -MD -MP -MF src/$(DEPDIR)/process_shortreads-process_shortreads.Tpo -c -o src/process_shortreads-process_shortreads.obj `if test -f 'src/process_shortreads.cc'; then $(CYGPATH_W) 'src/process_shortreads.cc'; else $(CYGPATH_W) '$(srcdir)/src/process_shortreads.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_shortreads-process_shortreads.Tpo src/$(DEPDIR)/process_shortreads-process_shortreads.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/process_shortreads.cc' object='src/process_shortreads-process_shortreads.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -c -o src/process_shortreads-process_shortreads.obj `if test -f 'src/process_shortreads.cc'; then $(CYGPATH_W) 'src/process_shortreads.cc'; else $(CYGPATH_W) '$(srcdir)/src/process_shortreads.cc'; fi` src/process_shortreads-clean.o: src/clean.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -MT src/process_shortreads-clean.o -MD -MP -MF src/$(DEPDIR)/process_shortreads-clean.Tpo -c -o src/process_shortreads-clean.o `test -f 'src/clean.cc' || echo '$(srcdir)/'`src/clean.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_shortreads-clean.Tpo src/$(DEPDIR)/process_shortreads-clean.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/clean.cc' object='src/process_shortreads-clean.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -c -o src/process_shortreads-clean.o `test -f 'src/clean.cc' || echo '$(srcdir)/'`src/clean.cc src/process_shortreads-clean.obj: src/clean.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -MT src/process_shortreads-clean.obj -MD -MP -MF src/$(DEPDIR)/process_shortreads-clean.Tpo -c -o src/process_shortreads-clean.obj `if test -f 'src/clean.cc'; then $(CYGPATH_W) 'src/clean.cc'; else $(CYGPATH_W) '$(srcdir)/src/clean.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_shortreads-clean.Tpo src/$(DEPDIR)/process_shortreads-clean.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/clean.cc' object='src/process_shortreads-clean.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -c -o src/process_shortreads-clean.obj `if test -f 'src/clean.cc'; then $(CYGPATH_W) 'src/clean.cc'; else $(CYGPATH_W) '$(srcdir)/src/clean.cc'; fi` src/process_shortreads-file_io.o: src/file_io.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -MT src/process_shortreads-file_io.o -MD -MP -MF src/$(DEPDIR)/process_shortreads-file_io.Tpo -c -o src/process_shortreads-file_io.o `test -f 'src/file_io.cc' || echo '$(srcdir)/'`src/file_io.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_shortreads-file_io.Tpo src/$(DEPDIR)/process_shortreads-file_io.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/file_io.cc' object='src/process_shortreads-file_io.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -c -o src/process_shortreads-file_io.o `test -f 'src/file_io.cc' || echo '$(srcdir)/'`src/file_io.cc src/process_shortreads-file_io.obj: src/file_io.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -MT src/process_shortreads-file_io.obj -MD -MP -MF src/$(DEPDIR)/process_shortreads-file_io.Tpo -c -o src/process_shortreads-file_io.obj `if test -f 'src/file_io.cc'; then $(CYGPATH_W) 'src/file_io.cc'; else $(CYGPATH_W) '$(srcdir)/src/file_io.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_shortreads-file_io.Tpo src/$(DEPDIR)/process_shortreads-file_io.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/file_io.cc' object='src/process_shortreads-file_io.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -c -o src/process_shortreads-file_io.obj `if test -f 'src/file_io.cc'; then $(CYGPATH_W) 'src/file_io.cc'; else $(CYGPATH_W) '$(srcdir)/src/file_io.cc'; fi` src/process_shortreads-utils.o: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -MT src/process_shortreads-utils.o -MD -MP -MF src/$(DEPDIR)/process_shortreads-utils.Tpo -c -o src/process_shortreads-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_shortreads-utils.Tpo src/$(DEPDIR)/process_shortreads-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/process_shortreads-utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -c -o src/process_shortreads-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc src/process_shortreads-utils.obj: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -MT src/process_shortreads-utils.obj -MD -MP -MF src/$(DEPDIR)/process_shortreads-utils.Tpo -c -o src/process_shortreads-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_shortreads-utils.Tpo src/$(DEPDIR)/process_shortreads-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/process_shortreads-utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -c -o src/process_shortreads-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` src/process_shortreads-log_utils.o: src/log_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -MT src/process_shortreads-log_utils.o -MD -MP -MF src/$(DEPDIR)/process_shortreads-log_utils.Tpo -c -o src/process_shortreads-log_utils.o `test -f 'src/log_utils.cc' || echo '$(srcdir)/'`src/log_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_shortreads-log_utils.Tpo src/$(DEPDIR)/process_shortreads-log_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/log_utils.cc' object='src/process_shortreads-log_utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -c -o src/process_shortreads-log_utils.o `test -f 'src/log_utils.cc' || echo '$(srcdir)/'`src/log_utils.cc src/process_shortreads-log_utils.obj: src/log_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -MT src/process_shortreads-log_utils.obj -MD -MP -MF src/$(DEPDIR)/process_shortreads-log_utils.Tpo -c -o src/process_shortreads-log_utils.obj `if test -f 'src/log_utils.cc'; then $(CYGPATH_W) 'src/log_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/log_utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_shortreads-log_utils.Tpo src/$(DEPDIR)/process_shortreads-log_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/log_utils.cc' object='src/process_shortreads-log_utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -c -o src/process_shortreads-log_utils.obj `if test -f 'src/log_utils.cc'; then $(CYGPATH_W) 'src/log_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/log_utils.cc'; fi` src/process_shortreads-write.o: src/write.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -MT src/process_shortreads-write.o -MD -MP -MF src/$(DEPDIR)/process_shortreads-write.Tpo -c -o src/process_shortreads-write.o `test -f 'src/write.cc' || echo '$(srcdir)/'`src/write.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_shortreads-write.Tpo src/$(DEPDIR)/process_shortreads-write.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/write.cc' object='src/process_shortreads-write.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -c -o src/process_shortreads-write.o `test -f 'src/write.cc' || echo '$(srcdir)/'`src/write.cc src/process_shortreads-write.obj: src/write.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -MT src/process_shortreads-write.obj -MD -MP -MF src/$(DEPDIR)/process_shortreads-write.Tpo -c -o src/process_shortreads-write.obj `if test -f 'src/write.cc'; then $(CYGPATH_W) 'src/write.cc'; else $(CYGPATH_W) '$(srcdir)/src/write.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_shortreads-write.Tpo src/$(DEPDIR)/process_shortreads-write.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/write.cc' object='src/process_shortreads-write.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -c -o src/process_shortreads-write.obj `if test -f 'src/write.cc'; then $(CYGPATH_W) 'src/write.cc'; else $(CYGPATH_W) '$(srcdir)/src/write.cc'; fi` src/process_shortreads-kmers.o: src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -MT src/process_shortreads-kmers.o -MD -MP -MF src/$(DEPDIR)/process_shortreads-kmers.Tpo -c -o src/process_shortreads-kmers.o `test -f 'src/kmers.cc' || echo '$(srcdir)/'`src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_shortreads-kmers.Tpo src/$(DEPDIR)/process_shortreads-kmers.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/kmers.cc' object='src/process_shortreads-kmers.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -c -o src/process_shortreads-kmers.o `test -f 'src/kmers.cc' || echo '$(srcdir)/'`src/kmers.cc src/process_shortreads-kmers.obj: src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -MT src/process_shortreads-kmers.obj -MD -MP -MF src/$(DEPDIR)/process_shortreads-kmers.Tpo -c -o src/process_shortreads-kmers.obj `if test -f 'src/kmers.cc'; then $(CYGPATH_W) 'src/kmers.cc'; else $(CYGPATH_W) '$(srcdir)/src/kmers.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_shortreads-kmers.Tpo src/$(DEPDIR)/process_shortreads-kmers.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/kmers.cc' object='src/process_shortreads-kmers.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -c -o src/process_shortreads-kmers.obj `if test -f 'src/kmers.cc'; then $(CYGPATH_W) 'src/kmers.cc'; else $(CYGPATH_W) '$(srcdir)/src/kmers.cc'; fi` src/process_shortreads-input.o: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -MT src/process_shortreads-input.o -MD -MP -MF src/$(DEPDIR)/process_shortreads-input.Tpo -c -o src/process_shortreads-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_shortreads-input.Tpo src/$(DEPDIR)/process_shortreads-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/process_shortreads-input.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -c -o src/process_shortreads-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc src/process_shortreads-input.obj: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -MT src/process_shortreads-input.obj -MD -MP -MF src/$(DEPDIR)/process_shortreads-input.Tpo -c -o src/process_shortreads-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/process_shortreads-input.Tpo src/$(DEPDIR)/process_shortreads-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/process_shortreads-input.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(process_shortreads_CXXFLAGS) $(CXXFLAGS) -c -o src/process_shortreads-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` src/pstacks-pstacks.o: src/pstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -MT src/pstacks-pstacks.o -MD -MP -MF src/$(DEPDIR)/pstacks-pstacks.Tpo -c -o src/pstacks-pstacks.o `test -f 'src/pstacks.cc' || echo '$(srcdir)/'`src/pstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/pstacks-pstacks.Tpo src/$(DEPDIR)/pstacks-pstacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/pstacks.cc' object='src/pstacks-pstacks.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/pstacks-pstacks.o `test -f 'src/pstacks.cc' || echo '$(srcdir)/'`src/pstacks.cc src/pstacks-pstacks.obj: src/pstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -MT src/pstacks-pstacks.obj -MD -MP -MF src/$(DEPDIR)/pstacks-pstacks.Tpo -c -o src/pstacks-pstacks.obj `if test -f 'src/pstacks.cc'; then $(CYGPATH_W) 'src/pstacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/pstacks.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/pstacks-pstacks.Tpo src/$(DEPDIR)/pstacks-pstacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/pstacks.cc' object='src/pstacks-pstacks.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/pstacks-pstacks.obj `if test -f 'src/pstacks.cc'; then $(CYGPATH_W) 'src/pstacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/pstacks.cc'; fi` src/pstacks-stacks.o: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -MT src/pstacks-stacks.o -MD -MP -MF src/$(DEPDIR)/pstacks-stacks.Tpo -c -o src/pstacks-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/pstacks-stacks.Tpo src/$(DEPDIR)/pstacks-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/pstacks-stacks.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/pstacks-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc src/pstacks-stacks.obj: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -MT src/pstacks-stacks.obj -MD -MP -MF src/$(DEPDIR)/pstacks-stacks.Tpo -c -o src/pstacks-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/pstacks-stacks.Tpo src/$(DEPDIR)/pstacks-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/pstacks-stacks.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/pstacks-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` src/pstacks-mstack.o: src/mstack.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -MT src/pstacks-mstack.o -MD -MP -MF src/$(DEPDIR)/pstacks-mstack.Tpo -c -o src/pstacks-mstack.o `test -f 'src/mstack.cc' || echo '$(srcdir)/'`src/mstack.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/pstacks-mstack.Tpo src/$(DEPDIR)/pstacks-mstack.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/mstack.cc' object='src/pstacks-mstack.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/pstacks-mstack.o `test -f 'src/mstack.cc' || echo '$(srcdir)/'`src/mstack.cc src/pstacks-mstack.obj: src/mstack.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -MT src/pstacks-mstack.obj -MD -MP -MF src/$(DEPDIR)/pstacks-mstack.Tpo -c -o src/pstacks-mstack.obj `if test -f 'src/mstack.cc'; then $(CYGPATH_W) 'src/mstack.cc'; else $(CYGPATH_W) '$(srcdir)/src/mstack.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/pstacks-mstack.Tpo src/$(DEPDIR)/pstacks-mstack.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/mstack.cc' object='src/pstacks-mstack.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/pstacks-mstack.obj `if test -f 'src/mstack.cc'; then $(CYGPATH_W) 'src/mstack.cc'; else $(CYGPATH_W) '$(srcdir)/src/mstack.cc'; fi` src/pstacks-DNANSeq.o: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -MT src/pstacks-DNANSeq.o -MD -MP -MF src/$(DEPDIR)/pstacks-DNANSeq.Tpo -c -o src/pstacks-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/pstacks-DNANSeq.Tpo src/$(DEPDIR)/pstacks-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/pstacks-DNANSeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/pstacks-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc src/pstacks-DNANSeq.obj: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -MT src/pstacks-DNANSeq.obj -MD -MP -MF src/$(DEPDIR)/pstacks-DNANSeq.Tpo -c -o src/pstacks-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/pstacks-DNANSeq.Tpo src/$(DEPDIR)/pstacks-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/pstacks-DNANSeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/pstacks-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` src/pstacks-DNASeq.o: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -MT src/pstacks-DNASeq.o -MD -MP -MF src/$(DEPDIR)/pstacks-DNASeq.Tpo -c -o src/pstacks-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/pstacks-DNASeq.Tpo src/$(DEPDIR)/pstacks-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/pstacks-DNASeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/pstacks-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc src/pstacks-DNASeq.obj: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -MT src/pstacks-DNASeq.obj -MD -MP -MF src/$(DEPDIR)/pstacks-DNASeq.Tpo -c -o src/pstacks-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/pstacks-DNASeq.Tpo src/$(DEPDIR)/pstacks-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/pstacks-DNASeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/pstacks-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` src/pstacks-models.o: src/models.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -MT src/pstacks-models.o -MD -MP -MF src/$(DEPDIR)/pstacks-models.Tpo -c -o src/pstacks-models.o `test -f 'src/models.cc' || echo '$(srcdir)/'`src/models.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/pstacks-models.Tpo src/$(DEPDIR)/pstacks-models.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/models.cc' object='src/pstacks-models.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/pstacks-models.o `test -f 'src/models.cc' || echo '$(srcdir)/'`src/models.cc src/pstacks-models.obj: src/models.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -MT src/pstacks-models.obj -MD -MP -MF src/$(DEPDIR)/pstacks-models.Tpo -c -o src/pstacks-models.obj `if test -f 'src/models.cc'; then $(CYGPATH_W) 'src/models.cc'; else $(CYGPATH_W) '$(srcdir)/src/models.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/pstacks-models.Tpo src/$(DEPDIR)/pstacks-models.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/models.cc' object='src/pstacks-models.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/pstacks-models.obj `if test -f 'src/models.cc'; then $(CYGPATH_W) 'src/models.cc'; else $(CYGPATH_W) '$(srcdir)/src/models.cc'; fi` src/pstacks-utils.o: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -MT src/pstacks-utils.o -MD -MP -MF src/$(DEPDIR)/pstacks-utils.Tpo -c -o src/pstacks-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/pstacks-utils.Tpo src/$(DEPDIR)/pstacks-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/pstacks-utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/pstacks-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc src/pstacks-utils.obj: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -MT src/pstacks-utils.obj -MD -MP -MF src/$(DEPDIR)/pstacks-utils.Tpo -c -o src/pstacks-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/pstacks-utils.Tpo src/$(DEPDIR)/pstacks-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/pstacks-utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/pstacks-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` src/pstacks-input.o: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -MT src/pstacks-input.o -MD -MP -MF src/$(DEPDIR)/pstacks-input.Tpo -c -o src/pstacks-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/pstacks-input.Tpo src/$(DEPDIR)/pstacks-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/pstacks-input.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/pstacks-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc src/pstacks-input.obj: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -MT src/pstacks-input.obj -MD -MP -MF src/$(DEPDIR)/pstacks-input.Tpo -c -o src/pstacks-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/pstacks-input.Tpo src/$(DEPDIR)/pstacks-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/pstacks-input.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(pstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/pstacks-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` src/rxstacks-rxstacks.o: src/rxstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-rxstacks.o -MD -MP -MF src/$(DEPDIR)/rxstacks-rxstacks.Tpo -c -o src/rxstacks-rxstacks.o `test -f 'src/rxstacks.cc' || echo '$(srcdir)/'`src/rxstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-rxstacks.Tpo src/$(DEPDIR)/rxstacks-rxstacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/rxstacks.cc' object='src/rxstacks-rxstacks.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-rxstacks.o `test -f 'src/rxstacks.cc' || echo '$(srcdir)/'`src/rxstacks.cc src/rxstacks-rxstacks.obj: src/rxstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-rxstacks.obj -MD -MP -MF src/$(DEPDIR)/rxstacks-rxstacks.Tpo -c -o src/rxstacks-rxstacks.obj `if test -f 'src/rxstacks.cc'; then $(CYGPATH_W) 'src/rxstacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/rxstacks.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-rxstacks.Tpo src/$(DEPDIR)/rxstacks-rxstacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/rxstacks.cc' object='src/rxstacks-rxstacks.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-rxstacks.obj `if test -f 'src/rxstacks.cc'; then $(CYGPATH_W) 'src/rxstacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/rxstacks.cc'; fi` src/rxstacks-stacks.o: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-stacks.o -MD -MP -MF src/$(DEPDIR)/rxstacks-stacks.Tpo -c -o src/rxstacks-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-stacks.Tpo src/$(DEPDIR)/rxstacks-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/rxstacks-stacks.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc src/rxstacks-stacks.obj: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-stacks.obj -MD -MP -MF src/$(DEPDIR)/rxstacks-stacks.Tpo -c -o src/rxstacks-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-stacks.Tpo src/$(DEPDIR)/rxstacks-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/rxstacks-stacks.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` src/rxstacks-locus.o: src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-locus.o -MD -MP -MF src/$(DEPDIR)/rxstacks-locus.Tpo -c -o src/rxstacks-locus.o `test -f 'src/locus.cc' || echo '$(srcdir)/'`src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-locus.Tpo src/$(DEPDIR)/rxstacks-locus.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/locus.cc' object='src/rxstacks-locus.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-locus.o `test -f 'src/locus.cc' || echo '$(srcdir)/'`src/locus.cc src/rxstacks-locus.obj: src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-locus.obj -MD -MP -MF src/$(DEPDIR)/rxstacks-locus.Tpo -c -o src/rxstacks-locus.obj `if test -f 'src/locus.cc'; then $(CYGPATH_W) 'src/locus.cc'; else $(CYGPATH_W) '$(srcdir)/src/locus.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-locus.Tpo src/$(DEPDIR)/rxstacks-locus.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/locus.cc' object='src/rxstacks-locus.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-locus.obj `if test -f 'src/locus.cc'; then $(CYGPATH_W) 'src/locus.cc'; else $(CYGPATH_W) '$(srcdir)/src/locus.cc'; fi` src/rxstacks-DNANSeq.o: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-DNANSeq.o -MD -MP -MF src/$(DEPDIR)/rxstacks-DNANSeq.Tpo -c -o src/rxstacks-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-DNANSeq.Tpo src/$(DEPDIR)/rxstacks-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/rxstacks-DNANSeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc src/rxstacks-DNANSeq.obj: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-DNANSeq.obj -MD -MP -MF src/$(DEPDIR)/rxstacks-DNANSeq.Tpo -c -o src/rxstacks-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-DNANSeq.Tpo src/$(DEPDIR)/rxstacks-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/rxstacks-DNANSeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` src/rxstacks-DNASeq.o: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-DNASeq.o -MD -MP -MF src/$(DEPDIR)/rxstacks-DNASeq.Tpo -c -o src/rxstacks-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-DNASeq.Tpo src/$(DEPDIR)/rxstacks-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/rxstacks-DNASeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc src/rxstacks-DNASeq.obj: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-DNASeq.obj -MD -MP -MF src/$(DEPDIR)/rxstacks-DNASeq.Tpo -c -o src/rxstacks-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-DNASeq.Tpo src/$(DEPDIR)/rxstacks-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/rxstacks-DNASeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` src/rxstacks-mst.o: src/mst.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-mst.o -MD -MP -MF src/$(DEPDIR)/rxstacks-mst.Tpo -c -o src/rxstacks-mst.o `test -f 'src/mst.cc' || echo '$(srcdir)/'`src/mst.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-mst.Tpo src/$(DEPDIR)/rxstacks-mst.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/mst.cc' object='src/rxstacks-mst.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-mst.o `test -f 'src/mst.cc' || echo '$(srcdir)/'`src/mst.cc src/rxstacks-mst.obj: src/mst.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-mst.obj -MD -MP -MF src/$(DEPDIR)/rxstacks-mst.Tpo -c -o src/rxstacks-mst.obj `if test -f 'src/mst.cc'; then $(CYGPATH_W) 'src/mst.cc'; else $(CYGPATH_W) '$(srcdir)/src/mst.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-mst.Tpo src/$(DEPDIR)/rxstacks-mst.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/mst.cc' object='src/rxstacks-mst.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-mst.obj `if test -f 'src/mst.cc'; then $(CYGPATH_W) 'src/mst.cc'; else $(CYGPATH_W) '$(srcdir)/src/mst.cc'; fi` src/rxstacks-models.o: src/models.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-models.o -MD -MP -MF src/$(DEPDIR)/rxstacks-models.Tpo -c -o src/rxstacks-models.o `test -f 'src/models.cc' || echo '$(srcdir)/'`src/models.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-models.Tpo src/$(DEPDIR)/rxstacks-models.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/models.cc' object='src/rxstacks-models.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-models.o `test -f 'src/models.cc' || echo '$(srcdir)/'`src/models.cc src/rxstacks-models.obj: src/models.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-models.obj -MD -MP -MF src/$(DEPDIR)/rxstacks-models.Tpo -c -o src/rxstacks-models.obj `if test -f 'src/models.cc'; then $(CYGPATH_W) 'src/models.cc'; else $(CYGPATH_W) '$(srcdir)/src/models.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-models.Tpo src/$(DEPDIR)/rxstacks-models.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/models.cc' object='src/rxstacks-models.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-models.obj `if test -f 'src/models.cc'; then $(CYGPATH_W) 'src/models.cc'; else $(CYGPATH_W) '$(srcdir)/src/models.cc'; fi` src/rxstacks-aln_utils.o: src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-aln_utils.o -MD -MP -MF src/$(DEPDIR)/rxstacks-aln_utils.Tpo -c -o src/rxstacks-aln_utils.o `test -f 'src/aln_utils.cc' || echo '$(srcdir)/'`src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-aln_utils.Tpo src/$(DEPDIR)/rxstacks-aln_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/aln_utils.cc' object='src/rxstacks-aln_utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-aln_utils.o `test -f 'src/aln_utils.cc' || echo '$(srcdir)/'`src/aln_utils.cc src/rxstacks-aln_utils.obj: src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-aln_utils.obj -MD -MP -MF src/$(DEPDIR)/rxstacks-aln_utils.Tpo -c -o src/rxstacks-aln_utils.obj `if test -f 'src/aln_utils.cc'; then $(CYGPATH_W) 'src/aln_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/aln_utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-aln_utils.Tpo src/$(DEPDIR)/rxstacks-aln_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/aln_utils.cc' object='src/rxstacks-aln_utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-aln_utils.obj `if test -f 'src/aln_utils.cc'; then $(CYGPATH_W) 'src/aln_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/aln_utils.cc'; fi` src/rxstacks-utils.o: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-utils.o -MD -MP -MF src/$(DEPDIR)/rxstacks-utils.Tpo -c -o src/rxstacks-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-utils.Tpo src/$(DEPDIR)/rxstacks-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/rxstacks-utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc src/rxstacks-utils.obj: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-utils.obj -MD -MP -MF src/$(DEPDIR)/rxstacks-utils.Tpo -c -o src/rxstacks-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-utils.Tpo src/$(DEPDIR)/rxstacks-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/rxstacks-utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` src/rxstacks-input.o: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-input.o -MD -MP -MF src/$(DEPDIR)/rxstacks-input.Tpo -c -o src/rxstacks-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-input.Tpo src/$(DEPDIR)/rxstacks-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/rxstacks-input.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc src/rxstacks-input.obj: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-input.obj -MD -MP -MF src/$(DEPDIR)/rxstacks-input.Tpo -c -o src/rxstacks-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-input.Tpo src/$(DEPDIR)/rxstacks-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/rxstacks-input.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` src/rxstacks-sql_utilities.o: src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-sql_utilities.o -MD -MP -MF src/$(DEPDIR)/rxstacks-sql_utilities.Tpo -c -o src/rxstacks-sql_utilities.o `test -f 'src/sql_utilities.cc' || echo '$(srcdir)/'`src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-sql_utilities.Tpo src/$(DEPDIR)/rxstacks-sql_utilities.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/sql_utilities.cc' object='src/rxstacks-sql_utilities.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-sql_utilities.o `test -f 'src/sql_utilities.cc' || echo '$(srcdir)/'`src/sql_utilities.cc src/rxstacks-sql_utilities.obj: src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-sql_utilities.obj -MD -MP -MF src/$(DEPDIR)/rxstacks-sql_utilities.Tpo -c -o src/rxstacks-sql_utilities.obj `if test -f 'src/sql_utilities.cc'; then $(CYGPATH_W) 'src/sql_utilities.cc'; else $(CYGPATH_W) '$(srcdir)/src/sql_utilities.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-sql_utilities.Tpo src/$(DEPDIR)/rxstacks-sql_utilities.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/sql_utilities.cc' object='src/rxstacks-sql_utilities.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-sql_utilities.obj `if test -f 'src/sql_utilities.cc'; then $(CYGPATH_W) 'src/sql_utilities.cc'; else $(CYGPATH_W) '$(srcdir)/src/sql_utilities.cc'; fi` src/rxstacks-MetaPopInfo.o: src/MetaPopInfo.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-MetaPopInfo.o -MD -MP -MF src/$(DEPDIR)/rxstacks-MetaPopInfo.Tpo -c -o src/rxstacks-MetaPopInfo.o `test -f 'src/MetaPopInfo.cc' || echo '$(srcdir)/'`src/MetaPopInfo.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-MetaPopInfo.Tpo src/$(DEPDIR)/rxstacks-MetaPopInfo.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/MetaPopInfo.cc' object='src/rxstacks-MetaPopInfo.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-MetaPopInfo.o `test -f 'src/MetaPopInfo.cc' || echo '$(srcdir)/'`src/MetaPopInfo.cc src/rxstacks-MetaPopInfo.obj: src/MetaPopInfo.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -MT src/rxstacks-MetaPopInfo.obj -MD -MP -MF src/$(DEPDIR)/rxstacks-MetaPopInfo.Tpo -c -o src/rxstacks-MetaPopInfo.obj `if test -f 'src/MetaPopInfo.cc'; then $(CYGPATH_W) 'src/MetaPopInfo.cc'; else $(CYGPATH_W) '$(srcdir)/src/MetaPopInfo.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/rxstacks-MetaPopInfo.Tpo src/$(DEPDIR)/rxstacks-MetaPopInfo.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/MetaPopInfo.cc' object='src/rxstacks-MetaPopInfo.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(rxstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/rxstacks-MetaPopInfo.obj `if test -f 'src/MetaPopInfo.cc'; then $(CYGPATH_W) 'src/MetaPopInfo.cc'; else $(CYGPATH_W) '$(srcdir)/src/MetaPopInfo.cc'; fi` src/sstacks-sstacks.o: src/sstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-sstacks.o -MD -MP -MF src/$(DEPDIR)/sstacks-sstacks.Tpo -c -o src/sstacks-sstacks.o `test -f 'src/sstacks.cc' || echo '$(srcdir)/'`src/sstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-sstacks.Tpo src/$(DEPDIR)/sstacks-sstacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/sstacks.cc' object='src/sstacks-sstacks.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-sstacks.o `test -f 'src/sstacks.cc' || echo '$(srcdir)/'`src/sstacks.cc src/sstacks-sstacks.obj: src/sstacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-sstacks.obj -MD -MP -MF src/$(DEPDIR)/sstacks-sstacks.Tpo -c -o src/sstacks-sstacks.obj `if test -f 'src/sstacks.cc'; then $(CYGPATH_W) 'src/sstacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/sstacks.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-sstacks.Tpo src/$(DEPDIR)/sstacks-sstacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/sstacks.cc' object='src/sstacks-sstacks.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-sstacks.obj `if test -f 'src/sstacks.cc'; then $(CYGPATH_W) 'src/sstacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/sstacks.cc'; fi` src/sstacks-stacks.o: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-stacks.o -MD -MP -MF src/$(DEPDIR)/sstacks-stacks.Tpo -c -o src/sstacks-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-stacks.Tpo src/$(DEPDIR)/sstacks-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/sstacks-stacks.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc src/sstacks-stacks.obj: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-stacks.obj -MD -MP -MF src/$(DEPDIR)/sstacks-stacks.Tpo -c -o src/sstacks-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-stacks.Tpo src/$(DEPDIR)/sstacks-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/sstacks-stacks.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` src/sstacks-locus.o: src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-locus.o -MD -MP -MF src/$(DEPDIR)/sstacks-locus.Tpo -c -o src/sstacks-locus.o `test -f 'src/locus.cc' || echo '$(srcdir)/'`src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-locus.Tpo src/$(DEPDIR)/sstacks-locus.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/locus.cc' object='src/sstacks-locus.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-locus.o `test -f 'src/locus.cc' || echo '$(srcdir)/'`src/locus.cc src/sstacks-locus.obj: src/locus.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-locus.obj -MD -MP -MF src/$(DEPDIR)/sstacks-locus.Tpo -c -o src/sstacks-locus.obj `if test -f 'src/locus.cc'; then $(CYGPATH_W) 'src/locus.cc'; else $(CYGPATH_W) '$(srcdir)/src/locus.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-locus.Tpo src/$(DEPDIR)/sstacks-locus.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/locus.cc' object='src/sstacks-locus.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-locus.obj `if test -f 'src/locus.cc'; then $(CYGPATH_W) 'src/locus.cc'; else $(CYGPATH_W) '$(srcdir)/src/locus.cc'; fi` src/sstacks-kmers.o: src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-kmers.o -MD -MP -MF src/$(DEPDIR)/sstacks-kmers.Tpo -c -o src/sstacks-kmers.o `test -f 'src/kmers.cc' || echo '$(srcdir)/'`src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-kmers.Tpo src/$(DEPDIR)/sstacks-kmers.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/kmers.cc' object='src/sstacks-kmers.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-kmers.o `test -f 'src/kmers.cc' || echo '$(srcdir)/'`src/kmers.cc src/sstacks-kmers.obj: src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-kmers.obj -MD -MP -MF src/$(DEPDIR)/sstacks-kmers.Tpo -c -o src/sstacks-kmers.obj `if test -f 'src/kmers.cc'; then $(CYGPATH_W) 'src/kmers.cc'; else $(CYGPATH_W) '$(srcdir)/src/kmers.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-kmers.Tpo src/$(DEPDIR)/sstacks-kmers.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/kmers.cc' object='src/sstacks-kmers.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-kmers.obj `if test -f 'src/kmers.cc'; then $(CYGPATH_W) 'src/kmers.cc'; else $(CYGPATH_W) '$(srcdir)/src/kmers.cc'; fi` src/sstacks-aln_utils.o: src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-aln_utils.o -MD -MP -MF src/$(DEPDIR)/sstacks-aln_utils.Tpo -c -o src/sstacks-aln_utils.o `test -f 'src/aln_utils.cc' || echo '$(srcdir)/'`src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-aln_utils.Tpo src/$(DEPDIR)/sstacks-aln_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/aln_utils.cc' object='src/sstacks-aln_utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-aln_utils.o `test -f 'src/aln_utils.cc' || echo '$(srcdir)/'`src/aln_utils.cc src/sstacks-aln_utils.obj: src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-aln_utils.obj -MD -MP -MF src/$(DEPDIR)/sstacks-aln_utils.Tpo -c -o src/sstacks-aln_utils.obj `if test -f 'src/aln_utils.cc'; then $(CYGPATH_W) 'src/aln_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/aln_utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-aln_utils.Tpo src/$(DEPDIR)/sstacks-aln_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/aln_utils.cc' object='src/sstacks-aln_utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-aln_utils.obj `if test -f 'src/aln_utils.cc'; then $(CYGPATH_W) 'src/aln_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/aln_utils.cc'; fi` src/sstacks-DNASeq.o: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-DNASeq.o -MD -MP -MF src/$(DEPDIR)/sstacks-DNASeq.Tpo -c -o src/sstacks-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-DNASeq.Tpo src/$(DEPDIR)/sstacks-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/sstacks-DNASeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc src/sstacks-DNASeq.obj: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-DNASeq.obj -MD -MP -MF src/$(DEPDIR)/sstacks-DNASeq.Tpo -c -o src/sstacks-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-DNASeq.Tpo src/$(DEPDIR)/sstacks-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/sstacks-DNASeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` src/sstacks-DNANSeq.o: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-DNANSeq.o -MD -MP -MF src/$(DEPDIR)/sstacks-DNANSeq.Tpo -c -o src/sstacks-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-DNANSeq.Tpo src/$(DEPDIR)/sstacks-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/sstacks-DNANSeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc src/sstacks-DNANSeq.obj: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-DNANSeq.obj -MD -MP -MF src/$(DEPDIR)/sstacks-DNANSeq.Tpo -c -o src/sstacks-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-DNANSeq.Tpo src/$(DEPDIR)/sstacks-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/sstacks-DNANSeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` src/sstacks-utils.o: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-utils.o -MD -MP -MF src/$(DEPDIR)/sstacks-utils.Tpo -c -o src/sstacks-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-utils.Tpo src/$(DEPDIR)/sstacks-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/sstacks-utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc src/sstacks-utils.obj: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-utils.obj -MD -MP -MF src/$(DEPDIR)/sstacks-utils.Tpo -c -o src/sstacks-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-utils.Tpo src/$(DEPDIR)/sstacks-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/sstacks-utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` src/sstacks-input.o: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-input.o -MD -MP -MF src/$(DEPDIR)/sstacks-input.Tpo -c -o src/sstacks-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-input.Tpo src/$(DEPDIR)/sstacks-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/sstacks-input.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc src/sstacks-input.obj: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-input.obj -MD -MP -MF src/$(DEPDIR)/sstacks-input.Tpo -c -o src/sstacks-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-input.Tpo src/$(DEPDIR)/sstacks-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/sstacks-input.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` src/sstacks-sql_utilities.o: src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-sql_utilities.o -MD -MP -MF src/$(DEPDIR)/sstacks-sql_utilities.Tpo -c -o src/sstacks-sql_utilities.o `test -f 'src/sql_utilities.cc' || echo '$(srcdir)/'`src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-sql_utilities.Tpo src/$(DEPDIR)/sstacks-sql_utilities.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/sql_utilities.cc' object='src/sstacks-sql_utilities.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-sql_utilities.o `test -f 'src/sql_utilities.cc' || echo '$(srcdir)/'`src/sql_utilities.cc src/sstacks-sql_utilities.obj: src/sql_utilities.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -MT src/sstacks-sql_utilities.obj -MD -MP -MF src/$(DEPDIR)/sstacks-sql_utilities.Tpo -c -o src/sstacks-sql_utilities.obj `if test -f 'src/sql_utilities.cc'; then $(CYGPATH_W) 'src/sql_utilities.cc'; else $(CYGPATH_W) '$(srcdir)/src/sql_utilities.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/sstacks-sql_utilities.Tpo src/$(DEPDIR)/sstacks-sql_utilities.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/sql_utilities.cc' object='src/sstacks-sql_utilities.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(sstacks_CXXFLAGS) $(CXXFLAGS) -c -o src/sstacks-sql_utilities.obj `if test -f 'src/sql_utilities.cc'; then $(CYGPATH_W) 'src/sql_utilities.cc'; else $(CYGPATH_W) '$(srcdir)/src/sql_utilities.cc'; fi` src/ustacks-ustacks.o: src/ustacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-ustacks.o -MD -MP -MF src/$(DEPDIR)/ustacks-ustacks.Tpo -c -o src/ustacks-ustacks.o `test -f 'src/ustacks.cc' || echo '$(srcdir)/'`src/ustacks.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-ustacks.Tpo src/$(DEPDIR)/ustacks-ustacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/ustacks.cc' object='src/ustacks-ustacks.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-ustacks.o `test -f 'src/ustacks.cc' || echo '$(srcdir)/'`src/ustacks.cc src/ustacks-ustacks.obj: src/ustacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-ustacks.obj -MD -MP -MF src/$(DEPDIR)/ustacks-ustacks.Tpo -c -o src/ustacks-ustacks.obj `if test -f 'src/ustacks.cc'; then $(CYGPATH_W) 'src/ustacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/ustacks.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-ustacks.Tpo src/$(DEPDIR)/ustacks-ustacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/ustacks.cc' object='src/ustacks-ustacks.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-ustacks.obj `if test -f 'src/ustacks.cc'; then $(CYGPATH_W) 'src/ustacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/ustacks.cc'; fi` src/ustacks-stacks.o: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-stacks.o -MD -MP -MF src/$(DEPDIR)/ustacks-stacks.Tpo -c -o src/ustacks-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-stacks.Tpo src/$(DEPDIR)/ustacks-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/ustacks-stacks.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-stacks.o `test -f 'src/stacks.cc' || echo '$(srcdir)/'`src/stacks.cc src/ustacks-stacks.obj: src/stacks.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-stacks.obj -MD -MP -MF src/$(DEPDIR)/ustacks-stacks.Tpo -c -o src/ustacks-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-stacks.Tpo src/$(DEPDIR)/ustacks-stacks.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/stacks.cc' object='src/ustacks-stacks.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-stacks.obj `if test -f 'src/stacks.cc'; then $(CYGPATH_W) 'src/stacks.cc'; else $(CYGPATH_W) '$(srcdir)/src/stacks.cc'; fi` src/ustacks-mstack.o: src/mstack.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-mstack.o -MD -MP -MF src/$(DEPDIR)/ustacks-mstack.Tpo -c -o src/ustacks-mstack.o `test -f 'src/mstack.cc' || echo '$(srcdir)/'`src/mstack.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-mstack.Tpo src/$(DEPDIR)/ustacks-mstack.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/mstack.cc' object='src/ustacks-mstack.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-mstack.o `test -f 'src/mstack.cc' || echo '$(srcdir)/'`src/mstack.cc src/ustacks-mstack.obj: src/mstack.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-mstack.obj -MD -MP -MF src/$(DEPDIR)/ustacks-mstack.Tpo -c -o src/ustacks-mstack.obj `if test -f 'src/mstack.cc'; then $(CYGPATH_W) 'src/mstack.cc'; else $(CYGPATH_W) '$(srcdir)/src/mstack.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-mstack.Tpo src/$(DEPDIR)/ustacks-mstack.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/mstack.cc' object='src/ustacks-mstack.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-mstack.obj `if test -f 'src/mstack.cc'; then $(CYGPATH_W) 'src/mstack.cc'; else $(CYGPATH_W) '$(srcdir)/src/mstack.cc'; fi` src/ustacks-mst.o: src/mst.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-mst.o -MD -MP -MF src/$(DEPDIR)/ustacks-mst.Tpo -c -o src/ustacks-mst.o `test -f 'src/mst.cc' || echo '$(srcdir)/'`src/mst.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-mst.Tpo src/$(DEPDIR)/ustacks-mst.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/mst.cc' object='src/ustacks-mst.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-mst.o `test -f 'src/mst.cc' || echo '$(srcdir)/'`src/mst.cc src/ustacks-mst.obj: src/mst.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-mst.obj -MD -MP -MF src/$(DEPDIR)/ustacks-mst.Tpo -c -o src/ustacks-mst.obj `if test -f 'src/mst.cc'; then $(CYGPATH_W) 'src/mst.cc'; else $(CYGPATH_W) '$(srcdir)/src/mst.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-mst.Tpo src/$(DEPDIR)/ustacks-mst.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/mst.cc' object='src/ustacks-mst.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-mst.obj `if test -f 'src/mst.cc'; then $(CYGPATH_W) 'src/mst.cc'; else $(CYGPATH_W) '$(srcdir)/src/mst.cc'; fi` src/ustacks-cmb.o: src/cmb.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-cmb.o -MD -MP -MF src/$(DEPDIR)/ustacks-cmb.Tpo -c -o src/ustacks-cmb.o `test -f 'src/cmb.cc' || echo '$(srcdir)/'`src/cmb.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-cmb.Tpo src/$(DEPDIR)/ustacks-cmb.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/cmb.cc' object='src/ustacks-cmb.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-cmb.o `test -f 'src/cmb.cc' || echo '$(srcdir)/'`src/cmb.cc src/ustacks-cmb.obj: src/cmb.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-cmb.obj -MD -MP -MF src/$(DEPDIR)/ustacks-cmb.Tpo -c -o src/ustacks-cmb.obj `if test -f 'src/cmb.cc'; then $(CYGPATH_W) 'src/cmb.cc'; else $(CYGPATH_W) '$(srcdir)/src/cmb.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-cmb.Tpo src/$(DEPDIR)/ustacks-cmb.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/cmb.cc' object='src/ustacks-cmb.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-cmb.obj `if test -f 'src/cmb.cc'; then $(CYGPATH_W) 'src/cmb.cc'; else $(CYGPATH_W) '$(srcdir)/src/cmb.cc'; fi` src/ustacks-DNASeq.o: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-DNASeq.o -MD -MP -MF src/$(DEPDIR)/ustacks-DNASeq.Tpo -c -o src/ustacks-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-DNASeq.Tpo src/$(DEPDIR)/ustacks-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/ustacks-DNASeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-DNASeq.o `test -f 'src/DNASeq.cc' || echo '$(srcdir)/'`src/DNASeq.cc src/ustacks-DNASeq.obj: src/DNASeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-DNASeq.obj -MD -MP -MF src/$(DEPDIR)/ustacks-DNASeq.Tpo -c -o src/ustacks-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-DNASeq.Tpo src/$(DEPDIR)/ustacks-DNASeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNASeq.cc' object='src/ustacks-DNASeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-DNASeq.obj `if test -f 'src/DNASeq.cc'; then $(CYGPATH_W) 'src/DNASeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNASeq.cc'; fi` src/ustacks-DNANSeq.o: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-DNANSeq.o -MD -MP -MF src/$(DEPDIR)/ustacks-DNANSeq.Tpo -c -o src/ustacks-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-DNANSeq.Tpo src/$(DEPDIR)/ustacks-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/ustacks-DNANSeq.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-DNANSeq.o `test -f 'src/DNANSeq.cc' || echo '$(srcdir)/'`src/DNANSeq.cc src/ustacks-DNANSeq.obj: src/DNANSeq.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-DNANSeq.obj -MD -MP -MF src/$(DEPDIR)/ustacks-DNANSeq.Tpo -c -o src/ustacks-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-DNANSeq.Tpo src/$(DEPDIR)/ustacks-DNANSeq.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/DNANSeq.cc' object='src/ustacks-DNANSeq.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-DNANSeq.obj `if test -f 'src/DNANSeq.cc'; then $(CYGPATH_W) 'src/DNANSeq.cc'; else $(CYGPATH_W) '$(srcdir)/src/DNANSeq.cc'; fi` src/ustacks-models.o: src/models.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-models.o -MD -MP -MF src/$(DEPDIR)/ustacks-models.Tpo -c -o src/ustacks-models.o `test -f 'src/models.cc' || echo '$(srcdir)/'`src/models.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-models.Tpo src/$(DEPDIR)/ustacks-models.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/models.cc' object='src/ustacks-models.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-models.o `test -f 'src/models.cc' || echo '$(srcdir)/'`src/models.cc src/ustacks-models.obj: src/models.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-models.obj -MD -MP -MF src/$(DEPDIR)/ustacks-models.Tpo -c -o src/ustacks-models.obj `if test -f 'src/models.cc'; then $(CYGPATH_W) 'src/models.cc'; else $(CYGPATH_W) '$(srcdir)/src/models.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-models.Tpo src/$(DEPDIR)/ustacks-models.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/models.cc' object='src/ustacks-models.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-models.obj `if test -f 'src/models.cc'; then $(CYGPATH_W) 'src/models.cc'; else $(CYGPATH_W) '$(srcdir)/src/models.cc'; fi` src/ustacks-utils.o: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-utils.o -MD -MP -MF src/$(DEPDIR)/ustacks-utils.Tpo -c -o src/ustacks-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-utils.Tpo src/$(DEPDIR)/ustacks-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/ustacks-utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-utils.o `test -f 'src/utils.cc' || echo '$(srcdir)/'`src/utils.cc src/ustacks-utils.obj: src/utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-utils.obj -MD -MP -MF src/$(DEPDIR)/ustacks-utils.Tpo -c -o src/ustacks-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-utils.Tpo src/$(DEPDIR)/ustacks-utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/utils.cc' object='src/ustacks-utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-utils.obj `if test -f 'src/utils.cc'; then $(CYGPATH_W) 'src/utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/utils.cc'; fi` src/ustacks-kmers.o: src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-kmers.o -MD -MP -MF src/$(DEPDIR)/ustacks-kmers.Tpo -c -o src/ustacks-kmers.o `test -f 'src/kmers.cc' || echo '$(srcdir)/'`src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-kmers.Tpo src/$(DEPDIR)/ustacks-kmers.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/kmers.cc' object='src/ustacks-kmers.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-kmers.o `test -f 'src/kmers.cc' || echo '$(srcdir)/'`src/kmers.cc src/ustacks-kmers.obj: src/kmers.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-kmers.obj -MD -MP -MF src/$(DEPDIR)/ustacks-kmers.Tpo -c -o src/ustacks-kmers.obj `if test -f 'src/kmers.cc'; then $(CYGPATH_W) 'src/kmers.cc'; else $(CYGPATH_W) '$(srcdir)/src/kmers.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-kmers.Tpo src/$(DEPDIR)/ustacks-kmers.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/kmers.cc' object='src/ustacks-kmers.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-kmers.obj `if test -f 'src/kmers.cc'; then $(CYGPATH_W) 'src/kmers.cc'; else $(CYGPATH_W) '$(srcdir)/src/kmers.cc'; fi` src/ustacks-aln_utils.o: src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-aln_utils.o -MD -MP -MF src/$(DEPDIR)/ustacks-aln_utils.Tpo -c -o src/ustacks-aln_utils.o `test -f 'src/aln_utils.cc' || echo '$(srcdir)/'`src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-aln_utils.Tpo src/$(DEPDIR)/ustacks-aln_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/aln_utils.cc' object='src/ustacks-aln_utils.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-aln_utils.o `test -f 'src/aln_utils.cc' || echo '$(srcdir)/'`src/aln_utils.cc src/ustacks-aln_utils.obj: src/aln_utils.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-aln_utils.obj -MD -MP -MF src/$(DEPDIR)/ustacks-aln_utils.Tpo -c -o src/ustacks-aln_utils.obj `if test -f 'src/aln_utils.cc'; then $(CYGPATH_W) 'src/aln_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/aln_utils.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-aln_utils.Tpo src/$(DEPDIR)/ustacks-aln_utils.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/aln_utils.cc' object='src/ustacks-aln_utils.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-aln_utils.obj `if test -f 'src/aln_utils.cc'; then $(CYGPATH_W) 'src/aln_utils.cc'; else $(CYGPATH_W) '$(srcdir)/src/aln_utils.cc'; fi` src/ustacks-input.o: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-input.o -MD -MP -MF src/$(DEPDIR)/ustacks-input.Tpo -c -o src/ustacks-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-input.Tpo src/$(DEPDIR)/ustacks-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/ustacks-input.o' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-input.o `test -f 'src/input.cc' || echo '$(srcdir)/'`src/input.cc src/ustacks-input.obj: src/input.cc @am__fastdepCXX_TRUE@ $(AM_V_CXX)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -MT src/ustacks-input.obj -MD -MP -MF src/$(DEPDIR)/ustacks-input.Tpo -c -o src/ustacks-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` @am__fastdepCXX_TRUE@ $(AM_V_at)$(am__mv) src/$(DEPDIR)/ustacks-input.Tpo src/$(DEPDIR)/ustacks-input.Po @AMDEP_TRUE@@am__fastdepCXX_FALSE@ $(AM_V_CXX)source='src/input.cc' object='src/ustacks-input.obj' libtool=no @AMDEPBACKSLASH@ @AMDEP_TRUE@@am__fastdepCXX_FALSE@ DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) @AMDEPBACKSLASH@ @am__fastdepCXX_FALSE@ $(AM_V_CXX@am__nodep@)$(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(ustacks_CXXFLAGS) $(CXXFLAGS) -c -o src/ustacks-input.obj `if test -f 'src/input.cc'; then $(CYGPATH_W) 'src/input.cc'; else $(CYGPATH_W) '$(srcdir)/src/input.cc'; fi` install-nobase_pkgdataDATA: $(nobase_pkgdata_DATA) @$(NORMAL_INSTALL) @list='$(nobase_pkgdata_DATA)'; test -n "$(pkgdatadir)" || list=; \ if test -n "$$list"; then \ echo " $(MKDIR_P) '$(DESTDIR)$(pkgdatadir)'"; \ $(MKDIR_P) "$(DESTDIR)$(pkgdatadir)" || exit 1; \ fi; \ $(am__nobase_list) | while read dir files; do \ xfiles=; for file in $$files; do \ if test -f "$$file"; then xfiles="$$xfiles $$file"; \ else xfiles="$$xfiles $(srcdir)/$$file"; fi; done; \ test -z "$$xfiles" || { \ test "x$$dir" = x. || { \ echo " $(MKDIR_P) '$(DESTDIR)$(pkgdatadir)/$$dir'"; \ $(MKDIR_P) "$(DESTDIR)$(pkgdatadir)/$$dir"; }; \ echo " $(INSTALL_DATA) $$xfiles '$(DESTDIR)$(pkgdatadir)/$$dir'"; \ $(INSTALL_DATA) $$xfiles "$(DESTDIR)$(pkgdatadir)/$$dir" || exit $$?; }; \ done uninstall-nobase_pkgdataDATA: @$(NORMAL_UNINSTALL) @list='$(nobase_pkgdata_DATA)'; test -n "$(pkgdatadir)" || list=; \ $(am__nobase_strip_setup); files=`$(am__nobase_strip)`; \ dir='$(DESTDIR)$(pkgdatadir)'; $(am__uninstall_files_from_dir) # This directory's subdirectories are mostly independent; you can cd # into them and run 'make' without going through this Makefile. # To change the values of 'make' variables: instead of editing Makefiles, # (1) if the variable is set in 'config.status', edit 'config.status' # (which will cause the Makefiles to be regenerated when you run 'make'); # (2) otherwise, pass the desired values on the 'make' command line. $(am__recursive_targets): @fail=; \ if $(am__make_keepgoing); then \ failcom='fail=yes'; \ else \ failcom='exit 1'; \ fi; \ dot_seen=no; \ target=`echo $@ | sed s/-recursive//`; \ case "$@" in \ distclean-* | maintainer-clean-*) list='$(DIST_SUBDIRS)' ;; \ *) list='$(SUBDIRS)' ;; \ esac; \ for subdir in $$list; do \ echo "Making $$target in $$subdir"; \ if test "$$subdir" = "."; then \ dot_seen=yes; \ local_target="$$target-am"; \ else \ local_target="$$target"; \ fi; \ ($(am__cd) $$subdir && $(MAKE) $(AM_MAKEFLAGS) $$local_target) \ || eval $$failcom; \ done; \ if test "$$dot_seen" = "no"; then \ $(MAKE) $(AM_MAKEFLAGS) "$$target-am" || exit 1; \ fi; test -z "$$fail" ID: $(am__tagged_files) $(am__define_uniq_tagged_files); mkid -fID $$unique tags: tags-recursive TAGS: tags tags-am: $(TAGS_DEPENDENCIES) $(am__tagged_files) set x; \ here=`pwd`; \ if ($(ETAGS) --etags-include --version) >/dev/null 2>&1; then \ include_option=--etags-include; \ empty_fix=.; \ else \ include_option=--include; \ empty_fix=; \ fi; \ list='$(SUBDIRS)'; for subdir in $$list; do \ if test "$$subdir" = .; then :; else \ test ! -f $$subdir/TAGS || \ set "$$@" "$$include_option=$$here/$$subdir/TAGS"; \ fi; \ done; \ $(am__define_uniq_tagged_files); \ shift; \ if test -z "$(ETAGS_ARGS)$$*$$unique"; then :; else \ test -n "$$unique" || unique=$$empty_fix; \ if test $$# -gt 0; then \ $(ETAGS) $(ETAGSFLAGS) $(AM_ETAGSFLAGS) $(ETAGS_ARGS) \ "$$@" $$unique; \ else \ $(ETAGS) $(ETAGSFLAGS) $(AM_ETAGSFLAGS) $(ETAGS_ARGS) \ $$unique; \ fi; \ fi ctags: ctags-recursive CTAGS: ctags ctags-am: $(TAGS_DEPENDENCIES) $(am__tagged_files) $(am__define_uniq_tagged_files); \ test -z "$(CTAGS_ARGS)$$unique" \ || $(CTAGS) $(CTAGSFLAGS) $(AM_CTAGSFLAGS) $(CTAGS_ARGS) \ $$unique GTAGS: here=`$(am__cd) $(top_builddir) && pwd` \ && $(am__cd) $(top_srcdir) \ && gtags -i $(GTAGS_ARGS) "$$here" cscope: cscope.files test ! -s cscope.files \ || $(CSCOPE) -b -q $(AM_CSCOPEFLAGS) $(CSCOPEFLAGS) -i cscope.files $(CSCOPE_ARGS) clean-cscope: -rm -f cscope.files cscope.files: clean-cscope cscopelist cscopelist: cscopelist-recursive cscopelist-am: $(am__tagged_files) list='$(am__tagged_files)'; \ case "$(srcdir)" in \ [\\/]* | ?:[\\/]*) sdir="$(srcdir)" ;; \ *) sdir=$(subdir)/$(srcdir) ;; \ esac; \ for i in $$list; do \ if test -f "$$i"; then \ echo "$(subdir)/$$i"; \ else \ echo "$$sdir/$$i"; \ fi; \ done >> $(top_builddir)/cscope.files distclean-tags: -rm -f TAGS ID GTAGS GRTAGS GSYMS GPATH tags -rm -f cscope.out cscope.in.out cscope.po.out cscope.files # Recover from deleted '.trs' file; this should ensure that # "rm -f foo.log; make foo.trs" re-run 'foo.test', and re-create # both 'foo.log' and 'foo.trs'. Break the recipe in two subshells # to avoid problems with "make -n". .log.trs: rm -f $< $@ $(MAKE) $(AM_MAKEFLAGS) $< # Leading 'am--fnord' is there to ensure the list of targets does not # expand to empty, as could happen e.g. with make check TESTS=''. am--fnord $(TEST_LOGS) $(TEST_LOGS:.log=.trs): $(am__force_recheck) am--force-recheck: @: $(TEST_SUITE_LOG): $(TEST_LOGS) @$(am__set_TESTS_bases); \ am__f_ok () { test -f "$$1" && test -r "$$1"; }; \ redo_bases=`for i in $$bases; do \ am__f_ok $$i.trs && am__f_ok $$i.log || echo $$i; \ done`; \ if test -n "$$redo_bases"; then \ redo_logs=`for i in $$redo_bases; do echo $$i.log; done`; \ redo_results=`for i in $$redo_bases; do echo $$i.trs; done`; \ if $(am__make_dryrun); then :; else \ rm -f $$redo_logs && rm -f $$redo_results || exit 1; \ fi; \ fi; \ if test -n "$$am__remaking_logs"; then \ echo "fatal: making $(TEST_SUITE_LOG): possible infinite" \ "recursion detected" >&2; \ elif test -n "$$redo_logs"; then \ am__remaking_logs=yes $(MAKE) $(AM_MAKEFLAGS) $$redo_logs; \ fi; \ if $(am__make_dryrun); then :; else \ st=0; \ errmsg="fatal: making $(TEST_SUITE_LOG): failed to create"; \ for i in $$redo_bases; do \ test -f $$i.trs && test -r $$i.trs \ || { echo "$$errmsg $$i.trs" >&2; st=1; }; \ test -f $$i.log && test -r $$i.log \ || { echo "$$errmsg $$i.log" >&2; st=1; }; \ done; \ test $$st -eq 0 || exit 1; \ fi @$(am__sh_e_setup); $(am__tty_colors); $(am__set_TESTS_bases); \ ws='[ ]'; \ results=`for b in $$bases; do echo $$b.trs; done`; \ test -n "$$results" || results=/dev/null; \ all=` grep "^$$ws*:test-result:" $$results | wc -l`; \ pass=` grep "^$$ws*:test-result:$$ws*PASS" $$results | wc -l`; \ fail=` grep "^$$ws*:test-result:$$ws*FAIL" $$results | wc -l`; \ skip=` grep "^$$ws*:test-result:$$ws*SKIP" $$results | wc -l`; \ xfail=`grep "^$$ws*:test-result:$$ws*XFAIL" $$results | wc -l`; \ xpass=`grep "^$$ws*:test-result:$$ws*XPASS" $$results | wc -l`; \ error=`grep "^$$ws*:test-result:$$ws*ERROR" $$results | wc -l`; \ if test `expr $$fail + $$xpass + $$error` -eq 0; then \ success=true; \ else \ success=false; \ fi; \ br='==================='; br=$$br$$br$$br$$br; \ result_count () \ { \ if test x"$$1" = x"--maybe-color"; then \ maybe_colorize=yes; \ elif test x"$$1" = x"--no-color"; then \ maybe_colorize=no; \ else \ echo "$@: invalid 'result_count' usage" >&2; exit 4; \ fi; \ shift; \ desc=$$1 count=$$2; \ if test $$maybe_colorize = yes && test $$count -gt 0; then \ color_start=$$3 color_end=$$std; \ else \ color_start= color_end=; \ fi; \ echo "$${color_start}# $$desc $$count$${color_end}"; \ }; \ create_testsuite_report () \ { \ result_count $$1 "TOTAL:" $$all "$$brg"; \ result_count $$1 "PASS: " $$pass "$$grn"; \ result_count $$1 "SKIP: " $$skip "$$blu"; \ result_count $$1 "XFAIL:" $$xfail "$$lgn"; \ result_count $$1 "FAIL: " $$fail "$$red"; \ result_count $$1 "XPASS:" $$xpass "$$red"; \ result_count $$1 "ERROR:" $$error "$$mgn"; \ }; \ { \ echo "$(PACKAGE_STRING): $(subdir)/$(TEST_SUITE_LOG)" | \ $(am__rst_title); \ create_testsuite_report --no-color; \ echo; \ echo ".. contents:: :depth: 2"; \ echo; \ for b in $$bases; do echo $$b; done \ | $(am__create_global_log); \ } >$(TEST_SUITE_LOG).tmp || exit 1; \ mv $(TEST_SUITE_LOG).tmp $(TEST_SUITE_LOG); \ if $$success; then \ col="$$grn"; \ else \ col="$$red"; \ test x"$$VERBOSE" = x || cat $(TEST_SUITE_LOG); \ fi; \ echo "$${col}$$br$${std}"; \ echo "$${col}Testsuite summary for $(PACKAGE_STRING)$${std}"; \ echo "$${col}$$br$${std}"; \ create_testsuite_report --maybe-color; \ echo "$$col$$br$$std"; \ if $$success; then :; else \ echo "$${col}See $(subdir)/$(TEST_SUITE_LOG)$${std}"; \ if test -n "$(PACKAGE_BUGREPORT)"; then \ echo "$${col}Please report to $(PACKAGE_BUGREPORT)$${std}"; \ fi; \ echo "$$col$$br$$std"; \ fi; \ $$success || exit 1 check-TESTS: @list='$(RECHECK_LOGS)'; test -z "$$list" || rm -f $$list @list='$(RECHECK_LOGS:.log=.trs)'; test -z "$$list" || rm -f $$list @test -z "$(TEST_SUITE_LOG)" || rm -f $(TEST_SUITE_LOG) @set +e; $(am__set_TESTS_bases); \ log_list=`for i in $$bases; do echo $$i.log; done`; \ trs_list=`for i in $$bases; do echo $$i.trs; done`; \ log_list=`echo $$log_list`; trs_list=`echo $$trs_list`; \ $(MAKE) $(AM_MAKEFLAGS) $(TEST_SUITE_LOG) TEST_LOGS="$$log_list"; \ exit $$?; recheck: all @test -z "$(TEST_SUITE_LOG)" || rm -f $(TEST_SUITE_LOG) @set +e; $(am__set_TESTS_bases); \ bases=`for i in $$bases; do echo $$i; done \ | $(am__list_recheck_tests)` || exit 1; \ log_list=`for i in $$bases; do echo $$i.log; done`; \ log_list=`echo $$log_list`; \ $(MAKE) $(AM_MAKEFLAGS) $(TEST_SUITE_LOG) \ am__force_recheck=am--force-recheck \ TEST_LOGS="$$log_list"; \ exit $$? tests/process_radtags.t.log: tests/process_radtags.t @p='tests/process_radtags.t'; \ b='tests/process_radtags.t'; \ $(am__check_pre) $(LOG_DRIVER) --test-name "$$f" \ --log-file $$b.log --trs-file $$b.trs \ $(am__common_driver_flags) $(AM_LOG_DRIVER_FLAGS) $(LOG_DRIVER_FLAGS) -- $(LOG_COMPILE) \ "$$tst" $(AM_TESTS_FD_REDIRECT) tests/kmer_filter.t.log: tests/kmer_filter.t @p='tests/kmer_filter.t'; \ b='tests/kmer_filter.t'; \ $(am__check_pre) $(LOG_DRIVER) --test-name "$$f" \ --log-file $$b.log --trs-file $$b.trs \ $(am__common_driver_flags) $(AM_LOG_DRIVER_FLAGS) $(LOG_DRIVER_FLAGS) -- $(LOG_COMPILE) \ "$$tst" $(AM_TESTS_FD_REDIRECT) tests/ustacks.t.log: tests/ustacks.t @p='tests/ustacks.t'; \ b='tests/ustacks.t'; \ $(am__check_pre) $(LOG_DRIVER) --test-name "$$f" \ --log-file $$b.log --trs-file $$b.trs \ $(am__common_driver_flags) $(AM_LOG_DRIVER_FLAGS) $(LOG_DRIVER_FLAGS) -- $(LOG_COMPILE) \ "$$tst" $(AM_TESTS_FD_REDIRECT) tests/pstacks.t.log: tests/pstacks.t @p='tests/pstacks.t'; \ b='tests/pstacks.t'; \ $(am__check_pre) $(LOG_DRIVER) --test-name "$$f" \ --log-file $$b.log --trs-file $$b.trs \ $(am__common_driver_flags) $(AM_LOG_DRIVER_FLAGS) $(LOG_DRIVER_FLAGS) -- $(LOG_COMPILE) \ "$$tst" $(AM_TESTS_FD_REDIRECT) .test.log: @p='$<'; \ $(am__set_b); \ $(am__check_pre) $(TEST_LOG_DRIVER) --test-name "$$f" \ --log-file $$b.log --trs-file $$b.trs \ $(am__common_driver_flags) $(AM_TEST_LOG_DRIVER_FLAGS) $(TEST_LOG_DRIVER_FLAGS) -- $(TEST_LOG_COMPILE) \ "$$tst" $(AM_TESTS_FD_REDIRECT) @am__EXEEXT_TRUE@.test$(EXEEXT).log: @am__EXEEXT_TRUE@ @p='$<'; \ @am__EXEEXT_TRUE@ $(am__set_b); \ @am__EXEEXT_TRUE@ $(am__check_pre) $(TEST_LOG_DRIVER) --test-name "$$f" \ @am__EXEEXT_TRUE@ --log-file $$b.log --trs-file $$b.trs \ @am__EXEEXT_TRUE@ $(am__common_driver_flags) $(AM_TEST_LOG_DRIVER_FLAGS) $(TEST_LOG_DRIVER_FLAGS) -- $(TEST_LOG_COMPILE) \ @am__EXEEXT_TRUE@ "$$tst" $(AM_TESTS_FD_REDIRECT) distdir: $(DISTFILES) $(am__remove_distdir) test -d "$(distdir)" || mkdir "$(distdir)" @srcdirstrip=`echo "$(srcdir)" | sed 's/[].[^$$\\*]/\\\\&/g'`; \ topsrcdirstrip=`echo "$(top_srcdir)" | sed 's/[].[^$$\\*]/\\\\&/g'`; \ list='$(DISTFILES)'; \ dist_files=`for file in $$list; do echo $$file; done | \ sed -e "s|^$$srcdirstrip/||;t" \ -e "s|^$$topsrcdirstrip/|$(top_builddir)/|;t"`; \ case $$dist_files in \ */*) $(MKDIR_P) `echo "$$dist_files" | \ sed '/\//!d;s|^|$(distdir)/|;s,/[^/]*$$,,' | \ sort -u` ;; \ esac; \ for file in $$dist_files; do \ if test -f $$file || test -d $$file; then d=.; else d=$(srcdir); fi; \ if test -d $$d/$$file; then \ dir=`echo "/$$file" | sed -e 's,/[^/]*$$,,'`; \ if test -d "$(distdir)/$$file"; then \ find "$(distdir)/$$file" -type d ! -perm -700 -exec chmod u+rwx {} \;; \ fi; \ if test -d $(srcdir)/$$file && test $$d != $(srcdir); then \ cp -fpR $(srcdir)/$$file "$(distdir)$$dir" || exit 1; \ find "$(distdir)/$$file" -type d ! -perm -700 -exec chmod u+rwx {} \;; \ fi; \ cp -fpR $$d/$$file "$(distdir)$$dir" || exit 1; \ else \ test -f "$(distdir)/$$file" \ || cp -p $$d/$$file "$(distdir)/$$file" \ || exit 1; \ fi; \ done @list='$(DIST_SUBDIRS)'; for subdir in $$list; do \ if test "$$subdir" = .; then :; else \ $(am__make_dryrun) \ || test -d "$(distdir)/$$subdir" \ || $(MKDIR_P) "$(distdir)/$$subdir" \ || exit 1; \ dir1=$$subdir; dir2="$(distdir)/$$subdir"; \ $(am__relativize); \ new_distdir=$$reldir; \ dir1=$$subdir; dir2="$(top_distdir)"; \ $(am__relativize); \ new_top_distdir=$$reldir; \ echo " (cd $$subdir && $(MAKE) $(AM_MAKEFLAGS) top_distdir="$$new_top_distdir" distdir="$$new_distdir" \\"; \ echo " am__remove_distdir=: am__skip_length_check=: am__skip_mode_fix=: distdir)"; \ ($(am__cd) $$subdir && \ $(MAKE) $(AM_MAKEFLAGS) \ top_distdir="$$new_top_distdir" \ distdir="$$new_distdir" \ am__remove_distdir=: \ am__skip_length_check=: \ am__skip_mode_fix=: \ distdir) \ || exit 1; \ fi; \ done -test -n "$(am__skip_mode_fix)" \ || find "$(distdir)" -type d ! -perm -755 \ -exec chmod u+rwx,go+rx {} \; -o \ ! -type d ! -perm -444 -links 1 -exec chmod a+r {} \; -o \ ! -type d ! -perm -400 -exec chmod a+r {} \; -o \ ! -type d ! -perm -444 -exec $(install_sh) -c -m a+r {} {} \; \ || chmod -R a+r "$(distdir)" dist-gzip: distdir tardir=$(distdir) && $(am__tar) | GZIP=$(GZIP_ENV) gzip -c >$(distdir).tar.gz $(am__post_remove_distdir) dist-bzip2: distdir tardir=$(distdir) && $(am__tar) | BZIP2=$${BZIP2--9} bzip2 -c >$(distdir).tar.bz2 $(am__post_remove_distdir) dist-lzip: distdir tardir=$(distdir) && $(am__tar) | lzip -c $${LZIP_OPT--9} >$(distdir).tar.lz $(am__post_remove_distdir) dist-xz: distdir tardir=$(distdir) && $(am__tar) | XZ_OPT=$${XZ_OPT--e} xz -c >$(distdir).tar.xz $(am__post_remove_distdir) dist-tarZ: distdir @echo WARNING: "Support for distribution archives compressed with" \ "legacy program 'compress' is deprecated." >&2 @echo WARNING: "It will be removed altogether in Automake 2.0" >&2 tardir=$(distdir) && $(am__tar) | compress -c >$(distdir).tar.Z $(am__post_remove_distdir) dist-shar: distdir @echo WARNING: "Support for shar distribution archives is" \ "deprecated." >&2 @echo WARNING: "It will be removed altogether in Automake 2.0" >&2 shar $(distdir) | GZIP=$(GZIP_ENV) gzip -c >$(distdir).shar.gz $(am__post_remove_distdir) dist-zip: distdir -rm -f $(distdir).zip zip -rq $(distdir).zip $(distdir) $(am__post_remove_distdir) dist dist-all: $(MAKE) $(AM_MAKEFLAGS) $(DIST_TARGETS) am__post_remove_distdir='@:' $(am__post_remove_distdir) # This target untars the dist file and tries a VPATH configuration. Then # it guarantees that the distribution is self-contained by making another # tarfile. distcheck: dist case '$(DIST_ARCHIVES)' in \ *.tar.gz*) \ GZIP=$(GZIP_ENV) gzip -dc $(distdir).tar.gz | $(am__untar) ;;\ *.tar.bz2*) \ bzip2 -dc $(distdir).tar.bz2 | $(am__untar) ;;\ *.tar.lz*) \ lzip -dc $(distdir).tar.lz | $(am__untar) ;;\ *.tar.xz*) \ xz -dc $(distdir).tar.xz | $(am__untar) ;;\ *.tar.Z*) \ uncompress -c $(distdir).tar.Z | $(am__untar) ;;\ *.shar.gz*) \ GZIP=$(GZIP_ENV) gzip -dc $(distdir).shar.gz | unshar ;;\ *.zip*) \ unzip $(distdir).zip ;;\ esac chmod -R a-w $(distdir) chmod u+w $(distdir) mkdir $(distdir)/_build $(distdir)/_build/sub $(distdir)/_inst chmod a-w $(distdir) test -d $(distdir)/_build || exit 0; \ dc_install_base=`$(am__cd) $(distdir)/_inst && pwd | sed -e 's,^[^:\\/]:[\\/],/,'` \ && dc_destdir="$${TMPDIR-/tmp}/am-dc-$$$$/" \ && am__cwd=`pwd` \ && $(am__cd) $(distdir)/_build/sub \ && ../../configure \ $(AM_DISTCHECK_CONFIGURE_FLAGS) \ $(DISTCHECK_CONFIGURE_FLAGS) \ --srcdir=../.. --prefix="$$dc_install_base" \ && $(MAKE) $(AM_MAKEFLAGS) \ && $(MAKE) $(AM_MAKEFLAGS) dvi \ && $(MAKE) $(AM_MAKEFLAGS) check \ && $(MAKE) $(AM_MAKEFLAGS) install \ && $(MAKE) $(AM_MAKEFLAGS) installcheck \ && $(MAKE) $(AM_MAKEFLAGS) uninstall \ && $(MAKE) $(AM_MAKEFLAGS) distuninstallcheck_dir="$$dc_install_base" \ distuninstallcheck \ && chmod -R a-w "$$dc_install_base" \ && ({ \ (cd ../.. && umask 077 && mkdir "$$dc_destdir") \ && $(MAKE) $(AM_MAKEFLAGS) DESTDIR="$$dc_destdir" install \ && $(MAKE) $(AM_MAKEFLAGS) DESTDIR="$$dc_destdir" uninstall \ && $(MAKE) $(AM_MAKEFLAGS) DESTDIR="$$dc_destdir" \ distuninstallcheck_dir="$$dc_destdir" distuninstallcheck; \ } || { rm -rf "$$dc_destdir"; exit 1; }) \ && rm -rf "$$dc_destdir" \ && $(MAKE) $(AM_MAKEFLAGS) dist \ && rm -rf $(DIST_ARCHIVES) \ && $(MAKE) $(AM_MAKEFLAGS) distcleancheck \ && cd "$$am__cwd" \ || exit 1 $(am__post_remove_distdir) @(echo "$(distdir) archives ready for distribution: "; \ list='$(DIST_ARCHIVES)'; for i in $$list; do echo $$i; done) | \ sed -e 1h -e 1s/./=/g -e 1p -e 1x -e '$$p' -e '$$x' distuninstallcheck: @test -n '$(distuninstallcheck_dir)' || { \ echo 'ERROR: trying to run $@ with an empty' \ '$$(distuninstallcheck_dir)' >&2; \ exit 1; \ }; \ $(am__cd) '$(distuninstallcheck_dir)' || { \ echo 'ERROR: cannot chdir into $(distuninstallcheck_dir)' >&2; \ exit 1; \ }; \ test `$(am__distuninstallcheck_listfiles) | wc -l` -eq 0 \ || { echo "ERROR: files left after uninstall:" ; \ if test -n "$(DESTDIR)"; then \ echo " (check DESTDIR support)"; \ fi ; \ $(distuninstallcheck_listfiles) ; \ exit 1; } >&2 distcleancheck: distclean @if test '$(srcdir)' = . ; then \ echo "ERROR: distcleancheck can only run from a VPATH build" ; \ exit 1 ; \ fi @test `$(distcleancheck_listfiles) | wc -l` -eq 0 \ || { echo "ERROR: files left in build directory after distclean:" ; \ $(distcleancheck_listfiles) ; \ exit 1; } >&2 check-am: all-am $(MAKE) $(AM_MAKEFLAGS) check-TESTS check: check-recursive all-am: Makefile $(PROGRAMS) $(SCRIPTS) $(DATA) config.h installdirs: installdirs-recursive installdirs-am: for dir in "$(DESTDIR)$(bindir)" "$(DESTDIR)$(bindir)" "$(DESTDIR)$(pkgdatadir)"; do \ test -z "$$dir" || $(MKDIR_P) "$$dir"; \ done install: install-recursive install-exec: install-exec-recursive install-data: install-data-recursive uninstall: uninstall-recursive install-am: all-am @$(MAKE) $(AM_MAKEFLAGS) install-exec-am install-data-am installcheck: installcheck-recursive install-strip: if test -z '$(STRIP)'; then \ $(MAKE) $(AM_MAKEFLAGS) INSTALL_PROGRAM="$(INSTALL_STRIP_PROGRAM)" \ install_sh_PROGRAM="$(INSTALL_STRIP_PROGRAM)" INSTALL_STRIP_FLAG=-s \ install; \ else \ $(MAKE) $(AM_MAKEFLAGS) INSTALL_PROGRAM="$(INSTALL_STRIP_PROGRAM)" \ install_sh_PROGRAM="$(INSTALL_STRIP_PROGRAM)" INSTALL_STRIP_FLAG=-s \ "INSTALL_PROGRAM_ENV=STRIPPROG='$(STRIP)'" install; \ fi mostlyclean-generic: -test -z "$(TEST_LOGS)" || rm -f $(TEST_LOGS) -test -z "$(TEST_LOGS:.log=.trs)" || rm -f $(TEST_LOGS:.log=.trs) -test -z "$(TEST_SUITE_LOG)" || rm -f $(TEST_SUITE_LOG) clean-generic: distclean-generic: -test -z "$(CONFIG_CLEAN_FILES)" || rm -f $(CONFIG_CLEAN_FILES) -test . = "$(srcdir)" || test -z "$(CONFIG_CLEAN_VPATH_FILES)" || rm -f $(CONFIG_CLEAN_VPATH_FILES) -rm -f src/$(DEPDIR)/$(am__dirstamp) -rm -f src/$(am__dirstamp) maintainer-clean-generic: @echo "This command is intended for maintainers to use" @echo "it deletes files that may require special tools to rebuild." clean: clean-recursive clean-am: clean-binPROGRAMS clean-generic mostlyclean-am distclean: distclean-recursive -rm -f $(am__CONFIG_DISTCLEAN_FILES) -rm -rf src/$(DEPDIR) -rm -f Makefile distclean-am: clean-am distclean-compile distclean-generic \ distclean-hdr distclean-tags dvi: dvi-recursive dvi-am: html: html-recursive html-am: info: info-recursive info-am: install-data-am: install-nobase_pkgdataDATA @$(NORMAL_INSTALL) $(MAKE) $(AM_MAKEFLAGS) install-data-hook install-dvi: install-dvi-recursive install-dvi-am: install-exec-am: install-binPROGRAMS install-dist_binSCRIPTS install-html: install-html-recursive install-html-am: install-info: install-info-recursive install-info-am: install-man: install-pdf: install-pdf-recursive install-pdf-am: install-ps: install-ps-recursive install-ps-am: installcheck-am: maintainer-clean: maintainer-clean-recursive -rm -f $(am__CONFIG_DISTCLEAN_FILES) -rm -rf $(top_srcdir)/autom4te.cache -rm -rf src/$(DEPDIR) -rm -f Makefile maintainer-clean-am: distclean-am maintainer-clean-generic mostlyclean: mostlyclean-recursive mostlyclean-am: mostlyclean-compile mostlyclean-generic pdf: pdf-recursive pdf-am: ps: ps-recursive ps-am: uninstall-am: uninstall-binPROGRAMS uninstall-dist_binSCRIPTS \ uninstall-nobase_pkgdataDATA @$(NORMAL_INSTALL) $(MAKE) $(AM_MAKEFLAGS) uninstall-hook .MAKE: $(am__recursive_targets) all check-am install-am \ install-data-am install-strip uninstall-am .PHONY: $(am__recursive_targets) CTAGS GTAGS TAGS all all-am \ am--refresh check check-TESTS check-am clean clean-binPROGRAMS \ clean-cscope clean-generic cscope cscopelist-am ctags ctags-am \ dist dist-all dist-bzip2 dist-gzip dist-lzip dist-shar \ dist-tarZ dist-xz dist-zip distcheck distclean \ distclean-compile distclean-generic distclean-hdr \ distclean-tags distcleancheck distdir distuninstallcheck dvi \ dvi-am html html-am info info-am install install-am \ install-binPROGRAMS install-data install-data-am \ install-data-hook install-dist_binSCRIPTS install-dvi \ install-dvi-am install-exec install-exec-am install-html \ install-html-am install-info install-info-am install-man \ install-nobase_pkgdataDATA install-pdf install-pdf-am \ install-ps install-ps-am install-strip installcheck \ installcheck-am installdirs installdirs-am maintainer-clean \ maintainer-clean-generic mostlyclean mostlyclean-compile \ mostlyclean-generic pdf pdf-am ps ps-am recheck tags tags-am \ uninstall uninstall-am uninstall-binPROGRAMS \ uninstall-dist_binSCRIPTS uninstall-hook \ uninstall-nobase_pkgdataDATA .PRECIOUS: Makefile debug: $(MAKE) all "CXXFLAGS=-g -Wall -DDEBUG -std=gnu++0x" install-data-hook: sed -e 's,_VERSION_,$(VERSION),' -e 's,_BINDIR_,$(bindir)/,g' -e 's,_PKGDATADIR_,$(pkgdatadir)/,g' $(DESTDIR)$(bindir)/denovo_map.pl > $(DESTDIR)$(bindir)/denovo_map.pl.subst mv $(DESTDIR)$(bindir)/denovo_map.pl.subst $(DESTDIR)$(bindir)/denovo_map.pl chmod +x $(DESTDIR)$(bindir)/denovo_map.pl sed -e 's,_VERSION_,$(VERSION),' -e 's,_BINDIR_,$(bindir)/,g' -e 's,_PKGDATADIR_,$(pkgdatadir)/,g' $(DESTDIR)$(bindir)/ref_map.pl > $(DESTDIR)$(bindir)/ref_map.pl.subst mv $(DESTDIR)$(bindir)/ref_map.pl.subst $(DESTDIR)$(bindir)/ref_map.pl chmod +x $(DESTDIR)$(bindir)/ref_map.pl sed -e 's,_VERSION_,$(VERSION),' -e 's,_PKGDATADIR_,$(pkgdatadir)/,g' $(DESTDIR)$(bindir)/export_sql.pl > $(DESTDIR)$(bindir)/export_sql.pl.subst mv $(DESTDIR)$(bindir)/export_sql.pl.subst $(DESTDIR)$(bindir)/export_sql.pl chmod +x $(DESTDIR)$(bindir)/export_sql.pl sed -e 's,_VERSION_,$(VERSION),' -e 's,_PKGDATADIR_,$(pkgdatadir)/,g' $(DESTDIR)$(bindir)/index_radtags.pl > $(DESTDIR)$(bindir)/index_radtags.pl.subst mv $(DESTDIR)$(bindir)/index_radtags.pl.subst $(DESTDIR)$(bindir)/index_radtags.pl chmod +x $(DESTDIR)$(bindir)/index_radtags.pl sed -e 's,_VERSION_,$(VERSION),' -e 's,_PKGDATADIR_,$(pkgdatadir)/,g' $(DESTDIR)$(bindir)/load_radtags.pl > $(DESTDIR)$(bindir)/load_radtags.pl.subst mv $(DESTDIR)$(bindir)/load_radtags.pl.subst $(DESTDIR)$(bindir)/load_radtags.pl chmod +x $(DESTDIR)$(bindir)/load_radtags.pl sed -e 's,_VERSION_,$(VERSION),' $(DESTDIR)$(bindir)/sort_read_pairs.pl > $(DESTDIR)$(bindir)/sort_read_pairs.pl.subst mv $(DESTDIR)$(bindir)/sort_read_pairs.pl.subst $(DESTDIR)$(bindir)/sort_read_pairs.pl chmod +x $(DESTDIR)$(bindir)/sort_read_pairs.pl sed -e 's,_VERSION_,$(VERSION),' $(DESTDIR)$(bindir)/exec_velvet.pl > $(DESTDIR)$(bindir)/exec_velvet.pl.subst mv $(DESTDIR)$(bindir)/exec_velvet.pl.subst $(DESTDIR)$(bindir)/exec_velvet.pl chmod +x $(DESTDIR)$(bindir)/exec_velvet.pl sed -e 's,_VERSION_,$(VERSION),' -e 's,_PKGDATADIR_,$(pkgdatadir)/,g' $(DESTDIR)$(bindir)/load_sequences.pl > $(DESTDIR)$(bindir)/load_sequences.pl.subst mv $(DESTDIR)$(bindir)/load_sequences.pl.subst $(DESTDIR)$(bindir)/load_sequences.pl chmod +x $(DESTDIR)$(bindir)/load_sequences.pl sed -e 's,_VERSION_,$(VERSION),' -e 's,_PKGDATADIR_,$(pkgdatadir)/,g' -e 's,_BINDIR_,$(bindir)/,g' \ $(DESTDIR)$(bindir)/stacks_export_notify.pl > $(DESTDIR)$(bindir)/stacks_export_notify.pl.subst mv $(DESTDIR)$(bindir)/stacks_export_notify.pl.subst $(DESTDIR)$(bindir)/stacks_export_notify.pl chmod +x $(DESTDIR)$(bindir)/stacks_export_notify.pl sed -e 's,_PKGDATADIR_,$(pkgdatadir)/,g' -e 's,_BINDIR_,$(bindir)/,g' \ $(DESTDIR)$(pkgdatadir)/php/constants.php.dist > $(DESTDIR)$(pkgdatadir)/php/constants.php.dist.subst mv $(DESTDIR)$(pkgdatadir)/php/constants.php.dist.subst $(DESTDIR)$(pkgdatadir)/php/constants.php.dist echo $(VERSION) > $(DESTDIR)$(pkgdatadir)/php/version.php $(install_sh) -d -m 755 $(DESTDIR)$(pkgdatadir)/php/export uninstall-hook: rm -rf $(DESTDIR)$(pkglocalstatedir) # Tell versions [3.59,3.63) of GNU make to not export all variables. # Otherwise a system limit (for SysV at least) may be exceeded. .NOEXPORT: stacks-1.44/php/000755 000765 000024 00000000000 13001012452 014255 5ustar00catchenstaff000000 000000 stacks-1.44/README000644 000765 000024 00000016511 12750771761 014403 0ustar00catchenstaff000000 000000 The Stacks Pipeline ------------------- 0. Prerequisites Stacks should build on any standard UNIX-like environment (Apple OS X, Linux, etc.) Stacks is an independent pipeline and can be run without any additional external software. Note: Apple OS X does not use the GNU Compiler Collection, which is standard on Linux-based systems. Instead, Apple distributes CLANG, which is a nice compiler but does not yet support the OpenMP library which Stacks relies on for parallel processing. Stacks can still be built and run on an Apple system, however, you will have to disable building with OpenMP (supply the --disable-openmp flag to configure) and use non-parallelized code. If you want to install a parallelized version of Stacks, you can install GCC by hand, or using a package system such as Homebrew (http://brew.sh/) or MacPorts (http://www.macports.org/). Several Perl scripts are distributed with Stacks to run the pipeline components and upload pipeline output to the MySQL database serer. For these to work, you must have the Perl DBI module installed with the MySQL driver. Most Linux distributions will include the Perl DBI module, but if not Perl modules are easily installed with the cpan tool. 1. Install optional components: If you would like to lower memory usage install Google's SparseHash class. http://code.google.com/p/sparsehash/ If you are running a version of Linux, the above software can be installed via the package manager. If you are using Ubuntu, you can install the following packages: % sudo apt-get install libdbd-mysql-perl % sudo apt-get install libsparsehash-dev A similar set of commands can be executed on Debian using apt-get, or on a RedHat derived Linux system using yum, or another package manager on other Linux distributions. 2. Build the software. Stacks uses the standard autotools install: % tar xfvz stacks-x.xx.tar.gz % cd stacks-x.xx % ./configure You can change the root of the install location (/usr/local/ on most operating systems) by specifying the --prefix command line option to the configure script. % ./configure --prefix=/home/smith/local You can enable Sparsehash by adding the following options: % ./configure --enable-sparsehash % make You can speed up the build if you have more than one processor: % make -j 8 3. Install the software. % sudo make install A default Stacks install will install files in the following way: /usr/local/bin - stacks executables and perl scripts /usr/local/share/stacks - PHP files for the web interface and SQL files for creating the MySQL database The pipeline is now ready to run. The remaining install instructions are to get the web interface up and running. The web interface is very useful for visualization and more or less required for building genetic maps. However, Stacks does not depend on the web interface to run. The Stacks Web Interface ------------------------ To visualize data, Stacks uses a web-based interface (written in PHP) that interacts with a MySQL database server. MySQL provides various functions to store, sort, and export data from a database. 0. Prerequisites Most server installations will provide Apache, MySQL, Perl, and PHP by default. If you want to export data in Microsoft Excel Spreadsheets, you will need the Spreadsheet::WriteExcel Perl module. While installing these components is beyond these instructions, here are some links that might be useful: 1. MySQL Database: http://dev.mysql.com/downloads/mysql/ 2. Spreadsheet Perl Module: http://search.cpan.org/~jmcnamara/Spreadsheet-WriteExcel-2.40/ If you are running a version of Linux, the above software can be installed via the package manager. If you are using Ubuntu, you can install the following packages: % sudo apt-get install mysql-server mysql-client % sudo apt-get install php5 php5-mysqlnd % sudo apt-get install libspreadsheet-writeexcel-perl A similar set of commands can be executed on Debian using apt-get, or on a RedHat derived Linux system using yum, or another package manager on other Linux distributions. 1. Edit the MySQL configuration file, installed in /usr/local/share/stacks/sql/mysql.cnf.dist, to enable access to the database from the Stacks scripts. % cd /usr/local/share/stacks/sql/ % cp mysql.cnf.dist mysql.cnf Edit the file to reflect the proper username, password, and host to use to access MySQL. The various scripts that access the database will search for a MySQL configuration file in your home directory before using the Stacks-distributed copy. If you already have a personal account set up and configured (in ~/.my.cnf) you can continue to use these credentials instead of setting up new, common ones. If you just installed MySQL and have not added any users, you can do so with these commands: % mysql mysql> GRANT ALL ON *.* TO 'stacks_user'@'localhost' IDENTIFIED BY 'stackspassword'; Edit /usr/local/share/stacks/sql/mysql.cnf to contain the username and password you specified to MySQL. (This information was taken from: http://dev.mysql.com/doc/refman/5.1/en/grant.html) 2. Enable the Stacks web interface in the Apache webserver. Add the following lines to your Apache configuration to make the Stacks PHP files visible to the web server and to provide a easily readable URL to access them: Order deny,allow Deny from all Allow from all Require all granted Alias /stacks "/usr/local/share/stacks/php" A sensible way to do this is to create the file stacks.conf with the above lines. If you are using Apache 2.3 or earlier: --------------------------------------- Place the stacks.conf file in either /etc/apache2/conf.d/ or /etc/httpd/conf.d/ directory (depending on your Linux distro) and restart the apache server: # vi /etc/apache2/conf.d/stacks.conf # apachectl restart (See the Apache configuration for more information on what these do: http://httpd.apache.org/docs/2.0/mod/core.html#directory) If you are using Apache 2.4 or later: --------------------------------------- Place the stacks.conf file in /etc/apache2/conf-available directory and then create a symlink to it in the /etc/apache2/conf-enabled directory. Then restart Apache. Like so: # vi /etc/apache2/conf-available/stacks.conf # ln -s /etc/apache2/conf-available/stacks.conf /etc/apache2/conf-enabled/stacks.conf # apachectl restart 3. Provide access to the MySQL database from the web interface Edit the PHP configuration file (constants.php.dist) to allow it access to the MySQL database. Change the file to include the proper database username ($db_user), password ($db_pass), and hostname ($db_host). Rename the distribution file so it is active. % cp /usr/local/share/stacks/php/constants.php.dist /usr/local/share/stacks/php/constants.php % vi /usr/local/share/stacks/php/constants.php You may find it advantageous to create a specific MySQL user with limited permissions - SELECT, UPDATE, and DELETE to allow users to interact with the database through the web interface. 4. Enable web-based exporting from the MySQL database. Edit the stacks_export_notify.pl script to specify the email and SMTP server to use in notification messages. Ensure that the permissions of the php/export directory allow the webserver to write to it. Assuming your web server user is 'www': % chown www /usr/local/share/stacks/php/export stacks-1.44/scripts/000755 000765 000024 00000000000 13001012452 015155 5ustar00catchenstaff000000 000000 stacks-1.44/sql/000755 000765 000024 00000000000 13001012452 014265 5ustar00catchenstaff000000 000000 stacks-1.44/src/000755 000765 000024 00000000000 13001012452 014255 5ustar00catchenstaff000000 000000 stacks-1.44/tests/000755 000765 000024 00000000000 13001012452 014630 5ustar00catchenstaff000000 000000 stacks-1.44/tests/kmer_filter.t000644 000765 000024 00000001363 12335173442 017343 0ustar00catchenstaff000000 000000 #!/usr/bin/env bash # Preamble test_path=$(cd `dirname "${BASH_SOURCE[0]}"` && pwd) test_data_path="$test_path/"$(basename "${BASH_SOURCE[0]}" | sed -e 's@\.t$@@') source $test_path/setup.sh plan 5 #kmer_filter tests ok_ "filter out rare kmers" \ 000_rare \ "kmer_filter -f %in/in.fastq -o %out --rare" ok_ "filter out overly abundant kmers" \ 001_abundant \ "kmer_filter -f %in/in.fastq -o %out --abundant" ok_ "set max kmer frequency for abundance filtering" \ 002_mkf \ "kmer_filter -f %in/in.fastq -o %out --abundant --max_k_freq 10" skip_ "input gzfastq" \ 003_ingzfastq \ "kmer_filter -i gzfastq -f %in/in.fastq.gz -o out" ok_ "output fastq" \ 004_outfasta \ "kmer_filter -y fasta -i $freq_in -o %out"stacks-1.44/tests/process_radtags.t000644 000765 000024 00000005737 12441417455 020237 0ustar00catchenstaff000000 000000 #!/usr/bin/env bash # Preamble test_path=$(cd `dirname "${BASH_SOURCE[0]}"` && pwd) test_data_path="$test_path/"$(basename "${BASH_SOURCE[0]}" | sed -e 's@\.t$@@') source $test_path/setup.sh # Setup barcodes=$test_data_path/frequent_data/Barcodes.txt freq_in=$test_data_path/frequent_data/in.fastq.gz freq_in2=$test_data_path/frequent_data/in.fastq plan 15 # # Example libtap tests. Uncomment to run. # ok "This test will pass" true # ok "This test will fail" ls -al /this/file/does/not/exist # diag 'I just love word plays ...' # ok "This test is expected to fail# TODO fix this" ls -al /neither/does/this/file # skip "This command doesn't make sense:" more /dev/null # process_radtags tests ok_ 'input gzfastq' \ 000_input_gzfastq \ "process_radtags -i gzfastq -f $freq_in -o %out -b $barcodes -E phred33 -e sbfI" ok_ 'input fastq' \ 001_input_fastq \ "process_radtags -i fastq -f $freq_in2 -o %out -b $barcodes -E phred33 -e sbfI" diag 'FIXME: Input files for this test are NOT actaully phred64 encoded! This is just an example test...' ok_ 'input phred64' \ 002_input_phred64 \ "process_radtags -i gzfastq -f $freq_in -o %out -b $barcodes -E phred64 -e sbfI" ok_ 'clean' \ 003_clean_data \ "process_radtags -i gzfastq -f $freq_in -o %out -b $barcodes -E phred33 -e sbfI -c" ok_ 'discarded reads' \ 004_discarded_reads \ "process_radtags -i gzfastq -f $freq_in -o %out -b $barcodes -E phred33 -e sbfI -D" ok_ 'fasta output' \ 005_output_fasta \ "process_radtags -i gzfastq -f $freq_in -o %out -b $barcodes -E phred33 -e sbfI -y fasta" ok_ 'discard low quality reads' \ 006_discard_lq \ "process_radtags -i gzfastq -f $freq_in -o %out -b $barcodes -E phred33 -e sbfI -q" ok_ 'rescue barcodes and radtags' \ 007_rescue_bcrt \ "process_radtags -i gzfastq -p %in -o %out -b $barcodes -E phred33 -e sbfI -r" ok_ 'truncate final read length' \ 008_truncate \ "process_radtags -i gzfastq -f $freq_in -o %out -b $barcodes -E phred33 -e sbfI -t 50" ok_ 'set window length' \ 009_winlen \ "process_radtags -i gzfastq -p %in -o %out -b $barcodes -E phred33 -e sbfI -q -w .12" ok_ 'minimum window score' \ 010_minscore \ "process_radtags -i gzfastq -f $freq_in -o %out -b $barcodes -E phred33 -e sbfI -q -s 15" ok_ 'merge output' \ 011_merge \ "process_radtags -i gzfastq -f %in/in.fastq.gz -o %out -E phred33 -e sbfI --merge" ok_ 'Remove sequences marked by Illumina as failing chastity/purity filter' \ 012_filt_ill \ "process_radtags -i gzfastq -f %in/in.fastq.gz -o %out -E phred33 -e sbfI -b $barcodes --filter_illumina" ok_ 'Disable checking for RAD site' \ 013_disrc \ "process_radtags -i gzfastq -f %in/in.fastq.gz -o %out -E phred33 -e sbfI -b $barcodes --disable_rad_check" ok_ 'Provide distance between barcodes for rescue' \ 014_bcdist \ "process_radtags -i gzfastq -f %in/in.fastq.gz -o %out -E phred33 -e sbfI -b $barcodes --barcode_dist 1 -r" # I'm not sure yet what finish() does. finish stacks-1.44/tests/pstacks.t000644 000765 000024 00000003656 12441417455 016522 0ustar00catchenstaff000000 000000 #!/usr/bin/env bash # Preamble test_path=$(cd `dirname "${BASH_SOURCE[0]}"` && pwd) test_data_path="$test_path/"$(basename "${BASH_SOURCE[0]}" | sed -e 's@\.t$@@') source $test_path/setup.sh freq_in=$test_data_path/frequent_input/in.bam freq_in2=$test_data_path/frequent_input/in2.bam freq_in3=$test_data_path/frequent_input/in.sam plan 12 ok_ 'input bam' \ 000_inbam \ "pstacks -t bam -f $freq_in -o %out" ok_ 'input sam' \ 001_insam \ "pstacks -t sam -f $freq_in3 -o %out" skip_ 'input bowtie - need data input file type' \ 002_inbowtie \ "pstacks -t bowtie -f %in/??? -o %out" ok_ 'minimum coverage depth for contig report' \ 003_mincovdepth \ "pstacks -t bam -f $freq_in -o %out -m 2" ok_ 'R^2 significance level of 0.1 for calling homozygote/heterozygote' \ 004_alpha0.1 \ "pstacks -t bam -f %in/in.bam -o %out --alpha 0.1" ok_ 'R^2 significance level of 0.05 for calling homozygote/heterozygote' \ 005_alpha0.05 \ "pstacks -t bam -f %in/in.bam -o %out --alpha 0.05" ok_ 'R^2 significance level of 0.01 for calling homozygote/heterozygote' \ 006_alpha0.01 \ "pstacks -t bam -f %in/in.bam -o %out --alpha 0.01" ok_ 'R^2 significance level of 0.001 for calling homozygote/heterozygote' \ 007_alpha0.001 \ "pstacks -t bam -f %in/in.bam -o %out --alpha 0.001" ok_ 'For bounded model, specify upper bound' \ 008_bound_high \ "pstacks -t bam -f $freq_in2 -o %out --model_type bounded --bound_high 0.5" ok_ 'For bounded model, specify lower bound' \ 009_bound_low \ "pstacks -t bam -f $freq_in2 -o %out --model_type bounded --bound_low 0.2" ok_ 'For bounded model, specify upper and lower bounds' \ 010_bounded \ "pstacks -t bam -f $freq_in2 -o %out --model_type bounded --bound_high 0.5 --bound_low 0.2" ok_ 'For fixed model, specify barcode error frequency' \ 011_fixed \ "pstacks -t bam -f %in/in.bam -o %out --model_type fixed --bc_err_freq 0.39" finishstacks-1.44/tests/ustacks.t000644 000765 000024 00000006321 12441417455 016517 0ustar00catchenstaff000000 000000 #!/usr/bin/env bash # Preamble test_path=$(cd `dirname "${BASH_SOURCE[0]}"` && pwd) test_data_path="$test_path/"$(basename "${BASH_SOURCE[0]}" | sed -e 's@\.t$@@') source $test_path/setup.sh freq_in=$test_data_path/frequent_inputs/in.fastq.gz freq_in2=$test_data_path/frequent_inputs/in2.fastq.gz freq_in3=$test_data_path/frequent_inputs/in.fasta freq_in4=$test_data_path/frequent_inputs/in.fasta.gz freq_in5=$test_data_path/frequent_inputs/in.fastq plan 21 ref_map ok_ 'input gzipped fastq' \ 000_ingzfastq \ "ustacks -t gzfastq -f $freq_in -o %out" \ ok_ 'input fastq' \ 001_infastq \ "ustacks -t fastq -f $freq_in5 -o %out" ok_ 'input fasta' \ 002_infasta \ "ustacks -t fasta -f $freq_in3 -o %out" ok_ 'inpus gzipped fasta' \ 003_ingzfasta \ "ustacks -t gzfasta -f $freq_in4 -o %out" ok_ 'set sample ID=1 (AKA MySQL column 2)' \ 004_sqlid \ "ustacks -t gzfastq -f $freq_in -o %out -i 1" ok_ 'specify minimum depth of coverage required to call a stack' \ 005_mindepcov \ "ustacks -t gzfastq -f $freq_in -o %out -m 3" ok_ 'specify maximum distance between stacks' \ 006_maxdistbtw \ "ustacks -t gzfastq -f $freq_in -o %out -M 3" ok_ 'specify max distance to align secondary reads to primary stacks' \ 007_maxdist_srps \ "ustacks -t gzfastq -f $freq_in -o %out -N 5" skip_ 'retain unused reads' \ 008_retain \ "ustacks -t gzfastq -f %in -o %out -R" ok_ 'disable haplotype calling from secondary reads' \ 009_hapcall \ "ustacks -t gzfastq -f $freq_in -o %out -H" ok_ 'remove highly repetative (likely error) reads' \ 010_remrep \ "ustacks -t gzfastq -f %in/in.fastq.gz -o %out -r" ok_ 'enable deleveraging algorithm' \ 011_deleverage \ "ustacks -t gzfastq -f %in/in.fastq.gz -o %out -d" \ zip_test ok_ 'specify max number of stacks at a de novo locus' \ 012_maxlocus \ "ustacks -t gzfastq -f %in/in.fastq.gz -o %out --max_locus_stacks 4" ok_ 'specify chi square significance level for calling heteroz/homozygote' \ 013_alpha0.1 \ "ustacks -t gzfastq -f %in/in.fastq.gz -o %out --alpha 0.1" ok_ 'specify chi square significance level for calling heteroz/homozygote' \ 014_alpha0.05 \ "ustacks -t gzfastq -f %in/in.fastq.gz -o %out --alpha 0.05" ok_ 'specify chi square significance level for calling heteroz/homozygote' \ 015_alpha0.01 \ "ustacks -t gzfastq -f %in/in.fastq.gz -o %out --alpha 0.01" ok_ 'specify chi square significance level for calling heteroz/homozygote' \ 016_alpha0.001 \ "ustacks -t gzfastq -f %in/in.fastq.gz -o %out --alpha 0.001" ok_ 'For bounded model, specify upper bound' \ 017_bound_high \ "ustacks -t gzfastq -f $freq_in2 -o %out --model_type bounded --bound_high 0.01" ok_ 'For bounded model, specify lower bound' \ 018_bound_low \ "ustacks -t gzfastq -f $freq_in2 -o %out --model_type bounded --bound_low 0.001" ok_ 'For bounded model, specify upper and lower bounds' \ 019_bounded \ "ustacks -t gzfastq -f $freq_in2 -o %out --model_type bounded --bound_low 0.001 --bound_high 0.01" ok_ 'For fixed model, specify barcode error frequency rate' \ 020_fixed \ "ustacks -t gzfastq -f %in/in.fastq.gz -o %out --model_type fixed --bc_err_freq 0.98" finishstacks-1.44/src/aln_utils.cc000644 000765 000024 00000024442 12764047443 016613 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // aln_utils.cc -- common routines needed for dealing with gapped alignments. // #include "aln_utils.h" string invert_cigar(string cigar) { for (uint i = 0; i < cigar.length(); i++) { if (cigar[i] == 'I') cigar[i] = 'D'; else if (cigar[i] == 'D') cigar[i] = 'I'; } return cigar; } int parse_cigar(const char *cigar_str, vector > &cigar) { char buf[id_len]; int dist; const char *p, *q; uint seqlen = 0; cigar.clear(); p = cigar_str; while (*p != '\0') { q = p + 1; while (*q != '\0' && isdigit(*q)) q++; strncpy(buf, p, q - p); buf[q-p] = '\0'; dist = is_integer(buf); cigar.push_back(make_pair(*q, dist)); seqlen += dist; p = q + 1; } return seqlen; } string apply_cigar_to_seq(const char *seq, vector > &cigar) { uint size = cigar.size(); char op; uint dist, bp, edited_bp, stop; string edited_seq; // // Calculate the overall sequence length. // uint seqlen = 0; for (uint i = 0; i < size; i++) seqlen += cigar[i].second; bp = 0; edited_seq.reserve(seqlen); for (uint i = 0; i < size; i++) { op = cigar[i].first; dist = cigar[i].second; switch(op) { case 'S': stop = bp + dist; while (bp < stop) { edited_seq.push_back('N'); bp++; } break; case 'D': edited_bp = 0; while (edited_bp < dist) { edited_seq.push_back('N'); edited_bp++; } break; case 'I': case 'M': stop = bp + dist; while (bp < stop) { edited_seq.push_back(seq[bp]); bp++; } break; default: break; } } return edited_seq; } string apply_cigar_to_model_seq(const char *seq, vector > &cigar) { uint size = cigar.size(); char op; uint dist, bp, edited_bp, stop; string edited_seq; // // Calculate the overall sequence length. // uint seqlen = 0; for (uint i = 0; i < size; i++) seqlen += cigar[i].second; bp = 0; edited_seq.reserve(seqlen); for (uint i = 0; i < size; i++) { op = cigar[i].first; dist = cigar[i].second; switch(op) { case 'S': stop = bp + dist; while (bp < stop) { edited_seq.push_back('U'); bp++; } break; case 'D': edited_bp = 0; while (edited_bp < dist) { edited_seq.push_back('U'); edited_bp++; } break; case 'I': case 'M': stop = bp + dist; while (bp < stop) { edited_seq.push_back(seq[bp]); bp++; } break; default: break; } } return edited_seq; } int apply_cigar_to_seq(char *seq, uint seq_len, const char *old_seq, vector > &cigar) { uint size = cigar.size(); char op; uint dist, bp, seq_bp, stop; bp = 0; seq_bp = 0; for (uint i = 0; i < size; i++) { op = cigar[i].first; dist = cigar[i].second; switch(op) { case 'S': stop = seq_bp + dist; stop = stop > seq_len ? seq_len : stop; while (seq_bp < stop) { seq[seq_bp] = 'N'; seq_bp++; bp++; } break; case 'D': stop = seq_bp + dist; stop = stop > seq_len ? seq_len : stop; while (seq_bp < stop) { seq[seq_bp] = 'N'; seq_bp++; } break; case 'I': case 'M': stop = bp + dist; stop = stop > seq_len ? seq_len : stop; while (bp < stop) { seq[seq_bp] = old_seq[bp]; seq_bp++; bp++; } break; default: break; } } seq[seq_len] = '\0'; return 0; } int apply_cigar_to_model_seq(char *seq, uint seq_len, const char *model, vector > &cigar) { uint size = cigar.size(); char op; uint dist, model_bp, seq_bp, stop; model_bp = 0; seq_bp = 0; for (uint i = 0; i < size; i++) { op = cigar[i].first; dist = cigar[i].second; switch(op) { case 'S': stop = seq_bp + dist; stop = stop > seq_len ? seq_len : stop; while (seq_bp < stop) { seq[seq_bp] = 'U'; seq_bp++; model_bp++; } break; case 'D': stop = seq_bp + dist; stop = stop > seq_len ? seq_len : stop; while (seq_bp < stop) { seq[seq_bp] = 'U'; seq_bp++; } break; case 'I': case 'M': stop = model_bp + dist; stop = stop > seq_len ? seq_len : stop; while (model_bp < stop) { seq[seq_bp] = model[model_bp]; seq_bp++; model_bp++; } break; default: break; } } seq[seq_len] = '\0'; return 0; } string remove_cigar_from_seq(const char *seq, vector > &cigar) { uint size = cigar.size(); char op; uint dist, bp, edited_bp, stop; string edited_seq; // // Calculate the overall sequence length. // uint seqlen = 0; for (uint i = 0; i < size; i++) seqlen += cigar[i].first != 'D' ? cigar[i].second : 0; bp = 0; edited_seq.reserve(seqlen); for (uint i = 0; i < size; i++) { op = cigar[i].first; dist = cigar[i].second; switch(op) { case 'D': edited_bp = 0; while (edited_bp < dist) { edited_bp++; bp++; } break; case 'I': case 'M': stop = bp + dist; while (bp < stop) { edited_seq.push_back(seq[bp]); bp++; } break; default: break; } } return edited_seq; } int adjust_snps_for_gaps(vector > &cigar, Locus *loc) { uint size = cigar.size(); char op; uint dist, bp, stop, offset, snp_index; bp = 0; offset = 0; snp_index = 0; for (uint i = 0; i < size; i++) { op = cigar[i].first; dist = cigar[i].second; switch(op) { case 'D': offset += dist; break; case 'I': case 'M': case 'S': stop = bp + dist; while (bp < stop && snp_index < loc->snps.size()) { if (loc->snps[snp_index]->col == bp) { loc->snps[snp_index]->col += offset; snp_index++; } bp++; } break; default: break; } } return 0; } int adjust_and_add_snps_for_gaps(vector > &cigar, Locus *loc) { uint size = cigar.size(); char op; uint dist, bp, new_bp, stop, snp_cnt; SNP *s; bp = 0; new_bp = 0; snp_cnt = loc->snps.size(); vector snps; for (uint i = 0; i < size; i++) { op = cigar[i].first; dist = cigar[i].second; switch(op) { case 'D': stop = new_bp + dist; while (new_bp < stop) { s = new SNP; s->col = new_bp; s->type = snp_type_unk; s->rank_1 = 'N'; snps.push_back(s); new_bp++; } break; case 'I': case 'M': case 'S': stop = bp + dist > snp_cnt ? snp_cnt : bp + dist; while (bp < stop) { loc->snps[bp]->col = new_bp; snps.push_back(loc->snps[bp]); bp++; new_bp++; } break; default: break; } } loc->snps.clear(); for (uint i = 0; i < snps.size(); i++) loc->snps.push_back(snps[i]); return 0; } int remove_snps_from_gaps(vector > &cigar, Locus *loc) { uint size = cigar.size(); char op; uint dist, bp, new_bp, stop, snp_cnt; bp = 0; new_bp = 0; snp_cnt = loc->snps.size(); vector snps; for (uint i = 0; i < size; i++) { op = cigar[i].first; dist = cigar[i].second; switch(op) { case 'D': stop = bp + dist; while (bp < stop) { delete loc->snps[bp]; bp++; } break; case 'I': case 'M': case 'S': stop = bp + dist > snp_cnt ? snp_cnt : bp + dist; while (bp < stop) { loc->snps[bp]->col = new_bp; snps.push_back(loc->snps[bp]); bp++; new_bp++; } break; default: break; } } loc->snps.clear(); for (uint i = 0; i < snps.size(); i++) loc->snps.push_back(snps[i]); return 0; } stacks-1.44/src/aln_utils.h000644 000765 000024 00000003324 12706750425 016446 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __ALN_UTILS_H__ #define __ALN_UTILS_H__ #include using std::pair; using std::make_pair; #include using std::string; #include using std::vector; #include "locus.h" #include "constants.h" #include "utils.h" string invert_cigar(string); int parse_cigar(const char *, vector > &); string apply_cigar_to_seq(const char *, vector > &); string remove_cigar_from_seq(const char *, vector > &); string apply_cigar_to_model_seq(const char *, vector > &); int apply_cigar_to_seq(char *, uint, const char *, vector > &); int apply_cigar_to_model_seq(char *, uint, const char *, vector > &); int adjust_snps_for_gaps(vector > &, Locus *); int adjust_and_add_snps_for_gaps(vector > &, Locus *); int remove_snps_from_gaps(vector > &, Locus *); #endif // __ALN_UTILS_H__ stacks-1.44/src/BamI.h000644 000765 000024 00000026152 13001012336 015245 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2013-2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __BAMI_H__ #define __BAMI_H__ // // Code to parse binary BAM format. This format is created for // reads that have been aligned to a reference genome. // #ifdef HAVE_BAM #include "input.h" #include "sam.h" class Bam: public Input { htsFile *bam_fh; bam_hdr_t *bamh; bam1_t *aln; map chrs; int parse_header(); int parse_cigar(const char *, vector > &, bool); int parse_bam_cigar(vector > &, bool); int find_start_bp_pos(int, vector > &); int find_start_bp_neg(int, vector > &); int edit_gaps(vector > &, char *); public: Bam(const char *path) : Input() { this->path = string(path); this->bam_fh = hts_open(path, "r"); this->aln = bam_init1(); this->parse_header(); }; ~Bam() { hts_close(this->bam_fh); bam_hdr_destroy(this->bamh); bam_destroy1(this->aln); }; Seq *next_seq(); int next_seq(Seq&); }; int Bam::parse_header() { this->bamh = bam_hdr_init(); this->bamh = sam_hdr_read(this->bam_fh); for (uint j = 0; j < (uint) this->bamh->n_targets; j++) { // // Record the mapping from integer ID to chromosome name that we will see in BAM records. // this->chrs[j] = string(this->bamh->target_name[j]); } return 0; } Seq * Bam::next_seq() { Seq* s = new Seq(); if(next_seq(*s) != 1) { delete s; s = NULL; } return s; } int Bam::next_seq(Seq& s) { int bytes_read = 0; int sflag = 0; int flag = 0; // // Read a record from the file, skipping unmapped reads, and place it in a Seq object. // do { bytes_read = sam_read1(this->bam_fh, this->bamh, this->aln); if (bytes_read <= 0) return 0; flag = ((this->aln->core.flag & BAM_FUNMAP) != 0); } while (flag == 1); // // Check which strand this is aligned to: // SAM reference: FLAG bit 0x10 - sequence is reverse complemented // sflag = ((this->aln->core.flag & BAM_FREVERSE) != 0); // // If the read was aligned on the reverse strand (and is therefore reverse complemented) // alter the start point of the alignment to reflect the right-side of the read, at the // end of the RAD cut site. // // To accomplish this, we must parse the alignment CIGAR string // vector > cigar; this->parse_bam_cigar(cigar, sflag); uint bp = sflag ? this->find_start_bp_neg(this->aln->core.pos, cigar) : this->find_start_bp_pos(this->aln->core.pos, cigar); // // Check if this is the primary or secondary alignment. // alnt aln_type = pri_aln; flag = ((this->aln->core.flag & BAM_FSECONDARY) != 0); if (flag) aln_type = sec_aln; // // Check if this is a supplemenatry (chimeric) alignment (not yet defined in Bam.h). // flag = ((this->aln->core.flag & 2048) != 0); if (flag) aln_type = sup_aln; // // Fetch the sequence. // string seq; uint8_t j; seq.reserve(this->aln->core.l_qseq); for (int i = 0; i < this->aln->core.l_qseq; i++) { j = bam_seqi(bam_get_seq(this->aln), i); switch(j) { case 1: seq += 'A'; break; case 2: seq += 'C'; break; case 4: seq += 'G'; break; case 8: seq += 'T'; break; case 15: seq += 'N'; break; } } // // Fetch the quality score. // string qual; uint8_t *q = bam_get_qual(this->aln); for (int i = 0; i < this->aln->core.l_qseq; i++) { qual += char(int(q[i]) + 33); } string chr = this->chrs[this->aln->core.tid]; // // Calculate the percentage of the sequence that was aligned to the reference. // double len = 0.0; for (uint i = 0; i < cigar.size(); i++) switch (cigar[i].first) { case 'M': case 'I': case '=': case 'X': len += cigar[i].second; break; case 'D': case 'S': case 'H': case 'N': break; default: cerr << "Error parsing CIGAR string '" << cigar[i].second << cigar[i].first << "'.\n"; break; } double pct_aln = len / double(seq.length()); s = Seq((const char *) bam_get_qname(this->aln), seq.c_str(), qual.c_str(), chr.c_str(), bp, sflag ? strand_minus : strand_plus, aln_type, pct_aln); if (cigar.size() > 0) this->edit_gaps(cigar, s.seq); return 1; } int Bam::parse_bam_cigar(vector > &cigar, bool orientation) { int op, len; char c; uint32_t *cgr = bam_get_cigar(this->aln); for (int k = 0; k < this->aln->core.n_cigar; k++) { op = cgr[k] & BAM_CIGAR_MASK; len = cgr[k] >> BAM_CIGAR_SHIFT; switch(op) { case BAM_CMATCH: c = 'M'; break; case BAM_CEQUAL: c = '='; break; case BAM_CDIFF: c = 'X'; break; case BAM_CINS: c = 'I'; break; case BAM_CDEL: c = 'D'; break; case BAM_CREF_SKIP: c = 'N'; break; case BAM_CSOFT_CLIP: c = 'S'; break; case BAM_CHARD_CLIP: c = 'H'; break; case BAM_CPAD: c = 'P'; break; default: cerr << "Unknown operation present in CIGAR string.\n"; break; } // // If aligned to the negative strand, sequence has been reverse complemented and // CIGAR string should be interpreted in reverse. // if (orientation == strand_plus) cigar.push_back(make_pair(c, len)); else cigar.insert(cigar.begin(), make_pair(c, len)); } return 0; } int Bam::parse_cigar(const char *cigar_str, vector > &cigar, bool orientation) { char buf[id_len]; int dist; const char *p, *q; p = cigar_str; if (*p == '*') return 0; while (*p != '\0') { q = p + 1; while (*q != '\0' && isdigit(*q)) q++; strncpy(buf, p, q - p); buf[q-p] = '\0'; dist = atoi(buf); // // If aligned to the negative strand, sequence has been reverse complemented and // CIGAR string should be interpreted in reverse. // if (orientation == strand_plus) cigar.push_back(make_pair(*q, dist)); else cigar.insert(cigar.begin(), make_pair(*q, dist)); p = q + 1; } return 0; } int Bam::find_start_bp_neg(int aln_bp, vector > &cigar) { uint size = cigar.size(); char op; uint dist; for (uint i = 0; i < size; i++) { op = cigar[i].first; dist = cigar[i].second; switch(op) { case 'I': case 'H': break; case 'S': if (i < size - 1) aln_bp += dist; break; case 'M': case '=': case 'X': case 'D': case 'N': aln_bp += dist; break; default: break; } } return aln_bp - 1; } int Bam::find_start_bp_pos(int aln_bp, vector > &cigar) { char op; uint dist; op = cigar[0].first; dist = cigar[0].second; if (op == 'S') aln_bp -= dist; return aln_bp; } int Bam::edit_gaps(vector > &cigar, char *seq) { char *buf; uint size = cigar.size(); char op; uint dist, bp, len, buf_len, buf_size, j, k, stop; len = strlen(seq); bp = 0; buf = new char[len + 1]; buf_size = len + 1; for (uint i = 0; i < size; i++) { op = cigar[i].first; dist = cigar[i].second; switch(op) { case 'S': stop = bp + dist; stop = stop > len ? len : stop; while (bp < stop) { seq[bp] = 'N'; bp++; } break; case 'D': // // A deletion has occured in the read relative to the reference genome. // Pad the read with sufficent Ns to match the deletion, shifting the existing // sequence down. Trim the final length to keep the read length consistent. // k = bp >= len ? len : bp; strncpy(buf, seq + k, buf_size - 1); buf[buf_size - 1] = '\0'; buf_len = strlen(buf); stop = bp + dist; stop = stop > len ? len : stop; while (bp < stop) { seq[bp] = 'N'; bp++; } j = bp; k = 0; while (j < len && k < buf_len) { seq[j] = buf[k]; k++; j++; } break; case 'I': // // An insertion has occurred in the read relative to the reference genome. Delete the // inserted bases and pad the end of the read with Ns. // if (bp >= len) break; k = bp + dist > len ? len : bp + dist; strncpy(buf, seq + k, buf_size - 1); buf[buf_size - 1] = '\0'; buf_len = strlen(buf); j = bp; k = 0; while (j < len && k < buf_len) { seq[j] = buf[k]; k++; j++; } stop = j + dist; stop = stop > len ? len : stop; while (j < stop) { seq[j] = 'N'; j++; } break; case 'M': case '=': case 'X': bp += dist; break; default: break; } } delete [] buf; return 0; } #else // If HAVE_BAM is undefined and BAM library is not present. #include "input.h" class Bam: public Input { public: Bam(const char *path) : Input() { cerr << "BAM support was not enabled when Stacks was compiled.\n"; }; ~Bam() {}; Seq *next_seq() { return NULL; }; int next_seq(Seq &) { return 0; }; }; #endif // HAVE_BAM #endif // __BAMI_H__ stacks-1.44/src/BamUnalignedI.h000644 000765 000024 00000007636 12764047443 017130 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __BAMUNALIGNEDI_H__ #define __BAMUNALIGNEDI_H__ // // Code to parse binary BAM format. This format is created for // reads that have NOT been aligned to a reference genome. // #ifdef HAVE_BAM #include "input.h" #include "sam.h" class BamUnAln: public Input { htsFile *bam_fh; bam_hdr_t *bamh; bam1_t *aln; map chrs; int parse_header(); public: BamUnAln(const char *path) : Input() { this->path = string(path); this->bam_fh = hts_open(path, "r"); this->aln = bam_init1(); this->parse_header(); }; BamUnAln(string path) : Input() { this->path = path; this->bam_fh = hts_open(path.c_str(), "r"); this->aln = bam_init1(); this->parse_header(); }; ~BamUnAln() { hts_close(this->bam_fh); bam_hdr_destroy(this->bamh); bam_destroy1(this->aln); }; Seq *next_seq(); int next_seq(Seq &); }; int BamUnAln::parse_header() { this->bamh = bam_hdr_init(); this->bamh = sam_hdr_read(this->bam_fh); for (uint j = 0; j < (uint) this->bamh->n_targets; j++) { // // Record the mapping from integer ID to chromosome name that we will see in BAM records. // this->chrs[j] = string(this->bamh->target_name[j]); } return 0; } Seq * BamUnAln::next_seq() { Seq* s = new Seq(); if(next_seq(*s) != 1) { delete s; s = NULL; } return s; } int BamUnAln::next_seq(Seq& s) { int bytes_read = 0; // // Read a record from the file and place it in a Seq object. // bytes_read = sam_read1(this->bam_fh, this->bamh, this->aln); if (bytes_read <= 0) return 0; // // Fetch the sequence. // string seq; uint8_t j; seq.reserve(this->aln->core.l_qseq); for (int i = 0; i < this->aln->core.l_qseq; i++) { j = bam_seqi(bam_get_seq(this->aln), i); switch(j) { case 1: seq += 'A'; break; case 2: seq += 'C'; break; case 4: seq += 'G'; break; case 8: seq += 'T'; break; case 15: seq += 'N'; break; } } // // Fetch the quality score. // string qual; uint8_t *q = bam_get_qual(this->aln); for (int i = 0; i < this->aln->core.l_qseq; i++) { qual += char(int(q[i]) + 33); } string chr = this->chrs[this->aln->core.tid]; // // Attempt to parse the query name for this read. // s = Seq((const char *) bam_get_qname(this->aln), seq.c_str(), qual.c_str()); return 1; } #else // If HAVE_BAM is undefined and BAM library is not present. #include "input.h" class BamUnAln: public Input { public: BamUnAln(const char *path) : Input() { cerr << "BAM support was not enabled when Stacks was compiled.\n"; }; BamUnAln(string path) : Input() { cerr << "BAM support was not enabled when Stacks was compiled.\n"; }; ~BamUnAln() {}; Seq *next_seq() { return NULL; }; int next_seq(Seq &) { return 0; }; }; #endif // HAVE_BAM #endif // __BAMUNALIGNEDI_H__ stacks-1.44/src/bootstrap.h000644 000765 000024 00000025142 12764047443 016476 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2014, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __BOOTSTRAP_H__ #define __BOOTSTRAP_H__ #include #include using std::vector; #include "smoothing_utils.h" extern double sigma; extern int bootstrap_reps; extern bool bootstrap_wl; extern set bootstraplist; // // Bootstrap resamplign structure. // class BSample { public: int bp; int alleles; bool fixed; double stat[PopStatSize]; BSample() { this->bp = 0; this->alleles = 0; this->fixed = false; for (uint i = 0; i < PopStatSize; i++) this->stat[i] = 0.0; } }; template class Bootstrap { double *weights; // Weight matrix to apply while smoothing. vector > stats; uint num_stats; public: Bootstrap(uint size) { this->num_stats = size; this->weights = calc_weights(); this->stats.resize(size, vector()); } ~Bootstrap() { delete [] this->weights; } int add_data(vector &); int execute(vector &); int execute_mixed(vector &); double pval(double, vector &); }; template int Bootstrap::add_data(vector &sites) { for (uint i = 0; i < sites.size(); i++) { if (sites[i] != NULL && sites[i]->fixed == false) for (uint j = 0; j < this->num_stats; j++) this->stats[j].push_back(sites[i]->stat[j]); } return 0; } template int Bootstrap::execute(vector &sites) { #pragma omp parallel { PopStat *c; double final_weight, sum, weighted_stat[PopStatSize]; int dist, index; uint pos_l = 0; uint pos_u = 0; //#pragma omp for schedule(dynamic, 1) for (uint pos_c = 0; pos_c < sites.size(); pos_c++) { c = sites[pos_c]; if (c == NULL) continue; if (bootstrap_wl && bootstraplist.count(c->loc_id) == 0) continue; // cerr << "Bootstrapping " << c->loc_id << "; pos_c: " << pos_c << "; bp: " << c->bp << "\n"; determine_window_limits(sites, c->bp, pos_l, pos_u); int size = 0; for (uint i = pos_l; i < pos_u; i++) if (sites[i] != NULL) size++; // // Allocate an array of bootstrap resampling objects. // BSample *bs = new BSample[size]; // // Populate the BSample objects. // int j = 0; for (uint i = pos_l; i < pos_u; i++) { if (sites[i] == NULL) continue; bs[j].bp = sites[i]->bp; bs[j].alleles = sites[i]->alleles; j++; } vector > resampled_stats(this->num_stats, vector()); for (uint i = 0; i < this->num_stats; i++) resampled_stats[i].reserve(bootstrap_reps); // // Bootstrap this bitch. // for (int i = 0; i < bootstrap_reps; i++) { // if (i % 100 == 0) cerr << " Bootsrap rep " << i << "\n"; for (uint k = 0; k < this->num_stats; k++) weighted_stat[k] = 0.0; sum = 0.0; for (j = 0; j < size; j++) { // // Distance from center of window. // dist = bs[j].bp > c->bp ? bs[j].bp - c->bp : c->bp - bs[j].bp; // // Resample for this round of bootstrapping. // index = (int) (this->stats[0].size() * (random() / (RAND_MAX + 1.0))); for (uint k = 0; k < this->num_stats; k++) bs[j].stat[k] = this->stats[k][index]; final_weight = (bs[j].alleles - 1) * this->weights[dist]; for (uint k = 0; k < this->num_stats; k++) weighted_stat[k] += bs[j].stat[k] * final_weight; sum += final_weight; } // cerr << " New weighted Fst value: " << weighted_fst / sum << "\n"; for (uint k = 0; k < this->num_stats; k++) resampled_stats[k].push_back(weighted_stat[k] / sum); } // // Cacluate the p-value for this window based on the empirical Fst distribution. // for (uint k = 0; k < this->num_stats; k++) { sort(resampled_stats[k].begin(), resampled_stats[k].end()); c->bs[k] = this->pval(c->smoothed[k], resampled_stats[k]); } delete [] bs; } } return 0; } template int Bootstrap::execute_mixed(vector &sites) { #pragma omp parallel { PopStat *c; double final_weight, sum, weighted_stat[PopStatSize]; int dist, index; uint pos_l = 0; uint pos_u = 0; //#pragma omp for schedule(dynamic, 1) for (uint pos_c = 0; pos_c < sites.size(); pos_c++) { c = sites[pos_c]; if (c == NULL || c->fixed == true) continue; if (bootstrap_wl && bootstraplist.count(c->loc_id) == 0) continue; // cerr << "Bootstrapping " << c->loc_id << "; pos_c: " << pos_c << "; bp: " << c->bp << "\n"; determine_window_limits(sites, c->bp, pos_l, pos_u); int size = 0; for (uint i = pos_l; i < pos_u; i++) if (sites[i] != NULL) size++; // // Allocate an array of bootstrap resampling objects. // BSample *bs = new BSample[size]; // // Populate the BSample objects. // int j = 0; for (uint i = pos_l; i < pos_u; i++) { if (sites[i] == NULL) continue; bs[j].bp = sites[i]->bp; bs[j].alleles = sites[i]->alleles; bs[j].fixed = sites[i]->fixed; for (uint k = 0; k < this->num_stats; k++) bs[j].stat[k] = sites[i]->stat[k]; j++; } // // Precompute the fraction of the window that will not change during resampling. // double partial_weighted_stat[this->num_stats]; double partial_sum = 0.0; memset(partial_weighted_stat, 0, this->num_stats); for (j = 0; j < size; j++) { if (bs[j].fixed == false) continue; dist = bs[j].bp > c->bp ? bs[j].bp - c->bp : c->bp - bs[j].bp; final_weight = (bs[j].alleles - 1.0) * this->weights[dist]; partial_sum += final_weight; for (uint k = 0; k < this->num_stats; k++) partial_weighted_stat[k] += bs[j].stat[k] * final_weight; } vector > resampled_stats(this->num_stats, vector()); for (uint i = 0; i < this->num_stats; i++) resampled_stats[i].reserve(bootstrap_reps); // cerr << "Window starts at " << bs[0].bp << "; centered on " << c->bp << "\n"; // // Bootstrap this bitch. // for (int i = 0; i < bootstrap_reps; i++) { // if (i % 100 == 0) cerr << " Bootsrap rep " << i << "\n"; for (uint k = 0; k < this->num_stats; k++) weighted_stat[k] = partial_weighted_stat[k]; sum = partial_sum; for (j = 0; j < size; j++) { if (bs[j].fixed == true) continue; dist = bs[j].bp > c->bp ? bs[j].bp - c->bp : c->bp - bs[j].bp; // // Resample for this round of bootstrapping. // index = (int) (this->stats[0].size() * (random() / (RAND_MAX + 1.0))); for (uint k = 0; k < this->num_stats; k++) bs[j].stat[k] = this->stats[k][index]; final_weight = (bs[j].alleles - 1) * this->weights[dist]; for (uint k = 0; k < this->num_stats; k++) weighted_stat[k] += bs[j].stat[k] * final_weight; sum += final_weight; } // cerr << " New weighted value: " << (weighted_stat[0] / sum) << "\n"; for (uint k = 0; k < this->num_stats; k++) resampled_stats[k].push_back(weighted_stat[k] / sum); } // // Cacluate the p-value for this window based on the empirical Fst distribution. // for (uint k = 0; k < this->num_stats; k++) { sort(resampled_stats[k].begin(), resampled_stats[k].end()); c->bs[k] = this->pval(c->smoothed[k], resampled_stats[k]); } delete [] bs; } } return 0; } template double Bootstrap::pval(double stat, vector &dist) { vector::iterator up; double pos; up = upper_bound(dist.begin(), dist.end(), stat); if (up == dist.begin()) pos = 1; else if (up == dist.end()) pos = dist.size(); else pos = up - dist.begin() + 1; double res = 1.0 - (pos / (double) dist.size()); // cerr << "Generated Smoothed Fst Distribution:\n"; // for (uint n = 0; n < dist.size(); n++) // cerr << " n: " << n << "; Fst: " << dist[n] << "\n"; // cerr << "Comparing Fst value: " << stat // << " at position " << (up - dist.begin()) << " out of " // << dist.size() << " positions (converted position: " << pos << "); pvalue: " << res << ".\n"; return res; } #endif // __BOOTSTRAP_H__ stacks-1.44/src/BowtieI.h000644 000765 000024 00000004335 12764047443 016024 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __BOWTIEI_H__ #define __BOWTIEI_H__ // // Code to parse Bowtie's alignment format. This format is created for // reads that have been aligned to a reference genome. It takes the tab-separated form: // // // // One record per line. // #include "input.h" class Bowtie: public Input { public: Bowtie(const char *path) : Input(path) {}; ~Bowtie() {}; Seq *next_seq(); int next_seq(Seq &); }; Seq *Bowtie::next_seq() { Seq* s = new Seq(); if(next_seq(*s) != 1) { delete s; s = NULL; } return s; } int Bowtie::next_seq(Seq& s) { vector parts; // // Read a record from the file and place it in a Seq object // this->fh.getline(this->line, max_len); if (!this->fh.good()) { return 0; } parse_tsv(this->line, parts); strand_type strand = parts[1] == "+" ? strand_plus : strand_minus; // // If the read was aligned on the reverse strand (and is therefore reverse complemented) // alter the start point of the alignment to reflect the right-side of the read, at the // end of the RAD cut site. // int bp = strand == strand_plus ? atoi(parts[3].c_str()) : atoi(parts[3].c_str()) + parts[4].length(); s = Seq(parts[0].c_str(), parts[4].c_str(), parts[5].c_str(), parts[2].c_str(), bp, strand); return 1; } #endif // __BOWTIEI_H__ stacks-1.44/src/BustardI.h000644 000765 000024 00000005657 12761620410 016173 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2013, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // Code to parse Illumina's Bustard file format. It takes the tab-separated form: // // // // One record per line. // #ifndef __BUSTARDI_H__ #define __BUSTARDI_H__ #include "input.h" const uint num_bustd_fields = 11; class Bustard: public Input { public: Bustard(const char *path) : Input(path) {}; Bustard(string path) : Input(path.c_str()) {}; ~Bustard() {}; Seq *next_seq(); int next_seq(Seq &); }; Seq *Bustard::next_seq() { vector parts; // // Read a record from the file and place it in a Seq object // this->fh.getline(this->line, max_len); if (!this->fh.good()) { return NULL; } parse_tsv(this->line, parts); if (parts.size() != num_bustd_fields) { cerr << "Error parsing '" << this->path.c_str() << "' found " << parts.size() << " fields, but expecting " << num_bustd_fields << "). " << "Perhaps you should specify the input file type (-i)?\n"; return NULL; } Seq *s = new Seq; s->seq = new char[parts[8].length() + 1]; strcpy(s->seq, parts[8].c_str()); s->qual = new char[parts[9].length() + 1]; strcpy(s->qual, parts[9].c_str()); s->id = new char[id_len]; sprintf(s->id, "@%s:%s:%s:%s:%s#%s/%s", parts[0].c_str(), parts[2].c_str(), parts[3].c_str(), parts[4].c_str(), parts[5].c_str(), parts[6].c_str(), parts[7].c_str()); return s; } int Bustard::next_seq(Seq &s) { vector parts; // // Read a record from the file and place it in a Seq object // this->fh.getline(this->line, max_len); if (!this->fh.good()) { return 0; } parse_tsv(this->line, parts); strcpy(s.seq, parts[2].c_str()); strcpy(s.qual, parts[3].c_str()); sprintf(s.id, "@%s:%s:%s:%s:%s#%s/%s", parts[0].c_str(), parts[1].c_str(), parts[2].c_str(), parts[3].c_str(), parts[4].c_str(), parts[5].c_str(), parts[6].c_str()); return 1; } #endif // __BUSTARDI_H__ stacks-1.44/src/catalog_utils.cc000644 000765 000024 00000027515 12764047443 017457 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2013, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // catalog_utils.cc -- common routines for manipulating catalog objects. // // Julian Catchen // jcatchen@uoregon.edu // University of Oregon // #include "catalog_utils.h" int reduce_catalog(map &catalog, set &whitelist, set &blacklist) { map list; map::iterator it; CSLocus *loc; if (whitelist.size() == 0 && blacklist.size() == 0) return 0; int i = 0; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (whitelist.size() > 0 && whitelist.count(loc->id) == 0) continue; if (blacklist.count(loc->id)) continue; list[it->first] = it->second; i++; } catalog = list; return i; } int implement_single_snp_whitelist(map &catalog, PopSum *psum, map > &whitelist) { map > new_wl; CSLocus *loc; LocTally *t; if (whitelist.size() > 0) { map >::iterator it; for (it = whitelist.begin(); it != whitelist.end(); it++) { loc = catalog[it->first]; t = psum->locus_tally(loc->id); // // If no specific SNPs are specified in the whitelist all SNPs are included, choose the first variant. // if (it->second.size() == 0) { for (uint i = 0; i < loc->snps.size(); i++) if (t->nucs[loc->snps[i]->col].fixed == false) { new_wl[loc->id].insert(loc->snps[i]->col); break; } } else { // // Otherwise, choose the first SNP that is already in the whitelist. // for (uint i = 0; i < loc->snps.size(); i++) { if (it->second.count(loc->snps[i]->col) == 0 || t->nucs[loc->snps[i]->col].fixed == true) continue; new_wl[loc->id].insert(loc->snps[i]->col); break; } } } } else { map::iterator it; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; t = psum->locus_tally(loc->id); for (uint i = 0; i < loc->snps.size(); i++) if (t->nucs[loc->snps[i]->col].fixed == false) { new_wl[loc->id].insert(loc->snps[i]->col); break; } } } whitelist = new_wl; return 0; } int implement_random_snp_whitelist(map &catalog, PopSum *psum, map > &whitelist) { map > new_wl; CSLocus *loc; uint index; if (whitelist.size() > 0) { map >::iterator it; for (it = whitelist.begin(); it != whitelist.end(); it++) { loc = catalog[it->first]; if (loc->snps.size() == 0) continue; if (it->second.size() == 0) { index = rand() % loc->snps.size(); new_wl[loc->id].insert(loc->snps[index]->col); } else { do { index = rand() % loc->snps.size(); } while (it->second.count(loc->snps[index]->col) == 0); new_wl[loc->id].insert(loc->snps[index]->col); } } } else { map::iterator it; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (loc->snps.size() > 0) { index = rand() % loc->snps.size(); new_wl[loc->id].insert(loc->snps[index]->col); } } } whitelist = new_wl; return 0; } int check_whitelist_integrity(map &catalog, map > &whitelist) { if (whitelist.size() == 0) return 0; int rm_snps = 0; int rm_loci = 0; CSLocus *loc; map >::iterator it; set::iterator sit; map > new_wl; cerr << "Checking the integrity of the whitelist..."; for (it = whitelist.begin(); it != whitelist.end(); it++) { if (catalog.count(it->first) == 0) { rm_loci++; cerr << "\n Removing locus " << it->first << " from whitelist as it does not exist in the catalog."; } else { loc = catalog[it->first]; if (it->second.size() == 0) { new_wl.insert(make_pair(it->first, std::set())); continue; } set cat_snps; for (uint i = 0; i < loc->snps.size(); i++) cat_snps.insert(loc->snps[i]->col); for (sit = it->second.begin(); sit != it->second.end(); sit++) if (cat_snps.count(*sit)) { new_wl[it->first].insert(*sit); } else { rm_snps++; cerr << "\n Removing SNP at column " << *sit << " in locus " << it->first << " from whitelist as it does not exist in the catalog."; } } } whitelist = new_wl; if (rm_loci > 0 || rm_snps > 0) cerr << "\n"; cerr << "done.\n" << "Removed " << rm_loci << " loci and " << rm_snps << " SNPs from the whitelist that were not found in the catalog.\n"; return 0; } int reduce_catalog(map &catalog, map > &whitelist, set &blacklist) { map list; map::iterator it; CSLocus *loc; if (whitelist.size() == 0 && blacklist.size() == 0) return 0; int i = 0; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (whitelist.size() > 0 && whitelist.count(loc->id) == 0) continue; if (blacklist.count(loc->id)) continue; list[it->first] = it->second; i++; } catalog = list; return i; } int reduce_catalog_snps(map &catalog, map > &whitelist, PopMap *pmap) { map::iterator it; CSLocus *loc; Datum **d; if (whitelist.size() == 0) return 0; // // We want to prune out SNP objects that are not in the whitelist. // int pos; vector tmp; vector cols; map obshaps; map::iterator sit; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (whitelist[loc->id].size() == 0) continue; tmp.clear(); cols.clear(); d = pmap->locus(loc->id); for (uint i = 0; i < loc->snps.size(); i++) { if (whitelist[loc->id].count(loc->snps[i]->col) > 0) { tmp.push_back(loc->snps[i]); cols.push_back(i); } else { // // Change the model calls in the samples to no longer contain this SNP. // pos = loc->snps[i]->col; for (int j = 0; j < pmap->sample_cnt(); j++) { if (d[j] == NULL || pos >= d[j]->len) continue; if (d[j]->model != NULL) { d[j]->model[pos] = 'U'; } } delete loc->snps[i]; } } loc->snps.clear(); for (uint i = 0; i < tmp.size(); i++) loc->snps.push_back(tmp[i]); map::iterator it; char allele_old[id_len], allele_new[id_len]; // // We need to adjust the catalog's list of haplotypes/alleles // for this locus to account for the pruned SNPs. // for (it = loc->alleles.begin(); it != loc->alleles.end(); it++) { strncpy(allele_old, it->first.c_str(), id_len - 1); allele_old[id_len - 1] = '\0'; for (uint k = 0; k < cols.size(); k++) allele_new[k] = allele_old[cols[k]]; allele_new[cols.size()] = '\0'; obshaps[string(allele_new)] += it->second; } loc->alleles.clear(); for (sit = obshaps.begin(); sit != obshaps.end(); sit++) { loc->alleles[sit->first] = sit->second; } obshaps.clear(); loc->populate_alleles(); // // Now we need to adjust the matched haplotypes to sync to // the SNPs left in the catalog. // // Reducing the lengths of the haplotypes may create // redundant (shorter) haplotypes, we need to remove these. // for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] == NULL) continue; for (uint j = 0; j < d[i]->obshap.size(); j++) { for (uint k = 0; k < cols.size(); k++) d[i]->obshap[j][k] = d[i]->obshap[j][cols[k]]; d[i]->obshap[j][cols.size()] = '\0'; obshaps[d[i]->obshap[j]] += d[i]->depth[j]; } uint j = 0; for (sit = obshaps.begin(); sit != obshaps.end(); sit++) { strcpy(d[i]->obshap[j], sit->first.c_str()); d[i]->depth[j] = sit->second; j++; } while (j < d[i]->obshap.size()) { delete [] d[i]->obshap[j]; j++; } d[i]->obshap.resize(obshaps.size()); d[i]->depth.resize(obshaps.size()); obshaps.clear(); } } return 0; } map create_catalog(const vector& records) { map catalog; for (size_t i = 0; i < records.size(); ++i) { const VcfRecord& rec = records[i]; CSLocus* loc = catalog.insert(make_pair(i, new CSLocus())).first->second; loc->sample_id = 0; loc->id = i; loc->len = 1; loc->con = new char[2]; strcpy(loc->con, rec.alleles[0].c_str()); loc->loc.set(rec.chrom.c_str(), (uint)rec.pos, strand_plus); for (const string& a : rec.alleles) { if (a=="*") continue; loc->alleles.insert({a, 0}); } loc->depth = 0; loc->lnl = 0; loc->snps.push_back(new SNP()); SNP& snp = *loc->snps.back(); snp.col = 0; vector snp_alleles = {&snp.rank_1, &snp.rank_2, &snp.rank_3, &snp.rank_4}; try { size_t s=0; for (const string& a : rec.alleles) { if (a=="*") continue; *snp_alleles.at(s) = a.at(0); ++s; } } catch (out_of_range& e) { cerr << "Warning: Skipping malformed VCF SNP record '" << rec.chrom << ":" << rec.pos << "'." << " Alleles were:"; for (const string& a : rec.alleles) cerr << " '" << a << "';"; cerr << ".\n"; delete loc->snps[0]; delete loc->con; delete loc; catalog.erase(i); continue; } snp.type = *snp_alleles[2] == 0 ? snp_type_hom : snp_type_het; loc->populate_alleles(); } return catalog; } stacks-1.44/src/catalog_utils.h000644 000765 000024 00000007045 12750704270 017306 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2013-2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __CATALOG_UTILS_H__ #define __CATALOG_UTILS_H__ #include using std::map; #include using std::set; #include "constants.h" #include "stacks.h" #include "locus.h" #include "PopMap.h" #include "PopSum.h" #include "Vcf.h" int check_whitelist_integrity(map &, map > &); int reduce_catalog(map &, set &, set &); int reduce_catalog(map &, map > &, set &); int reduce_catalog_snps(map &, map > &, PopMap *); int implement_single_snp_whitelist(map &, PopSum *, map > &); int implement_random_snp_whitelist(map &, PopSum *, map > &); /* * create_catalog(vector&): * Creates a catalog based on VCF SNP records. * * We observe the following rules to create the catalog loci : * [sample_id] (batch number) Always set to 0. * [id] VCF records do not intrinsically have locus ids; we use the SNP records indexes. * [len] Always set to 1. * [con] We use the reference nucleotide as the consensus. * [loc] Use the chromosome and position given by each record (n.b. the * VCF format requires these field), and strand "strand_plus". * [snps] Use the ref+alt alleles. * [col] Always set to 0 (first nucleotide in the consensus). * [type] "snp_type_het" if the alt field is not empty, otherwise "snp_type_hom". * [lratio] Always set to 0. * [rank_1] The ref allele. * [rank_2], [rank_3], [rank_4] The alt allele(s). * [alleles] Use the ref+alt alleles in the order they appear, skipping * the special '*' allele ('variant is irrelevant in certain context * because of an neighboring structural polymorphism') if present. * [strings] We fill this by calling Locus::populate_alleles(). * [cnt] Set to the approriate value when filling the PopMap. * [hcnt] (Same as above.) * [confounded_cnt] (Same as above.) * [gmap] This is filled by tabulate_haplotypes() (in "populations.cc"). * [gcnt] (Same as above.) * [marker] (Same as above.) * * When no depth information is available, the [depth] and the depths * of the [alleles] are set to 0. (n.b. the parsing of depth information in * VCF is not implemented as of Mar 21, 2016.) * * When no likelihood information is available, [lnl] is set to 0. (n.b. * the parsing of likelihood information in VCF is not implemented as of * Mar 21, 2016.) * * The following members are left unset, on the premise that * "populations" does not use them : * model, blacklisted, deleveraged, lumberjack, components, reads, comp_cnt, * comp_type, annotation, uncor_marker, hap_cnts, f, trans_gcnt, chisq. */ map create_catalog(const vector& vcf_records); #endif // __CATALOG_UTILS_H__ stacks-1.44/src/clean.cc000644 000765 000024 00000044465 12764047443 015712 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2011-2014, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // clean.cc -- common routines for processing and cleaning raw seqeunce data. // // Julian Catchen // jcatchen@uoregon.edu // University of Oregon // #include "clean.h" int parse_illumina_v1(const char *file) { const char *p, *q; // // Parse a file name that looks like: s_7_1_0001_qseq.txt ... s_7_1_0120_qseq.txt // but exclude the paired-end files: s_7_2_0001_qseq.txt ... s_7_2_0120_qseq.txt // if (file[0] != 's') return 0; int underscore_cnt = 0; for (p = file; *p != '\0'; p++) { if (*p == '_') { underscore_cnt++; q = p; } } if (underscore_cnt != 4) return 0; // Check the file suffix. if (strncmp(q, "_qseq.txt", 8) != 0) return 0; // Make sure it is not the paired-end file p = file; p += 3; if (strncmp(p, "_1_", 3) != 0) return 0; // // Return the position of the paired-end number, so the other file name can be generated. // return (p + 1 - file); } int parse_illumina_v2(const char *file) { const char *p, *q; // // Parse a file name that looks like: lane6_NoIndex_L006_R1_003.fastq // but exclude the paired-end files: lane6_NoIndex_L006_R2_003.fastq // // Another example could be: GfddRAD1_001_ATCACG_L008_R1_001.fastq.gz // and excluding the paired-end file: GfddRAD1_001_ATCACG_L008_R2_001.fastq.gz // // Make sure it ends in "fastq" or "fastq.gz" // for (q = file; *q != '\0'; q++); for (p = q; *p != '.' && p > file; p--); if (strncmp(p, ".gz", 3) == 0) for (p--; *p != '.' && p > file; p--); if (strncmp(p, ".fastq", 6) != 0) return 0; // // Find the part of the name marking the pair, "_R1_", make sure it is not the paired-end file. // p = file; while (*p != '\0') { for (; *p != '_' && *p != '\0'; p++); if (*p == '\0') return 0; if (strncmp(p, "_R1_", 4) == 0) { // // Return the position of the paired-end number, so the other file name can be generated. // return (p + 2 - file); } p++; } return 0; } int parse_input_record(Seq *s, Read *r) { char *p, *q, *z; uint lim; // // Count the number of colons to differentiate Illumina version. // CASAVA 1.8+ has a FASTQ header like this: // @HWI-ST0747:155:C01WHABXX:8:1101:6455:26332 1:N:0: // Or, with the embedded barcode: // @HWI-ST1233:67:D12GNACXX:7:2307:14604:78978 1:N:0:ATCACG // // // Or, parse FASTQ header from previous versions that looks like this: // @HWI-ST0747_0141:4:1101:1240:2199#0/1 // @HWI-ST0747_0143:2:2208:21290:200914#0/1 // char *stop = s->id + strlen(s->id); int colon_cnt = 0; int hash_cnt = 0; for (p = s->id, q = p; q < stop; q++) { colon_cnt += *q == ':' ? 1 : 0; hash_cnt += *q == '#' ? 1 : 0; } if (colon_cnt == 9 && hash_cnt == 0) { r->fastq_type = illv2_fastq; // // According to Illumina manual, "CASAVA v1.8 User Guide" page 41: // @:::::: ::: // for (p = s->id, q = p; *q != ':' && q < stop; q++); if (q < stop) { *q = '\0'; strcpy(r->machine, p); *q = ':'; } // Run number. for (p = q+1, q = p; *q != ':' && q < stop; q++); //*q = '\0'; // Flowcell ID. for (p = q+1, q = p; *q != ':' && q < stop; q++); //*q = '\0'; for (p = q+1, q = p; *q != ':' && q < stop; q++); if (q < stop) { *q = '\0'; r->lane = atoi(p); *q = ':'; } for (p = q+1, q = p; *q != ':' && q < stop; q++); if (q < stop) { *q = '\0'; r->tile = atoi(p); *q = ':'; } for (p = q+1, q = p; *q != ':' && q < stop; q++); if (q < stop) { *q = '\0'; r->x = atoi(p); *q = ':'; } for (p = q+1, q = p; *q != ' ' && q < stop; q++); if (q < stop) { *q = '\0'; r->y = atoi(p); *q = ' '; } for (p = q+1, q = p; *q != ':' && q < stop; q++); if (q < stop) { *q = '\0'; // r->read = atoi(p); *q = ':'; } for (p = q+1, q = p; *q != ':' && q < stop; q++); if (q < stop) { *q = '\0'; r->filter = *p == 'Y' ? true : false; *q = ':'; } // Control Number. for (p = q+1, q = p; *q != ':' && q < stop; q++); //*q = '\0'; // // Index barcode // // The index barcode appears identically in both single-end and paired-end reads. // If the barcode type is index_index, the barcode will appear as NNNNNN+NNNNNN // in both reads. If the specified barcode type is null_index we want to read only // the second half of the index, if the type is index_null, we want to read // only the first half, or the full string if there is no '+' character. // if (q < stop) for (p = q+1, q = p; q < stop; q++); else p = q; if (*p != '\0') { // // Check if there is a '+' character. // for (z = p; *z != '+' && *z != '\0'; z++); if (r->read == 1) { lim = z - p; switch (barcode_type) { case index_null: case index_index: case index_inline: lim = lim < max_bc_size_1 ? lim : max_bc_size_1; strncpy(r->index_bc, p, lim); r->index_bc[lim] = '\0'; break; case inline_index: lim = lim < max_bc_size_2 ? lim : max_bc_size_2; strncpy(r->index_bc, p, lim); r->index_bc[lim] = '\0'; break; default: break; } } else if (r->read == 2) { if (*z == '+') p = z + 1; switch (barcode_type) { case null_index: case index_index: case inline_index: strncpy(r->index_bc, p, max_bc_size_2); r->index_bc[max_bc_size_2] = '\0'; break; default: break; } } } } else if (colon_cnt == 4 && hash_cnt == 1) { r->fastq_type = illv1_fastq; for (p = s->id, q = p; *q != ':' && q < stop; q++); if (q < stop) { *q = '\0'; strcpy(r->machine, p); *q = ':'; } for (p = q+1, q = p; *q != ':' && q < stop; q++); if (q < stop) { *q = '\0'; r->lane = atoi(p); *q = ':'; } for (p = q+1, q = p; *q != ':' && q < stop; q++); if (q < stop) { *q = '\0'; r->tile = atoi(p); *q = ':'; } for (p = q+1, q = p; *q != ':' && q < stop; q++); if (q < stop) { *q = '\0'; r->x = atoi(p); *q = ':'; } for (p = q+1, q = p; *q != '#' && q < stop; q++); if (q < stop) { *q = '\0'; r->y = atoi(p); *q = '#'; } for (p = q+1, q = p; *q != '/' && q < stop; q++); if (q < stop) { *q = '\0'; r->index = atoi(p); *q = '/'; } for (p = q+1, q = p; *q != '\0' && q < stop; q++); // r->read = atoi(p); } else { r->fastq_type = generic_fastq; strncpy(r->machine, s->id, id_len); r->machine[id_len] = '\0'; } uint len = strlen(s->seq); // // Resize the sequence/phred buffers if necessary. // if (len > r->size - 1) r->resize(len + 1); strncpy(r->seq, s->seq, r->size - 1); strncpy(r->phred, s->qual, r->size - 1); r->seq[r->size - 1] = '\0'; r->phred[r->size - 1] = '\0'; r->len = len; if (r->read == 1) { switch (barcode_type) { case inline_null: case inline_inline: case inline_index: strncpy(r->inline_bc, r->seq, max_bc_size_1); r->inline_bc[max_bc_size_1] = '\0'; break; case index_inline: strncpy(r->inline_bc, r->seq, max_bc_size_2); r->inline_bc[max_bc_size_2] = '\0'; break; default: break; } } else if (r->read == 2 && (barcode_type == inline_inline || barcode_type == index_inline)) { strncpy(r->inline_bc, r->seq, max_bc_size_2); r->inline_bc[max_bc_size_2] = '\0'; } r->retain = 1; return 0; } int rev_complement(char *seq, int offset, bool overhang) { char *p, *q; offset += overhang ? 1 : 0; q = seq + offset; int len = strlen(q); int j = 0; char *com = new char[len + 1]; for (p = q + len - 1; p >= q; p--) { switch (*p) { case 'A': case 'a': com[j] = 'T'; break; case 'C': case 'c': com[j] = 'G'; break; case 'G': case 'g': com[j] = 'C'; break; case 'T': case 't': com[j] = 'A'; break; } j++; } com[len] = '\0'; for (j = 0; j < len; j++) q[j] = com[j]; delete [] com; return 0; } int reverse_qual(char *qual, int offset, bool overhang) { char *p, *q; offset += overhang ? 1 : 0; q = qual + offset; int len = strlen(q); int j = 0; char *com = new char[len + 1]; for (p = q + len - 1; p >= q; p--) { com[j] = *p; j++; } com[len] = '\0'; for (j = 0; j < len; j++) q[j] = com[j]; delete [] com; return 0; } // // Functions for quality filtering based on phred scores. // int check_quality_scores(Read *href, int qual_offset, int score_limit, int len_limit, int offset) { // // Phred quality scores are discussed here: // http://en.wikipedia.org/wiki/FASTQ_format // // Illumina 1.3+ encodes phred scores between ASCII values 64 (0 quality) and 104 (40 quality) // // @ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefgh // | | | | | // 64 74 84 94 104 // 0 10(90%) 20(99%) 30(99.9%) 40(99.99%) // // // Drop sequence if the average phred quality score drops below a threshold within a sliding window. // // // Convert the encoded quality scores to their integer values // // cerr << "Integer scores: "; for (uint j = 0; j < href->len; j++) { href->int_scores[j] = href->phred[j] - qual_offset; // cerr << href->int_scores[j] << " "; } // cerr << "\n"; double mean = 0.0; double working_sum = 0.0; int *p, *q, j; // cerr << "Window length: " << href->win_len << "; Stop position: " << href->stop_pos << "\n"; // // Populate the sliding window. // for (j = offset; j < href->win_len + offset; j++) working_sum += href->int_scores[j]; // cerr << "Populating the sliding window using position " << offset << " to " << href->win_len + offset - 1 << "; initial working sum: " << working_sum << "\n"; // // Set p pointer to the first element in the window, and q to one element past the last element in the window. // p = href->int_scores + offset; q = p + (int) href->win_len; j = offset; // cerr << "Setting pointers; P: " << (href->int_scores + offset) - href->int_scores << "; Q: " << p + (int) href->win_len - p << "; J: " << j << "\n"; do { mean = working_sum / href->win_len; // cerr << "J: " << j << "; Window contents: "; // for (int *r = p; r < q; r++) // cerr << *r << " "; // cerr << "\n"; // cerr << " Mean: " << mean << "\n"; if (mean < score_limit) { if (j < len_limit) { return 0; } else { href->len = j + 1; href->seq[j] = '\0'; href->phred[j] = '\0'; return -1; } } // // Advance the window: // Add the score from the front edge of the window, subtract the score // from the back edge of the window. // working_sum -= (double) *p; working_sum += (double) *q; // cerr << " Removing value of p: " << *p << " (position: " << p - (href->int_scores) << ")\n"; // cerr << " Adding value of q: " << *q << " (position: " << q - (href->int_scores) << ")\n"; p++; q++; j++; } while (j <= href->stop_pos); return 1; } bool correct_barcode(set &bcs, Read *href, seqt type, int num_errs) { if (recover == false) return false; // // The barcode_dist variable specifies how far apart in sequence space barcodes are. If barcodes // are off by two nucleotides in sequence space, than we can correct barcodes that have a single // sequencing error. // // If the barcode sequence is off by no more than barcodes_dist-1 nucleotides, correct it. We will // search the whole possible space of barcodes if more than one length of barcode was specified. // const char *p; char *q; char bc[id_len]; int d, close; string b; set::iterator it; close = 0; for (it = bcs.begin(); it != bcs.end(); it++) { // // Copy the proper subset of the barcode to match the length of the barcode in the bcs set. // strncpy(bc, type == single_end ? href->se_bc : href->pe_bc, it->length()); bc[it->length()] = '\0'; d = 0; for (p = it->c_str(), q = bc; *p != '\0'; p++, q++) if (*p != *q) d++; if (d <= num_errs) { close++; b = *it; break; } } if (close == 1) { // // Correct the barcode. // if (type == single_end) { strcpy(href->se_bc, b.c_str()); if (barcode_type == inline_null || barcode_type == inline_index || barcode_type == inline_inline) href->inline_bc_len = b.length(); } else { strcpy(href->pe_bc, b.c_str()); if (barcode_type == index_inline || barcode_type == inline_inline) href->inline_bc_len = b.length(); } return true; } return false; } // // Functions for filtering adapter sequence // int init_adapter_seq(int kmer_size, char *adapter, int &adp_len, AdapterHash &kmers) { string kmer; adp_len = strlen(adapter); int num_kmers = adp_len - kmer_size + 1; char *p = adapter; for (int i = 0; i < num_kmers; i++) { kmer.assign(p, kmer_size); kmers[kmer].push_back(i); p++; } return 0; } int filter_adapter_seq(Read *href, char *adapter, int adp_len, AdapterHash &adp_kmers, int kmer_size, int distance, int len_limit) { vector > hits; int num_kmers = href->len - kmer_size + 1; const char *p = href->seq; string kmer; // // Identify matching kmers and their locations of occurance. // for (int i = 0; i < num_kmers; i++) { kmer.assign(p, kmer_size); if (adp_kmers.count(kmer) > 0) { for (uint j = 0; j < adp_kmers[kmer].size(); j++) { // cerr << "Kmer hit " << kmer << " at query position " << i << " at hit position " << adp_kmers[kmer][j] << "\n"; hits.push_back(make_pair(i, adp_kmers[kmer][j])); } } p++; } // // Scan backwards from the position of the k-mer and then scan forwards // counting the number of mismatches. // int mismatches, i, j, start_pos; for (uint k = 0; k < hits.size(); k++) { mismatches = 0; i = hits[k].first; // Position in query sequence j = hits[k].second; // Position in adapter hit // cerr << "Starting comparison at i: "<< i << "; j: " << j << "\n"; while (i >= 0 && j >= 0) { if (href->seq[i] != adapter[j]) mismatches++; i--; j--; } if (mismatches > distance) continue; start_pos = i + 1; i = hits[k].first; j = hits[k].second; while (i < (int) href->len && j < adp_len && mismatches <= distance) { if (href->seq[i] != adapter[j]) mismatches++; i++; j++; } // cerr << "Starting position: " << start_pos << "; Query end (i): " << i << "; adapter end (j): " << j // << "; number of mismatches: " << mismatches << "; Seq Len: " << href->len << "; SeqSeq Len: " << strlen(href->seq) << "\n"; if (mismatches <= distance && (i == (int) href->len || j == adp_len)) { // cerr << " Trimming or dropping.\n"; if (start_pos < len_limit) { return 0; } else { href->len = start_pos + 1; href->seq[start_pos] = '\0'; href->phred[start_pos] = '\0'; return -1; } } } return 1; } stacks-1.44/src/clean.h000644 000765 000024 00000032045 12764047443 015543 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2011, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __CLEAN_H__ #define __CLEAN_H__ #include using std::string; #include using std::map; #include #include using std::ofstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include using std::unordered_map; #include "input.h" #include "kmers.h" enum fastqt {generic_fastq, illv1_fastq, illv2_fastq}; enum barcodet {null_null, null_index, inline_null, index_null, inline_inline, index_index, inline_index, index_inline}; enum seqt {single_end, paired_end}; typedef unordered_map, std::hash > AdapterHash; extern uint min_bc_size_1, max_bc_size_1, min_bc_size_2, max_bc_size_2; extern int barcode_dist_1, barcode_dist_2; extern barcodet barcode_type; extern uint truncate_seq; extern double win_size; extern bool paired; extern bool recover; class BarcodePair { public: string se; // Single-end barcode. string pe; // Paired-end barcode. string name; // Filename to open for this barcode combination. BarcodePair() { this->se = ""; this->pe = ""; } BarcodePair(char *p) { this->se = string(p); this->pe = ""; } BarcodePair(char *p, char *q, char *n) { if (p != NULL) this->se = string(p); if (q != NULL) this->pe = string(q); if (n != NULL) this->name = string(n); } BarcodePair(string se, string pe, string name) { this->se = se; this->pe = pe; this->name = name; } BarcodePair(string se) { this->se = se; this->pe = ""; } void set(char *p, char *q) { this->se = string(p); this->pe = string(q); } void set(char *p) { this->se = string(p); this->pe = ""; } void set(string p, string q) { this->se = p; this->pe = q; } void set(string p) { this->se = p; this->pe = ""; } string str() { if (this->pe.length() > 0) return string(this->se + "-" + this->pe); else return this->se; } bool name_exists() { if (this->name.length() > 0) return true; return false; } friend bool operator<(const BarcodePair &lhs, const BarcodePair &rhs) { if (lhs.se < rhs.se) return true; else if (lhs.se == rhs.se && lhs.pe < rhs.pe) return true; else return false; } friend bool operator==(const BarcodePair &lhs, const BarcodePair &rhs) { return (lhs.se == rhs.se && lhs.pe == rhs.pe); } friend ofstream& operator<<(ofstream &out, const BarcodePair &bp) { if (bp.pe.length() > 0) out << bp.se << "-" << bp.pe; else out << bp.se; return out; } }; class Read { public: fastqt fastq_type; char *inline_bc; char *index_bc; char *se_bc; char *pe_bc; char *machine; int lane; int tile; int x; int y; int index; int read; char *seq; char *phred; int *int_scores; bool filter; int inline_bc_len; int retain; uint size; uint len; double win_len; double stop_pos; Read(uint buf_len, int read, int barcode_size, double win_size) { this->inline_bc = new char[id_len + 1]; this->index_bc = new char[id_len + 1]; this->machine = new char[id_len + 1]; this->seq = new char[buf_len + 1]; this->phred = new char[buf_len + 1]; this->int_scores = new int[buf_len]; this->size = buf_len + 1; this->read = read; this->retain = 1; this->inline_bc_len = 0; this->tile = 0; this->lane = 0; this->x = 0; this->y = 0; this->index = 0; this->len = 0; this->inline_bc[0] = '\0'; this->index_bc[0] = '\0'; this->machine[0] = '\0'; this->seq[0] = '\0'; this->phred[0] = '\0'; this->set_len(buf_len); this->se_bc = NULL; this->pe_bc = NULL; if (this->read == 1) { switch(barcode_type) { case index_inline: this->se_bc = this->index_bc; this->pe_bc = this->inline_bc; break; case inline_index: this->se_bc = this->inline_bc; this->pe_bc = this->index_bc; this->inline_bc_len = barcode_size; break; case inline_null: case inline_inline: this->se_bc = this->inline_bc; this->inline_bc_len = barcode_size; break; case index_null: case index_index: this->se_bc = this->index_bc; break; default: break; } } else if (this->read == 2) { switch(barcode_type) { case inline_inline: case index_inline: this->pe_bc = this->inline_bc; this->inline_bc_len = barcode_size; break; case index_index: case inline_index: this->pe_bc = this->index_bc; break; default: break; } } } ~Read() { delete [] this->inline_bc; delete [] this->index_bc; delete [] this->machine; delete [] this->seq; delete [] this->phred; delete [] this->int_scores; } int resize(int size) { delete [] this->seq; delete [] this->phred; delete [] this->int_scores; this->size = size; this->seq = new char[this->size]; this->phred = new char[this->size]; this->int_scores = new int[this->size - 1]; this->set_len(size - 1); return 0; } int set_len(uint buf_len) { if (buf_len == this->len) return 0; if (buf_len > this->size - 1) buf_len = this->size - 1; this->seq[buf_len] = '\0'; this->phred[buf_len] = '\0'; // // Set the parameters for checking read quality later in processing. // Window length is 15% (rounded) of the sequence length. // this->len = buf_len - this->inline_bc_len; this->win_len = round((double) this->len * win_size); if (this->win_len < 1) this->win_len = 1; this->len += this->inline_bc_len; this->stop_pos = this->len - this->win_len; return 0; } }; int parse_illumina_v1(const char *); int parse_illumina_v2(const char *); int parse_input_record(Seq *, Read *); int rev_complement(char *, int, bool); int reverse_qual(char *, int, bool); bool correct_barcode(set &, Read *, seqt, int); int filter_adapter_seq(Read *, char *, int, AdapterHash &, int, int, int); int init_adapter_seq(int, char *, int &, AdapterHash &); int check_quality_scores(Read *, int, int, int, int); // // Templated function to process barcodes. // template int process_barcode(Read *href_1, Read *href_2, BarcodePair &bc, map &fhs, set &se_bc, set &pe_bc, map > &barcode_log, map &counter) { if (barcode_type == null_null) return 0; // // Is this a legitimate barcode? The barcode passed into this function is the maximally long // barcode. If we fail to find a match at maximum length, step down to minimum length and // continue to search for a match. // char *p; char bc_1[id_len]; char bc_2[id_len]; strcpy(bc_1, bc.se.c_str()); strcpy(bc_2, bc.pe.c_str()); bool valid_se_bc = false; bool valid_pe_bc = false; p = bc_1 + max_bc_size_1; // Point p at the end of string NULL. for (uint i = max_bc_size_1; i >= min_bc_size_1; i--) if (se_bc.count(bc_1) > 0) { valid_se_bc = true; break; } else { p--; *p = '\0'; } if (pe_bc.size() > 0) { p = bc_2 + max_bc_size_2; // Point p at the end of string NULL. for (uint i = max_bc_size_2; i >= min_bc_size_2; i--) if (pe_bc.count(bc_2) > 0) { valid_pe_bc = true; break; } else { p--; *p = '\0'; } } if (valid_se_bc == true && valid_pe_bc == true) bc.set(bc_1, bc_2); else if (valid_se_bc == true) bc.se = bc_1; else if (valid_pe_bc == true) bc.pe = bc_2; // // Log the barcodes we receive. // if (barcode_log.count(bc) == 0) { barcode_log[bc]["noradtag"] = 0; barcode_log[bc]["total"] = 0; barcode_log[bc]["low_qual"] = 0; barcode_log[bc]["retained"] = 0; } barcode_log[bc]["total"] += paired ? 2 : 1; // // If we have a perfectly matching barcode, set the barcode and length in the right places. // if (pe_bc.size() > 0 && valid_se_bc == true && valid_pe_bc == true) { if (fhs.count(bc) > 0) { if (paired) { strcpy(href_1->se_bc, bc_1); strcpy(href_2->pe_bc, bc_2); } else { strcpy(href_1->se_bc, bc_1); strcpy(href_1->pe_bc, bc_2); } if (barcode_type == inline_index || barcode_type == inline_inline) href_1->inline_bc_len = strlen(bc_1); if (barcode_type == index_inline || barcode_type == inline_inline) href_2->inline_bc_len = strlen(bc_2); return 0; } } else if (valid_se_bc == true) { strcpy(href_1->se_bc, bc_1); if (barcode_type == inline_null || barcode_type == inline_index || barcode_type == inline_inline) href_1->inline_bc_len = strlen(bc_1); } else if (valid_pe_bc == true) { if (paired) strcpy(href_2->pe_bc, bc_2); else strcpy(href_1->pe_bc, bc_2); if (barcode_type == index_inline || barcode_type == inline_inline) href_2->inline_bc_len = strlen(bc_2); } // // Try to correct the barcode. // BarcodePair old_barcode = bc; bool se_correct = false; bool pe_correct = false; if (paired) { if (se_bc.count(bc.se) == 0) se_correct = correct_barcode(se_bc, href_1, single_end, barcode_dist_1); if (pe_bc.size() > 0 && pe_bc.count(bc.pe) == 0) pe_correct = correct_barcode(pe_bc, href_2, paired_end, barcode_dist_2); if (se_correct) bc.se = string(href_1->se_bc); if (pe_bc.size() > 0 && pe_correct) bc.pe = string(href_2->pe_bc); // // After correcting the individual barcodes, check if the combination is valid. // if (fhs.count(bc) == 0) { counter["ambiguous"] += 2; href_1->retain = 0; href_2->retain = 0; } } else { if (se_bc.count(bc.se) == 0) se_correct = correct_barcode(se_bc, href_1, single_end, barcode_dist_1); if (pe_bc.size() > 0 && pe_bc.count(bc.pe) == 0) pe_correct = correct_barcode(pe_bc, href_1, paired_end, barcode_dist_2); if (se_correct) bc.se = string(href_1->se_bc); if (pe_bc.size() > 0 && pe_correct) bc.pe = string(href_1->pe_bc); if (fhs.count(bc) == 0) { counter["ambiguous"]++; href_1->retain = 0; } } if (href_1->retain && (se_correct || pe_correct)) { counter["recovered"] += paired ? 2 : 1; barcode_log[old_barcode]["total"] -= paired ? 2 : 1; if (barcode_log.count(bc) == 0) { barcode_log[bc]["total"] = 0; barcode_log[bc]["retained"] = 0; barcode_log[bc]["low_qual"] = 0; barcode_log[bc]["noradtag"] = 0; } barcode_log[bc]["total"] += paired ? 2 : 1; } return 0; } #endif // __CLEAN_H__ stacks-1.44/src/clone_filter.cc000644 000765 000024 00000122230 12764047443 017260 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2011-2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // clone_filter -- find duplicate read pairs and reduce them to one representative // pair of sequences in the data set. These reads are assumed to be the product of // PCR amplification. // #include "clone_filter.h" // // Global variables to hold command-line options. // FileT in_file_type = FileT::unknown; FileT out_file_type = FileT::unknown; string in_file; string in_file_p1; string in_file_p2; string in_path_1; string in_path_2; string out_path; bool discards = false; bool interleaved = false; bool merge = false; bool paired = false; bool retain_oligo = false; barcodet barcode_type = null_null; int oligo_len_1 = 0; int oligo_len_2 = 0; // // These variables are required for other linked objects, but we won't use them in clone_filter. // int barcode_size; uint truncate_seq; bool ill_barcode; bool recover; uint min_bc_size_1; uint max_bc_size_1; uint min_bc_size_2; uint max_bc_size_2; double win_size; int main (int argc, char* argv[]) { parse_command_line(argc, argv); if (oligo_len_2 == 0) oligo_len_2 = oligo_len_1; min_bc_size_1 = oligo_len_1; max_bc_size_1 = oligo_len_1; min_bc_size_2 = oligo_len_2; max_bc_size_2 = oligo_len_2; // // If input files are gzipped, output gziped files, unless the user chooses an output type. // if (out_file_type == FileT::unknown) { if (in_file_type == FileT::gzfastq || in_file_type == FileT::bam) out_file_type = FileT::gzfastq; else out_file_type = FileT::fastq; } if (paired) cerr << "Processing paired-end data.\n"; else cerr << "Processing single-end data.\n"; switch(barcode_type) { case null_null: cerr << "No oligo sequence specified, will use single and paired-end reads to determine clones.\n"; break; case null_index: cerr << "Searching for index oligo (i7 Illumina read).\n"; break; case index_null: cerr << "Searching for index oligo (i5 Illumina read).\n"; break; case inline_null: cerr << "Searching for inline oligo on single-end read.\n"; break; case index_index: cerr << "Searching for index oligos (i5 and i7 Illumina reads).\n"; break; case inline_inline: cerr << "Searching for inline oligos on single and paired-end read.\n"; break; case inline_index: cerr << "Searching for inline oligo on single-end read and index oligo (i5 or i7 Illumina read).\n"; break; case index_inline: if (paired) cerr << "Searching for inline oligo on paired-end read and index oligo (i5 or i7 Illumina read).\n"; else cerr << "Searching for inline oligo on single-end read and index oligo (i5 or i7 Illumina read).\n"; break; } map counters; counters["total"] = 0; counters["red_reads"] = 0; counters["dis_reads"] = 0; vector > files; build_file_list(files); CloneHash clone_map; OligoHash oligo_map; map clone_dist; vector clone_map_keys; int result = 1; for (uint i = 0; i < files.size(); i++) { cerr << "Processing file " << i+1 << " of " << files.size() << " [" << files[i].first.c_str() << "]\n"; result = 1; if (paired) { if (barcode_type == null_null) result = process_paired_reads_by_sequence(files[i].first, files[i].second, counters, clone_map, clone_map_keys); else result = process_paired_reads(files[i].first, files[i].second, counters, oligo_map); } else { result = process_reads(files[i].first, counters, oligo_map); } if (!result) { cerr << "Error processing reads.\n"; break; } } if (barcode_type == null_null && result) { write_clonereduced_sequence(files[0].first, files[0].second, clone_map, clone_dist, counters); } else { for (OligoHash::iterator i = oligo_map.begin(); i != oligo_map.end(); i++) for (map::iterator j = i->second.begin(); j != i->second.end(); j++) clone_dist[j->second]++; } if (clone_map_keys.size() > 0) { cerr << "Freeing hash key memory..."; free_hash(clone_map_keys); cerr << "done.\n"; } // // Determine and print the distribution of read clones. // cerr << "Calculating the distribution of cloned read pairs...\n"; vector bins; map::iterator it; for (it = clone_dist.begin(); it != clone_dist.end(); it++) bins.push_back(it->first); sort(bins.begin(), bins.end()); cout << "Num Clones\tCount\n"; for (uint i = 0; i < bins.size(); i++) cout << bins[i] << "\t" << clone_dist[bins[i]] << "\n"; char buf[32]; sprintf(buf, "%0.2f%%", ((double) (counters["total"] - counters["red_reads"]) / (double) counters["total"]) * 100); cerr << counters["total"] << " pairs of reads input. " << counters["red_reads"] << " pairs of reads output, discarded " << counters["dis_reads"] << " pairs of reads, " << buf << " clone reads.\n"; return 0; } int process_paired_reads_by_sequence(string prefix_1, string prefix_2, map &counters, CloneHash &clone_map, vector &clone_map_keys) { Input *fh_1, *fh_2; int return_val = 1; string path_1 = in_path_1 + prefix_1; string path_2 = in_path_2 + prefix_2; cerr << "Reading data from:\n " << path_1 << " and\n " << path_2 << "\n"; switch (in_file_type) { case FileT::fastq: fh_1 = new Fastq(path_1); fh_2 = interleaved ? fh_1 : new Fastq(path_2); break; case FileT::gzfastq: fh_1 = new GzFastq(path_1); fh_2 = interleaved ? fh_1 : new GzFastq(path_2); break; case FileT::fasta: fh_1 = new Fasta(path_1); fh_2 = interleaved ? fh_1 : new Fasta(path_2); break; case FileT::gzfasta: fh_1 = new GzFasta(path_1); fh_2 = interleaved ? fh_1 : new GzFasta(path_2); break; case FileT::bam: fh_1 = new BamUnAln(path_1); fh_2 = fh_1; break; case FileT::bustard: fh_1 = new Bustard(path_1); fh_2 = interleaved ? fh_1 : new Bustard(path_2); default: fh_1 = NULL; fh_2 = NULL; break; } // // Read in the first records, initializing the Seq objects, then loop, using the same objects. // Seq *s_1 = fh_1->next_seq(); Seq *s_2 = fh_2->next_seq(); if (s_1 == NULL || s_2 == NULL) { cerr << "Unable to allocate Seq object.\n"; return 0; } long i = 1; bool exists; char *hash_key; uint seq_len = strlen(s_1->seq); do { if (i % 10000 == 0) cerr << "Processing short read " << i << " \r"; counters["total"]++; exists = clone_map.count(s_1->seq) == 0 ? false : true; if (exists) { hash_key = s_1->seq; } else { hash_key = new char [seq_len + 1]; strcpy(hash_key, s_1->seq); clone_map_keys.push_back(hash_key); } if (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) clone_map[hash_key][s_2->seq].push_back(Pair(s_1->id, s_2->id, s_1->qual, s_2->qual)); else if (out_file_type == FileT::fasta || out_file_type == FileT::gzfasta) clone_map[hash_key][s_2->seq].push_back(Pair(s_1->id, s_2->id)); delete s_1; delete s_2; i++; } while ((s_1 = fh_1->next_seq()) != NULL && (s_2 = fh_2->next_seq()) != NULL); cerr << "\n"; delete fh_1; delete fh_2; return return_val; } int write_clonereduced_sequence(string prefix_1, string prefix_2, CloneHash &clone_map, map &clone_dist, map &counters) { ofstream out_fh_1, out_fh_2, discard_fh_1, discard_fh_2; gzFile out_gzfh_1, out_gzfh_2, discard_gzfh_1, discard_gzfh_2; int return_val = 1; // // Open the input files. // string path_1; string path_2; // // Open the output files. // string suffix_1, suffix_2; if (out_file_type == FileT::gzfastq) { suffix_1 = ".1.fq.gz"; suffix_2 = ".2.fq.gz"; } else if (out_file_type == FileT::fastq) { suffix_1 = ".1.fq"; suffix_2 = ".2.fq"; } else if (out_file_type == FileT::gzfasta) { suffix_1 = ".1.fa.gz"; suffix_2 = ".2.fa.gz"; } else if (out_file_type == FileT::fasta) { suffix_1 = ".1.fa"; suffix_2 = ".2.fa"; } string file_1 = remove_suffix(in_file_type, prefix_1); path_1 = out_path + file_1 + suffix_1; if (out_file_type == FileT::gzfastq || out_file_type == FileT::gzfasta) { out_gzfh_1 = gzopen(path_1.c_str(), "wb"); if (!(out_gzfh_1)) { cerr << "Error opening output file '" << path_1 << "'\n"; return -1; } } else { out_fh_1.open(path_1.c_str(), ifstream::out); if (out_fh_1.fail()) { cerr << "Error opening output file '" << path_1 << "'\n"; return -1; } } string file_2 = remove_suffix(in_file_type, prefix_2); path_2 = out_path + file_2 + suffix_2; if (out_file_type == FileT::gzfastq || out_file_type == FileT::gzfasta) { out_gzfh_2 = gzopen(path_2.c_str(), "wb"); if (!(out_gzfh_2)) { cerr << "Error opening output file '" << path_2 << "'\n"; return -1; } } else { out_fh_2.open(path_2.c_str(), ifstream::out); if (out_fh_2.fail()) { cerr << "Error opening output file '" << path_2 << "'\n"; return -1; } } // // Open files for recording discarded reads. // if (discards) { path_1 = out_path + file_1 + ".discards" + suffix_1; if (out_file_type == FileT::gzfastq || out_file_type == FileT::gzfasta) { discard_gzfh_1 = gzopen(path_1.c_str(), "wb"); if (!(discard_gzfh_1)) { cerr << "Error opening output file '" << path_1 << "'\n"; return -1; } } else { discard_fh_1.open(path_1.c_str(), ifstream::out); if (discard_fh_1.fail()) { cerr << "Error opening discard output file '" << path_1 << "'\n"; return -1; } } path_2 = out_path + file_2 + ".discards" + suffix_2; if (out_file_type == FileT::gzfastq || out_file_type == FileT::gzfasta) { discard_gzfh_2 = gzopen(path_2.c_str(), "wb"); if (!(discard_gzfh_2)) { cerr << "Error opening output file '" << path_2 << "'\n"; return -1; } } else { discard_fh_2.open(path_2.c_str(), ifstream::out); if (discard_fh_2.fail()) { cerr << "Error opening discard output file '" << path_2 << "'\n"; return -1; } } } CloneHash::iterator hash_it; map >::iterator map_it; stringstream sstr_1, sstr_2; cerr << "Writing filtered data...\n"; for (hash_it = clone_map.begin(); hash_it != clone_map.end(); hash_it++) { for (map_it = hash_it->second.begin(); map_it != hash_it->second.end(); map_it++) { sstr_1.str(""); sstr_2.str(""); if (out_file_type == FileT::gzfastq || out_file_type == FileT::fastq) { sstr_1 << "@" << map_it->second[0].p1_id << "\n" << hash_it->first << "\n" << "+\n" << map_it->second[0].p1_qual << "\n"; sstr_2 << "@" << map_it->second[0].p2_id << "\n" << map_it->first << "\n" << "+\n" << map_it->second[0].p2_qual << "\n"; } else { sstr_1 << ">" << map_it->second[0].p1_id << "\n" << hash_it->first << "\n"; sstr_2 << ">" << map_it->second[0].p2_id << "\n" << map_it->first << "\n"; } switch(out_file_type) { case FileT::gzfastq: case FileT::gzfasta: gzputs(out_gzfh_1, sstr_1.str().c_str()); gzputs(out_gzfh_2, sstr_2.str().c_str()); break; case FileT::fastq: case FileT::fasta: default: out_fh_1 << sstr_1.str(); out_fh_2 << sstr_2.str(); } counters["dis_reads"] += map_it->second.size() - 1; clone_dist[map_it->second.size()]++; // // Write cloned read pairs that we are discarding // if (discards) { for (uint i = 1; i < map_it->second.size(); i++) { sstr_1.str(""); sstr_2.str(""); if (out_file_type == FileT::gzfastq || out_file_type == FileT::fastq) { sstr_1 << "@" << map_it->second[i].p1_id << "\n" << hash_it->first << "\n" << "+\n" << map_it->second[i].p1_qual << "\n"; sstr_2 << "@" << map_it->second[i].p2_id << "\n" << map_it->first << "\n" << "+\n" << map_it->second[i].p2_qual << "\n"; } else { sstr_1 << ">" << map_it->second[i].p1_id << "\n" << hash_it->first << "\n"; sstr_2 << ">" << map_it->second[i].p2_id << "\n" << map_it->first << "\n"; } switch(out_file_type) { case FileT::gzfastq: case FileT::gzfasta: gzputs(discard_gzfh_1, sstr_1.str().c_str()); gzputs(discard_gzfh_2, sstr_2.str().c_str()); break; case FileT::fastq: case FileT::fasta: default: discard_fh_1 << sstr_1.str(); discard_fh_2 << sstr_2.str(); } } } counters["red_reads"]++; } } cerr << "done.\n"; if (out_file_type == FileT::gzfastq || out_file_type == FileT::gzfasta) { gzclose(out_gzfh_1); gzclose(out_gzfh_2); if (discards) { gzclose(discard_gzfh_1); gzclose(discard_gzfh_2); } } else { out_fh_1.close(); out_fh_2.close(); if (discards) { discard_fh_1.close(); discard_fh_2.close(); } } return return_val; } int process_paired_reads(string prefix_1, string prefix_2, map &counters, OligoHash &oligo_map) { Input *fh_1, *fh_2; Read *r_1, *r_2; ofstream out_fh_1, out_fh_2, discard_fh_1, discard_fh_2; gzFile out_gzfh_1, out_gzfh_2, discard_gzfh_1, discard_gzfh_2; int return_val = 1; // // Open the input files. // string path_1 = in_path_1 + prefix_1; string path_2 = in_path_2 + prefix_2; if (interleaved) cerr << " Reading data from:\n " << path_1 << "\n"; else cerr << " Reading data from:\n " << path_1 << " and\n " << path_2 << "\n"; switch (in_file_type) { case FileT::fastq: fh_1 = new Fastq(path_1); fh_2 = interleaved ? fh_1 : new Fastq(path_2); break; case FileT::gzfastq: fh_1 = new GzFastq(path_1); fh_2 = interleaved ? fh_1 : new GzFastq(path_2); break; case FileT::fasta: fh_1 = new Fasta(path_1); fh_2 = interleaved ? fh_1 : new Fasta(path_2); break; case FileT::gzfasta: fh_1 = new GzFasta(path_1); fh_2 = interleaved ? fh_1 : new GzFasta(path_2); break; case FileT::bam: fh_1 = new BamUnAln(path_1); fh_2 = fh_1; break; case FileT::bustard: fh_1 = new Bustard(path_1); fh_2 = interleaved ? fh_1 : new Bustard(path_2); default: fh_1 = NULL; fh_2 = NULL; break; } // // Open the output files. // string suffix_1, suffix_2; if (out_file_type == FileT::gzfastq) { suffix_1 = ".1.fq.gz"; suffix_2 = ".2.fq.gz"; } else if (out_file_type == FileT::fastq) { suffix_1 = ".1.fq"; suffix_2 = ".2.fq"; } else if (out_file_type == FileT::gzfasta) { suffix_1 = ".1.fa.gz"; suffix_2 = ".2.fa.gz"; } else if (out_file_type == FileT::fasta) { suffix_1 = ".1.fa"; suffix_2 = ".2.fa"; } string file_1 = remove_suffix(in_file_type, prefix_1); path_1 = out_path + file_1 + suffix_1; if (in_file_type == FileT::gzfastq || in_file_type == FileT::gzfasta) { out_gzfh_1 = gzopen(path_1.c_str(), "wb"); if (!(out_gzfh_1)) { cerr << "Error opening output file '" << path_1 << "'\n"; return -1; } } else { out_fh_1.open(path_1.c_str(), ifstream::out); if (out_fh_1.fail()) { cerr << "Error opening output file '" << path_1 << "'\n"; return -1; } } string file_2 = remove_suffix(in_file_type, prefix_2); path_2 = out_path + file_2 + suffix_2; if (in_file_type == FileT::gzfastq || in_file_type == FileT::gzfasta) { out_gzfh_2 = gzopen(path_2.c_str(), "wb"); if (!(out_gzfh_2)) { cerr << "Error opening output file '" << path_2 << "'\n"; return -1; } } else { out_fh_2.open(path_2.c_str(), ifstream::out); if (out_fh_2.fail()) { cerr << "Error opening output file '" << path_2 << "'\n"; return -1; } } // // Open files for recording discarded reads. // if (discards) { path_1 = out_path + file_1 + ".discards" + suffix_1; if (in_file_type == FileT::gzfastq || in_file_type == FileT::gzfasta) { discard_gzfh_1 = gzopen(path_1.c_str(), "wb"); if (!(discard_gzfh_1)) { cerr << "Error opening discard file '" << path_1 << "'\n"; return -1; } } else { discard_fh_1.open(path_1.c_str(), ifstream::out); if (discard_fh_1.fail()) { cerr << "Error opening discard file '" << path_1 << "'\n"; return -1; } } path_2 = out_path + file_2 + ".discards" + suffix_2; if (in_file_type == FileT::gzfastq || in_file_type == FileT::gzfasta) { discard_gzfh_2 = gzopen(path_2.c_str(), "wb"); if (!(discard_gzfh_2)) { cerr << "Error opening discard file '" << path_2 << "'\n"; return -1; } } else { discard_fh_2.open(path_2.c_str(), ifstream::out); if (discard_fh_2.fail()) { cerr << "Error opening discard file '" << path_2 << "'\n"; return -1; } } } // // Determine how much sequence we need to trim to remove the oligo seqeunce before printing. // int offset_1, offset_2; switch (barcode_type) { case inline_null: case inline_index: offset_1 = oligo_len_1; offset_2 = 0; break; case index_null: case null_index: case index_index: offset_1 = 0; offset_2 = 0; break; case inline_inline: offset_1 = oligo_len_1; offset_2 = oligo_len_2; break; case index_inline: offset_1 = 0; offset_2 = oligo_len_2; default: break; } // // Read in the first record, initializing the Seq object s. Then // initialize the Read object r, then loop, using the same objects. // Seq *s_1 = fh_1->next_seq(); Seq *s_2 = fh_2->next_seq(); if (s_1 == NULL || s_2 == NULL) { cerr << "Attempting to read first pair of input records, unable to allocate " << "Seq object (Was the correct input type specified?).\n"; exit(1); } r_1 = new Read(strlen(s_1->seq), 1, min_bc_size_1, win_size); r_2 = new Read(strlen(s_2->seq), 2, min_bc_size_2, win_size); long i = 1; int result_1 = 1; int result_2 = 1; bool clone = false; string oligo_1, oligo_2, key, oligo; do { if (i % 10000 == 0) cerr << " Processing RAD-Tag " << i << " \r"; parse_input_record(s_1, r_1); parse_input_record(s_2, r_2); counters["total"]++; result_1 = 1; result_2 = 1; clone = false; // // Fetch the randomized oligo sequence from the proper position in the reads. // switch (barcode_type) { case inline_null: oligo_1 = r_1->inline_bc; break; case index_null: oligo_1 = r_1->index_bc; break; case null_index: oligo_1 = r_2->index_bc; break; case inline_inline: oligo_1 = r_1->inline_bc; oligo_2 = r_2->inline_bc; break; case index_index: oligo_1 = r_1->index_bc; oligo_2 = r_2->index_bc; break; case inline_index: oligo_1 = r_1->inline_bc; oligo_2 = r_2->index_bc; break; case index_inline: oligo_1 = r_1->index_bc; oligo_2 = r_2->inline_bc; default: break; } // // Have we seen this combination of oligos before for this read? // oligo = oligo_1 + oligo_2; key = string(s_1->seq + offset_1) + string(s_2->seq + offset_2); // cerr << "Oligo: '" << oligo << "'\n" // << "Seq: '" << s_1->seq << "'\n" // << "Key: '" << key << "'\n"; if (oligo_map.count(key) == 0) oligo_map[key] = map(); if (oligo_map[key].count(oligo) == 0) { oligo_map[key][oligo] = 1; clone = false; } else { oligo_map[key][oligo]++; clone = true; } if (clone == false) { counters["red_reads"]++; switch (out_file_type) { case FileT::fastq: result_1 = write_fastq(&out_fh_1, s_1, retain_oligo ? 0 : offset_1); result_2 = write_fastq(&out_fh_2, s_2, retain_oligo ? 0 : offset_2); break; case FileT::gzfastq: result_1 = write_fastq(&out_gzfh_1, s_1, retain_oligo ? 0 : offset_1); result_2 = write_fastq(&out_gzfh_2, s_2, retain_oligo ? 0 : offset_2); break; case FileT::fasta: result_1 = write_fasta(&out_fh_1, s_1, retain_oligo ? 0 : offset_1); result_2 = write_fasta(&out_fh_2, s_2, retain_oligo ? 0 : offset_2); break; case FileT::gzfasta: result_1 = write_fasta(&out_gzfh_1, s_1, retain_oligo ? 0 : offset_1); result_2 = write_fasta(&out_gzfh_2, s_2, retain_oligo ? 0 : offset_2); default: break; } if (!result_1 || !result_2) { cerr << "Error writing to output file for '" << file_1 << " / " << file_2 << "'\n"; return_val = -1; break; } } else if (clone == true && discards) { counters["dis_reads"]++; switch (out_file_type) { case FileT::fastq: result_1 = write_fastq(&discard_fh_1, s_1); result_2 = write_fastq(&discard_fh_2, s_2); break; case FileT::gzfastq: result_1 = write_fastq(&discard_gzfh_1, s_1); result_2 = write_fastq(&discard_gzfh_2, s_2); break; case FileT::fasta: result_1 = write_fasta(&discard_fh_1, s_1); result_2 = write_fasta(&discard_fh_2, s_2); break; case FileT::gzfasta: result_1 = write_fasta(&discard_gzfh_1, s_1); result_2 = write_fasta(&discard_gzfh_2, s_2); default: break; } if (!result_1 || !result_2) { cerr << "Error writing to discard file for '" << file_1 << " / " << file_2 << "'\n"; return_val = -1; break; } } delete s_1; delete s_2; i++; } while ((s_1 = fh_1->next_seq()) != NULL && (s_2 = fh_2->next_seq()) != NULL); if (out_file_type == FileT::gzfastq || out_file_type == FileT::gzfasta) { gzclose(out_gzfh_1); gzclose(out_gzfh_2); if (discards) { gzclose(discard_gzfh_1); gzclose(discard_gzfh_2); } } else { out_fh_1.close(); out_fh_2.close(); if (discards) { discard_fh_1.close(); discard_fh_2.close(); } } delete fh_1; if (interleaved == false) delete fh_2; delete r_1; delete r_2; return return_val; } int process_reads(string prefix_1, map &counters, OligoHash &oligo_map) { Input *fh_1; Read *r_1; ofstream out_fh_1, discard_fh_1; gzFile out_gzfh_1, discard_gzfh_1; int return_val = 1; // // Open the input file. // string path_1 = in_path_1 + prefix_1; cerr << " Reading data from:\n " << path_1 << "\n"; switch(in_file_type) { case FileT::fastq: fh_1 = new Fastq(path_1); break; case FileT::gzfastq: fh_1 = new GzFastq(path_1.c_str()); break; case FileT::fasta: fh_1 = new Fasta(path_1); break; case FileT::gzfasta: fh_1 = new GzFasta(path_1); break; case FileT::bam: fh_1 = new BamUnAln(path_1); break; case FileT::bustard: fh_1 = new Bustard(path_1); default: break; } // // Open the output files. // string suffix_1; if (out_file_type == FileT::gzfastq) suffix_1 = ".fq.gz"; else if (out_file_type == FileT::fastq) suffix_1 = ".fq"; else if (out_file_type == FileT::gzfasta) suffix_1 = ".fa.gz"; else if (out_file_type == FileT::fasta) suffix_1 = ".fa"; string file_1 = prefix_1; int pos = file_1.find_last_of("."); if ((in_file_type == FileT::gzfastq || in_file_type == FileT::gzfasta) && file_1.substr(pos) == ".gz") { file_1 = file_1.substr(0, pos); pos = file_1.find_last_of("."); } path_1 = out_path + file_1.substr(0, pos) + suffix_1; if (in_file_type == FileT::gzfastq || in_file_type == FileT::gzfasta) { out_gzfh_1 = gzopen(path_1.c_str(), "wb"); if (!(out_gzfh_1)) { cerr << "Error opening output file '" << path_1 << "'\n"; return -1; } } else { out_fh_1.open(path_1.c_str(), ifstream::out); if (out_fh_1.fail()) { cerr << "Error opening output file '" << path_1 << "'\n"; return -1; } } // // Open files for recording discarded reads. // if (discards) { path_1 = out_path + file_1 + ".discards" + suffix_1; if (in_file_type == FileT::gzfastq || in_file_type == FileT::gzfasta) { discard_gzfh_1 = gzopen(path_1.c_str(), "wb"); if (!(discard_gzfh_1)) { cerr << "Error opening discard file '" << path_1 << "'\n"; return -1; } } else { discard_fh_1.open(path_1.c_str(), ifstream::out); if (discard_fh_1.fail()) { cerr << "Error opening discard file '" << path_1 << "'\n"; return -1; } } } // // Determine how much sequence we need to trim to remove the oligo seqeunce before printing. // int offset_1; switch (barcode_type) { case inline_null: case inline_index: case index_inline: offset_1 = oligo_len_1; break; default: offset_1 = 0; break; } // // Read in the first record, initializing the Seq object s. Then // initialize the Read object r, then loop, using the same objects. // Seq *s_1 = fh_1->next_seq(); if (s_1 == NULL) { cerr << "Attempting to read first pair of input records, unable to allocate " << "Seq object (Was the correct input type specified?).\n"; exit(1); } r_1 = new Read(strlen(s_1->seq), 1, min_bc_size_1, win_size); long i = 1; int result_1 = 1; bool clone = false; string key, oligo_1; do { if (i % 10000 == 0) cerr << " Processing RAD-Tag " << i << " \r"; parse_input_record(s_1, r_1); counters["total"]++; result_1 = 1; clone = false; // // Fetch the randomized oligo sequence from the proper position in the reads. // if (barcode_type == inline_null) oligo_1 = r_1->inline_bc; else if (barcode_type == index_null) oligo_1 = r_1->index_bc; // // Have we seen this combination of oligos before for this read? // key = string(s_1->seq + offset_1); if (oligo_map.count(key) == 0) oligo_map[key] = map(); if (oligo_map[key].count(oligo_1) == 0) { oligo_map[key][oligo_1] = 1; clone = false; } else { oligo_map[key][oligo_1]++; clone = true; } if (clone == false) { counters["red_reads"]++; switch (out_file_type) { case FileT::fastq: result_1 = write_fastq(&out_fh_1, s_1, retain_oligo ? 0 : offset_1); break; case FileT::gzfastq: result_1 = write_fastq(&out_gzfh_1, s_1, retain_oligo ? 0 : offset_1); break; case FileT::fasta: result_1 = write_fasta(&out_fh_1, s_1, retain_oligo ? 0 : offset_1); break; case FileT::gzfasta: result_1 = write_fasta(&out_gzfh_1, s_1, retain_oligo ? 0 : offset_1); default: break; } if (!result_1) { cerr << "Error writing to output file for '" << file_1 << "'\n"; return_val = -1; break; } } else if (clone == true && discards) { counters["dis_reads"]++; switch (out_file_type) { case FileT::fastq: result_1 = write_fastq(&discard_fh_1, s_1); break; case FileT::gzfastq: result_1 = write_fastq(&discard_gzfh_1, s_1); break; case FileT::fasta: result_1 = write_fasta(&discard_fh_1, s_1); break; case FileT::gzfasta: result_1 = write_fasta(&discard_gzfh_1, s_1); default: break; } if (!result_1) { cerr << "Error writing to discard file for '" << file_1 << "'\n"; return_val = -1; break; } } delete s_1; i++; } while ((s_1 = fh_1->next_seq()) != NULL); if (out_file_type == FileT::gzfastq || out_file_type == FileT::gzfasta) { gzclose(out_gzfh_1); if (discards) gzclose(discard_gzfh_1); } else { out_fh_1.close(); if (discards) discard_fh_1.close(); } delete fh_1; delete r_1; return return_val; } int free_hash(vector &keys) { for (uint i = 0; i < keys.size(); i++) { delete [] keys[i]; } keys.clear(); return 0; } int parse_command_line(int argc, char* argv[]) { FileT ftype; int c; while (1) { static struct option long_options[] = { {"help", no_argument, NULL, 'h'}, {"version", no_argument, NULL, 'v'}, {"discards", no_argument, NULL, 'D'}, {"paired", no_argument, NULL, 'P'}, {"null_index", no_argument, NULL, 'U'}, {"index_null", no_argument, NULL, 'u'}, {"inline_null", no_argument, NULL, 'V'}, {"index_index", no_argument, NULL, 'W'}, {"inline_inline", no_argument, NULL, 'x'}, {"index_inline", no_argument, NULL, 'Y'}, {"inline_index", no_argument, NULL, 'Z'}, {"infile_type", required_argument, NULL, 'i'}, {"outfile_type", required_argument, NULL, 'y'}, {"file", required_argument, NULL, 'f'}, {"path", required_argument, NULL, 'p'}, {"file_p1", required_argument, NULL, '1'}, {"file_p2", required_argument, NULL, '2'}, {"outpath", required_argument, NULL, 'o'}, {"oligo_len_1", required_argument, NULL, 'O'}, {"oligo_len_2", required_argument, NULL, 'L'}, {"retain_oligo", required_argument, NULL, 'R'}, {0, 0, 0, 0} }; // getopt_long stores the option index here. int option_index = 0; c = getopt_long(argc, argv, "hvDPuUVWxYZi:y:f:p:1:2:o:O:L:R:", long_options, &option_index); // Detect the end of the options. if (c == -1) break; switch (c) { case 'h': help(); break; case 'i': if (strcasecmp(optarg, "bustard") == 0) in_file_type = FileT::bustard; else if (strcasecmp(optarg, "fasta") == 0) in_file_type = FileT::fasta; else if (strcasecmp(optarg, "gzfasta") == 0) in_file_type = FileT::gzfasta; else if (strcasecmp(optarg, "gzfastq") == 0) in_file_type = FileT::gzfastq; else in_file_type = FileT::fastq; break; case 'y': if (strcasecmp(optarg, "fastq") == 0) out_file_type = FileT::fastq; else if (strcasecmp(optarg, "gzfastq") == 0) out_file_type = FileT::gzfastq; else if (strcasecmp(optarg, "fasta") == 0) out_file_type = FileT::fasta; else if (strcasecmp(optarg, "gzfasta") == 0) out_file_type = FileT::gzfasta; break; case 'D': discards = true; break; case 'f': in_file = optarg; ftype = FileT::fastq; break; case 'p': in_path_1 = optarg; in_path_2 = in_path_1; ftype = FileT::fastq; break; case '1': paired = true; in_file_p1 = optarg; ftype = FileT::fastq; break; case '2': paired = true; in_file_p2 = optarg; ftype = FileT::fastq; break; case 'P': paired = true; break; case 'o': out_path = optarg; break; case 'U': barcode_type = null_index; break; case 'u': barcode_type = index_null; break; case 'V': barcode_type = inline_null; break; case 'W': barcode_type = index_index; break; case 'x': barcode_type = inline_inline; break; case 'Y': barcode_type = index_inline; break; case 'Z': barcode_type = inline_index; break; case 'O': oligo_len_1 = is_integer(optarg); break; case 'L': oligo_len_2 = is_integer(optarg); break; case 'R': retain_oligo = true; break; case 'v': version(); break; case '?': // getopt_long already printed an error message. help(); break; default: cerr << "Unknown command line option '" << (char) c << "'\n"; help(); abort(); } } if (in_file.length() == 0 && in_path_1.length() == 0 && in_file_p1.length() == 0) { cerr << "You must specify an input file of a directory path to a set of input files.\n"; help(); } if (in_file.length() > 0 && in_path_1.length() > 0) { cerr << "You must specify either a single input file (-f) or a directory path (-p), not both.\n"; help(); } if (in_file.length() > 0 && (in_file_p1.length() > 0 || in_file_p2.length() > 0)) { cerr << "You must specify either a single input file (-f) or a set of paired files (-1, -2), not both.\n"; help(); } if (in_path_1.length() > 0 && (in_file_p1.length() > 0 || in_file_p2.length() > 0)) { cerr << "You must specify either a file path (-p) or a set of paired files (-1, -2), not both.\n"; help(); } if (in_path_1.length() > 0 && in_path_1.at(in_path_1.length() - 1) != '/') in_path_1 += "/"; if (in_path_2.length() > 0 && in_path_2.at(in_path_2.length() - 1) != '/') in_path_2 += "/"; if (out_path.length() == 0) out_path = "."; if (out_path.at(out_path.length() - 1) != '/') out_path += "/"; if (in_file_type == FileT::unknown) in_file_type = ftype; if (paired == false && barcode_type == null_null) { cerr << "You must specify paired-end data if you do not have oligo sequences to differentiate cloned reads.\n"; help(); } if (barcode_type != null_null && oligo_len_1 == 0 && oligo_len_2 == 0) { cerr << "You must specify the length of the oligo sequences (--oligo_len_1 / --oligo_len_2).\n"; help(); } return 0; } void version() { std::cerr << "clone_filter " << VERSION << "\n\n"; exit(0); } void help() { std::cerr << "clone_filter " << VERSION << "\n" << "clone_filter [-f in_file | -p in_dir [-P] [-I] | -1 pair_1 -2 pair_2] -o out_dir [-i type] [-y type] [-D] [-h]\n" << " f: path to the input file if processing single-end sequences.\n" << " p: path to a directory of files.\n" << " P: files contained within directory specified by '-p' are paired.\n" << " 1: first input file in a set of paired-end sequences.\n" << " 2: second input file in a set of paired-end sequences.\n" << " i: input file type, either 'bustard', 'fastq', 'fasta', 'gzfasta', or 'gzfastq' (default 'fastq').\n" << " o: path to output the processed files.\n" << " y: output type, either 'fastq', 'fasta', 'gzfasta', or 'gzfastq' (default same as input type).\n" << " D: capture discarded reads to a file.\n" << " h: display this help messsage.\n" << " --oligo_len_1 len: length of the single-end oligo sequence in data set.\n" << " --oligo_len_2 len: length of the paired-end oligo sequence in data set.\n" << " --retain_oligo: do not trim off the random oligo sequence (if oligo is inline).\n\n" << " Oligo sequence options:\n" << " --inline_null: random oligo is inline with sequence, occurs only on single-end read (default).\n" << " --null_index: random oligo is provded in FASTQ header (Illumina i7 read if both i5 and i7 read are provided).\n" << " --index_null: random oligo is provded in FASTQ header (Illumina i5 or i7 read).\n" << " --inline_inline: random oligo is inline with sequence, occurs on single and paired-end read.\n" << " --index_index: random oligo is provded in FASTQ header (Illumina i5 and i7 read).\n" << " --inline_index: random oligo is inline with sequence on single-end read and second oligo occurs in FASTQ header.\n" << " --index_inline: random oligo occurs in FASTQ header (Illumina i5 or i7 read) and is inline with sequence on single-end read (if single read data) or paired-end read (if paired data).\n\n"; exit(0); } stacks-1.44/src/clone_filter.h000644 000765 000024 00000006252 12764047443 017127 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2011-2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __CLONE_FILTER_H__ #define __CLONE_FILTER_H__ #include "constants.h" #include #include // Process command-line options #include // Open/Read contents of a directory #include #include #include #include #include #include using std::stringstream; using std::istream; using std::ofstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include using std::vector; #include using std::map; #include using std::set; #include using std::pair; #include using std::unordered_map; #ifdef HAVE_SPARSEHASH #include using google::sparse_hash_map; #endif #include "kmers.h" #include "BustardI.h" // Reading input files in Tab-separated Bustard format #include "FastqI.h" // Reading input files in FASTQ format #include "FastaI.h" // Reading input files in FASTA format #include "gzFasta.h" // Reading gzipped input files in FASTA format #include "gzFastq.h" // Reading gzipped input files in FASTQ format #include "BamUnalignedI.h" #include "clean.h" #include "file_io.h" #include "write.h" class Pair { public: string p1_id; string p2_id; string p1_qual; string p2_qual; Pair(string p1_id, string p2_id, string p1_qual, string p2_qual) { this->p1_id = p1_id; this->p2_id = p2_id; this->p1_qual = p1_qual; this->p2_qual = p2_qual; } Pair(string p1_id, string p2_id) { this->p1_id = p1_id; this->p2_id = p2_id; } }; #ifdef HAVE_SPARSEHASH typedef sparse_hash_map >, hash_charptr, eqstr> CloneHash; typedef sparse_hash_map > OligoHash; #else typedef unordered_map >, hash_charptr, eqstr> CloneHash; typedef unordered_map > OligoHash; #endif int process_paired_reads(string, string, map &, OligoHash &); int process_reads(string, map &, OligoHash &); int process_paired_reads_by_sequence(string, string, map &, CloneHash &, vector &); int write_clonereduced_sequence(string, string, CloneHash &, map &, map &); int free_hash(vector &); void help( void ); void version( void ); int parse_command_line(int, char**); #endif // __CLONE_FILTER_H__ stacks-1.44/src/cmb.cc000644 000765 000024 00000026664 12764047443 015372 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // cmb.cc -- routines to implement the Combination generating class: CombSet. // // Julian Catchen // jcatchen@uoregon.edu // University of Oregon // // $Id: cmb.cc 1990 2010-11-03 04:49:22Z catchen $ // #include "cmb.h" // // A cache to store combinations generated as (N choose k) // for all set sizes encountered // map > _cmbs; CombSet::~CombSet() { int num_sets, set, i; Cmb **c; num_sets = this->compound_comb.size(); for (set = 0; set < num_sets; set++) { c = this->compound_comb[set]; this->destroy(c); } // // Delete edge map // for (i = 0; i < (int) this->node_list.size(); i++) delete [] this->edges[i]; delete [] this->edges; } CombSet::CombSet(int n, int k, MinSpanTree *tree) { this->max_set_size = n >= k ? k : n; this->num_elements = n - 1; this->index = 0; this->mst = tree; int set_size = this->max_set_size - 1; int size; int **comb; cerr << " Generating combinations for a set of " << n << " elements, with a maximum subset size of " << k << "\n"; // // Add the initial combination: the empty set // if (_cmbs.count(this->num_elements) == 0 && _cmbs[this->num_elements].count(0) == 0) { //cerr << " N: " << this->num_elements << "; K: 0; Total elements: 0\n"; comb = new int * [2]; comb[0] = new int[1]; comb[0][0] = -1; comb[1] = NULL; _cmbs[this->num_elements][0] = comb; } while (set_size > 0) { // // Check if this set of combinations is already cached. // if (_cmbs.count(this->num_elements) > 0 && _cmbs[this->num_elements].count(set_size) > 0) { set_size--; continue; } // // How many combinations will we make? // size = (int) this->num_combinations(this->num_elements, set_size); cerr << " N: " << this->num_elements << "; K: " << set_size << "; Total elements: " << size << "\n"; // // Generate all combinations, N choose K; N=num_elements, K=set_size // comb = this->generate_combinations(this->num_elements, set_size, size); // // Cache this set of combinations // _cmbs[this->num_elements][set_size] = comb; set_size--; } this->catalog_tree(); // // Finally, generate all combinations of nodes in the tree. // int max = this->max_set_size < this->num_elements ? this->max_set_size : this->num_elements + 1; for (set_size = 0; set_size < max; set_size++) this->partition_tree(set_size); cerr << " Total compound combinations for sets of size " << n << ": " << this->compound_comb.size() << "\n"; } int CombSet::catalog_tree() { set visited; queue q; uint i, n_1, n_2; // // Create a two-dimensional array to represent edges between nodes in the tree // uint cnt = this->mst->node_count(); //cerr << "Creating a two-dimensional array of size: " << cnt << " x " << cnt << "\n"; this->edges = new int * [cnt]; for (i = 0; i < cnt; i++) this->edges[i] = new int[cnt]; Node *n = this->mst->head(); q.push(n); cnt = 0; while (!q.empty()) { n = q.front(); q.pop(); visited.insert(n->id); this->node_list.push_back(n); this->node_map[n->id] = cnt; cnt++; for (i = 0; i < n->min_adj_list.size(); i++) if (visited.count(n->min_adj_list[i]->id) == false) q.push(n->min_adj_list[i]); } n = this->mst->head(); q.push(n); visited.clear(); while (!q.empty()) { n = q.front(); q.pop(); visited.insert(n->id); for (i = 0; i < n->min_adj_list.size(); i++) { if (visited.count(n->min_adj_list[i]->id) == false) { n_1 = this->node_map[n->id]; n_2 = this->node_map[n->min_adj_list[i]->id]; // Create a list of min spanning edges this->edge_list.push_back(make_pair(n->id, n->min_adj_list[i]->id)); // Mark the nodes as connected in our edge array this->edges[n_1][n_2] = 1; this->edges[n_2][n_1] = 1; // Queue this node to be visited next q.push(n->min_adj_list[i]); } } } return 0; } int CombSet::partition_tree(uint set_size) { uint i, j; set visited; queue q; Node *n; int n_1, n_2, node_cnt, cmb_cnt; Cmb **new_comb, *cmb; list nlist_work; int **comb = _cmbs[this->num_elements][set_size]; int *subgraph = new int[this->node_list.size()]; // // We want to methodically remove every set of branches of set_size size. The // subgraphs represent the combinations we want to generate. // for (i = 0; comb[i] != NULL; ++i) { // // This compound combination will consist of set_size+1 subgraphs // new_comb = new Cmb * [set_size + 2]; new_comb[set_size + 1] = NULL; // // Initialize working node list. // nlist_work = this->node_list; // // Remove edges // for (j = 0; j < set_size; j++) { n_1 = this->edge_list[comb[i][j]].first; n_2 = this->edge_list[comb[i][j]].second; this->edges[this->node_map[n_1]][this->node_map[n_2]] = 0; this->edges[this->node_map[n_2]][this->node_map[n_1]] = 0; } // // Traverse the subgraphs of the tree and record combinations. // visited.clear(); cmb_cnt = 0; while (nlist_work.size() > 0) { node_cnt = 0; n = nlist_work.front(); q.push(n); nlist_work.pop_front(); //subgraph[node_cnt] = n->id; while (!q.empty()) { n = q.front(); q.pop(); visited.insert(n->id); subgraph[node_cnt] = n->id; node_cnt++; nlist_work.remove(n); for (j = 0; j < n->min_adj_list.size(); j++) { n_1 = this->node_map[n->id]; n_2 = this->node_map[n->min_adj_list[j]->id]; if (visited.count(n->min_adj_list[j]->id) == false && edges[n_1][n_2] == 1) { q.push(n->min_adj_list[j]); } } } // // Package up this combination. // cmb = new Cmb; cmb->size = node_cnt; cmb->elem = new int[cmb->size]; for (j = 0; j < cmb->size; j++) cmb->elem[j] = subgraph[j]; new_comb[cmb_cnt] = cmb; cmb_cnt++; } this->compound_comb.push_back(new_comb); // // Reset the edges. // for (j = 0; j < set_size; j++) { n_1 = this->edge_list[comb[i][j]].first; n_2 = this->edge_list[comb[i][j]].second; this->edges[this->node_map[n_1]][this->node_map[n_2]] = 1; this->edges[this->node_map[n_2]][this->node_map[n_1]] = 1; } } delete [] subgraph; return 0; } int **CombSet::generate_combinations(int n, int k, int total) { int **comb; // // Generate an int pointer for each combination, terminate the list with // a NULL pointer. // comb = new int * [total + 1]; for (int i = 0; i < total; i++) comb[i] = new int[k]; comb[total] = NULL; // // Setup the initial combination // int comb_num = 0; for (int i = 0; i < k; i++) comb[comb_num][i] = i; comb_num++; // // Generate each successive combination // while (comb_num < total) { for (int i = 0; i < k; i++) comb[comb_num][i] = comb[comb_num - 1][i]; this->next_combination(comb[comb_num], n, k); comb_num++; } return comb; } int CombSet::next_combination(int *comb, int n, int k) { int i; // // The zero'th position has been incremented to its maximal value, // it's not possible to further increment values in the set. // if (comb[0] > n - k) return 0; // // Increment the last position in the set. // i = k - 1; comb[i]++; // // Check if the last position has reached its maximal possible value, // if so, move back one position, and increment it. // while ((i > 0) && (comb[i] >= n - k + 1 + i)) { i--; comb[i]++; } // // Move from the position we incremented above back out to the final position // for (i = i + 1; i < k; i++) comb[i] = comb[i - 1] + 1; return 1; } long int CombSet::num_combinations(int n, int k) { // // Compute the binomial coefficient using the method of: // Y. Manolopoulos, "Binomial coefficient computation: recursion or iteration?", // ACM SIGCSE Bulletin, 34(4):65-67, 2002. // long int r = 1; long int s = (k < n - k) ? n - k + 1 : k + 1; for (long int i = n; i >= s; i--) r = r * i / (n - i + 1); return r; } // // Return a variable length array of Cmb objects, terminated by a NULL pointer. // Cmb **CombSet::next(int map[]) { if (this->index >= (int) this->compound_comb.size()) return NULL; // int index, i, j, k, n; // int size = this->compound_comb[this->index]->size; // int *e = this->compound_comb[this->index]->elem; // Cmb **c = new Cmb * [size + 1]; // for (i = 0; i < size; i++) { // index = e[i]; // // sets vector index number // k = this->compound_set[index].first; // // combination number // n = this->compound_set[index].second; // c[i] = new Cmb; // c[i]->size = this->size[k]; // c[i]->elem = new int[this->size[k]]; // for (j = 0; j < this->size[k]; j++) // c[i]->elem[j] = (map == NULL) ? // this->sets[k][n][j] : // map[this->sets[k][n][j]]; // } // c[size] = NULL; Cmb **c = this->compound_comb[this->index]; this->index++; return c; } void CombSet::reset() { this->index = 0; } void CombSet::destroy(Cmb **cmb) { for (uint j = 0; cmb[j] != NULL; j++) { delete [] cmb[j]->elem; delete cmb[j]; } delete [] cmb; } void write_cmb(int *comb, int size) { stringstream s; string t; s << "{"; for (int i = 0; i < size; i++) s << comb[i] << ", "; t = s.str().substr(0, s.str().length() - 2); t += "}"; cerr << t << "\n"; } stacks-1.44/src/cmb.h000644 000765 000024 00000005321 12764047443 015217 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __CMB_H__ #define __CMB_H__ #include #include #include using std::stringstream; using std::cin; using std::cout; using std::cerr; #include using std::list; #include using std::vector; #include using std::string; #include using std::set; #include using std::pair; using std::make_pair; #include "constants.h" #include "mst.h" #include "utils.h" typedef unsigned int uint; void write_cmb(int *, int); typedef struct cmb { uint size; int *elem; } Cmb; class CombSet { // // Given these two variables, we will select N choose K combinations. // This combination will be stored in sets, and we will then decrement K by 1 // and continue to generate sets. // // Once we have generated all the combinations of a particular size, K, we // will partition the minimum spanning tree by dropping combinations of edges // from the graph. The selection of edges to drop is provided by the combinations // generated first. Finally, each set of disconnected subgraphs makes for one // possible combination. // int num_elements; // N elements from which we wish to produce combinations int max_set_size; // maximum set size, K, the largest subset we wish to select. map node_map; // Convert non-contiguous IDs from the MST into array indexes for this->edges list node_list; vector > edge_list; int **edges; int index; vector compound_comb; MinSpanTree *mst; int catalog_tree(); int partition_tree(uint); int **generate_combinations(int, int, int); int next_combination(int *, int, int); long int num_combinations(int, int); void destroy(Cmb **); public: CombSet(int, int, MinSpanTree *); ~CombSet(); Cmb **next(int map[] = NULL); void reset(); }; #endif // __CMB_H__ stacks-1.44/src/constants.h000644 000765 000024 00000002766 12764047443 016504 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __CONSTANTS_H__ #define __CONSTANTS_H__ // // Pull in the configuration variables from the configure script // #if HAVE_CONFIG_H #include "config.h" #endif typedef unsigned int uint; // // // const unsigned int fieldw = 4; // // Maximum line length for parsing input files. // const int max_len = 1024; // // Maximum length of idetifiers, such as sequence IDs and chromosome names. // const int id_len = 255; // // Size to use for internal buffer size for gzipped files being read with libz. // const int libz_buffer_size = 1048576; // // Supported file types // enum class FileT {unknown, sql, gzsql, fasta, gzfasta, fastq, gzfastq, bowtie, sam, bam, tsv, bustard, phase, fastphase, beagle}; #endif stacks-1.44/src/cstacks.cc000644 000765 000024 00000172672 12775212561 016262 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // cstacks -- Create a catalog of Stacks. // #include "cstacks.h" // Global variables to hold command-line options. queue > samples; string out_path; string catalog_path; FileT in_file_type = FileT::sql; int batch_id = 0; int ctag_dist = 1; bool set_kmer_len = true; int kmer_len = 0; searcht search_type = sequence; int num_threads = 1; bool mult_matches = false; bool report_mmatches = false; bool require_uniq_haplotypes = false; bool gapped_alignments = false; double min_match_len = 0.80; double max_gaps = 2.0; int main (int argc, char* argv[]) { parse_command_line(argc, argv); uint sample_cnt = samples.size(); cerr << "cstacks paramters selected:\n" << " Loci matched based on " << (search_type == sequence ? "sequence identity" : "genomic location") << ".\n"; if (search_type == sequence) cerr << " Number of mismatches allowed between stacks: " << ctag_dist << "\n"; cerr << " Gapped alignments: " << (gapped_alignments ? "enabled" : "disabled") << "\n" << "Constructing catalog from " << sample_cnt << " samples.\n"; // // Set the number of OpenMP parallel threads to execute. // #ifdef _OPENMP omp_set_num_threads(num_threads); #endif map catalog; map::iterator cat_it; map::iterator query_it; pair s; bool compressed = false; int i; if (catalog_path.length() > 0) { cerr << "Initializing existing catalog...\n"; if (!initialize_existing_catalog(catalog_path, catalog)) { cerr << "Failed to initialize the catalog.\n"; return 1; } i = 1; } else { s = samples.front(); samples.pop(); cerr << "Initializing new catalog...\n"; if (!initialize_new_catalog(s, catalog)) { cerr << "Failed to initialize the catalog.\n"; return 1; } i = 2; } // // Build an index of the catalog // map cat_index; if (search_type == genomic_loc) { cerr << "Building an index of the catalog.\n"; update_catalog_index(catalog, cat_index); } while (!samples.empty()) { map sample; cerr << "Processing sample " << s.second << " [" << i << " of " << sample_cnt << "]\n"; s = samples.front(); samples.pop(); if (!load_loci(s.second, sample, false, false, compressed)) { cerr << "Failed to load sample " << i << "\n"; continue; } // // Assign the ID for this sample data. // s.first = sample.begin()->second->sample_id; //dump_loci(sample); if (search_type == sequence) { cerr << "Searching for sequence matches...\n"; find_kmer_matches_by_sequence(catalog, sample, ctag_dist); if (gapped_alignments) { cerr << "Searching for gapped alignments...\n"; search_for_gaps(catalog, sample, min_match_len, ctag_dist); } } else if (search_type == genomic_loc) { cerr << "Searching for matches by genomic location...\n"; find_matches_by_genomic_loc(cat_index, sample); } cerr << "Merging matches into catalog...\n"; uint mmatches = 0; uint gmatches = 0; uint umatches = 0; uint nmatches = 0; merge_matches(catalog, sample, s, ctag_dist, nmatches, umatches, gmatches, mmatches); cerr << " " << umatches << " loci were matched to a catalog locus.\n" << " " << gmatches << " loci were matched to a catalog locus using gapped alignments.\n" << " " << nmatches << " loci were newly added to the catalog.\n" << " " << mmatches << " loci matched more than one catalog locus and were excluded.\n"; // // Regenerate the alleles for the catalog tags after merging the new sample into the catalog. // for (cat_it = catalog.begin(); cat_it != catalog.end(); cat_it++) { cat_it->second->populate_alleles(); cat_it->second->match_cnt = 0; } if (search_type == genomic_loc) { cerr << " Updating catalog index...\n"; update_catalog_index(catalog, cat_index); } i++; for (query_it = sample.begin(); query_it != sample.end(); query_it++) delete query_it->second; sample.clear(); } cerr << "Writing catalog to '" << out_path << "..."; write_catalog(catalog); cerr << " done.\n"; // // Free memory associated with the catalog. // for (cat_it = catalog.begin(); cat_it != catalog.end(); cat_it++) delete cat_it->second; catalog.clear(); return 0; } int update_catalog_index(map &catalog, map &cat_index) { map::iterator j; char id[id_len]; for (j = catalog.begin(); j != catalog.end(); j++) { snprintf(id, id_len - 1, "%s|%d|%c", j->second->loc.chr, j->second->loc.bp, j->second->loc.strand == strand_plus ? '+' : '-'); if (cat_index.count(id) == 0) { cat_index[id] = j->first; } else { if (cat_index[id] != j->first) cerr << "Error: Catalog index mismatch, key: '" << id << "'.\n"; } } return 0; } int characterize_mismatch_snps(CLocus *catalog_tag, QLocus *query_tag) { set snp_cols; uint i; for (i = 0; i < catalog_tag->snps.size(); i++) snp_cols.insert(catalog_tag->snps[i]->col); for (i = 0; i < query_tag->snps.size(); i++) snp_cols.insert(query_tag->snps[i]->col); // // For each mismatch found, create a SNP object // const char *c = catalog_tag->con; const char *c_beg = c; const char *c_end = c + strlen(c); const char *q = query_tag->con; const char *q_beg = q; const char *q_end = q + strlen(q); i = 0; while (c < c_end && q < q_end) { if (snp_cols.count(i) == 0 && (*c != *q) && (*c != 'N' && *q != 'N')) { // cerr << "Adding a new SNP at position " << c - c_beg << ", " << *c << "/" << *q << "\n"; SNP *s = new SNP; s->type = snp_type_het; s->col = c - c_beg; s->lratio = 0; s->rank_1 = *c; s->rank_2 = *q; merge_allele(catalog_tag, s); merge_allele(query_tag, s); catalog_tag->snps.push_back(s); s = new SNP; s->type = snp_type_het; s->col = q - q_beg; s->lratio = 0; s->rank_1 = *q; s->rank_2 = *c; query_tag->snps.push_back(s); } c++; q++; i++; } return 1; } int merge_matches(map &catalog, map &sample, pair &sample_file, int ctag_dist, uint &new_matches, uint &unique_matches, uint &gapped_matches, uint &multiple_matches) { map::iterator i; CLocus *ctag; QLocus *qtag; string cseq, qseq, cigar_str; int cseq_len, match_index; vector > cigar; GappedAln *aln = new GappedAln(); for (i = sample.begin(); i != sample.end(); i++) { qtag = i->second; // // If this stack didn't match an existing catalog stack, add this stack to the // catalog as a new stack. // if (qtag->matches.size() == 0) { add_unique_tag(sample_file, catalog, qtag); new_matches++; continue; } // // Check for multiple matches. We will reduce the list of Match objects, which // contain matches to multiple alleles for a single locus, to the smallest distance // for a locus. // map local_matches; for (uint k = 0; k < qtag->matches.size(); k++) { if (local_matches.count(qtag->matches[k]->cat_id) == 0) local_matches[qtag->matches[k]->cat_id] = qtag->matches[k]->dist; else if (qtag->matches[k]->dist < local_matches[qtag->matches[k]->cat_id]) local_matches[qtag->matches[k]->cat_id] = qtag->matches[k]->dist; } uint min_dist = 1000; uint num_matches = 0; int min_cat_id = -1; // // Find the minimum distance and then check how many matches have that distance. // for (map::iterator j = local_matches.begin(); j != local_matches.end(); j++) min_dist = j->second < min_dist ? j->second : min_dist; for (map::iterator j = local_matches.begin(); j != local_matches.end(); j++) if (j->second == min_dist) { num_matches++; min_cat_id = j->first; } // // Emit a warning if the query tag matches more than one tag in the catalog. // if (num_matches > 1) { multiple_matches++; if (report_mmatches) { cerr << " Warning: sample " << sample_file.second << ", tag " << qtag->id << ", matches more than one tag in the catalog and was excluded: "; for (map::iterator j = local_matches.begin(); j != local_matches.end(); j++) cerr << j->first << " "; cerr << "\n"; } // // Don't record matches to multiple catalog entries unless instructed // to do so by the command line option. // if (!mult_matches) continue; } ctag = catalog[min_cat_id]; if (ctag == NULL) cerr << " Unable to locate catalog tag " << min_cat_id << "\n"; cigar_str = ""; for (uint k = 0; k < qtag->matches.size(); k++) if ((int)qtag->matches[k]->cat_id == min_cat_id) { cigar_str = qtag->matches[k]->cigar; match_index = k; break; } // // If we have already matched a query locus to this catalog locus for the current // sample, we must re-align the sequences in case changes have been made to the // sequence by the previous matching sequence. // if (gapped_alignments && ctag->match_cnt > 0) { string query_allele, query_seq, cat_allele, cat_seq; // cerr << " Warning: Catalog locus " << ctag->id // << ", Sample " << qtag->sample_id << ", locus " << qtag->id // << "; sequence length has changed since original alignment: " // << qseq_len << " <-> " << ctag->len // << "; re-aligning.\n"; // // Find the proper query allele to align against the catalog. We can align // against the catalog consensus because the allele strings may have changed. // query_allele = qtag->matches[match_index]->query_type; for (uint k = 0; k < qtag->strings.size(); k++) if (qtag->strings[k].first == query_allele) { query_seq = qtag->strings[k].second; break; } aln->init(ctag->len, qtag->len); aln->align(ctag->con, query_seq); cigar_str = invert_cigar(aln->result().cigar); } bool gapped_aln = false; if (cigar_str.length() > 0) gapped_aln = true; // // If the match was a gapped alignment, adjust the lengths of the consensus sequences. // Adjust the postition of any SNPs that were shifted down sequence due to a gap. // if (gapped_aln) { cseq_len = parse_cigar(cigar_str.c_str(), cigar); qseq = apply_cigar_to_seq(qtag->con, cigar); adjust_snps_for_gaps(cigar, qtag); cigar_str = invert_cigar(cigar_str); cseq_len = parse_cigar(cigar_str.c_str(), cigar); cseq = apply_cigar_to_seq(ctag->con, cigar); adjust_snps_for_gaps(cigar, ctag); if (qseq.length() != cseq.length()) cerr << "Sample ID: " << qtag->sample_id << "; Query ID: " << qtag->id << "; catalog ID: " << ctag->id << ";\n" << "qloc: " << qtag->con << "\n" << "qseq: " << qseq << "\n" << "cloc: " << ctag->con << " [" << cigar_str << "]\n" << "cseq: " << cseq << "\n"; // // If the alignment modified the catalog locus, record it so we can re-align // any other matching sequences from this sample. // if ((uint)cseq_len > ctag->len) ctag->match_cnt++; // // Adjust the consensus sequences for both loci. // ctag->add_consensus(cseq.c_str()); qtag->add_consensus(qseq.c_str()); gapped_matches++; } else { unique_matches++; ctag->match_cnt++; } // // If mismatches are allowed between query and catalog tags, identify the // mismatches and convert them into SNP objects to be merged into the catalog tag. // if ((ctag_dist > 0 || search_type == genomic_loc) && !characterize_mismatch_snps(ctag, qtag)) cerr << " Error characterizing mismatch SNPs " << sample_file.second << ", tag " << qtag->id << " with catalog tag " << ctag->id << "\n"; // // Merge the SNPs and alleles from the sample into the catalog tag. // if (!ctag->merge_snps(qtag)) { cerr << "Error merging " << sample_file.second << ", tag " << qtag->id << " with catalog tag " << ctag->id << "\n"; } // // Add any new sequence information into the catalog consensus. // if (gapped_aln) { for (uint k = 0; k < ctag->len && k < qtag->len; k++) if (qtag->con[k] != 'N' && ctag->con[k] == 'N') ctag->con[k] = qtag->con[k]; } else if (strlen(ctag->con) < strlen(qtag->con)) { ctag->add_consensus(qtag->con); } ctag->sources.push_back(make_pair(sample_file.first, qtag->id)); } delete aln; return 0; } int add_unique_tag(pair &sample_file, map &catalog, QLocus *qloc) { vector::iterator i; map::iterator j; int cid = catalog.size(); CLocus *c = new CLocus; c->id = cid + 1; c->add_consensus(qloc->con); // // Record the source of this catalog tag. // c->sources.push_back(make_pair(sample_file.first, qloc->id)); // // Add the physical genome location of this locus. // c->loc.set(qloc->loc.chr, qloc->loc.bp, qloc->loc.strand); catalog[c->id] = c; // cerr << "Adding sample: " << qloc->id << " to the catalog as ID: " << c->id << "\n"; for (i = qloc->snps.begin(); i != qloc->snps.end(); i++) { SNP *snp = new SNP; snp->col = (*i)->col; snp->type = (*i)->type; snp->lratio = (*i)->lratio; snp->rank_1 = (*i)->rank_1; snp->rank_2 = (*i)->rank_2; c->snps.push_back(snp); } for (j = qloc->alleles.begin(); j != qloc->alleles.end(); j++) { c->alleles[j->first] = j->second; } c->populate_alleles(); return 0; } int find_kmer_matches_by_sequence(map &catalog, map &sample, int ctag_dist) { // // Calculate the distance (number of mismatches) between each pair // of Radtags. We expect all radtags to be the same length; // KmerHashMap kmer_map; map > allele_map; vector kmer_map_keys; map::iterator it; vector >::iterator allele; QLocus *tag_1; CLocus *tag_2; // OpenMP can't parallelize random access iterators, so we convert // our map to a vector of integer keys. vector keys; for (it = sample.begin(); it != sample.end(); it++) keys.push_back(it->first); // // Calculate the number of k-mers we will generate. If kmer_len == 0, // determine the optimal length for k-mers. // int con_len = strlen(sample[keys[0]]->con); if (set_kmer_len) kmer_len = determine_kmer_length(con_len, ctag_dist); int num_kmers = con_len - kmer_len + 1; // // Calculate the minimum number of matching k-mers required for a possible sequence match. // int min_hits = calc_min_kmer_matches(kmer_len, ctag_dist, con_len, set_kmer_len ? true : false); cerr << " Distance allowed between stacks: " << ctag_dist << "; searching with a k-mer length of " << kmer_len << " (" << num_kmers << " k-mers per read); " << min_hits << " k-mer hits required.\n"; // clock_t time_1, time_2, time_3, time_4; // double per_locus = 0.0; // time_1 = clock(); populate_kmer_hash(catalog, kmer_map, kmer_map_keys, allele_map, kmer_len); // time_2 = clock(); cerr << " " << catalog.size() << " loci in the catalog, " << kmer_map.size() << " kmers in the catalog hash.\n"; #pragma omp parallel private(tag_1, tag_2, allele) { KmerHashMap::iterator h; vector kmers; set uniq_kmers; vector hits; vector > ordered_hits; uint hit_cnt, index, prev_id, allele_id, hits_size; int d; pair cat_hit; initialize_kmers(kmer_len, num_kmers, kmers); #pragma omp for for (uint i = 0; i < keys.size(); i++) { tag_1 = sample[keys[i]]; // time_3 = clock(); for (allele = tag_1->strings.begin(); allele != tag_1->strings.end(); allele++) { generate_kmers_lazily(allele->second.c_str(), kmer_len, num_kmers, kmers); // // We want to create a list of unique kmers to search with; otherwise, repetitive kmers will // generate, multiple, spurious hits in sequences with multiple copies of the same kmer. // uniq_kmers.clear(); for (int j = 0; j < num_kmers; j++) uniq_kmers.insert(kmers[j]); hits.clear(); ordered_hits.clear(); // // Lookup the occurances of each k-mer in the kmer_map // for (set::iterator j = uniq_kmers.begin(); j != uniq_kmers.end(); j++) { h = kmer_map.find(j->c_str()); if (h != kmer_map.end()) for (uint k = 0; k < h->second.size(); k++) hits.push_back(h->second[k]); } // // Sort the vector of indexes; provides the number of hits to each allele/locus // and orders them largest to smallest. // sort(hits.begin(), hits.end()); // // Iterate through the list of hits and collapse them down by number of kmer hits per allele. // hits_size = hits.size(); if (hits_size == 0) continue; prev_id = hits[0]; index = 0; do { hit_cnt = 0; allele_id = prev_id; while (index < hits_size && (uint) hits[index] == prev_id) { hit_cnt++; index++; } if (index < hits_size) prev_id = hits[index]; if (hit_cnt >= (uint) min_hits) ordered_hits.push_back(make_pair(allele_id, hit_cnt)); } while (index < hits_size); for (uint j = 0; j < ordered_hits.size(); j++) { cat_hit = allele_map.at(ordered_hits[j].first); hit_cnt = ordered_hits[j].second; tag_2 = catalog[cat_hit.second]; d = dist(allele->second.c_str(), tag_2, cat_hit.first); if (d < 0) cerr << "Unknown error calculating distance between " << tag_1->id << " and " << tag_2->id << "; query allele: " << allele->first << "\n"; // // Add a match to the query sequence: catalog ID, catalog allele, query allele, distance // if (d <= ctag_dist) tag_1->add_match(tag_2->id, cat_hit.first, allele->first, d); } } // Sort the vector of distances. sort(tag_1->matches.begin(), tag_1->matches.end(), compare_matches); // time_4 = clock(); // per_locus += (time_4 - time_3); } // // Free the allocated k-mers. // for (uint j = 0; j < kmers.size(); j++) delete [] kmers[j]; kmers.clear(); } // cerr << "Time to kmerize catalog: " << time_2 - time_1 << "\n" // << "Average time per locus: " << per_locus / (double) keys.size() << "\n"; free_kmer_hash(kmer_map, kmer_map_keys); return 0; } int search_for_gaps(map &catalog, map &sample, double min_match_len, double ctag_dist) { // // Search for loci that can be merged with a gapped alignment. // KmerHashMap kmer_map; map > allele_map; vector kmer_map_keys; map::iterator it; QLocus *tag_1; CLocus *tag_2; // // OpenMP can't parallelize random access iterators, so we convert // our map to a vector of integer keys. // vector keys; for (it = sample.begin(); it != sample.end(); it++) keys.push_back(it->first); // // Calculate the number of k-mers we will generate. If kmer_len == 0, // determine the optimal length for k-mers. // int con_len = strlen(sample[keys[0]]->con); int kmer_len = 19; int num_kmers = con_len - kmer_len + 1; // // Calculate the minimum number of matching k-mers required for a possible sequence match. // int min_hits = (round((double) con_len * min_match_len) - (kmer_len * max_gaps)) - kmer_len + 1; cerr << " Searching with a k-mer length of " << kmer_len << " (" << num_kmers << " k-mers per read); " << min_hits << " k-mer hits required.\n"; // clock_t time_1, time_2, time_3, time_4; // double per_locus = 0.0; // time_1 = clock(); populate_kmer_hash(catalog, kmer_map, kmer_map_keys, allele_map, kmer_len); // time_2 = clock(); #pragma omp parallel private(tag_1, tag_2) { KmerHashMap::iterator h; AlignRes aln_res; vector kmers; set uniq_kmers; vector hits; vector > ordered_hits; uint hit_cnt, index, prev_id, allele_id, hits_size, stop, top_hit; int d; vector > cigar; pair cat_hit; string cat_seq; GappedAln *aln = new GappedAln(); initialize_kmers(kmer_len, num_kmers, kmers); #pragma omp for schedule(dynamic) for (uint i = 0; i < keys.size(); i++) { tag_1 = sample[keys[i]]; // // If we already matched this locus to the catalog without using gapped alignments, skip it now. // if (tag_1->matches.size() > 0) continue; // time_3 = clock(); for (vector >::iterator allele = tag_1->strings.begin(); allele != tag_1->strings.end(); allele++) { generate_kmers_lazily(allele->second.c_str(), kmer_len, num_kmers, kmers); // // We want to create a list of unique kmers to search with; otherwise, repetitive kmers will // generate, multiple, spurious hits in sequences with multiple copies of the same kmer. // uniq_kmers.clear(); for (int j = 0; j < num_kmers; j++) uniq_kmers.insert(kmers[j]); hits.clear(); ordered_hits.clear(); // // Lookup the occurances of each k-mer in the kmer_map // for (set::iterator j = uniq_kmers.begin(); j != uniq_kmers.end(); j++) { h = kmer_map.find(j->c_str()); if (h != kmer_map.end()) for (uint k = 0; k < h->second.size(); k++) hits.push_back(h->second[k]); } // // Sort the vector of indexes; provides the number of hits to each allele/locus // and orders them largest to smallest. // sort(hits.begin(), hits.end()); // // Iterate through the list of hits and collapse them down by number of kmer hits per allele. // hits_size = hits.size(); if (hits_size == 0) continue; prev_id = hits[0]; index = 0; do { hit_cnt = 0; allele_id = prev_id; while (index < hits_size && (uint) hits[index] == prev_id) { hit_cnt++; index++; } if (index < hits_size) prev_id = hits[index]; if (hit_cnt >= (uint) min_hits) ordered_hits.push_back(make_pair(allele_id, hit_cnt)); } while (index < hits_size); if (ordered_hits.size() == 0) continue; // // Process the hits from most kmer hits to least kmer hits. // sort(ordered_hits.begin(), ordered_hits.end(), compare_pair_intint); // // Only try to align the sequences with the most kmers in common. // top_hit = ordered_hits[0].second; stop = 1; for (uint j = 1; j < ordered_hits.size(); j++) if ((uint)ordered_hits[j].second < top_hit) { stop = j; break; } for (uint j = 0; j < stop; j++) { cat_hit = allele_map.at(ordered_hits[j].first); hit_cnt = ordered_hits[j].second; tag_2 = catalog[cat_hit.second]; cat_seq = ""; for (uint k = 0; k < tag_2->strings.size(); k++) if (tag_2->strings[k].first == cat_hit.first) { cat_seq = tag_2->strings[k].second; break; } aln->init(tag_2->len, tag_1->len); if (aln->align(cat_seq, allele->second)) { cigar.clear(); aln->parse_cigar(cigar); aln_res = aln->result(); d = dist(cat_seq.c_str(), allele->second.c_str(), cigar); // // If the alignment has too many gaps, skip it. // if (aln_res.gap_cnt <= (max_gaps + 1)) { // // If the alignment doesn't span enough of the two sequences, skip it. // if (aln_res.pct_id >= min_match_len) { if (d <= ctag_dist) tag_1->add_match(tag_2->id, cat_hit.first, allele->first, d, invert_cigar(aln_res.cigar)); } } } } } // time_4 = clock(); // per_locus += (time_4 - time_3); } // // Free the k-mers we generated for this query // for (uint j = 0; j < kmers.size(); j++) delete [] kmers[j]; kmers.clear(); delete aln; } // cerr << "Time to kmerize catalog: " << time_2 - time_1 << "\n" // << "Average time per locus: " << per_locus / (double) keys.size() << "\n"; free_kmer_hash(kmer_map, kmer_map_keys); return 0; } bool compare_matches(Match *a, Match *b) { return (a->dist < b->dist); } int find_matches_by_sequence(map &catalog, map &sample) { // // Calculate the distance (number of mismatches) between each pair // of Radtags. We expect all radtags to be the same length; // map::iterator i; map::iterator j; int k; // OpenMP can't parallelize random access iterators, so we convert // our map to a vector of integer keys. vector keys; for (i = sample.begin(); i != sample.end(); i++) keys.push_back(i->first); #pragma omp parallel private(i, j, k) { #pragma omp for schedule(dynamic) for (k = 0; k < (int) keys.size(); k++) { i = sample.find(keys[k]); vector >::iterator r, s; // // Iterate through the possible SAMPLE alleles // for (r = i->second->strings.begin(); r != i->second->strings.end(); r++) { for (j = catalog.begin(); j != catalog.end(); j++) { // // Iterate through the possible CATALOG alleles // for (s = j->second->strings.begin(); s != j->second->strings.end(); s++) { if (r->second == s->second) { //cerr << "Found a match between " << i->first << " (" << r->first << ") and " << j->first << " (" << s->first << ")\n"; i->second->add_match(j->second->id, s->first, r->first, 0); } } } } } } return 0; } int find_matches_by_genomic_loc(map &cat_index, map &sample) { map::iterator i; map::iterator j; // // OpenMP can't parallelize random access iterators, so we convert // our map to a vector of integer keys. // vector keys; for (i = sample.begin(); i != sample.end(); i++) keys.push_back(i->first); #pragma omp parallel private(i, j) { char id[id_len]; #pragma omp for for (int k = 0; k < (int) keys.size(); k++) { i = sample.find(keys[k]); snprintf(id, id_len - 1, "%s|%d|%c", i->second->loc.chr, i->second->loc.bp, i->second->loc.strand == strand_plus ? '+' : '-'); if (cat_index.count(id) > 0) i->second->add_match(cat_index[id], "", "", 0); } } return 0; } int write_catalog(map &catalog) { map::iterator i; CLocus *tag; set matches; bool gzip = (in_file_type == FileT::gzsql) ? true : false; // // Parse the input file names to create the output file // stringstream prefix; prefix << out_path << "batch_" << batch_id; string tag_file = prefix.str() + ".catalog.tags.tsv"; string snp_file = prefix.str() + ".catalog.snps.tsv"; string all_file = prefix.str() + ".catalog.alleles.tsv"; if (gzip) { tag_file += ".gz"; snp_file += ".gz"; all_file += ".gz"; } // // Open the output files for writing. // gzFile gz_tags, gz_snps, gz_alle; ofstream tags, snps, alle; if (gzip) { gz_tags = gzopen(tag_file.c_str(), "wb"); if (!gz_tags) { cerr << "Error: Unable to open gzipped catalog tag file '" << tag_file << "': " << strerror(errno) << ".\n"; exit(1); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_tags, libz_buffer_size); #endif gz_snps = gzopen(snp_file.c_str(), "wb"); if (!gz_snps) { cerr << "Error: Unable to open gzipped catalog snps file '" << snp_file << "': " << strerror(errno) << ".\n"; exit(1); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_snps, libz_buffer_size); #endif gz_alle = gzopen(all_file.c_str(), "wb"); if (!gz_alle) { cerr << "Error: Unable to open gzipped catalog alleles file '" << all_file << "': " << strerror(errno) << ".\n"; exit(1); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_alle, libz_buffer_size); #endif } else { tags.open(tag_file.c_str()); if (tags.fail()) { cerr << "Error: Unable to open catalog tag file for writing.\n"; exit(1); } snps.open(snp_file.c_str()); if (snps.fail()) { cerr << "Error: Unable to open catalog SNPs file for writing.\n"; exit(1); } alle.open(all_file.c_str()); if (alle.fail()) { cerr << "Error: Unable to open catalog alleles file for writing.\n"; exit(1); } } // // Record the version of Stacks used and the date generated as a comment in the catalog. // // Obtain the current date. // stringstream log; time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%F %T", timeinfo); log << "# cstacks version " << VERSION << "; catalog generated on " << date << "\n"; if (gzip) { gzputs(gz_tags, log.str().c_str()); gzputs(gz_snps, log.str().c_str()); gzputs(gz_alle, log.str().c_str()); } else { tags << log.str(); snps << log.str(); alle << log.str(); } for (i = catalog.begin(); i != catalog.end(); i++) { tag = i->second; if (gzip) write_gzip_output(tag, gz_tags, gz_snps, gz_alle); else write_simple_output(tag, tags, snps, alle); } if (gzip) { gzclose(gz_tags); gzclose(gz_snps); gzclose(gz_alle); } else { tags.close(); snps.close(); alle.close(); } return 0; } int merge_allele(Locus *locus, SNP *snp) { map > columns; map >::iterator c; vector::iterator i; SNP *lsnp; for (i = locus->snps.begin(); i != locus->snps.end(); i++) columns[(*i)->col] = make_pair("sample", *i); if (columns.count(snp->col)) { lsnp = columns[snp->col].second; // // If this is a new allele for this nucleotide, add it to the catalog SNP. // bool rank_1_exists = false; bool rank_2_exists = false; if (snp->rank_1 == lsnp->rank_1 || snp->rank_1 == lsnp->rank_2 || snp->rank_1 == lsnp->rank_3 || snp->rank_1 == lsnp->rank_4) { rank_1_exists = true; } if (snp->rank_2 == lsnp->rank_1 || snp->rank_2 == lsnp->rank_2 || snp->rank_2 == lsnp->rank_3 || snp->rank_2 == lsnp->rank_4) { rank_2_exists = true; } if (rank_1_exists == false) { if (lsnp->rank_3 == 0) lsnp->rank_3 = snp->rank_1; else lsnp->rank_4 = snp->rank_1; } if (rank_2_exists == false) { if (lsnp->rank_3 == 0) lsnp->rank_3 = snp->rank_2; else lsnp->rank_4 = snp->rank_2; } columns[snp->col] = make_pair("both", lsnp); } else { columns[snp->col] = make_pair("merge", snp); } vector > merged_snps; for (c = columns.begin(); c != columns.end(); c++) merged_snps.push_back((*c).second); // // Sort the SNPs by column // sort(merged_snps.begin(), merged_snps.end(), compare_pair_snp); // // Modify any existing alleles to account for this new SNP. If there are not any alleles, // create new ones. // stringstream sallele; set merged_alleles; string allele, new_allele; int pos; if (locus->alleles.size() == 0) { sallele << locus->con[snp->col]; merged_alleles.insert(sallele.str()); } map::iterator j; vector >::iterator k; for (j = locus->alleles.begin(); j != locus->alleles.end(); j++) { allele = j->first; new_allele = ""; pos = 0; // cerr << "Allele length: " << allele.size() << "\n"; for (k = merged_snps.begin(); k != merged_snps.end(); k++) { // // If we inserted a SNP from the sample, add the proper nucleotide from the consensus // sequence to account for it in the allele string. // if ((*k).first == "merge") { new_allele += locus->con[(*k).second->col]; // cerr << " Adding char '" << locus->con[k->second->col] << "' from consensus position " << (*k).second->col << "\n"; } else { new_allele += allele[pos]; // cerr << " Adding char '" << allele[pos] << "' from allele position " << pos << "\n"; pos++; } } merged_alleles.insert(new_allele); } set::iterator s; locus->alleles.clear(); for (s = merged_alleles.begin(); s != merged_alleles.end(); s++) { locus->alleles[*s] = 0; } return 1; } int CLocus::merge_snps(QLocus *matched_tag) { vector::iterator i; map::iterator j; vector >::iterator k; map > columns; map >::iterator c; vector > merged_snps; set merged_alleles; set::iterator s; SNP *csnp; for (i = this->snps.begin(); i != this->snps.end(); i++) columns[(*i)->col] = make_pair("catalog", *i); for (i = matched_tag->snps.begin(); i != matched_tag->snps.end(); i++) { // // Is this column already represented from the previous sample? // if (columns.count((*i)->col)) { csnp = columns[(*i)->col].second; // // If this is a new allele for this nucleotide, add it to the catalog SNP. // bool rank_1_exists = false; bool rank_2_exists = false; if ((*i)->rank_1 == csnp->rank_1 || (*i)->rank_1 == csnp->rank_2 || (*i)->rank_1 == csnp->rank_3 || (*i)->rank_1 == csnp->rank_4) { rank_1_exists = true; } if ((*i)->rank_2 == csnp->rank_1 || (*i)->rank_2 == csnp->rank_2 || (*i)->rank_2 == csnp->rank_3 || (*i)->rank_2 == csnp->rank_4) { rank_2_exists = true; } if (rank_1_exists == false) { if (csnp->rank_3 == 0) csnp->rank_3 = (*i)->rank_1; else csnp->rank_4 = (*i)->rank_1; } if (rank_2_exists == false) { if (csnp->rank_3 == 0) csnp->rank_3 = (*i)->rank_2; else csnp->rank_4 = (*i)->rank_2; } columns[(*i)->col] = make_pair("both", csnp); } else { columns[(*i)->col] = make_pair("sample", *i); } } for (c = columns.begin(); c != columns.end(); c++) merged_snps.push_back((*c).second); // // Sort the SNPs by column // sort(merged_snps.begin(), merged_snps.end(), compare_pair_snp); // // If the catalog tag has no defined alleles, create a matching haplotype // from the consensus sequence before merging in the new alleles. // string allele, new_allele; int pos; if (this->alleles.size() == 0) { char c; new_allele = ""; for (k = merged_snps.begin(); k != merged_snps.end(); k++) { csnp = k->second; c = this->con[k->second->col]; new_allele += (csnp->col > this->len - 1) ? 'N' : c; if (csnp->col > this->len - 1) continue; if (c != csnp->rank_1 && c != csnp->rank_2 && c != csnp->rank_3 && c != csnp->rank_4) { if (csnp->rank_3 == 0) csnp->rank_3 = c; else csnp->rank_4 = c; } } if (new_allele.length() > 0) merged_alleles.insert(new_allele); } // // Merge the alleles accounting for any SNPs added from either of the two samples. // for (j = this->alleles.begin(); j != this->alleles.end(); j++) { allele = j->first; new_allele = ""; pos = 0; for (k = merged_snps.begin(); k != merged_snps.end(); k++) { // // If we inserted a SNP from the sample, add the proper nucleotide from the consensus // sequence to account for it in the allele string. // if (k->first == "sample") { new_allele += k->second->col > this->len - 1 ? 'N' : this->con[k->second->col]; } else { new_allele += allele[pos]; pos++; } } merged_alleles.insert(new_allele); } for (j = matched_tag->alleles.begin(); j != matched_tag->alleles.end(); j++) { allele = j->first; new_allele = ""; pos = 0; for (k = merged_snps.begin(); k != merged_snps.end(); k++) { if (k->first == "catalog") { new_allele += k->second->col > matched_tag->len - 1 ? 'N' : matched_tag->con[k->second->col]; } else { new_allele += allele[pos]; pos++; } } merged_alleles.insert(new_allele); } // // If the matching tag being merged into the catalog had no called SNPs // create alleles from the consensus sequence and check that catalog SNP // objects contain all the nucleoties. // if (matched_tag->alleles.size() == 0) { char c; new_allele = ""; for (k = merged_snps.begin(); k != merged_snps.end(); k++) { csnp = k->second; c = matched_tag->con[k->second->col]; new_allele += (csnp->col > matched_tag->len - 1) ? 'N' : c; if (csnp->col > matched_tag->len - 1) continue; if (c != csnp->rank_1 && c != csnp->rank_2 && c != csnp->rank_3 && c != csnp->rank_4) { if (csnp->rank_3 == 0) csnp->rank_3 = c; else csnp->rank_4 = c; } } if (new_allele.length() > 0) merged_alleles.insert(new_allele); } // // // // If the newly merged alleles contain Ns due to different sequence lengths, // // check if we can reduce the alleles as one of the longer allele haplotypes // // may fully encompass a shorter allele haplotype that has been padded with Ns. // // // if (require_uniq_haplotypes) this->reduce_alleles(merged_alleles); // // Update the catalog entry's list of SNPs and alleles // this->snps.clear(); for (k = merged_snps.begin(); k != merged_snps.end(); k++) { SNP *snp = new SNP; snp->col = (*k).second->col; snp->type = (*k).second->type; snp->lratio = 0.0; snp->rank_1 = (*k).second->rank_1; snp->rank_2 = (*k).second->rank_2; snp->rank_3 = (*k).second->rank_3; snp->rank_4 = (*k).second->rank_4; this->snps.push_back(snp); if (k->first == "catalog" || k->first == "both") delete k->second; } this->alleles.clear(); for (s = merged_alleles.begin(); s != merged_alleles.end(); s++) { this->alleles[*s] = 0; } return 1; } int CLocus::reduce_alleles(set &alleles) { set::iterator it; uint len, max_len, match, ncnt; vector haplotypes, cur, next; max_len = 0; for (it = alleles.begin(); it != alleles.end(); it++) { max_len = it->length() > max_len ? it->length() : max_len; haplotypes.push_back(*it); } len = alleles.size(); alleles.clear(); for (uint i = 0; i < len; i++) { //cerr << "Looking at haplotype[" << i << "]: " << haplotypes[i] << "\n"; // // We will only look at strings that contain Ns. // if (haplotypes[i].find('N') == string::npos) { alleles.insert(haplotypes[i]); //cerr << " No Ns, skipping...\n"; continue; } uint k = 0; uint j = i + 1; while (k < len - 1) { cur.push_back(haplotypes[j % len]); k++; j++; } // // Examine the haplotype alleles one SNP at a time. If we are able to uniquely // determine a second haplotype that encompasses the first // to, return it. // j = 0; while (cur.size() > 1 && j < max_len) { for (k = 0; k < cur.size(); k++) { cerr << "Comparing haplotypes[" << i << "]: '" << haplotypes[i] << "' to '" << cur[k] << " at position " << j << "'\n"; if (haplotypes[i][j] == cur[k][j] || haplotypes[i][j] == 'N') { cerr << " Keeping this haplotype.\n"; next.push_back(cur[k]); } else { cerr << " Discarding this haplotype.\n"; } } cur = next; next.clear(); j++; } // // If there is only one left, make sure what we have of the haplotype does match // and its not simply an erroneously called haplotype. If so, then this haplotype // is encompassed by another, longer haplotype and we do not need to keep it. // ncnt = 0; match = 0; if (cur.size() > 1) { cerr << "Discarding " << haplotypes[i] << "\n"; continue; } else if (cur.size() == 1) { for (k = 0; k < max_len; k++) if (haplotypes[i][k] != 'N') ncnt++; for (k = 0; k < max_len; k++) if (cur[0][k] == haplotypes[i][k]) match++; if (match == ncnt) { cerr << "Discarding " << haplotypes[i] << "\n"; continue; } } cerr << "Keeping " << haplotypes[i] << "\n"; alleles.insert(haplotypes[i]); } return 0; } int populate_kmer_hash(map &catalog, CatKmerHashMap &kmer_map, vector &kmer_map_keys, int kmer_len) { map::iterator it; vector >::iterator allele; vector kmers; CLocus *tag; char *hash_key; bool exists; int j; // // Break each stack down into k-mers and create a hash map of those k-mers // recording in which sequences they occur. // int num_kmers = strlen(catalog.begin()->second->con) - kmer_len + 1; for (it = catalog.begin(); it != catalog.end(); it++) { tag = it->second; // // Iterate through the possible Catalog alleles // for (allele = tag->strings.begin(); allele != tag->strings.end(); allele++) { // // Generate and hash the kmers for this allele string // generate_kmers(allele->second.c_str(), kmer_len, num_kmers, kmers); for (j = 0; j < num_kmers; j++) { hash_key = kmers[j]; exists = kmer_map.count(hash_key) == 0 ? false : true; kmer_map[hash_key].push_back(make_pair(allele->first, tag->id)); if (exists) delete [] kmers[j]; else kmer_map_keys.push_back(hash_key); } kmers.clear(); } } //dump_kmer_map(kmer_map); return 0; } int write_simple_output(CLocus *tag, ofstream &cat_file, ofstream &snp_file, ofstream &all_file) { vector::iterator snp_it; map::iterator all_it; vector >::iterator src_it; string sources; for (src_it = tag->sources.begin(); src_it != tag->sources.end(); src_it++) { stringstream s; s << (*src_it).first << "_" << (*src_it).second << ","; sources += s.str(); } sources = sources.substr(0, sources.length() - 1); cat_file << "0" << "\t" << batch_id << "\t" << tag->id << "\t" << tag->loc.chr << "\t" << tag->loc.bp << "\t" << (tag->loc.strand == strand_plus ? "+" : "-") << "\t" << "consensus" << "\t" << "0" << "\t" << sources << "\t" << tag->con << "\t" << 0 << "\t" << // These flags are unused in cstacks, but important in ustacks 0 << "\t" << 0 << "\t" << 0 << "\n"; // // Output the SNPs associated with the catalog tag // for (snp_it = tag->snps.begin(); snp_it != tag->snps.end(); snp_it++) { snp_file << "0" << "\t" << batch_id << "\t" << tag->id << "\t" << (*snp_it)->col << "\t"; switch((*snp_it)->type) { case snp_type_het: snp_file << "E\t"; break; case snp_type_hom: snp_file << "O\t"; break; default: snp_file << "U\t"; break; } snp_file << (*snp_it)->lratio << "\t" << (*snp_it)->rank_1 << "\t" << (*snp_it)->rank_2 << "\t" << ((*snp_it)->rank_3 == 0 ? '-' : (*snp_it)->rank_3) << "\t" << ((*snp_it)->rank_4 == 0 ? '-' : (*snp_it)->rank_4) << "\n"; } // // Output the alleles associated with the two matched tags // for (all_it = tag->alleles.begin(); all_it != tag->alleles.end(); all_it++) all_file << "0" << "\t" << batch_id << "\t" << tag->id << "\t" << all_it->first << "\t" << "0" << "\t" << // These two fields are used in the "0" << "\n"; // ustacks/pstacks output, not in cstacks. return 0; } int write_gzip_output(CLocus *tag, gzFile &cat_file, gzFile &snp_file, gzFile &all_file) { vector::iterator snp_it; map::iterator all_it; vector >::iterator src_it; string sources; stringstream sstr; for (src_it = tag->sources.begin(); src_it != tag->sources.end(); src_it++) { sstr << (*src_it).first << "_" << (*src_it).second << ","; } sources = sstr.str(); sources = sources.substr(0, sources.length() - 1); sstr.str(""); sstr << "0" << "\t" << batch_id << "\t" << tag->id << "\t" << tag->loc.chr << "\t" << tag->loc.bp << "\t" << (tag->loc.strand == strand_plus ? "+" : "-") << "\t" << "consensus" << "\t" << "0" << "\t" << sources << "\t" << tag->con << "\t" << 0 << "\t" << // These flags are unused in cstacks, but important in ustacks 0 << "\t" << 0 << "\t" << 0 << "\n"; gzputs(cat_file, sstr.str().c_str()); sstr.str(""); // // Output the SNPs associated with the catalog tag // for (snp_it = tag->snps.begin(); snp_it != tag->snps.end(); snp_it++) { sstr << "0" << "\t" << batch_id << "\t" << tag->id << "\t" << (*snp_it)->col << "\t"; switch((*snp_it)->type) { case snp_type_het: sstr << "E\t"; break; case snp_type_hom: sstr << "O\t"; break; default: sstr << "U\t"; break; } sstr << (*snp_it)->lratio << "\t" << (*snp_it)->rank_1 << "\t" << (*snp_it)->rank_2 << "\t" << ((*snp_it)->rank_3 == 0 ? '-' : (*snp_it)->rank_3) << "\t" << ((*snp_it)->rank_4 == 0 ? '-' : (*snp_it)->rank_4) << "\n"; } gzputs(snp_file, sstr.str().c_str()); sstr.str(""); // // Output the alleles associated with the two matched tags // for (all_it = tag->alleles.begin(); all_it != tag->alleles.end(); all_it++) sstr << "0\t" << batch_id << "\t" << tag->id << "\t" << all_it->first << "\t" << 0 << "\t" << 0 << "\n"; gzputs(all_file, sstr.str().c_str()); return 0; } int initialize_new_catalog(pair &sample, map &catalog) { map tmp_catalog; bool compressed = false; // // Parse the input files. // if (!load_loci(sample.second, tmp_catalog, false, false, compressed)) return 0; in_file_type = compressed == true ? FileT::gzsql : FileT::sql; sample.first = tmp_catalog.begin()->second->sample_id; // // Iterate over the catalog entires and renumber them after recording the source of // locus. // map::iterator j; int k = 1; for (j = tmp_catalog.begin(); j != tmp_catalog.end(); j++) { j->second->sources.push_back(make_pair(sample.first, j->second->id)); j->second->id = k; catalog[k] = j->second; k++; } cerr << " " << catalog.size() << " loci were newly added to the catalog.\n"; return 1; } int initialize_existing_catalog(string catalog_path, map &catalog) { bool compressed; // // Parse the input files. // if (!load_loci(catalog_path, catalog, false, false, compressed)) return 0; in_file_type = compressed == true ? FileT::gzsql : FileT::sql; // // Iterate over the catalog entires and convert the stack components // into source objects, to record what samples each locus came from. // map::iterator j; CLocus *loc; char *p, *q; int sample_id, locus_id; for (j = catalog.begin(); j != catalog.end(); j++) { loc = j->second; for (uint i = 0; i < loc->comp.size(); i++) { // // Parse the ID into sample ID / locus ID, given 43_1356, parse into // sample ID 43 and locus ID 1356. // for (p = loc->comp[i]; *p != '_' && *p != '\0'; p++); if (*p != '_') return 0; p++; sample_id = strtol(loc->comp[i], &q, 10); if (*q != '_') return 0; locus_id = strtol(p, &q, 10); if (*q != '\0') return 0; loc->sources.push_back(make_pair(sample_id, locus_id)); } } return 1; } int parse_command_line(int argc, char* argv[]) { int c; string sstr; while (1) { static struct option long_options[] = { {"help", no_argument, NULL, 'h'}, {"version", no_argument, NULL, 'v'}, {"mmatches", no_argument, NULL, 'm'}, {"genomic_loc", no_argument, NULL, 'g'}, {"uniq_haplotypes", no_argument, NULL, 'u'}, {"report_mmatches", no_argument, NULL, 'R'}, {"gapped", no_argument, NULL, 'G'}, {"max_gaps", required_argument, NULL, 'X'}, {"min_aln_len", required_argument, NULL, 'x'}, {"batch_id", required_argument, NULL, 'b'}, {"ctag_dist", required_argument, NULL, 'n'}, {"k_len", required_argument, NULL, 'k'}, {"catalog", required_argument, NULL, 'c'}, {"sample", required_argument, NULL, 's'}, {"outpath", required_argument, NULL, 'o'}, {"num_threads", required_argument, NULL, 'p'}, {0, 0, 0, 0} }; // getopt_long stores the option index here. int option_index = 0; c = getopt_long(argc, argv, "hgvuRmGX:x:o:s:c:b:p:n:k:", long_options, &option_index); // Detect the end of the options. if (c == -1) break; switch (c) { case 'h': help(); break; case 'b': batch_id = is_integer(optarg); if (batch_id < 0) { cerr << "Batch ID (-b) must be an integer, e.g. 1, 2, 3\n"; help(); } break; case 'n': ctag_dist = is_integer(optarg); break; case 'k': set_kmer_len = false; kmer_len = is_integer(optarg); break; case 'm': mult_matches = true; break; case 'R': report_mmatches = true; break; case 'g': search_type = genomic_loc; break; case 's': sstr = optarg; samples.push(make_pair(0, sstr)); break; case 'c': catalog_path = optarg; break; case 'o': out_path = optarg; break; case 'u': require_uniq_haplotypes = true; break; case 'G': gapped_alignments = true; break; case 'X': max_gaps = is_double(optarg); break; case 'x': min_match_len = is_double(optarg); break; case 'v': version(); break; case 'p': num_threads = is_integer(optarg); break; case '?': // getopt_long already printed an error message. help(); break; default: help(); abort(); } } if (set_kmer_len == false && (kmer_len < 5 || kmer_len > 31)) { cerr << "Kmer length must be between 5 and 31bp.\n"; help(); } if (samples.size() == 0) { cerr << "You must specify at least one sample file.\n"; help(); } if (out_path.length() == 0) out_path = "."; if (out_path.at(out_path.length() - 1) != '/') out_path += "/"; return 0; } void version() { std::cerr << "cstacks " << VERSION << "\n\n"; exit(0); } void help() { std::cerr << "cstacks " << VERSION << "\n" << "cstacks -b batch_id -s sample_file [-s sample_file_2 ...] [-o path] [-g] [-n num] [-p num_threads] [--catalog path] [-h]" << "\n" << " b: MySQL ID of this batch." << "\n" << " s: filename prefix from which to load loci into the catalog." << "\n" << " o: output path to write results." << "\n" << " g: base catalog matching on genomic location, not sequence identity." << "\n" << " m: include tags in the catalog that match to more than one entry (default false)." << "\n" << " n: number of mismatches allowed between sample tags when generating the catalog (default 1)." << "\n" << " p: enable parallel execution with num_threads threads.\n" << " h: display this help messsage." << "\n\n" << " Catalog editing:\n" << " --catalog : provide the path to an existing catalog. cstacks will add data to this existing catalog.\n\n" << " Gapped assembly options:\n" << " --gapped: preform gapped alignments between stacks.\n" << " --max_gaps: number of gaps allowed between stacks before merging (default: 2).\n" << " --min_aln_len: minimum length of aligned sequence in a gapped alignment (default: 0.80).\n\n" << " Advanced options:\n" << " --k_len : specify k-mer size for matching between between catalog loci (automatically calculated by default).\n" << " --report_mmatches: report query loci that match more than one catalog locus.\n"; exit(0); } stacks-1.44/src/cstacks.h000644 000765 000024 00000005623 12704260321 016100 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __CSTACKS_H__ #define __CSTACKS_H__ #ifdef _OPENMP #include // OpenMP library #endif #include #include // Support for gzipped output files. #include // Process command-line options #include #include #include #include using std::pair; using std::make_pair; #include using std::string; #include #include #include using std::ifstream; using std::ofstream; using std::stringstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include using std::vector; #include using std::map; #include using std::set; #include using std::queue; #include #include "constants.h" #include "stacks.h" #include "kmers.h" #include "locus.h" #include "GappedAln.h" #include "sql_utilities.h" #include "aln_utils.h" #include "utils.h" void help( void ); void version( void ); int parse_command_line(int, char**); int initialize_new_catalog(pair &, map &); int initialize_existing_catalog(string, map &); int update_catalog_index(map &, map &); int find_kmer_matches_by_sequence(map &, map &, int); int search_for_gaps(map &, map &, double, double); int find_matches_by_sequence(map &, map &); int find_matches_by_genomic_loc(map &, map &); int characterize_mismatch_snps(CLocus *, QLocus *); int merge_allele(Locus *, SNP *); int merge_matches(map &, map &, pair &, int, uint &, uint &, uint &, uint &); int add_unique_tag(pair &, map &, QLocus *); bool compare_dist(pair, pair); int write_catalog(map &); int write_simple_output(CLocus *, ofstream &, ofstream &, ofstream &); int write_gzip_output(CLocus *, gzFile &, gzFile &, gzFile &); bool compare_matches(Match *, Match *); int populate_kmer_hash(map &, CatKmerHashMap &, vector &, int); #endif // __CSTACKS_H__ stacks-1.44/src/DNANSeq.cc000644 000765 000024 00000010444 12775533564 016015 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2011, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // DNANSeq.cc // // Julian Catchen // jcatchen@uoregon.edu // University of Oregon // // $Id: DNANSeq.cc 2133 2011-06-07 04:07:41Z catchen $ // #include "DNANSeq.h" DNANSeq::DNANSeq(int size) { int bytes; this->bits = size * bits_per_nuc; bytes = BITNSLOTS(this->bits); this->s = new unsigned char[bytes]; memset(this->s, 0, bytes); } DNANSeq::DNANSeq(int size, unsigned char *seq) { unsigned int bytes; this->bits = size * bits_per_nuc; bytes = BITNSLOTS(this->bits); this->s = new unsigned char[bytes]; for (unsigned int i = 0; i < bytes; i++) this->s[i] = seq[i]; } DNANSeq::DNANSeq(int size, const char *seq) { this->bits = size * bits_per_nuc; int bytes = BITNSLOTS(this->bits); this->s = new unsigned char[bytes]; memset(this->s, 0, bytes); int bit = 0; for (int i = 0; i < size; i++) { switch (seq[i]) { case 'A': case 'a': // A == 000 bit += 3; break; case 'C': case 'c': // C == 001 bit += 2; BITSET(this->s, bit); bit++; break; case 'G': case 'g': // G == 010 bit++; BITSET(this->s, bit); bit++; bit++; break; case 'T': case 't': // T == 011 bit++; BITSET(this->s, bit); bit++; BITSET(this->s, bit); bit++; break; case 'N': case 'n': case '.': // N == 100 BITSET(this->s, bit); bit += 3; break; } } } DNANSeq::DNANSeq(const DNANSeq& other) : bits(other.bits) { const int n_bytes = BITNSLOTS(bits); s = new unsigned char[n_bytes]; memcpy(s, other.s, n_bytes); } DNANSeq& DNANSeq::operator=(const DNANSeq& other) { delete[] s; bits = other.bits; const int n_bytes = BITNSLOTS(bits); s = new unsigned char[n_bytes]; memcpy(s, other.s, n_bytes); return *this; } DNANSeq::~DNANSeq() { delete [] this->s; } char DNANSeq::operator[](int pos) { unsigned char c, base; int bit; if (pos > ((this->bits / bits_per_nuc) - 1)) return '\0'; bit = pos * bits_per_nuc; c = 0; base = 'X'; for (int i = bits_per_nuc - 1; i >= 0; i--) { if (BITTEST(this->s, bit)) c |= 1 << i; bit++; } switch (c) { case 0: base = 'A'; break; case 1: base = 'C'; break; case 2: base = 'G'; break; case 3: base = 'T'; break; case 4: base = 'N'; break; default: cerr << "Unknown character " << (int) c << "\n"; break; } //cerr << " Decoding character " << pos << ", '" << base << "'\n"; return base; } int DNANSeq::size() const { return this->bits / bits_per_nuc; } char *DNANSeq::subseq(char *seq, int start, int end) { int i; for (i = start; i <= end; i++) seq[i - start] = this->operator[](i); seq[i - start] = '\0'; return seq; } char *DNANSeq::seq(char *seq) { int i; int end = this->bits / bits_per_nuc; for (i = 0; i < end; i++) seq[i] = this->operator[](i); seq[i] = '\0'; return seq; } char *DNANSeq::seq() { int i; int size = this->bits / bits_per_nuc; char *seq = new char[size + 1]; for (i = 0; i < size; i++) seq[i] = this->operator[](i); seq[i] = '\0'; return seq; } stacks-1.44/src/DNANSeq.h000644 000765 000024 00000007645 12775533564 015670 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2011, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __DNANSeq_H__ #define __DNANSeq_H__ #include #include #include //std::hash #ifdef HAVE_SPARSEHASH #include #endif #define BITMASK(b) (1 << ((b) % CHAR_BIT)) #define BITSLOT(b) ((b) / CHAR_BIT) #define BITSET(a, b) ((a)[BITSLOT(b)] |= BITMASK(b)) #define BITCLEAR(a, b) ((a)[BITSLOT(b)] &= ~BITMASK(b)) #define BITTEST(a, b) ((a)[BITSLOT(b)] & BITMASK(b)) #define BITNSLOTS(nb) ((nb + CHAR_BIT - 1) / CHAR_BIT) // // We expect (and C++ defines) an unsigned char as 8 bits. // const unsigned short int bits_per_nuc = 3; const unsigned short int byte_size = 8; // // DNA Sequence Storage Class // // Three-bit compression, 2.667 bases per byte of storage: // A == 000 // C == 001 // G == 010 // T == 011 // N == 100 // class DNANSeq { public: // // The number of bits required to store string of DNA string // unsigned short int bits; // // Array of bytes to store DNA sequence. // unsigned char *s; #ifdef HAVE_SPARSEHASH DNANSeq() : bits(3), s(new unsigned char[1]) { *s = (1<<2); } // "N" #endif DNANSeq(int); DNANSeq(int, const char *); DNANSeq(int, unsigned char *); DNANSeq(const DNANSeq& other); DNANSeq& operator=(const DNANSeq& other); ~DNANSeq(); char operator[](int); int size() const; char *seq(char *); char *seq(); char *subseq(char *, int, int); bool operator== (const DNANSeq& other) const { if (bits != other.bits) return false; unsigned int bytes = BITNSLOTS(bits); for (unsigned int i = 0; i < bytes; i++) if (s[i] != other.s[i]) return false; return true; } }; #include #include #include using std::stringstream; using std::cin; using std::cout; using std::cerr; // Specialization for std::hash // Based on GCC namespace std { template<> struct hash { size_t operator()(const DNANSeq& seq) const { size_t __result = static_cast(14695981039346656037ULL); unsigned short int __bytes = BITNSLOTS(seq.bits); for (unsigned short int i = 0; i < __bytes; i++) { __result ^= static_cast(seq.s[i]); __result *= static_cast(1099511628211ULL); } return __result; } }; #ifdef HAVE_SPARSEHASH namespace tr1 { template<> struct hash { size_t operator()(const DNANSeq& seq) const { return std::hash()(seq); } }; } #endif } // namespace __gnu_cxx { // template<> // struct hash // { // size_t // operator()(DNANSeq *__s) const { // unsigned long __h = 0; // unsigned int bytes = BITNSLOTS(__s->bits); // for (unsigned int i = 0; i < bytes; i++) // __h = 5 * __h + __s->s[i]; // return size_t(__h); // } // }; // } /* struct hash_dnanseq { // Deprecated size_t operator()(const DNANSeq* seq) const {return std::hash{}(*seq);} }; struct dnanseq_eqstr { // Deprecated bool operator()(const DNANSeq *s1, const DNANSeq *s2) const {return *s1 == *s2;} }; */ #endif // __DNANSeq_H__ stacks-1.44/src/DNASeq.cc000644 000765 000024 00000016077 12761620410 015665 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // DNASeq.cc // // Julian Catchen // jcatchen@uoregon.edu // University of Oregon // // $Id: DNASeq.cc 2133 2011-06-07 04:07:41Z catchen $ // #include "DNASeq.h" DNASeq::DNASeq(int size) { int bytes; this->size = size; bytes = size / bases_per_byte; bytes += size % bases_per_byte > 0 ? 1 : 0; this->s = new unsigned char[bytes]; memset(this->s, 0, bytes); } DNASeq::DNASeq(int size, unsigned char *seq) { unsigned int bytes; this->size = size; bytes = size / bases_per_byte; bytes += size % bases_per_byte > 0 ? 1 : 0; this->s = new unsigned char[bytes]; for (unsigned int i = 0; i < bytes; i++) this->s[i] = seq[i]; } DNASeq::DNASeq(int size, const char *seq) { int bytes, rem; this->size = size; bytes = size / bases_per_byte; rem = size % bases_per_byte; bytes += rem > 0 ? 1 : 0; this->s = new unsigned char[bytes]; int index = 0; for (int i = 0; i < this->size; i++) { //cerr << "Encoding character " << i << ", '" << seq[i] << "'\n"; if (i > 0 && i % bases_per_byte == 0) index++; //cerr << " encoding '" << seq[i] << "' into byte " << index << ".\n"; this->s[index] <<= 2; switch (seq[i]) { case 'A': case 'a': // A == 00 break; case 'C': case 'c': // C == 01 this->s[index] |= 0x1; break; case 'G': case 'g': // G == 10 this->s[index] |= 0x2; break; case 'T': case 't': // T == 11 this->s[index] |= 0x3; break; } //cerr << " s[" << index << "," << i % bases_per_byte << "] == " << (int)this->s[index] << "\n"; } if (rem > 0) this->s[index] <<= (bases_per_byte - rem) * 2; } DNASeq::~DNASeq() { delete [] this->s; } char DNASeq::operator[](int pos) { unsigned char c, base; int index, rem; if (pos > (this->size - 1)) return '\0'; index = pos / bases_per_byte; rem = pos % bases_per_byte; //cerr << "s[" << index << "," << rem << "] == " << (int)this->s[index] << "\n"; switch (rem) { case 0: c = this->s[index] & 0xC0; // 11000000 c >>= 6; break; case 1: c = this->s[index] & 0x30; // 00110000 c >>= 4; break; case 2: c = this->s[index] & 0xC; // 00001100 c >>= 2; break; case 3: c = this->s[index] & 0x3; // 00000011 break; } switch (c) { case 0: base = 'A'; break; case 1: base = 'C'; break; case 2: base = 'G'; break; case 3: base = 'T'; break; } //cerr << " Decoding character " << pos << ", '" << base << "'\n"; return base; } char *DNASeq::subseq(char *seq, int start, int end) { unsigned char c; int i, index, rem; i = start; index = i / bases_per_byte; rem = i % bases_per_byte; for (; i <= end; i++) { rem = i % bases_per_byte; if (i > 0 && rem == 0) index++; //cerr << "s[" << index << "," << rem << "] == " << (int)this->s[index] << "\n"; switch (rem) { case 0: c = this->s[index] & 0xC0; // 11000000 c >>= 6; break; case 1: c = this->s[index] & 0x30; // 00110000 c >>= 4; break; case 2: c = this->s[index] & 0xC; // 00001100 c >>= 2; break; case 3: c = this->s[index] & 0x3; // 00000011 break; } switch (c) { case 0: seq[i - start] = 'A'; break; case 1: seq[i - start] = 'C'; break; case 2: seq[i - start] = 'G'; break; case 3: seq[i - start] = 'T'; break; } //cerr << " Decoding character " << i << ", '" << seq[i - start] << "'\n"; } seq[i - start] = '\0'; return seq; } char *DNASeq::seq(char *seq) { unsigned char c; int i; int index = 0; for (i = 0; i < this->size; i++) { if (i > 0 && i % bases_per_byte == 0) index++; //cerr << "s[" << index << "," << i % bases_per_byte << "] == " << (int)this->s[index] << "\n"; switch (i % bases_per_byte) { case 0: c = this->s[index] & 0xC0; // 11000000 c >>= 6; break; case 1: c = this->s[index] & 0x30; // 00110000 c >>= 4; break; case 2: c = this->s[index] & 0xC; // 00001100 c >>= 2; break; case 3: c = this->s[index] & 0x3; // 00000011 break; } switch (c) { case 0: seq[i] = 'A'; break; case 1: seq[i] = 'C'; break; case 2: seq[i] = 'G'; break; case 3: seq[i] = 'T'; break; } //cerr << " Decoding character " << i << ", '" << seq[i] << "'\n"; } seq[i] = '\0'; return seq; } char *DNASeq::seq() { unsigned char c; int i; int index = 0; char *seq = new char[this->size + 1]; for (i = 0; i < this->size; i++) { if (i > 0 && i % bases_per_byte == 0) index++; //cerr << "s[" << index << "," << i % bases_per_byte << "] == " << (int)this->s[index] << "\n"; switch (i % bases_per_byte) { case 0: c = this->s[index] & 0xC0; // 11000000 c >>= 6; break; case 1: c = this->s[index] & 0x30; // 00110000 c >>= 4; break; case 2: c = this->s[index] & 0xC; // 00001100 c >>= 2; break; case 3: c = this->s[index] & 0x3; // 00000011 break; } switch (c) { case 0: seq[i] = 'A'; break; case 1: seq[i] = 'C'; break; case 2: seq[i] = 'G'; break; case 3: seq[i] = 'T'; break; } //cerr << " Decoding character " << i << ", '" << seq[i] << "'\n"; } seq[i] = '\0'; return seq; } stacks-1.44/src/DNASeq.h000644 000765 000024 00000005327 12764047443 015537 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2011, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __DNASeq_H__ #define __DNASeq_H__ #include #include // #ifdef __GNUC__ // #include // using __gnu_cxx::hash_map; // using __gnu_cxx::hash; // #else // #include // #endif // // We expect (and C++ defines) an unsigned char as 8 bits, so we // should be able to store 4 nucleotide bases per byte of memory. // const unsigned short int bases_per_byte = CHAR_BIT / 2; // // DNA Sequence Storage Class // // Two-bit compression, four bases per byte of storage: // A == 00 // C == 01 // G == 10 // T == 11 // class DNASeq { public: // // The number of DNA bases we are storing // unsigned short int size; // // Array of bytes to store DNA sequence, one character per two bits, four per byte. // unsigned char *s; DNASeq(int); DNASeq(int, const char *); DNASeq(int, unsigned char *); ~DNASeq(); char len() { return this->size; } char operator[](int); char *seq(char *); char *seq(); char *subseq(char *, int, int); }; #include #include #include using std::stringstream; using std::cin; using std::cout; using std::cerr; struct hash_dnaseq { size_t operator()(DNASeq *__s) const { size_t __result = static_cast(14695981039346656037ULL); unsigned short int __bytes = (__s->size / bases_per_byte) + (__s->size % bases_per_byte > 0 ? 1 : 0); for (unsigned short int i = 0; i < __bytes; i++) { __result ^= static_cast(__s->s[i]); __result *= static_cast(1099511628211ULL); } return __result; } }; struct dnaseq_eqstr { bool operator()(DNASeq *s1, DNASeq *s2) const { unsigned int bytes = (s1->size / bases_per_byte) + (s1->size % bases_per_byte > 0 ? 1 : 0); for (unsigned int i = 0; i < bytes; i++) if (s1->s[i] != s2->s[i]) return false; return true; } }; #endif // __DNASeq_H__ stacks-1.44/src/estacks.cc000644 000765 000024 00000051551 12764047443 016257 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // estacks -- search an existing set of stacks for polymorphisms // #include "estacks.h" // // Global variables to hold command-line options. // FileT in_file_type; string in_file; string out_path; bool record_hom = false; int sql_id = 0; int min_stack_cov = 1; int num_threads = 1; // // For use with the multinomial model to call fixed nucleotides. // modelt model_type = snp; double bound_low = 0.0; double bound_high = 1.0; double p_freq = 0.5; double barcode_err_freq = 0.0; double heterozygote_limit = -3.84; double homozygote_limit = 3.84; int main (int argc, char* argv[]) { parse_command_line(argc, argv); // // Set the number of OpenMP parallel threads to execute. // #ifdef _OPENMP omp_set_num_threads(num_threads); #endif HashMap radtags; set merge_map; map unique; load_radtags(in_file, radtags); reduce_radtags(radtags, unique); //dump_stacks(unique); map merged; populate_merged_tags(unique, merged); //dump_merged_stacks(merged); // Call the consensus sequence again, now that remainder tags have been merged. call_consensus(merged, unique, true); count_raw_reads(unique, merged); cerr << "Writing results\n"; write_sql(merged, unique); return 0; } int call_alleles(MergedStack *mtag, vector &reads) { int row; int height = reads.size(); string allele; char base; vector::iterator snp; DNANSeq *d; if (mtag->snps.size() == 0) return 0; for (row = 0; row < height; row++) { allele.clear(); bool haplotype = true; for (snp = mtag->snps.begin(); snp != mtag->snps.end(); snp++) { d = reads[row]; base = (*d)[(*snp)->col]; // // Check to make sure the nucleotide at the location of this SNP is // of one of the two possible states the multinomial model called. // if (base == (*snp)->rank_1 || base == (*snp)->rank_2) allele += base; else haplotype = false; } if (haplotype && allele.size() == mtag->snps.size()) mtag->alleles[allele]++; } return 0; } int call_consensus(map &merged, map &unique, bool invoke_model) { // // OpenMP can't parallelize random access iterators, so we convert // our map to a vector of integer keys. // map::iterator it; vector keys; for (it = merged.begin(); it != merged.end(); it++) keys.push_back(it->first); int i; #pragma omp parallel private(i) { #pragma omp for schedule(dynamic) for (i = 0; i < (int) keys.size(); i++) { MergedStack *mtag; PStack *utag; mtag = merged[keys[i]]; // // Create a two-dimensional array, each row containing one read. For // each unique tag that has been merged together, add the sequence for // that tag into our array as many times as it originally occurred. // vector::iterator j; vector reads; for (j = mtag->utags.begin(); j != mtag->utags.end(); j++) { utag = unique[*j]; for (uint k = 0; k < utag->count; k++) { reads.push_back(utag->seq); } } // // Iterate over each column of the array and call the consensus base. // int row, col; int length = reads[0]->size(); int height = reads.size(); string con; map nuc; map::iterator max, n; DNANSeq *d; for (col = 0; col < length; col++) { nuc['A'] = 0; nuc['C'] = 0; nuc['G'] = 0; nuc['T'] = 0; for (row = 0; row < height; row++) { d = reads[row]; nuc[(*d)[col]]++; } // // Find the base with a plurality of occurances and call it. // max = nuc.end(); for (n = nuc.begin(); n != nuc.end(); n++) { if (max == nuc.end() || n->second > max->second) max = n; } con += max->second == 0 ? 'N' : max->first; // Search this column for the presence of a SNP if (invoke_model) model_type == snp ? call_multinomial_snp(mtag, col, nuc, record_hom) : call_multinomial_fixed(mtag, col, nuc); } if (invoke_model) { call_alleles(mtag, reads); if (model_type == fixed) { // // Mask nucleotides that are not fixed. // vector::iterator s; for (s = mtag->snps.begin(); s != mtag->snps.end(); s++) { con.replace((*s)->col, 1, "N"); } } } mtag->add_consensus(con.c_str()); } } return 0; } int count_raw_reads(map &unique, map &merged) { map::iterator it; vector::iterator k; PStack *tag; long int m = 0; for (it = merged.begin(); it != merged.end(); it++) { for (k = it->second->utags.begin(); k != it->second->utags.end(); k++) { tag = unique[*k]; m += tag->count; } m += it->second->remtags.size(); } cerr << " Number of utilized reads " << m << "\n"; return 0; } int write_sql(map &m, map &u) { map::iterator i; vector::iterator j; vector::iterator k; vector::iterator s; map::iterator t; MergedStack *tag_1; PStack *tag_2; // // Parse the input file name to create the output files // size_t pos_1 = in_file.find_last_of("/"); size_t pos_2 = in_file.find_last_of("."); string tag_file = out_path + in_file.substr(pos_1 + 1, (pos_2 - pos_1 - 1)) + ".tags.tsv"; string snp_file = out_path + in_file.substr(pos_1 + 1, (pos_2 - pos_1 - 1)) + ".snps.tsv"; string all_file = out_path + in_file.substr(pos_1 + 1, (pos_2 - pos_1 - 1)) + ".alleles.tsv"; string pil_file = out_path + in_file.substr(pos_1 + 1, (pos_2 - pos_1 - 1)) + ".pileup.tsv"; // Open the output files for writing. std::ofstream tags(tag_file.c_str()); std::ofstream snps(snp_file.c_str()); std::ofstream alle(all_file.c_str()); std::ofstream pile(pil_file.c_str()); int tag_id, comp_id; tag_id = 0; for (i = m.begin(); i != m.end(); i++) { tag_1 = i->second; // First write the consensus sequence for (s = tag_1->snps.begin(); s != tag_1->snps.end(); s++) { float total = 0; for (k = tag_1->utags.begin(); k != tag_1->utags.end(); k++) { //if (u[*k]->seq[(*s)->col] == 'N') continue; total += u[*k]->count; } if (total < min_stack_cov) continue; tags << "0" << "\t" << sql_id << "\t" << tag_id << "\t" << tag_1->loc.chr << "\t" << tag_1->loc.bp + (*s)->col << "\t" << "consensus\t" << "\t\t" << tag_1->con[(*s)->col] << "\t" << tag_1->deleveraged << "\t" << tag_1->blacklisted << "\t" << tag_1->lumberjackstack << "\n"; // Now write out the components of each unique tag merged into this one. comp_id = 0; for (k = tag_1->utags.begin(); k != tag_1->utags.end(); k++) { tag_2 = u[*k]; //if (tag_2->seq[(*s)->col] == 'N') continue; for (j = tag_2->map.begin(); j != tag_2->map.end(); j++) { tags << "0" << "\t" << sql_id << "\t" << tag_id << "\t\t\t" << "primary\t" << comp_id << "\t" << *j << "\t" << (*tag_2->seq)[(*s)->col] << "\t\t\t\n"; } comp_id++; } snps << "0" << "\t" << sql_id << "\t" << tag_id << "\t" << 0 << "\t" << (*s)->lratio << "\t" << (*s)->rank_1 << "\t" << (*s)->rank_2 << "\n"; // Write the expressed alleles seen for the recorded SNPs and // the percentage of tags a particular allele occupies. map allele; for (k = tag_1->utags.begin(); k != tag_1->utags.end(); k++) { if ((*u[*k]->seq)[(*s)->col] != (*s)->rank_1 && (*u[*k]->seq)[(*s)->col] != (*s)->rank_2) continue; allele[(*u[*k]->seq)[(*s)->col]] += u[*k]->count; } char pct[id_len]; map::iterator a; for (a = allele.begin(); a != allele.end(); a++) { sprintf(pct, "%.2f", ((a->second/total) * 100)); alle << "0" << "\t" << sql_id << "\t" << tag_id << "\t" << a->first << "\t" << pct << "\t" << a->second << "\n"; } tag_id++; } } for (i = m.begin(); i != m.end(); i++) { tag_1 = i->second; float total = 0; for (k = tag_1->utags.begin(); k != tag_1->utags.end(); k++) total += u[*k]->count; if (total < min_stack_cov) continue; // First write the consensus sequence pile << "0" << "\t" << sql_id << "\t" << tag_1->id << "\t" << tag_1->loc.chr << "\t" << tag_1->loc.bp << "\t" << "consensus\t" << "\t\t" << tag_1->con << "\n"; // Now write out the components of each unique tag merged into this one. comp_id = 0; for (k = tag_1->utags.begin(); k != tag_1->utags.end(); k++) { tag_2 = u[*k]; for (j = tag_2->map.begin(); j != tag_2->map.end(); j++) { pile << "0" << "\t" << sql_id << "\t" << tag_1->id << "\t\t\t" << "primary\t" << comp_id << "\t" << *j << "\t" << tag_2->seq << "\t\t\t\n"; } comp_id++; } } tags.close(); snps.close(); alle.close(); pile.close(); return 0; } int populate_merged_tags(map &unique, map &merged) { map::iterator i; map::iterator it_new, it_old; map > locations; map >::iterator k; set::iterator s; char id[id_len]; PStack *u; MergedStack *m; int global_id = 0; // // Create a map of each unique Stack that has been aligned to the same genomic location. // for (i = unique.begin(); i != unique.end(); i++) { snprintf(id, id_len - 1, "%s_%d", i->second->loc.chr, i->second->loc.bp); locations[id].insert(i->second->id); } it_old = merged.begin(); for (k = locations.begin(); k != locations.end(); k++) { m = new MergedStack; m->id = global_id; // // Record the consensus and physical location for this stack. // s = k->second.begin(); m->add_consensus(unique[*s]->seq); strncpy(m->loc.chr, unique[*s]->loc.chr, id_len - 1); m->loc.chr[id_len] = '\0'; m->loc.bp = unique[*s]->loc.bp; // // Record the individual stacks that were aligned together. // for (; s != k->second.end(); s++) { u = unique[*s]; m->count += u->count; m->utags.push_back(u->id); } // Insert the new MergedStack giving a hint as to which position // to insert it at. it_new = merged.insert(it_old, pair(global_id, m)); it_old = it_new; global_id++; } cerr << " Merged " << unique.size() << " unique Stacks into " << merged.size() << " loci.\n"; return 0; } // // This function assumes that there may be identical reads, mapped to multiple // places in the genome. In this case, reads are broken down by read ID // and split into different Stack objects. // int reduce_radtags(HashMap &radtags, map &unique) { HashMap::iterator it; vector::iterator sit; PStack *u; int global_id = 1; for (it = radtags.begin(); it != radtags.end(); it++) { // // Make sure there aren't any reads of identical sequence that have been mapped to // different genomic locations. // map locations; map::iterator lit; for (sit = (*it).second.begin(); sit != (*it).second.end(); sit++) locations[(*sit)->loc_str]++; for (lit = locations.begin(); lit != locations.end(); lit++) { // // Populate a PStack object for this unique radtag. // u = new PStack; u->id = global_id; u->count = lit->second; u->add_seq(it->first); // // Record the physical location of this stack. // for (sit = (*it).second.begin(); sit != (*it).second.end(); sit++) { if (strcmp((*sit)->loc_str, lit->first.c_str()) == 0) { u->add_id((*sit)->id); u->loc.set((*sit)->loc.chr, (*sit)->loc.bp, (*sit)->loc.strand); } } unique[u->id] = u; global_id++; } } return 0; } // // We expect tags to have already been aligned to a reference genome. Therefore, the tags // are identified by their chromosome and basepair location. // int load_radtags(string in_file, HashMap &radtags) { Input *fh = NULL; Seq *c; if (in_file_type == FileT::bowtie) fh = new Bowtie(in_file.c_str()); else if (in_file_type == FileT::sam) fh = new Sam(in_file.c_str()); else if (in_file_type == FileT::tsv) fh = new Tsv(in_file.c_str()); cerr << "Parsing " << in_file.c_str() << "\n"; int i = 1; while ((c = fh->next_seq()) != NULL) { if (i % 10000 == 0) cerr << "Loading aligned sequence " << i << " \r"; radtags[c->seq].push_back(c); i++; } if (i == 0) { cerr << "Error: Unable to load data from '" << in_file.c_str() << "'.\n"; exit(1); } cerr << " " << "Analyzed " << i - 1 << " sequence reads; " << "Identified " << radtags.size() << " unique Stacks from those reads.\n"; // // Close the file and delete the Input object. // delete fh; return 0; } int dump_stacks(map &u) { map::iterator it; vector::iterator fit; vector >::iterator pit; vector::iterator mit; for (it = u.begin(); it != u.end(); it++) { cerr << "Stack ID: " << (*it).second->id << "\n" << " Seq: " << (*it).second->seq << "\n" << " IDs: "; for (fit = (*it).second->map.begin(); fit != (*it).second->map.end(); fit++) cerr << *fit << " "; cerr << "\n\n"; } return 0; } int dump_merged_stacks(map &m) { map::iterator it; vector >::iterator pit; vector::iterator fit; for (it = m.begin(); it != m.end(); it++) { cerr << "MergedStack ID: " << it->second->id << "\n" << " Consensus: "; if (it->second->con != NULL) cerr << it->second->con << "\n"; else cerr << "\n"; cerr << " IDs: "; for (fit = it->second->utags.begin(); fit != it->second->utags.end(); fit++) cerr << (*fit) << " "; cerr << "\n" << " Distances: "; for (pit = it->second->dist.begin(); pit != it->second->dist.end(); pit++) cerr << (*pit).first << ": " << (*pit).second << ", "; cerr << "\n\n"; } return 0; } int parse_command_line(int argc, char* argv[]) { int c; while (1) { static struct option long_options[] = { {"help", no_argument, NULL, 'h'}, {"version", no_argument, NULL, 'v'}, {"rec_hom", no_argument, NULL, 'O'}, {"infile_type", required_argument, NULL, 't'}, {"outfile_type", required_argument, NULL, 'y'}, {"file", required_argument, NULL, 'f'}, {"outpath", required_argument, NULL, 'o'}, {"id", required_argument, NULL, 'i'}, {"min_cov", required_argument, NULL, 'm'}, {"num_threads", required_argument, NULL, 'p'}, {"bc_err_freq", required_argument, NULL, 'e'}, {0, 0, 0, 0} }; // getopt_long stores the option index here. int option_index = 0; c = getopt_long(argc, argv, "hvOf:o:i:e:p:m:s:f:t:", long_options, &option_index); // Detect the end of the options. if (c == -1) break; switch (c) { case 'h': help(); break; case 't': if (strcmp(optarg, "bowtie") == 0) in_file_type = FileT::bowtie; else if (strcmp(optarg, "sam") == 0) in_file_type = FileT::sam; else if (strcmp(optarg, "tsv") == 0) in_file_type = FileT::tsv; else in_file_type = FileT::unknown; break; case 'f': in_file = optarg; break; case 'o': out_path = optarg; break; case 'i': sql_id = atoi(optarg); break; case 'm': min_stack_cov = atoi(optarg); break; case 'e': barcode_err_freq = atof(optarg); break; case 'p': num_threads = atoi(optarg); break; case 'O': record_hom = true; break; case 'v': version(); break; case '?': // getopt_long already printed an error message. help(); break; default: cerr << "Unknown command line option '" << (char) c << "'\n"; help(); abort(); } } if (in_file.length() == 0 || in_file_type == FileT::unknown) { cerr << "You must specify an input file of a supported type.\n"; help(); } if (out_path.length() == 0) out_path = "."; if (out_path.at(out_path.length() - 1) != '/') out_path += "/"; if (model_type == fixed && barcode_err_freq == 0) { cerr << "You must specify the barcode error frequency.\n"; help(); } return 0; } void version() { std::cerr << "estacks " << VERSION << "\n\n"; exit(0); } void help() { std::cerr << "estacks " << VERSION << "\n" << "estacks -t file_type -f file_path [-o path] [-i id] [-m min_cov] [-r] [-e errfreq] [-p num_threads] [-h]" << "\n" << " p: enable parallel execution with num_threads threads.\n" << " t: input file Type. Supported types: bowtie, sam.\n" << " f: input file path.\n" << " o: output path to write results.\n" << " i: SQL ID to insert into the output to identify this sample.\n" << " O: record homozygotes along with heterozygote SNPs.\n" << " m: minimum depth of coverage to report a stack (default 1).\n" << " e: specify the barcode error frequency (0 < e < 1) if using the 'fixed' model.\n" << " h: display this help messsage." << "\n\n"; exit(0); } stacks-1.44/src/estacks.h000644 000765 000024 00000005546 12764047443 016124 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __PSTACKS_H__ #define __PSTACKS_H__ #ifdef _OPENMP #include // OpenMP library #endif #include // Process command-line options #include #include #include #include using std::stringstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include using std::vector; #include using std::map; #include using std::set; #include using std::pair; #include using std::unordered_map; #ifdef HAVE_SPARSEHASH #include using google::sparse_hash_map; #endif #include "constants.h" #include "stacks.h" // Major data structures for holding stacks #include "kmers.h" #include "mstack.h" #include "utils.h" #include "models.h" // Contains maximum likelihood statistical models. #include "Tsv.h" // Reading input files in Tab-separated values format #include "BowtieI.h" // Reading input files in Bowtie format #include "SamI.h" // Reading input files in SAM format #include "FastaI.h" // Reading input files in FASTA format #include "FastqI.h" // Reading input files in FASTQ format #include "DNANSeq.h" const int barcode_size = 5; #ifdef HAVE_SPARSEHASH typedef sparse_hash_map, hash_charptr, eqstr> HashMap; #else typedef unordered_map, hash_charptr, eqstr> HashMap; #endif void help( void ); void version( void ); int parse_command_line(int, char**); int load_radtags(string, HashMap &); int reduce_radtags(HashMap &, map &); int populate_merged_tags(map &, map &); int call_consensus(map &, map &, bool); int call_alleles(MergedStack *, vector &); int count_raw_reads(map &, map &); int write_sql(map &, map &); int write_sam(map &, map &); // // Debugging // int dump_stacks(map &); int dump_merged_stacks(map &); #endif // __ESTACKS_H__ stacks-1.44/src/export_formats.cc000644 000765 000024 00000337704 12764047443 017705 0ustar00catchenstaff000000 000000 #include #include #include "ordered.h" #include "sql_utilities.h" #include "MetaPopInfo.h" #include "export_formats.h" using namespace std; extern InputMode input_mode; extern int batch_id; extern string in_path; extern string out_path; extern string out_prefix; extern bool phylip_var; extern bool loci_ordered; extern bool merge_sites; extern string enz; extern set debug_flags; extern MetaPopInfo mpopi; extern map renz_olap; int write_sql(map &catalog, PopMap *pmap) { string file = out_path + out_prefix + ".markers.tsv"; cerr << "Writing SQL markers file to '" << file << "'\n"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening markers SQL file '" << file << "'\n"; exit(1); } fh.precision(fieldw); fh.setf(std::ios::fixed); fh << "# SQL ID" << "\t" << "Batch ID" << "\t" << "Catalog Locus ID" << "\t" << "\t" << "Total Genotypes" << "\t" << "Max" << "\t" << "Genotype Freqs" << "\t" << "F" << "\t" << "Mean Log Likelihood" << "\t" << "Genotype Map" << "\t" << "\n"; map::iterator it; CSLocus *loc; stringstream gtype_map; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; string freq = ""; double max = 0.0; int total = 0; gtype_map.str(""); if (loc->marker.length() > 0) { tally_haplotype_freq(loc, pmap, total, max, freq); // // Record the haplotype to genotype map. // map::iterator j; for (j = loc->gmap.begin(); j != loc->gmap.end(); j++) gtype_map << j->first << ":" << j->second << ";"; } fh << 0 << "\t" << batch_id << "\t" << loc->id << "\t" << "\t" // Marker << total << "\t" << max << "\t" << freq << "\t" << loc->f << "\t" << loc->lnl << "\t" << gtype_map.str() << "\t" << "\n"; } fh.close(); return 0; } int write_fasta(map &catalog, PopMap *pmap) { // // Write a FASTA file containing each allele from each locus from // each sample in the population. // string file = out_path + out_prefix + ".fa"; cerr << "Writing population alleles to FASTA file '" << file << "'\n"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening FASTA file '" << file << "'\n"; exit(1); } map >::iterator it; CSLocus *loc; Datum **d; char *seq; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; d = pmap->locus(loc->id); seq = new char[loc->len + 1]; strcpy(seq, loc->con); for (int j = 0; j < pmap->sample_cnt(); j++) { if (d[j] == NULL) continue; for (uint k = 0; k < d[j]->obshap.size(); k++) { for (uint i = 0; i < loc->snps.size(); i++) { uint col = loc->snps[i]->col; seq[col] = col < loc->len ? d[j]->obshap[k][i] : loc->con[col]; } fh << ">CLocus_" << loc->id << "_Sample_" << pmap->rev_sample_index(j) << "_Locus_" << d[j]->id << "_Allele_" << k << " [" << mpopi.samples()[j].name; if (strcmp(loc->loc.chr, "un") != 0) fh << "; " << loc->loc.chr << ", " << loc->sort_bp() + 1 << ", " << (loc->loc.strand == strand_plus ? "+" : "-"); fh << "]\n" << seq << "\n"; } } delete [] seq; } } fh.close(); return 0; } int write_strict_fasta(map &catalog, PopMap *pmap) { // // Write a FASTA file containing each allele from each locus from // each sample in the population. // string file = out_path + out_prefix + ".strict.fa"; cerr << "Writing strict population alleles to FASTA file '" << file << "'\n"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening strict FASTA file '" << file << "'\n"; exit(1); } map >::iterator it; CSLocus *loc; Datum **d; char *seq; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; d = pmap->locus(loc->id); seq = new char[loc->len + 1]; strcpy(seq, loc->con); for (int j = 0; j < pmap->sample_cnt(); j++) { if (d[j] == NULL) continue; if (d[j]->obshap.size() > 2) continue; if (d[j]->obshap.size() == 1) { for (uint i = 0; i < loc->snps.size(); i++) { uint col = loc->snps[i]->col; seq[col] = col < loc->len ? d[j]->obshap[0][i] : loc->con[col]; } fh << ">CLocus_" << loc->id << "_Sample_" << pmap->rev_sample_index(j) << "_Locus_" << d[j]->id << "_Allele_" << 0 << " [" << mpopi.samples()[j].name; if (strcmp(loc->loc.chr, "un") != 0) fh << "; " << loc->loc.chr << ", " << loc->sort_bp() + 1 << ", " << (loc->loc.strand == strand_plus ? "+" : "-"); fh << "]\n" << seq << "\n"; fh << ">CLocus_" << loc->id << "_Sample_" << pmap->rev_sample_index(j) << "_Locus_" << d[j]->id << "_Allele_" << 1 << " [" << mpopi.samples()[j].name; if (strcmp(loc->loc.chr, "un") != 0) fh << "; " << loc->loc.chr << ", " << loc->sort_bp() + 1 << ", " << (loc->loc.strand == strand_plus ? "+" : "-"); fh << "]\n" << seq << "\n"; } else { for (uint k = 0; k < d[j]->obshap.size(); k++) { for (uint i = 0; i < loc->snps.size(); i++) { uint col = loc->snps[i]->col; seq[col] = col < loc->len ? d[j]->obshap[k][i] : loc->con[col]; } fh << ">CLocus_" << loc->id << "_Sample_" << mpopi.samples()[j].id << "_Locus_" << d[j]->id << "_Allele_" << k << " [" << mpopi.samples()[j].name; if (strcmp(loc->loc.chr, "un") != 0) fh << "; " << loc->loc.chr << ", " << loc->sort_bp() + 1 << ", " << (loc->loc.strand == strand_plus ? "+" : "-"); fh << "]\n" << seq << "\n"; } } } delete [] seq; } } fh.close(); return 0; } int write_vcf_ordered(map &catalog, PopMap *pmap, PopSum *psum, map > &merge_map, ofstream &log_fh) { // // Write a VCF file as defined here: http://www.1000genomes.org/node/101 // Order SNPs by genomic position (and handle overlapping loci). // string file = out_path + out_prefix + ".vcf"; cerr << "Writing ordered population data to VCF file '" << file << "'\n"; log_fh << "\n#\n# Generating SNP-based VCF export.\n#\n"; VcfWriter writer (file); if (writer.fail()) { cerr << "Error opening VCF file '" << file << "'\n"; exit(-1); } bool gl = false; // Whether to include the GL genotype subfield. if (input_mode == InputMode::stacks && !debug_flags.count("VCFCOMP")) { gl=true; // Load SNP data so that model likelihoods can be output to VCF file. cerr << "In preparation for VCF export, loading SNP data for " << mpopi.samples().size() << " samples.\n"; populate_snp_calls(catalog, pmap, merge_map); } // // Obtain the current date. // time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%Y%m%d", timeinfo); VcfHeader header; header.init_meta(); header.add_meta(VcfMeta::predefined.at("INFO/NS")); header.add_meta(VcfMeta::predefined.at("INFO/AF")); header.add_meta(VcfMeta::predefined.at("FORMAT/GT")); header.add_meta(VcfMeta::predefined.at("FORMAT/DP")); header.add_meta(VcfMeta::predefined.at("FORMAT/AD")); header.add_meta(VcfMeta::predefined.at("FORMAT/GL")); for(auto& s : mpopi.samples()) { header.add_sample(s.name); } writer.write_header(header); // We need to order the SNPs taking into account overlapping loci. OLocTally *ord = new OLocTally(psum, log_fh); for (auto it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { vector sites; ord->order(sites, it->second); for (uint pos = 0; pos < sites.size(); pos++) { if (catalog.count(sites[pos]->loc_id) == 0) { cerr << "Unable to find locus id " << sites[pos]->loc_id << "\n"; continue; } CSLocus* loc = catalog[sites[pos]->loc_id]; uint16_t col = sites[pos]->col; int snp_index = loc->snp_index(col); if (snp_index < 0) { cerr << "Warning: unable to locate SNP call in column " << col << " for catalog locus " << loc->id << "\n"; continue; } Datum** d = pmap->locus(loc->id); const char ref = sites[pos]->p_allele; const char alt = sites[pos]->q_allele; char freq_alt[32]; sprintf(freq_alt, "%0.3f", 1 - sites[pos]->p_freq); VcfRecord rec; rec.type = Vcf::RType::expl; rec.chrom = loc->loc.chr; rec.pos = loc->sort_bp(col) + 1; rec.id = to_string(loc->id) + "_" + to_string(col); rec.alleles.push_back(string(1, loc->loc.strand == strand_plus ? ref : reverse(ref))); rec.alleles.push_back(string(1, loc->loc.strand == strand_plus ? alt : reverse(alt))); rec.qual = "."; rec.filter.push_back("PASS"); rec.info.push_back({"NS",to_string(sites[pos]->num_indv)}); rec.info.push_back({"AF",freq_alt}); rec.format.push_back("GT"); rec.format.push_back("DP"); rec.format.push_back("AD"); if (gl) rec.format.push_back("GL"); for (int j = 0; j < pmap->sample_cnt(); j++) { stringstream sample; if (d[j] == NULL || col >= d[j]->len) { // Data does not exist. sample << "./.:0:.,."; if (gl) sample << ":.,.,."; } else if (d[j]->model[col] == 'U') { // Data exists, but the model call was uncertain. sample << "./.:" << d[j]->tot_depth << ":.,."; if (gl) sample << ":.,.,."; } else { char allele1, allele2; tally_observed_haplotypes(d[j]->obshap, snp_index, allele1, allele2); if (allele1 == 0) { // More than two alleles in this sample sample << "./.:" << d[j]->tot_depth << ":.,."; if (gl) sample << ":.,.,."; } else { // Write the genotype. int dp1, dp2; find_datum_allele_depths(d[j], snp_index, allele1, allele2, dp1, dp2); if(gl && col >= d[j]->snps.size()) cerr << "Warning, unable to locate SNP call in column " << col << " for catalog locus " << loc->id << ", tag ID " << d[j]->id << "\n"; if (allele2 == 0) { // homozygote if (allele1 == ref) { sample << "0/0:" << d[j]->tot_depth << ":" << dp1 << "," << dp2; if (gl) sample << ":" << - d[j]->snps[col]->lratio << ",.,."; } else { sample << "1/1:" << d[j]->tot_depth << ":" << dp2 << "," << dp1; if (gl) sample << ":.,.," << - d[j]->snps[col]->lratio; } } else { // heterozygote sample << "0/1:" << d[j]->tot_depth << ":" << (allele1 == ref ? dp1 : dp2) << "," << (allele1 == ref ? dp2 : dp1); if (gl) sample << ":.," << - d[j]->snps[col]->lratio << ",."; } } } rec.samples.push_back(sample.str()); } writer.write_record(rec, header); } } return 0; } int write_vcf(map &catalog, PopMap *pmap, PopSum *psum, map > &merge_map) { // // Write a VCF file as defined here: http://www.1000genomes.org/node/101 // string file = out_path + out_prefix + ".vcf"; cerr << "Writing population data to VCF file '" << file << "'\n"; VcfWriter writer (file); if (writer.fail()) { cerr << "Error opening VCF file '" << file << "'\n"; exit(-1); } bool gl = false; // Whether to include the GL genotype subfield. if (input_mode == InputMode::stacks && !debug_flags.count("VCFCOMP")) { gl=true; // Load SNP data so that model likelihoods can be output to VCF file. cerr << "In preparation for VCF export, loading SNP data for " << mpopi.samples().size() << " samples.\n"; populate_snp_calls(catalog, pmap, merge_map); } // // Output the header. // // Obtain the current date. time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%Y%m%d", timeinfo); VcfHeader header; header.init_meta(); header.add_meta(VcfMeta::predefined.at("INFO/NS")); header.add_meta(VcfMeta::predefined.at("INFO/AF")); header.add_meta(VcfMeta::predefined.at("FORMAT/GT")); header.add_meta(VcfMeta::predefined.at("FORMAT/DP")); header.add_meta(VcfMeta::predefined.at("FORMAT/AD")); header.add_meta(VcfMeta::predefined.at("FORMAT/GL")); for(auto& s : mpopi.samples()) { header.add_sample(s.name); } writer.write_header(header); map >::iterator it; CSLocus *loc; Datum **d; LocTally *t; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { // We need to order the SNPs so negative and positive strand SNPs are properly ordered. vector ordered_loci; for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; t = psum->locus_tally(loc->id); for (uint i = 0; i < loc->snps.size(); i++) { uint16_t col = loc->snps[i]->col; if (t->nucs[col].allele_cnt == 2) ordered_loci.push_back(GenPos(loc->id, i, loc->sort_bp(col))); } } sort(ordered_loci.begin(), ordered_loci.end()); for (uint pos = 0; pos < ordered_loci.size(); pos++) { loc = catalog[ordered_loci[pos].id]; uint16_t col = loc->snps[ordered_loci[pos].snp_index]->col; int snp_index = loc->snp_index(col); if (snp_index < 0) { cerr << "Warning: unable to locate SNP call in column " << col << " for locus #" << loc->id << "\n"; continue; } t = psum->locus_tally(loc->id); d = pmap->locus(loc->id); const char ref = t->nucs[col].p_allele; const char alt = t->nucs[col].q_allele; char freq_alt[32]; sprintf(freq_alt, "%0.3f", 1 - t->nucs[col].p_freq); VcfRecord rec; rec.type = Vcf::RType::expl; rec.chrom = loc->loc.chr; rec.pos = loc->sort_bp(col) + 1; rec.id = to_string(loc->id) + "_" + to_string(col); rec.alleles.push_back(string(1, loc->loc.strand == strand_plus ? ref : reverse(ref))); rec.alleles.push_back(string(1, loc->loc.strand == strand_plus ? alt : reverse(alt))); rec.qual = "."; rec.filter.push_back("PASS"); rec.info.push_back({"NS",to_string(t->nucs[col].num_indv)}); rec.info.push_back({"AF",freq_alt}); rec.format.push_back("GT"); rec.format.push_back("DP"); rec.format.push_back("AD"); if (gl) rec.format.push_back("GL"); for (int j = 0; j < pmap->sample_cnt(); j++) { stringstream sample; if (d[j] == NULL || col >= uint(d[j]->len)) { // Data does not exist. sample << "./.:0:.,."; if (gl) sample << ":.,.,."; } else if (d[j]->model[col] == 'U') { // Data exists, but the model call was uncertain. sample << "./.:" << d[j]->tot_depth << ":.,."; if (gl) sample << ":.,.,."; } else { char allele1, allele2; tally_observed_haplotypes(d[j]->obshap, ordered_loci[pos].snp_index, allele1, allele2); if (allele1 == 0) { // More than two alleles in this sample sample << "./.:" << d[j]->tot_depth << ":.,."; if (gl) sample << ":.,.,."; } else { // Write the genotype. int dp1, dp2; find_datum_allele_depths(d[j], snp_index, allele1, allele2, dp1, dp2); if (allele2 == 0) { // homozygote if (allele1 == ref) { sample << "0/0:" << d[j]->tot_depth << ":" << dp1 << "," << dp2; if (gl) sample << ":" << - d[j]->snps[col]->lratio << ",.,."; } else { sample << "1/1:" << d[j]->tot_depth << ":" << dp2 << "," << dp1; if (gl) sample << ":.,.," << - d[j]->snps[col]->lratio; } } else { // heterozygote sample << "0/1:" << d[j]->tot_depth << ":" << (allele1 == ref ? dp1 : dp2) << "," << (allele1 == ref ? dp2 : dp1); if (gl) sample << ":.," << - d[j]->snps[col]->lratio << ",."; } } } rec.samples.push_back(sample.str()); } writer.write_record(rec, header); } } return 0; } int write_vcf_haplotypes(map &catalog, PopMap *pmap, PopSum *psum) { // // Write a VCF file as defined here: http://samtools.github.io/hts-specs/ //XXX Datum::obshap *is not* ordered, I think. See Bitbucket. @Nick (June 2016) // string file = out_path + out_prefix + ".haplotypes.vcf"; cerr << "Writing population data haplotypes to VCF file '" << file << "'\n"; VcfWriter writer (file); if (writer.fail()) { cerr << "Error opening VCF file '" << file << "'\n"; exit(-1); } // // Obtain the current date. // time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%Y%m%d", timeinfo); VcfHeader header; header.init_meta(); header.add_meta(VcfMeta::predefined.at("INFO/NS")); header.add_meta(VcfMeta::predefined.at("INFO/AF")); header.add_meta(VcfMeta::predefined.at("FORMAT/GT")); header.add_meta(VcfMeta::predefined.at("FORMAT/DP")); for(auto& s : mpopi.samples()) { header.add_sample(s.name); } writer.write_header(header); CSLocus *loc; Datum **d; char allele[id_len]; for (auto chr = pmap->ordered_loci.begin(); chr != pmap->ordered_loci.end(); chr++) { for (uint pos = 0; pos < chr->second.size(); pos++) { loc = chr->second[pos]; d = pmap->locus(loc->id); map hap_freq; const double n_alleles = count_haplotypes_at_locus(0, pmap->sample_cnt() - 1, d, hap_freq); if (hap_freq.size() <= 1) // Monomorphic locus. // XXX What does [hap_freq.size()==1] mean ? @Nick (July 2016) continue; for(auto& h : hap_freq) // Convert counts to freqs. h.second /= n_alleles; // // Order the haplotypes according to most frequent. Record the ordered position or each // haplotype and convert them from counts to frequencies. // VcfRecord rec; rec.type = Vcf::RType::expl; rec.chrom = loc->loc.chr; rec.pos = loc->sort_bp() + 1; rec.id = to_string(loc->id); //alleles vector > ordered_hap (hap_freq.begin(), hap_freq.end()); sort(ordered_hap.begin(), ordered_hap.end(), compare_pair_haplotype); map hap_index; for (size_t i = 0; i < ordered_hap.size(); i++) { string h = ordered_hap[i].first; rec.alleles.push_back(loc->loc.strand == strand_plus ? h : string(rev_comp(h.c_str()))); hap_index[h] = i; } rec.qual = "."; rec.filter.push_back("PASS"); //info stringstream ss; ss << n_alleles/2; rec.info.push_back({"NS",ss.str()}); rec.info.push_back({"AF",string()}); string& af=rec.info.back().second; sprintf(allele, "%0.3f", ordered_hap[1].second); //NB. hap_freq.size() >= 2 af += allele; for (auto h=ordered_hap.begin()+2; h!=ordered_hap.end(); ++h) { sprintf(allele, "%0.3f", h->second); af += string(",") + allele; } //format rec.format.push_back("GT"); rec.format.push_back("DP"); for (int j = 0; j < pmap->sample_cnt(); j++) { stringstream sample; if (d[j] == NULL) { // Data does not exist. sample << "./.:0"; } else if (d[j]->obshap.size() > 2) { // More than two alleles in this sample sample << "./.:" << d[j]->tot_depth; } else if (d[j]->obshap.size() == 1) { // Homozygote. char* h = d[j]->obshap[0]; int i = uncalled_haplotype(h) ? -1 : hap_index.at(h); if(i >= 0) sample << i << "/" << i << ":" << d[j]->tot_depth; else sample << "./.:" << d[j]->tot_depth; } else { // Heterozygote. char* h1 = d[j]->obshap[0]; char* h2 = d[j]->obshap[1]; int i1 = uncalled_haplotype(h1) ? -1 : hap_index.at(h1); int i2 = uncalled_haplotype(h2) ? -1 : hap_index.at(h2); if(i1 >= 0 && i2 >= 0) sample << (i1 < i2 ? i1 : i2) << "/" << (i1 < i2 ? i2 : i1) << ":" << d[j]->tot_depth; else if (i1 >= 0) sample << i1 << "/.:" << d[j]->tot_depth; else if (i2 >= 0) sample << i2 << "/.:" << d[j]->tot_depth; } rec.samples.push_back(sample.str()); } writer.write_record(rec, header); } } return 0; } int write_genepop(map &catalog, PopMap *pmap, PopSum *psum) { // // Write a GenePop file as defined here: http://kimura.univ-montp2.fr/~rousset/Genepop.htm // string file = out_path + out_prefix + ".genepop"; cerr << "Writing population data to GenePop file '" << file << "'\n"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening GenePop file '" << file << "'\n"; exit(1); } // // Obtain the current date. // time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%B %d, %Y", timeinfo); // // Output the header line. // fh << "Stacks version " << VERSION << "; Genepop version 4.1.3; " << date << "\n"; map::iterator it; CSLocus *loc; Datum **d; LocSum **s; LocTally *t; int col; char p_allele, q_allele; // // Determine how many loci will be output, then output all the loci on the second line, comma-separated. // uint cnt = 0; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; for (uint j = 0; j < loc->snps.size(); j++) { col = loc->snps[j]->col; t = psum->locus_tally(loc->id); if (t->nucs[col].allele_cnt != 2) continue; cnt++; } } uint i = 0; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; for (uint j = 0; j < loc->snps.size(); j++) { col = loc->snps[j]->col; t = psum->locus_tally(loc->id); // // If this site is fixed in all populations or has too many alleles don't output it. // if (t->nucs[col].allele_cnt != 2) continue; i++; fh << loc->id << "_" << col; if (i < cnt) fh << ","; } } fh << "\n"; map nuc_map; nuc_map['A'] = "01"; nuc_map['C'] = "02"; nuc_map['G'] = "03"; nuc_map['T'] = "04"; for (size_t p=0; psecond; d = pmap->locus(loc->id); s = psum->locus(loc->id); t = psum->locus_tally(loc->id); for (i = 0; i < loc->snps.size(); i++) { uint col = loc->snps[i]->col; if (t->nucs[col].allele_cnt != 2) continue; if (s[p]->nucs[col].incompatible_site || s[p]->nucs[col].filtered_site) { // // This site contains more than two alleles in this population or was filtered // due to a minor allele frequency that is too low. // fh << "\t0000"; } else if (d[j] == NULL || col >= uint(d[j]->len)) { // // Data does not exist. // fh << "\t0000"; } else if (d[j]->model[col] == 'U') { // // Data exists, but the model call was uncertain. // fh << "\t0000"; } else { // // Tally up the nucleotide calls. // tally_observed_haplotypes(d[j]->obshap, i, p_allele, q_allele); if (p_allele == 0 && q_allele == 0) { // More than two potential alleles. fh << "\t0000"; } else if (p_allele == 0) { fh << "\t" << nuc_map[q_allele] << nuc_map[q_allele]; } else if (q_allele == 0) { fh << "\t" << nuc_map[p_allele] << nuc_map[p_allele]; } else { fh << "\t" << nuc_map[p_allele] << nuc_map[q_allele]; } } } } fh << "\n"; } } fh.close(); return 0; } int write_genepop_ordered(map &catalog, PopMap *pmap, PopSum *psum, ofstream &log_fh) { // // Write a GenePop file as defined here: http://kimura.univ-montp2.fr/~rousset/Genepop.htm // string file = out_path + out_prefix + ".genepop"; cerr << "Writing population data to GenePop file '" << file << "'\n"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening GenePop file '" << file << "'\n"; exit(1); } // // Obtain the current date. // time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%B %d, %Y", timeinfo); // // Output the header line. // fh << "Stacks version " << VERSION << "; Genepop version 4.1.3; " << date << "\n"; map > genome_sites; map >::iterator it; CSLocus *loc; Datum **d; LocSum **s; uint col, snp_index; char p_allele, q_allele; // // We need to order the SNPs to take into account overlapping loci. // OLocTally *ord = new OLocTally(psum, log_fh); // // Output all the loci on the second line, comma-separated. // int chrs = pmap->ordered_loci.size(); int cnt = 0; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { vector &sites = genome_sites[it->first]; ord->order(sites, it->second); cnt++; for (uint pos = 0; pos < sites.size(); pos++) { fh << sites[pos]->loc_id << "_" << sites[pos]->col; if (cnt < chrs || pos < sites.size() - 1) fh << ","; } } fh << "\n"; map nuc_map; nuc_map['A'] = "01"; nuc_map['C'] = "02"; nuc_map['G'] = "03"; nuc_map['T'] = "04"; for (size_t p=0; pordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { vector &sites = genome_sites[it->first]; for (uint pos = 0; pos < sites.size(); pos++) { loc = catalog[sites[pos]->loc_id]; s = psum->locus(loc->id); d = pmap->locus(loc->id); col = sites[pos]->col; if (s[p]->nucs[col].incompatible_site || s[p]->nucs[col].filtered_site) { // // This site contains more than two alleles in this population or was filtered // due to a minor allele frequency that is too low. // fh << "\t0000"; } else if (d[j] == NULL || col >= uint(d[j]->len)) { // // Data does not exist. // fh << "\t0000"; } else if (d[j]->model[col] == 'U') { // // Data exists, but the model call was uncertain. // fh << "\t0000"; } else { snp_index = loc->snp_index(col); // // Tally up the nucleotide calls. // tally_observed_haplotypes(d[j]->obshap, snp_index, p_allele, q_allele); if (p_allele == 0 && q_allele == 0) { // More than two potential alleles. fh << "\t0000"; } else if (p_allele == 0) { fh << "\t" << nuc_map[q_allele] << nuc_map[q_allele]; } else if (q_allele == 0) { fh << "\t" << nuc_map[p_allele] << nuc_map[p_allele]; } else { fh << "\t" << nuc_map[p_allele] << nuc_map[q_allele]; } } } } fh << "\n"; } } fh.close(); return 0; } int write_structure(map &catalog, PopMap *pmap, PopSum *psum) { // // Write a Structure file as defined here: http://pritch.bsd.uchicago.edu/structure.html // // To avoid linked SNPs (which Structure can't handle), we will only output the first // SNP from each variable locus. // string file = out_path + out_prefix + ".structure.tsv"; cerr << "Writing population data to Structure file '" << file << "'..."; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening Structure file '" << file << "'\n"; exit(1); } // // Obtain the current date. // time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%B %d, %Y", timeinfo); // // Output the header. // fh << "# Stacks v" << VERSION << "; " << " Structure v2.3; " << date << "\n"; map >::iterator it; CSLocus *loc; Datum **d; LocSum **s; LocTally *t; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; t = psum->locus_tally(loc->id); for (uint i = 0; i < loc->snps.size(); i++) { uint col = loc->snps[i]->col; if (t->nucs[col].allele_cnt == 2) fh << "\t" << loc->id << "_" << col; } } } fh << "\n"; map nuc_map; nuc_map['A'] = "1"; nuc_map['C'] = "2"; nuc_map['G'] = "3"; nuc_map['T'] = "4"; char p_allele, q_allele; for (size_t p=0; pordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; s = psum->locus(loc->id); d = pmap->locus(loc->id); t = psum->locus_tally(loc->id); for (uint i = 0; i < loc->snps.size(); i++) { uint col = loc->snps[i]->col; // // If this site is fixed in all populations or has too many alleles don't output it. // if (t->nucs[col].allele_cnt != 2) continue; if (s[p]->nucs[col].incompatible_site || s[p]->nucs[col].filtered_site) { // // This site contains more than two alleles in this population or was filtered // due to a minor allele frequency that is too low. // fh << "\t" << "0"; } else if (d[j] == NULL || col >= uint(d[j]->len)) { // // Data does not exist. // fh << "\t" << "0"; } else if (d[j]->model[col] == 'U') { // // Data exists, but the model call was uncertain. // fh << "\t" << "0"; } else { // // Tally up the nucleotide calls. // tally_observed_haplotypes(d[j]->obshap, i, p_allele, q_allele); if (p_allele == 0 && q_allele == 0) fh << "\t" << "0"; else if (p_allele == 0) fh << "\t" << nuc_map[q_allele]; else fh << "\t" << nuc_map[p_allele]; } } } } fh << "\n"; // // Output all the loci for this sample again, now for the q allele // fh << mpopi.samples()[j].name << "\t" << pop.name; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; s = psum->locus(loc->id); d = pmap->locus(loc->id); t = psum->locus_tally(loc->id); for (uint i = 0; i < loc->snps.size(); i++) { uint col = loc->snps[i]->col; if (t->nucs[col].allele_cnt != 2) continue; if (s[p]->nucs[col].incompatible_site || s[p]->nucs[col].filtered_site) { fh << "\t" << "0"; } else if (d[j] == NULL || col >= uint(d[j]->len)) { fh << "\t" << "0"; } else if (d[j]->model[col] == 'U') { fh << "\t" << "0"; } else { tally_observed_haplotypes(d[j]->obshap, i, p_allele, q_allele); if (p_allele == 0 && q_allele == 0) fh << "\t" << "0"; else if (q_allele == 0) fh << "\t" << nuc_map[p_allele]; else fh << "\t" << nuc_map[q_allele]; } } } } fh << "\n"; } } fh.close(); cerr << "done.\n"; return 0; } int write_structure_ordered(map &catalog, PopMap *pmap, PopSum *psum, ofstream &log_fh) { // // Write a Structure file as defined here: http://pritch.bsd.uchicago.edu/structure.html // // To avoid linked SNPs (which Structure can't handle), we will only output the first // SNP from each variable locus. // string file = out_path + out_prefix + ".structure.tsv"; cerr << "Writing population data to Structure file '" << file << "'..."; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening Structure file '" << file << "'\n"; exit(1); } // // Obtain the current date. // time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%B %d, %Y", timeinfo); // // Output the header. // fh << "# Stacks v" << VERSION << "; " << " Structure v2.3; " << date << "\n"; map > genome_sites; map >::iterator it; CSLocus *loc; Datum **d; LocSum **s; // // We need to order the SNPs to take into account overlapping loci. // OLocTally *ord = new OLocTally(psum, log_fh); for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { vector &sites = genome_sites[it->first]; ord->order(sites, it->second); for (uint pos = 0; pos < sites.size(); pos++) fh << "\t" << sites[pos]->loc_id << "_" << sites[pos]->col; } fh << "\n"; map nuc_map; nuc_map['A'] = "1"; nuc_map['C'] = "2"; nuc_map['G'] = "3"; nuc_map['T'] = "4"; char p_allele, q_allele; uint col, snp_index; for (size_t p=0; pordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { vector &sites = genome_sites[it->first]; for (uint pos = 0; pos < sites.size(); pos++) { loc = catalog[sites[pos]->loc_id]; s = psum->locus(loc->id); d = pmap->locus(loc->id); col = sites[pos]->col; if (s[p]->nucs[col].incompatible_site || s[p]->nucs[col].filtered_site) { // // This site contains more than two alleles in this population or was filtered // due to a minor allele frequency that is too low. // fh << "\t" << "0"; } else if (d[j] == NULL || col >= uint(d[j]->len)) { // // Data does not exist. // fh << "\t" << "0"; } else if (d[j]->model[col] == 'U') { // // Data exists, but the model call was uncertain. // fh << "\t" << "0"; } else { snp_index = loc->snp_index(col); // // Tally up the nucleotide calls. // tally_observed_haplotypes(d[j]->obshap, snp_index, p_allele, q_allele); if (p_allele == 0 && q_allele == 0) fh << "\t" << "0"; else if (p_allele == 0) fh << "\t" << nuc_map[q_allele]; else fh << "\t" << nuc_map[p_allele]; } } } fh << "\n"; // // Output all the loci for this sample again, now for the q allele // fh << mpopi.samples()[j].name << "\t" << pop.name; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { vector &sites = genome_sites[it->first]; for (uint pos = 0; pos < sites.size(); pos++) { loc = catalog[sites[pos]->loc_id]; s = psum->locus(loc->id); d = pmap->locus(loc->id); col = sites[pos]->col; if (s[p]->nucs[col].incompatible_site || s[p]->nucs[col].filtered_site) { fh << "\t" << "0"; } else if (d[j] == NULL || col >= uint(d[j]->len)) { fh << "\t" << "0"; } else if (d[j]->model[col] == 'U') { fh << "\t" << "0"; } else { snp_index = loc->snp_index(col); tally_observed_haplotypes(d[j]->obshap, snp_index, p_allele, q_allele); if (p_allele == 0 && q_allele == 0) fh << "\t" << "0"; else if (q_allele == 0) fh << "\t" << nuc_map[p_allele]; else fh << "\t" << nuc_map[q_allele]; } } } fh << "\n"; } } fh.close(); cerr << "done.\n"; return 0; } int write_hzar(map &catalog, PopMap *pmap, PopSum *psum) { // // Write a Hybrid Zone Analysis using R (HZAR) file as defined here: // http://cran.r-project.org/web/packages/hzar/hzar.pdf // string file = out_path + out_prefix + ".hzar.csv"; cerr << "Writing population data to HZAR file '" << file << "'..."; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening HZAR file '" << file << "'\n"; exit(1); } // // Obtain the current date. // time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%B %d, %Y", timeinfo); // // Output the header. // fh << "# Stacks v" << VERSION << "; " << " HZAR v0.2-5; " << date << "\n" << "Population" << "," << "Distance"; map >::iterator it; CSLocus *loc; LocSum **s; LocTally *t; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; s = psum->locus(loc->id); t = psum->locus_tally(loc->id); for (uint i = 0; i < loc->snps.size(); i++) { uint col = loc->snps[i]->col; if (t->nucs[col].allele_cnt == 2) { fh << "," << loc->id << "_" << col << ".A" << "," << loc->id << "_" << col << ".B" << "," << loc->id << "_" << col << ".N"; } } } } fh << "\n"; for (size_t p=0; pordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; s = psum->locus(loc->id); t = psum->locus_tally(loc->id); for (uint i = 0; i < loc->snps.size(); i++) { uint col = loc->snps[i]->col; // // If this site is fixed in all populations or has too many alleles don't output it. // if (t->nucs[col].allele_cnt != 2) continue; if (s[p]->nucs[col].num_indv == 0 || s[p]->nucs[col].incompatible_site || s[p]->nucs[col].filtered_site) { fh << ",0,0,0"; continue; } if (t->nucs[col].p_allele == s[p]->nucs[col].p_nuc) fh << "," << s[p]->nucs[col].p << "," << 1 - s[p]->nucs[col].p << ","; else fh << "," << 1 - s[p]->nucs[col].p << "," << s[p]->nucs[col].p << ","; fh << s[p]->nucs[col].num_indv * 2; } } } fh << "\n"; } fh.close(); cerr << "done.\n"; return 0; } int write_treemix(map &catalog, PopMap *pmap, PopSum *psum) { // // Write a TreeMix file (Pickrell and Pritchard, 2012 PLoS Genetics) // https://bitbucket.org/nygcresearch/treemix/wiki/Home // string file = out_path + out_prefix + ".treemix"; cerr << "Writing population data to TreeMix file '" << file << "'; "; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening TreeMix file '" << file << "'\n"; exit(1); } file += ".log"; cerr << "logging nucleotide positions to '" << file << "'..."; ofstream log_fh(file.c_str(), ofstream::out); if (log_fh.fail()) { cerr << "Error opening Phylip Log file '" << file << "'\n"; exit(1); } // // Obtain the current date. // time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%B %d, %Y", timeinfo); log_fh << "# Stacks v" << VERSION << "; " << " TreeMix v1.1; " << date << "\n" << "# Line\tLocus ID\tColumn\tChr\tBasepair\n"; // // Output the header. // fh << "# Stacks v" << VERSION << "; " << " TreeMix v1.1; " << date << "\n"; map >::iterator it; CSLocus *loc; LocSum **s; LocTally *t; // // Output a space-separated list of the populations on the first line. // stringstream sstr; for (auto& pop : mpopi.pops()) sstr << pop.name << " "; fh << sstr.str().substr(0, sstr.str().length() - 1) << "\n"; double p_freq, p_cnt, q_cnt, allele_cnt; long int line = 1; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; s = psum->locus(loc->id); t = psum->locus_tally(loc->id); for (uint i = 0; i < loc->snps.size(); i++) { uint col = loc->snps[i]->col; sstr.str(""); // // If this site is fixed in all populations or has too many alleles don't output it. // if (t->nucs[col].allele_cnt != 2) continue; for (size_t p=0; pnucs[col].num_indv == 0 || s[p]->nucs[col].incompatible_site || s[p]->nucs[col].filtered_site) { sstr << "0,0 "; continue; } p_freq = (t->nucs[col].p_allele == s[p]->nucs[col].p_nuc) ? s[p]->nucs[col].p : 1 - s[p]->nucs[col].p; allele_cnt = s[p]->nucs[col].num_indv * 2; p_cnt = round(allele_cnt * p_freq); q_cnt = allele_cnt - p_cnt; sstr << (int) p_cnt << "," << (int) q_cnt << " "; } if (sstr.str().length() == 0) continue; fh << sstr.str().substr(0, sstr.str().length() - 1) << "\n"; log_fh << line << "\t" << loc->id << "\t" << col << "\t" << loc->loc.chr << "\t" << loc->sort_bp(col) + 1 << "\n"; line++; } } } fh.close(); log_fh.close(); cerr << "done.\n"; return 0; } int write_fastphase(map &catalog, PopMap *pmap, PopSum *psum) { // // Write a fastPHASE file as defined here: http://stephenslab.uchicago.edu/software.html // // Data will be written as independent, bi-allelic SNPs. We will write one file per chromosome. // cerr << "Writing population data to fastPHASE files..."; map >::iterator it; CSLocus *loc; Datum **d; LocSum **s; LocTally *t; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { string file = out_path + out_prefix + "." + it->first + ".fastphase.inp"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening fastPHASE file '" << file << "'\n"; exit(1); } // // Tally up the number of sites // int total_sites = 0; uint col; for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; t = psum->locus_tally(loc->id); for (uint i = 0; i < loc->snps.size(); i++) { col = loc->snps[i]->col; if (t->nucs[col].allele_cnt == 2) total_sites++; } } // // Output the total number of SNP sites and the number of individuals. // fh << mpopi.samples().size() << "\n" << total_sites << "\n"; // // We need to determine an ordering that can take into account overlapping RAD sites. // vector ordered_loci; for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; t = psum->locus_tally(loc->id); for (uint i = 0; i < loc->snps.size(); i++) { col = loc->snps[i]->col; if (t->nucs[col].allele_cnt == 2) ordered_loci.push_back(GenPos(loc->id, i, loc->sort_bp(col))); } } sort(ordered_loci.begin(), ordered_loci.end()); // // Output the position of each site according to its basepair. // fh << "P"; for (uint pos = 0; pos < ordered_loci.size(); pos++) { loc = catalog[ordered_loci[pos].id]; col = loc->snps[ordered_loci[pos].snp_index]->col; fh << " " << ordered_loci[pos].bp +1; } fh << "\n"; // // Output a line of 'S' characters, one per site, indicating that these are SNP markers. // string snp_markers, gtypes_str; snp_markers.assign(total_sites, 'S'); fh << snp_markers << '\n'; // // Now output each sample name followed by a new line, then all of the genotypes for that sample // on two lines. // char p_allele, q_allele; stringstream gtypes; for (size_t p=0; psnps[ordered_loci[pos].snp_index]->col; s = psum->locus(loc->id); d = pmap->locus(loc->id); t = psum->locus_tally(loc->id); // // If this site is fixed in all populations or has too many alleles don't output it. // if (t->nucs[col].allele_cnt != 2) continue; if (s[p]->nucs[col].incompatible_site || s[p]->nucs[col].filtered_site) { // // This site contains more than two alleles in this population or was filtered // due to a minor allele frequency that is too low. // gtypes << "? "; } else if (d[j] == NULL || col >= uint(d[j]->len)) { // // Data does not exist. // gtypes << "? "; } else if (d[j]->model[col] == 'U') { // // Data exists, but the model call was uncertain. // gtypes << "? "; } else { // // Tally up the nucleotide calls. // tally_observed_haplotypes(d[j]->obshap, ordered_loci[pos].snp_index, p_allele, q_allele); if (p_allele == 0 && q_allele == 0) gtypes << "? "; else if (p_allele == 0) gtypes << q_allele << " "; else gtypes << p_allele << " "; } } gtypes_str = gtypes.str(); fh << gtypes_str.substr(0, gtypes_str.length() - 1) << "\n"; // // Output all the loci for this sample again, now for the q allele // gtypes.str(""); for (uint pos = 0; pos < ordered_loci.size(); pos++) { loc = catalog[ordered_loci[pos].id]; col = loc->snps[ordered_loci[pos].snp_index]->col; s = psum->locus(loc->id); d = pmap->locus(loc->id); t = psum->locus_tally(loc->id); if (t->nucs[col].allele_cnt != 2) continue; if (s[p]->nucs[col].incompatible_site || s[p]->nucs[col].filtered_site) { gtypes << "? "; } else if (d[j] == NULL || col >= uint(d[j]->len)) { gtypes << "? "; } else if (d[j]->model[col] == 'U') { gtypes << "? "; } else { tally_observed_haplotypes(d[j]->obshap, ordered_loci[pos].snp_index, p_allele, q_allele); if (p_allele == 0 && q_allele == 0) gtypes << "? "; else if (q_allele == 0) gtypes << p_allele << " "; else gtypes << q_allele << " "; } } gtypes_str = gtypes.str(); fh << gtypes_str.substr(0, gtypes_str.length() - 1) << "\n"; } } fh.close(); } cerr << "done.\n"; return 0; } int write_phase(map &catalog, PopMap *pmap, PopSum *psum) { // // Write a PHASE file as defined here: http://stephenslab.uchicago.edu/software.html // // Data will be written as mixture of multiple allele, linked RAD sites // (SNPs within a single RAD locus are already phased), and bi-allelic SNPs. We // will write one file per chromosome. // cerr << "Writing population data to PHASE files..."; map >::iterator it; CSLocus *loc; Datum **d; LocSum **s; LocTally *t; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { string file = out_path + out_prefix + "." + it->first + ".phase.inp"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening PHASE file '" << file << "'\n"; exit(1); } // // We need to determine an ordering for all legitimate loci/SNPs. // uint col; vector ordered_loci; for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; t = psum->locus_tally(loc->id); if (loc->snps.size() == 0) continue; // // Will we output this locus as a haplotype or as a SNP? // if (loc->snps.size() > 1) { // // Check that there aren't too many haplotypes (PHASE has a max of 50). // if (loc->alleles.size() > 40) continue; // // Iterate over the population to determine that this subset of the population // has data at this locus. // d = pmap->locus(loc->id); for (int j = 0; j < pmap->sample_cnt(); j++) { if (d[j] != NULL && d[j]->obshap.size() > 0 && d[j]->obshap.size() <= 2) { // // Data exists, and there are the correct number of haplotypes. // ordered_loci.push_back(GenPos(loc->id, 0, loc->sort_bp(), haplotype)); break; } } } else { col = loc->snps[0]->col; if (t->nucs[col].allele_cnt == 2) ordered_loci.push_back(GenPos(loc->id, 0, loc->sort_bp(col), snp)); } } sort(ordered_loci.begin(), ordered_loci.end()); // // Output the total number of SNP sites and the number of individuals. // fh << mpopi.samples().size() << "\n" << ordered_loci.size() << "\n"; // // Output the position of each site according to its basepair. // fh << "P"; for (uint pos = 0; pos < ordered_loci.size(); pos++) fh << " " << ordered_loci[pos].bp +1; fh << "\n"; // // Output a line of 'S' characters for SNP markers, 'M' characters for multiallelic haplotypes. // for (uint pos = 0; pos < ordered_loci.size(); pos++) { if (pos > 0) fh << " "; fh << (ordered_loci[pos].type == snp ? "S" : "M"); } fh << "\n"; // // Now output each sample name followed by a new line, then all of the genotypes for that sample // on two lines. // string gtypes_str; bool found; char p_allele, q_allele; stringstream gtypes; for (size_t p=0; plocus(loc->id); d = pmap->locus(loc->id); t = psum->locus_tally(loc->id); // // Will we output this locus as a haplotype or as a SNP? // if (ordered_loci[pos].type == haplotype) { if (d[j] == NULL) { // // Data does not exist. // gtypes << "-1 "; } else { // // Data exists, output the first haplotype. We will assume the haplotypes are // numbered by their position in the loc->strings vector. // if (d[j]->obshap.size() > 2) { // cerr << "Warning: too many haplotypes, catalog locus: " << loc->id << "\n"; gtypes << "-1 "; } else { found = false; for (uint k = 0; k < loc->strings.size(); k++) if (d[j]->obshap[0] == loc->strings[k].first) { found = true; gtypes << k + 1 << " "; } if (found == false) cerr << "Unable to find haplotype " << d[j]->obshap[0] << " from individual " << mpopi.samples()[j].name << "; catalog locus: " << loc->id << "\n"; } } } else { col = loc->snps[ordered_loci[pos].snp_index]->col; if (s[p]->nucs[col].incompatible_site || s[p]->nucs[col].filtered_site) { // // This site contains more than two alleles in this population or was filtered // due to a minor allele frequency that is too low. // gtypes << "? "; } else if (d[j] == NULL || col >= uint(d[j]->len)) { // // Data does not exist. // gtypes << "? "; } else if (d[j]->model[col] == 'U') { // // Data exists, but the model call was uncertain. // gtypes << "? "; } else { // // Tally up the nucleotide calls. // tally_observed_haplotypes(d[j]->obshap, ordered_loci[pos].snp_index, p_allele, q_allele); if (p_allele == 0 && q_allele == 0) gtypes << "? "; else if (p_allele == 0) gtypes << q_allele << " "; else gtypes << p_allele << " "; } } } gtypes_str = gtypes.str(); fh << gtypes_str.substr(0, gtypes_str.length() - 1) << "\n"; // // Output all the loci for this sample again, now for the q allele // gtypes.str(""); for (uint pos = 0; pos < ordered_loci.size(); pos++) { loc = catalog[ordered_loci[pos].id]; s = psum->locus(loc->id); d = pmap->locus(loc->id); t = psum->locus_tally(loc->id); // // Will we output this locus as a haplotype or as a SNP? // if (ordered_loci[pos].type == haplotype) { if (d[j] == NULL) { // // Data does not exist. // gtypes << "-1 "; } else { // // Data exists, output the second haplotype. We will assume the haplotypes are // numbered by their position in the loc->strings vector. // if (d[j]->obshap.size() > 2) { // cerr << "Warning: too many haplotypes, catalog locus: " << loc->id << "\n"; gtypes << "-1 "; } else if (d[j]->obshap.size() > 1) { found = false; for (uint k = 0; k < loc->strings.size(); k++) if (d[j]->obshap[1] == loc->strings[k].first) { found = true; gtypes << k + 1 << " "; } if (found == false) cerr << "Unable to find haplotype " << d[j]->obshap[1] << " from individual " << mpopi.samples()[j].name << "; catalog locus: " << loc->id << "\n"; } else { found = false; for (uint k = 0; k < loc->strings.size(); k++) if (d[j]->obshap[0] == loc->strings[k].first) { found = true; gtypes << k + 1 << " "; } if (found == false) cerr << "Unable to find haplotype " << d[j]->obshap[0] << " from individual " << mpopi.samples()[j].name << "; catalog locus: " << loc->id << "\n"; } } } else { col = loc->snps[ordered_loci[pos].snp_index]->col; if (s[p]->nucs[col].incompatible_site || s[p]->nucs[col].filtered_site) { gtypes << "? "; } else if (d[j] == NULL || col >= uint(d[j]->len)) { gtypes << "? "; } else if (d[j]->model[col] == 'U') { gtypes << "? "; } else { tally_observed_haplotypes(d[j]->obshap, ordered_loci[pos].snp_index, p_allele, q_allele); if (p_allele == 0 && q_allele == 0) gtypes << "? "; else if (q_allele == 0) gtypes << p_allele << " "; else gtypes << q_allele << " "; } } } gtypes_str = gtypes.str(); fh << gtypes_str.substr(0, gtypes_str.length() - 1) << "\n"; } } fh.close(); } cerr << "done.\n"; return 0; } int write_plink(map &catalog, PopMap *pmap, PopSum *psum) { // // Write a PLINK file as defined here: http://pngu.mgh.harvard.edu/~purcell/plink/data.shtml // // We will write one file per chromosome. // cerr << "Writing population data to PLINK files..."; // // Obtain the current date. // time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%B %d, %Y", timeinfo); map >::iterator it; CSLocus *loc; Datum **d; LocSum **s; LocTally *t; string chr; // // First, write a markers file containing each marker, the chromosome it falls on, // an empty centiMorgan field, and finally its genomic position in basepairs. // string file = out_path + out_prefix + ".plink.map"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening PLINK markers file '" << file << "'\n"; exit(1); } // // Output the header. // fh << "# Stacks v" << VERSION << "; " << " PLINK v1.07; " << date << "\n"; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { chr = it->first; for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; t = psum->locus_tally(loc->id); for (uint i = 0; i < loc->snps.size(); i++) { uint col = loc->snps[i]->col; if (t->nucs[col].allele_cnt == 2) fh << chr << "\t" << loc->id << "_" << col << "\t" << "0\t" << loc->sort_bp(col) +1 << "\n"; } } } fh.close(); // // Now output the genotypes in a separate file. // file = out_path + out_prefix + ".plink.ped"; fh.open(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening PLINK markers file '" << file << "'\n"; exit(1); } fh << "# Stacks v" << VERSION << "; " << " PLINK v1.07; " << date << "\n"; char p_allele, q_allele; // // marker, output the genotypes for each sample in two successive columns. // for (size_t p=0; pordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; s = psum->locus(loc->id); d = pmap->locus(loc->id); t = psum->locus_tally(loc->id); for (uint i = 0; i < loc->snps.size(); i++) { uint col = loc->snps[i]->col; // // If this site is fixed in all populations or has too many alleles don't output it. // if (t->nucs[col].allele_cnt != 2) continue; // // Output the p and q alleles // if (s[p]->nucs[col].incompatible_site || s[p]->nucs[col].filtered_site) { // // This site contains more than two alleles in this population or was filtered // due to a minor allele frequency that is too low. // fh << "\t" << "0" << "\t" << "0"; } else if (d[j] == NULL || col >= uint(d[j]->len)) { // // Data does not exist. // fh << "\t" << "0" << "\t" << "0"; } else if (d[j]->model[col] == 'U') { // // Data exists, but the model call was uncertain. // fh << "\t" << "0" << "\t" << "0"; } else { // // Tally up the nucleotide calls. // tally_observed_haplotypes(d[j]->obshap, i, p_allele, q_allele); if (p_allele == 0 && q_allele == 0) fh << "\t" << "0" << "\t" << "0"; else if (p_allele == 0) fh << "\t" << q_allele << "\t" << q_allele; else if (q_allele == 0) fh << "\t" << p_allele << "\t" << p_allele; else fh << "\t" << p_allele << "\t" << q_allele; } } } } fh << "\n"; } } fh.close(); cerr << "done.\n"; return 0; } int write_beagle(map &catalog, PopMap *pmap, PopSum *psum) { // // Write a Beagle file as defined here: http://faculty.washington.edu/browning/beagle/beagle.html // // We will write one file per chromosome, per population. // cerr << "Writing population data to unphased Beagle files..."; // // Obtain the current date. // time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%B %d, %Y", timeinfo); map >::iterator it; CSLocus *loc; Datum **d; LocSum **s; LocTally *t; uint col; stringstream pop_name; string file; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { // // We need to determine an ordering that can take into account overlapping RAD sites. // vector ordered_loci; for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; t = psum->locus_tally(loc->id); for (uint i = 0; i < loc->snps.size(); i++) { col = loc->snps[i]->col; if (t->nucs[col].allele_cnt == 2) ordered_loci.push_back(GenPos(loc->id, i, loc->sort_bp(col))); } } sort(ordered_loci.begin(), ordered_loci.end()); // // Now output the genotypes in a separate file for each population. // for (size_t p=0; pfirst + ".unphased.bgl.markers"; ofstream mfh(file.c_str(), ofstream::out); if (mfh.fail()) { cerr << "Error opening Beagle markers file '" << file << "'\n"; exit(1); } mfh << "# Stacks v" << VERSION << "; " << " Beagle v3.3; " << date << "\n"; // // Open the genotypes file. // file = out_path + out_prefix + "." + pop.name + "-" + it->first + ".unphased.bgl"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening Beagle genotypes file '" << file << "'\n"; exit(1); } fh << "# Stacks v" << VERSION << "; " << " Beagle v3.3; " << date << "\n"; char p_allele, q_allele; // // Output a list of all the samples in this population. // fh << "I\tid"; for (size_t j = pop.first_sample; j <= pop.last_sample; j++) fh << "\t" << mpopi.samples()[j].name << "\t" << mpopi.samples()[j].name; fh << "\n"; // // Output population IDs for each sample. // fh << "S\tid"; for (size_t j = pop.first_sample; j <= pop.last_sample; j++) fh << "\t" << pop.name << "\t" << pop.name; fh << "\n"; // // For each marker, output the genotypes for each sample in two successive columns. // for (uint pos = 0; pos < ordered_loci.size(); pos++) { loc = catalog[ordered_loci[pos].id]; s = psum->locus(loc->id); d = pmap->locus(loc->id); t = psum->locus_tally(loc->id); col = loc->snps[ordered_loci[pos].snp_index]->col; // // If this site is fixed in all populations or has too many alleles don't output it. // if (t->nucs[col].allele_cnt != 2) continue; // // If this site is monomorphic in this population don't output it. // if (s[p]->nucs[col].pi == 0.0) continue; // // Output this locus to the markers file. // mfh << loc->id << "_" << col << "\t" << loc->sort_bp(col) +1 << "\t" << t->nucs[col].p_allele << "\t" << t->nucs[col].q_allele << "\n"; fh << "M" << "\t" << loc->id << "_" << col; for (size_t j = pop.first_sample; j <= pop.last_sample; j++) { // // Output the p allele // if (s[p]->nucs[col].incompatible_site || s[p]->nucs[col].filtered_site) { // // This site contains more than two alleles in this population or was filtered // due to a minor allele frequency that is too low. // fh << "\t" << "?"; } else if (d[j] == NULL || col >= uint(d[j]->len)) { // // Data does not exist. // fh << "\t" << "?"; } else if (d[j]->model[col] == 'U') { // // Data exists, but the model call was uncertain. // fh << "\t" << "?"; } else { // // Tally up the nucleotide calls. // tally_observed_haplotypes(d[j]->obshap, ordered_loci[pos].snp_index, p_allele, q_allele); if (p_allele == 0 && q_allele == 0) fh << "\t" << "?"; else if (p_allele == 0) fh << "\t" << q_allele; else fh << "\t" << p_allele; } // // Now output the q allele // if (s[p]->nucs[col].incompatible_site || s[p]->nucs[col].filtered_site) { fh << "\t" << "?"; } else if (d[j] == NULL || col >= uint(d[j]->len)) { fh << "\t" << "?"; } else if (d[j]->model[col] == 'U') { fh << "\t" << "?"; } else { if (p_allele == 0 && q_allele == 0) fh << "\t" << "?"; else if (q_allele == 0) fh << "\t" << p_allele; else fh << "\t" << q_allele; } } fh << "\n"; } fh.close(); mfh.close(); } } cerr << "done.\n"; return 0; } int write_beagle_phased(map &catalog, PopMap *pmap, PopSum *psum) { // // Write a Beagle file as a set of haplotpyes as defined here: // http://faculty.washington.edu/browning/beagle/beagle.html // // We will write one file per chromosome. // cerr << "Writing population data to phased Beagle files..."; // // Obtain the current date. // time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%B %d, %Y", timeinfo); map >::iterator it; CSLocus *loc; Datum **d; stringstream pop_name; string file; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { // // We need to determine an ordering for all legitimate loci/SNPs. // vector ordered_loci; for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; if (loc->snps.size() == 0) continue; // // Check that there aren't too many haplotypes (PHASE has a max of 50). // if (loc->alleles.size() > 40) continue; // // Iterate over the population to determine that this subset of the population // has data at this locus. // d = pmap->locus(loc->id); for (int j = 0; j < pmap->sample_cnt(); j++) { if (d[j] != NULL && d[j]->obshap.size() > 0 && d[j]->obshap.size() <= 2) { // // Data exists, and their are the corrent number of haplotypes. // ordered_loci.push_back(GenPos(loc->id, 0, loc->sort_bp(), haplotype)); break; } } } sort(ordered_loci.begin(), ordered_loci.end()); // // Now output the genotypes in a separate file for each population. // for (size_t i_pop=0; i_popfirst + ".phased.bgl.markers"; ofstream mfh(file.c_str(), ofstream::out); if (mfh.fail()) { cerr << "Error opening Beagle markers file '" << file << "'\n"; exit(1); } mfh << "# Stacks v" << VERSION << "; " << " Beagle v3.3; " << date << "\n"; // // Now output the haplotypes in a separate file. // file = out_path + out_prefix + "." + pop.name + "-" + it->first + ".phased.bgl"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening Beagle markers file '" << file << "'\n"; exit(1); } fh << "# Stacks v" << VERSION << "; " << " Beagle v3.3; " << date << "\n"; // // Output a list of all the samples in the data set. // fh << "I\tid"; for (size_t j = pop.first_sample; j <= pop.last_sample; j++) fh << "\t" << mpopi.samples()[j].name << "\t" << mpopi.samples()[j].name; fh << "\n"; // // Output population IDs for each sample. // fh << "S\tid"; for (size_t j = pop.first_sample; j <= pop.last_sample; j++) fh << "\t" << pop.name << "\t" << pop.name; fh << "\n"; for (uint pos = 0; pos < ordered_loci.size(); pos++) { loc = catalog[ordered_loci[pos].id]; d = pmap->locus(loc->id); // // If this locus is monomorphic in this population don't output it. // set haplotypes; for (size_t j = pop.first_sample; j <= pop.last_sample; j++) { if (d[j] == NULL) continue; if (d[j]->obshap.size() == 2) { haplotypes.insert(d[j]->obshap[0]); haplotypes.insert(d[j]->obshap[1]); } else { haplotypes.insert(d[j]->obshap[0]); } } if (haplotypes.size() == 1) continue; // // Output this locus to the markers file. // mfh << loc->id << "\t" << loc->sort_bp() +1; for (uint j = 0; j < loc->strings.size(); j++) mfh << "\t" << loc->strings[j].first; mfh << "\n"; // // For each marker, output the genotypes for each sample in two successive columns. // fh << "M" << "\t" << loc->id; for (size_t j = pop.first_sample; j <= pop.last_sample; j++) { // // Output the p and the q haplotype // if (d[j] == NULL) { // // Data does not exist. // fh << "\t" << "?" << "\t" << "?"; } else { // // Data exists, output the first haplotype. We will assume the haplotypes are // numbered by their position in the loc->strings vector. // if (d[j]->obshap.size() > 2) fh << "\t" << "?" << "\t" << "?"; else if (d[j]->obshap.size() == 2) fh << "\t" << d[j]->obshap[0] << "\t" << d[j]->obshap[1]; else fh << "\t" << d[j]->obshap[0] << "\t" << d[j]->obshap[0]; } } fh << "\n"; } fh.close(); mfh.close(); } } cerr << "done.\n"; return 0; } int write_phylip(map &catalog, PopMap *pmap, PopSum *psum) { // // We want to find loci where each locus is fixed within a population but variable between populations. // // We will write those loci to a Phylip file as defined here: // http://evolution.genetics.washington.edu/phylip/doc/main.html#inputfiles // string file = out_path + out_prefix + ".phylip"; cerr << "Writing population data to Phylip file '" << file << "'; "; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening Phylip file '" << file << "'\n"; exit(1); } file += ".log"; cerr << "logging nucleotide positions to '" << file << "'..."; ofstream log_fh(file.c_str(), ofstream::out); if (log_fh.fail()) { cerr << "Error opening Phylip Log file '" << file << "'\n"; exit(1); } // // Obtain the current date. // time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%B %d, %Y", timeinfo); log_fh << "# Stacks v" << VERSION << "; " << " Phylip sequential; " << date << "\n" << "# Seq Pos\tLocus ID\tColumn\tPopulation\n"; map >::iterator it; CSLocus *loc; LocSum **s; LocTally *t; int pop_cnt = psum->pop_cnt(); char nuc; // // A map storing, for each population, the concatenated list of interspecific nucleotides. // map interspecific_nucs; int index = 0; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; s = psum->locus(loc->id); t = psum->locus_tally(loc->id); for (uint i = 0; i < loc->snps.size(); i++) { uint col = loc->snps[i]->col; if (phylip_var == false) { // // We are looking for loci that are fixed within each population, but are // variable between one or more populations. // if (t->nucs[col].fixed == true || t->nucs[col].allele_cnt != 2 || t->nucs[col].pop_cnt < 2) continue; bool fixed_within = true; for (int j = 0; j < pop_cnt; j++) { if (s[j]->nucs[col].num_indv == 0) continue; if (s[j]->nucs[col].fixed == false) { fixed_within = false; break; } } if (fixed_within == false) continue; log_fh << index << "\t" << loc->id << "\t" << col << "\t"; for (int p=0; pnucs[col].num_indv > 0) { interspecific_nucs[p] += s[p]->nucs[col].p_nuc; log_fh << mpopi.pops()[p].name << ":" << s[p]->nucs[col].p_nuc << ","; } else { interspecific_nucs[p] += 'N'; log_fh << mpopi.pops()[p].name << ":N" << ","; } } log_fh << "\n"; index++; } else { // // Encode SNPs that are variable within a population as well, using IUPAC notation: // http://en.wikipedia.org/wiki/Nucleic_acid_notation#IUPAC_notation // if (t->nucs[col].allele_cnt != 2) continue; log_fh << index << "\t" << loc->id << "\t" << col << "\t"; for (int j=0; jnucs[col].p_nuc) { case 0: nuc = 'N'; break; case 'A': switch(s[j]->nucs[col].q_nuc) { case 'C': nuc = 'M'; break; case 'G': nuc = 'R'; break; case 'T': nuc = 'W'; break; case 0: nuc = 'A'; break; } break; case 'C': switch(s[j]->nucs[col].q_nuc) { case 'A': nuc = 'M'; break; case 'G': nuc = 'S'; break; case 'T': nuc = 'Y'; break; case 0: nuc = 'C'; break; } break; case 'G': switch(s[j]->nucs[col].q_nuc) { case 'A': nuc = 'R'; break; case 'C': nuc = 'S'; break; case 'T': nuc = 'K'; break; case 0: nuc = 'G'; break; } break; case 'T': switch(s[j]->nucs[col].q_nuc) { case 'A': nuc = 'W'; break; case 'C': nuc = 'Y'; break; case 'G': nuc = 'K'; break; case 0: nuc = 'T'; break; } break; } interspecific_nucs[j] += nuc; log_fh << mpopi.pops()[j].name << ":" << nuc << ","; } log_fh << "\n"; index++; } } } } if (interspecific_nucs.size() == 0) { cerr << " No data is available to write to the Phylip file.\n"; return 0; } fh << mpopi.pops().size() << " " << interspecific_nucs.begin()->second.length() << "\n"; for (size_t i_pop=0; i_pop &catalog, PopMap *pmap, PopSum *psum) { // // We want to write all variable loci in Phylip interleaved format. Polymorphic positions // will be encoded using IUPAC notation. // // We will write those loci to a Phylip file as defined here: // http://evolution.genetics.washington.edu/phylip/doc/main.html#inputfiles // string file = out_path + out_prefix + ".fullseq.phylip"; cerr << "Writing full sequence population data to Phylip file '" << file << "'; "; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening Phylip file '" << file << "'\n"; exit(1); } // // We will also write a file that allows us to specify each RAD locus as a separate partition // for use in phylogenetics programs. // file = out_path + out_prefix + ".fullseq.partitions.phylip"; ofstream par_fh(file.c_str(), ofstream::out); if (par_fh.fail()) { cerr << "Error opening Phylip partitions file '" << file << "'\n"; exit(1); } file = out_path + out_prefix + ".fullseq.phylip.log"; cerr << "logging nucleotide positions to '" << file << "'..."; ofstream log_fh(file.c_str(), ofstream::out); if (log_fh.fail()) { cerr << "Error opening Phylip Log file '" << file << "'\n"; exit(1); } // // Obtain the current date. // time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%B %d, %Y", timeinfo); log_fh << "# Stacks v" << VERSION << "; " << " Phylip interleaved; " << date << "\n" << "# Locus ID\tLine Number"; if (loci_ordered) log_fh << "\tChr\tBasepair"; log_fh << "\n"; map >::iterator it; CSLocus *loc; LocSum **s; LocTally *t; int pop_cnt = psum->pop_cnt(); char nuc = '\0'; bool include; uint len = 0; // // Determine the length of sequence we will output. // for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; t = psum->locus_tally(loc->id); include = true; for (uint i = 0; i < loc->snps.size(); i++) { uint col = loc->snps[i]->col; if (t->nucs[col].allele_cnt != 2) include = false; } if (include) len += strlen(loc->con); } } map outstrs; fh << mpopi.pops().size() << " " << len << "\n"; for (size_t i_pop=0; i_popordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; s = psum->locus(loc->id); t = psum->locus_tally(loc->id); include = true; for (uint i = 0; i < loc->snps.size(); i++) { uint col = loc->snps[i]->col; if (t->nucs[col].allele_cnt != 2) include = false; } if (!include) continue; seq = new char[loc->len + 1]; strcpy(seq, loc->con); for (int j = 0; j < pop_cnt; j++) { for (uint i = 0; i < loc->snps.size(); i++) { uint col = loc->snps[i]->col; // // Encode SNPs that are variable within a population using IUPAC notation: // http://en.wikipedia.org/wiki/Nucleic_acid_notation#IUPAC_notation // switch(s[j]->nucs[col].p_nuc) { case 0: nuc = 'N'; break; case 'A': switch(s[j]->nucs[col].q_nuc) { case 'C': nuc = 'M'; break; case 'G': nuc = 'R'; break; case 'T': nuc = 'W'; break; case 0: nuc = 'A'; break; } break; case 'C': switch(s[j]->nucs[col].q_nuc) { case 'A': nuc = 'M'; break; case 'G': nuc = 'S'; break; case 'T': nuc = 'Y'; break; case 0: nuc = 'C'; break; } break; case 'G': switch(s[j]->nucs[col].q_nuc) { case 'A': nuc = 'R'; break; case 'C': nuc = 'S'; break; case 'T': nuc = 'K'; break; case 0: nuc = 'G'; break; } break; case 'T': switch(s[j]->nucs[col].q_nuc) { case 'A': nuc = 'W'; break; case 'C': nuc = 'Y'; break; case 'G': nuc = 'K'; break; case 0: nuc = 'T'; break; } break; } seq[col] = nuc; } outstrs[j] += string(seq); } delete [] seq; log_fh << line << "\t" << loc->id; if (loci_ordered) log_fh << "\t" << loc->loc.chr << "\t" << loc->sort_bp() + 1; log_fh << "\n"; for (size_t i_pop=0; i_poplen - 1 << "\n"; index += loc->len; cnt++; } } // // Output the header. // fh << "# Stacks v" << VERSION << "; " << " Phylip interleaved; " << date << "\n"; fh.close(); par_fh.close(); log_fh.close(); cerr << "done.\n"; return 0; } int populate_snp_calls(map &catalog, PopMap *pmap, map > &merge_map) { map::iterator cit; map::iterator sit; CSLocus *loc; Datum *datum; SNPRes *snpr; SNP *snp; for (uint i = 0; i < mpopi.samples().size(); i++) { map snpres; load_snp_calls(in_path + mpopi.samples()[i].name, snpres); for (cit = catalog.begin(); cit != catalog.end(); cit++) { loc = cit->second; datum = pmap->datum(loc->id, mpopi.samples()[i].id); if (datum != NULL && snpres.count(datum->id)) { if (merge_sites && merge_map.count(loc->id)) { datum_adjust_snp_positions(merge_map, loc, datum, snpres); } else { // // Deep copy the SNP objects. // snpr = snpres[datum->id]; for (uint j = 0; j < snpr->snps.size(); j++) { snp = new SNP; snp->col = snpr->snps[j]->col; snp->lratio = snpr->snps[j]->lratio; snp->rank_1 = snpr->snps[j]->rank_1; snp->rank_2 = snpr->snps[j]->rank_2; snp->rank_3 = snpr->snps[j]->rank_3; snp->rank_4 = snpr->snps[j]->rank_4; datum->snps.push_back(snp); } } } } for (sit = snpres.begin(); sit != snpres.end(); sit++) delete sit->second; } return 0; } /* * Calculate the SNP-wise allelic depths by adding up the haplotype depths. */ int find_datum_allele_depths(Datum *d, int snp_index, char allele1, char allele2, int &dp1, int &dp2) { dp1 = 0; dp2 = 0; for(uint i = 0; i < d->obshap.size(); i++) { char nt = d->obshap[i][snp_index]; if(nt == allele1) dp1 += d->depth[i]; else if(nt == allele2) dp2 += d->depth[i]; else throw std::exception(); } return 0; } int tally_observed_haplotypes(vector &obshap, int snp_index, char &p_allele, char &q_allele) { int nucs[4] = {0}; char nuc; // // Pull each allele for this SNP from the observed haplotype. // for (uint j = 0; j < obshap.size(); j++) { nuc = obshap[j][snp_index]; switch(nuc) { case 'A': case 'a': nucs[0]++; break; case 'C': case 'c': nucs[1]++; break; case 'G': case 'g': nucs[2]++; break; case 'T': case 't': nucs[3]++; break; } } // // Determine how many alleles are present at this position in this population. // We cannot deal with more than two alternative alleles, if there are more than two // in a single population, print a warning and exclude this nucleotide position. // int i; int allele_cnt = 0; for (i = 0; i < 4; i++) if (nucs[i] > 0) allele_cnt++; if (allele_cnt > 2) { p_allele = 0; q_allele = 0; return -1; } // // Record which nucleotide is the P allele and which is the Q allele. // (The P allele is the first one alphabetically, and the Q allele the second // one, if any.) // p_allele = 0; q_allele = 0; i = 0; while (p_allele == 0 && i < 4) { if (nucs[i] > 0) { switch(i) { case 0: p_allele = 'A'; break; case 1: p_allele = 'C'; break; case 2: p_allele = 'G'; break; case 3: p_allele = 'T'; break; } } i++; } while (q_allele == 0 && i < 4) { if (nucs[i] > 0) { switch(i) { case 1: q_allele = 'C'; break; case 2: q_allele = 'G'; break; case 3: q_allele = 'T'; break; } } i++; } return 0; } int tally_haplotype_freq(CSLocus *locus, PopMap *pmap, int &total, double &max, string &freq_str) { map freq; Datum **d = pmap->locus(locus->id); total = 0; max = 0; for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] == NULL) continue; if (d[i]->gtype[0] != '-') { freq[d[i]->gtype]++; total++; } } if (total == 0) return 0; double frac; stringstream s; char f[id_len]; map::iterator it; for (it = freq.begin(); it != freq.end(); it++) { frac = (double) it->second / (double) total * 100; if (frac > max) max = frac; sprintf(f, "(%0.1f%%);", frac); s << it->first << ":" << it->second << f; } freq_str = s.str(); return 0; } int datum_adjust_snp_positions(map > &merge_map, CSLocus *loc, Datum *datum, map &snpres) { // // We will start with the 'sink' locus, which was originally on the negative strand: // 1. If the locus was shorter than the catalog locus, pad the difference. // 2. Convert to positive strand: Reverse the order, complement the alleles, // alter the internal column position. // SNP *snp; SNPRes *snpr = snpres[datum->id]; int index = 0; int stop_pos = renz_olap[enz] - 1; // // We know the catalog was padded since we already padded hte model call string // if it was necessary when originally merging. // while (datum->model[index] == 'N') { snp = new SNP; snp->col = index; snp->lratio = 0.0; snp->rank_1 = 'N'; snp->type = snp_type_unk; datum->snps.push_back(snp); index++; } for (int j = snpr->snps.size() - 1; j > stop_pos; j--) { snp = new SNP; snp->col = index; snp->lratio = snpr->snps[j]->lratio; snp->rank_1 = reverse(snpr->snps[j]->rank_1); snp->rank_2 = reverse(snpr->snps[j]->rank_2); snp->rank_3 = reverse(snpr->snps[j]->rank_3); snp->rank_4 = reverse(snpr->snps[j]->rank_4); datum->snps.push_back(snp); index++; } // // Now we fetch the former locus, the 'src', which was originally on the positive strand. // All we have to do is adjust the column position of each SNP. // snpr = snpres[datum->merge_partner]; for (uint j = 0; j < snpres[datum->id]->snps.size(); j++) { snp = new SNP; snp->col = index; snp->lratio = snpr->snps[j]->lratio; snp->rank_1 = snpr->snps[j]->rank_1; snp->rank_2 = snpr->snps[j]->rank_2; snp->rank_3 = snpr->snps[j]->rank_3; snp->rank_4 = snpr->snps[j]->rank_4; datum->snps.push_back(snp); index++; } return 0; } stacks-1.44/src/export_formats.h000644 000765 000024 00000006121 12750704270 017522 0ustar00catchenstaff000000 000000 #ifndef EXPORT_FORMATS_H #define EXPORT_FORMATS_H #include #include #include #include "locus.h" #include "PopMap.h" #include "PopSum.h" #include "ordered.h" // for "snp" #include "populations.h" // for "merget", "InputMode", "uncalled_haplotype()", "count_haplotypes_at_locus()" using std::ofstream; using std::pair; using std::map; class GenPos { public: uint id; uint bp; uint snp_index; loc_type type; GenPos(int id, int snp_index, int bp) { this->id = id; this->snp_index = snp_index; this->bp = bp; this->type = snp; } GenPos(int id, int snp_index, int bp, loc_type type) { this->id = id; this->snp_index = snp_index; this->bp = bp; this->type = type; } bool operator<(const GenPos& other) const {return bp < other.bp;} }; int write_sql(map &, PopMap *); int write_fst_stats(map &, PopMap *, PopSum *, ofstream &); int write_generic(map &, PopMap *, bool); int write_genomic(map &, PopMap *); int write_fasta(map &, PopMap *); int write_strict_fasta(map &, PopMap *); int write_vcf(map &, PopMap *, PopSum *, map > &); int write_vcf_ordered(map &, PopMap *, PopSum *, map > &, ofstream &); int write_vcf_haplotypes(map &, PopMap *, PopSum *); int write_genepop(map &, PopMap *, PopSum *); int write_genepop_ordered(map &, PopMap *, PopSum *, ofstream &); int write_structure(map &, PopMap *, PopSum *); int write_structure_ordered(map &, PopMap *, PopSum *, ofstream &); int write_phase(map &, PopMap *, PopSum *); int write_fastphase(map &, PopMap *, PopSum *); int write_beagle(map &, PopMap *, PopSum *); int write_beagle_phased(map &, PopMap *, PopSum *); int write_plink(map &, PopMap *, PopSum *); int write_hzar(map &, PopMap *, PopSum *); int write_treemix(map &, PopMap *, PopSum *); int write_phylip(map &, PopMap *, PopSum *); int write_fullseq_phylip(map &, PopMap *, PopSum *); int populate_snp_calls(map &, PopMap *, map > &); int find_datum_allele_depths(Datum *, int, char, char, int &, int &); int tally_observed_haplotypes(vector &, int, char &, char &); int tally_haplotype_freq(CSLocus *, PopMap *, int &, double &, string &); int datum_adjust_snp_positions(map > &, CSLocus *, Datum *, map &); #endif // EXPORT_FORMATS_H stacks-1.44/src/FastaI.h000644 000765 000024 00000007515 12761620410 015620 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2013, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __FASTAI_H__ #define __FASTAI_H__ #include "input.h" class Fasta: public Input { string buf; public: Fasta(const char *path) : Input(path) { }; Fasta(string path) : Input(path.c_str()) { }; ~Fasta() {}; Seq *next_seq(); int next_seq(Seq &); }; Seq *Fasta::next_seq() { // // Check the contents of the line buffer. When we finish reading a FASTA record // the buffer will either contain whitespace or the header of the next FAST // record. // while (this->line[0] != '>' && this->fh.good() ) { this->fh.getline(this->line, max_len); } if (!this->fh.good()) { return NULL; } // // Check if there is a carraige return in the buffer // uint len = strlen(this->line); if (this->line[len - 1] == '\r') this->line[len - 1] = '\0'; // // Initialize the Seq structure and store the FASTA ID // Seq *s = new Seq; s->id = new char[len + 1]; strcpy(s->id, this->line + 1); // // Read the sequence from the file -- keep reading lines until we reach the next // record or the end of file. // this->fh.getline(this->line, max_len); while (this->line[0] != '>' && this->fh.good()) { len = strlen(this->line); if (this->line[len - 1] == '\r') this->line[len - 1] = '\0'; this->buf += this->line; this->line[0] = '\0'; this->fh.getline(this->line, max_len); } if (this->fh.eof()) { len = strlen(this->line); if (this->line[len - 1] == '\r') this->line[len - 1] = '\0'; this->buf += this->line; } s->seq = new char[this->buf.length() + 1]; strcpy(s->seq, this->buf.c_str()); this->buf.clear(); return s; } int Fasta::next_seq(Seq &s) { // // Check the contents of the line buffer. When we finish reading a FASTA record // the buffer will either contain whitespace or the header of the next FAST // record. // while (this->line[0] != '>' && this->fh.good() ) { this->fh.getline(this->line, max_len); } if (!this->fh.good()) { return 0; } // // Check if there is a carraige return in the buffer // uint len = strlen(this->line); if (this->line[len - 1] == '\r') this->line[len - 1] = '\0'; // // Store the FASTA ID // strcpy(s.id, this->line + 1); // // Read the sequence from the file -- keep reading lines until we reach the next // record or the end of file. // this->fh.getline(this->line, max_len); while (this->line[0] != '>' && this->fh.good()) { len = strlen(this->line); if (len > 0 && this->line[len - 1] == '\r') this->line[len - 1] = '\0'; this->buf += this->line; this->line[0] = '\0'; this->fh.getline(this->line, max_len); } if (this->fh.eof()) { len = strlen(this->line); if (len > 0 && this->line[len - 1] == '\r') this->line[len - 1] = '\0'; this->buf += this->line; } strcpy(s.seq, this->buf.c_str()); this->buf.clear(); return 1; } #endif // __FASTAI_H__ stacks-1.44/src/FastqI.h000644 000765 000024 00000010631 12764047443 015645 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2013, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __FASTQI_H__ #define __FASTQI_H__ #include "input.h" class Fastq: public Input { public: Fastq(const char *path) : Input(path) { }; Fastq(string path) : Input(path.c_str()) { }; ~Fastq() {}; Seq *next_seq(); int next_seq(Seq &s); }; Seq *Fastq::next_seq() { // // Check the contents of the line buffer. When we finish reading a FASTQ record // the buffer will either contain whitespace or the header of the next FASTQ // record. // while (this->line[0] != '@' && this->fh.good() ) { this->fh.getline(this->line, max_len); } if (!this->fh.good()) { return NULL; } // // Check if there is a carraige return in the buffer // uint len = strlen(this->line); if (len > 0 && this->line[len - 1] == '\r') this->line[len - 1] = '\0'; // // Initialize the Seq structure and store the FASTQ ID // Seq *s = new Seq; s->id = new char[strlen(this->line) + 1]; strcpy(s->id, this->line + 1); // // Read the sequence from the file // this->fh.getline(this->line, max_len); if (!this->fh.good()) { return NULL; } len = strlen(this->line); if (len > 0 && this->line[len - 1] == '\r') this->line[len - 1] = '\0'; s->seq = new char[len + 1]; strcpy(s->seq, this->line); // // Read the repeat of the ID // this->fh.getline(this->line, max_len); if (this->line[0] != '+' || !this->fh.good()) { return NULL; } // // Read the quality score from the file // this->fh.getline(this->line, max_len); if (!this->fh.good() && !this->fh.eof()) { return NULL; } len = strlen(this->line); if (len > 0 && this->line[len - 1] == '\r') this->line[len - 1] = '\0'; s->qual = new char[len + 1]; strcpy(s->qual, this->line); // // Clear the line buffer so it is set up for the next record. If a '@' // appears in the quality scores read, it will break parsing next time // it is called. // this->line[0] = '\0'; return s; } int Fastq::next_seq(Seq &s) { // // Check the contents of the line buffer. When we finish reading a FASTQ record // the buffer will either contain whitespace or the header of the next FASTQ // record. // while (this->line[0] != '@' && this->fh.good() ) { this->fh.getline(this->line, max_len); } if (!this->fh.good()) { return 0; } // // Check if there is a carraige return in the buffer // uint len = strlen(this->line); if (this->line[len - 1] == '\r') this->line[len - 1] = '\0'; // // Store the FASTQ ID // strcpy(s.id, this->line + 1); // // Read the sequence from the file // this->fh.getline(this->line, max_len); if (!this->fh.good()) { return 0; } len = strlen(this->line); if (this->line[len - 1] == '\r') this->line[len - 1] = '\0'; strcpy(s.seq, this->line); // // Read the repeat of the ID // this->fh.getline(this->line, max_len); if (this->line[0] != '+' || !this->fh.good()) { return 0; } // // Read the quality score from the file // this->fh.getline(this->line, max_len); if (!this->fh.good() && !this->fh.eof()) { return 0; } len = strlen(this->line); if (this->line[len - 1] == '\r') this->line[len - 1] = '\0'; strcpy(s.qual, this->line); // // Clear the line buffer so it is set up for the next record. If a '@' // appears in the quality scores read, it will break parsing next time // it is called. // this->line[0] = '\0'; return 1; } #endif // __FASTQI_H__ stacks-1.44/src/file_io.cc000644 000765 000024 00000077323 12774470747 016245 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2012-2014, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // file_io.cc -- common routines for opening groups of files and processing barcode lists. // // Julian Catchen // jcatchen@uoregon.edu // University of Oregon // #include "file_io.h" int open_files(vector > &files, vector &barcodes, map &pair_1_fhs, map &pair_2_fhs, map &rem_1_fhs, map &rem_2_fhs, map > &counters) { string path, suffix_1, suffix_2, filepath, file; if (paired) { suffix_1 = ".1"; suffix_2 = ".2"; } if (out_file_type == FileT::fastq) { suffix_1 += ".fq"; suffix_2 += ".fq"; } else { suffix_1 += ".fa"; suffix_2 += ".fa"; } uint pos; ofstream *fh; BarcodePair bc; // // If the size of the barcodes vector is 0, then no barcodes // were submitted. In this case, we want to open output files // of the same name as input files, but in out_path. // if (barcodes.size() == 0 && merge == false) { struct stat sb_1, sb_2; for (uint i = 0; i < files.size(); i++) { bc.se = files[i].first; if (paired) bc.pe = files[i].second; path = out_path + files[i].first; // // Check that the file has a proper suffix for the output type. // pos = path.find_last_of("."); if (path.substr(pos) == ".bam") { path = path.substr(0, pos) + suffix_1; } else if (path.substr(pos) == ".gz") { path = path.substr(0, pos); pos = path.find_last_of("."); path = path.substr(0, pos) + suffix_1; } else { path = path.substr(0, pos) + suffix_1; } if (stat((in_path_1 + files[i].first).c_str(), &sb_1) == -1) { cerr << "Unable to stat input file '" << in_path_1 + files[i].first << "'\n"; exit(1); } if (stat(path.c_str(), &sb_2) == 0 && sb_2.st_dev == sb_1.st_dev && sb_2.st_ino == sb_1.st_ino) { cerr << "Input and output files ('" << path << "') are the same and will cause the input " << "file to be overwritten. Please specify a separate output directory using '-o'.\n"; help(); } fh = new ofstream(path.c_str(), ifstream::out); pair_1_fhs[bc] = fh; if (pair_1_fhs[bc]->fail()) { cerr << "Error opening output file '" << path << "'\n"; exit(1); } if (paired) { file = interleaved ? files[i].first : files[i].second; path = out_path + file; pos = path.find_last_of("."); if (path.substr(pos) == ".bam") { path = path.substr(0, pos) + suffix_2; } else if (path.substr(pos) == ".gz") { path = path.substr(0, pos); pos = path.find_last_of("."); path = path.substr(0, pos) + suffix_2; } else { path = path.substr(0, pos) + suffix_2; } if (stat((in_path_2 + file).c_str(), &sb_1) == -1) { cerr << "Unable to stat input file '" << in_path_2 + file << "'\n"; exit(1); } if (stat(path.c_str(), &sb_2) == 0 && sb_2.st_dev == sb_1.st_dev && sb_2.st_ino == sb_1.st_ino) { cerr << "Input and output file names ('" << path << "') are the same and will cause the input " << "file to be overwritten. Please specify a separate output directory using '-o'.\n"; help(); } fh = new ofstream(path.c_str(), ifstream::out); pair_2_fhs[bc] = fh; if (pair_2_fhs[bc]->fail()) { cerr << "Error opening output file '" << path << "'\n"; exit(1); } filepath = files[i].first; pos = filepath.find_last_of("."); if (filepath.substr(pos) == ".gz") { filepath = filepath.substr(0, pos); pos = filepath.find_last_of("."); filepath = filepath.substr(0, pos); } else if (filepath.substr(pos) == ".bam") { filepath = filepath.substr(0, pos); } path = out_path + filepath + ".rem" + suffix_1; fh = new ofstream(path.c_str(), ifstream::out); rem_1_fhs[bc] = fh; if (rem_1_fhs[bc]->fail()) { cerr << "Error opening remainder output file '" << path << "'\n"; exit(1); } filepath = file; pos = filepath.find_last_of("."); if (filepath.substr(pos) == ".gz") { filepath = filepath.substr(0, pos); pos = filepath.find_last_of("."); filepath = filepath.substr(0, pos); } else if (filepath.substr(pos) == ".bam") { filepath = filepath.substr(0, pos); } path = out_path + filepath + ".rem" + suffix_2; fh = new ofstream(path.c_str(), ifstream::out); rem_2_fhs[bc] = fh; if (rem_2_fhs[bc]->fail()) { cerr << "Error opening remainder output file '" << path << "'\n"; exit(1); } } } return 0; } else if (barcodes.size() == 0 && merge == true) { path = out_path + "sample_unbarcoded" + suffix_1; fh = new ofstream(path.c_str(), ifstream::out); if (fh->fail()) { cerr << "Error opening output file '" << path << "'\n"; exit(1); } for (uint i = 0; i < files.size(); i++) { bc.se = files[i].first; if (paired) bc.pe = files[i].second; pair_1_fhs[bc] = fh; } if (paired) { path = out_path + "sample_unbarcoded" + suffix_2; fh = new ofstream(path.c_str(), ifstream::out); if (fh->fail()) { cerr << "Error opening output file '" << path << "'\n"; exit(1); } for (uint i = 0; i < files.size(); i++) { bc.se = files[i].first; bc.pe = files[i].second; pair_2_fhs[bc] = fh; } path = out_path + "sample_unbarcoded.rem" + suffix_1; fh = new ofstream(path.c_str(), ifstream::out); if (fh->fail()) { cerr << "Error opening remainder output file '" << path << "'\n"; exit(1); } for (uint i = 0; i < files.size(); i++) { bc.se = files[i].first; bc.pe = files[i].second; rem_1_fhs[bc] = fh; } path = out_path + "sample_unbarcoded.rem" + suffix_2; fh = new ofstream(path.c_str(), ifstream::out); if (fh->fail()) { cerr << "Error opening remainder output file '" << path << "'\n"; exit(1); } for (uint i = 0; i < files.size(); i++) { bc.se = files[i].first; bc.pe = files[i].second; rem_2_fhs[bc] = fh; } } return 0; } string filename; for (uint i = 0; i < barcodes.size(); i++) { filename = barcodes[i].name_exists() ? barcodes[i].name : "sample_" + barcodes[i].str(); path = out_path + filename + suffix_1; fh = new ofstream(path.c_str(), ifstream::out); pair_1_fhs[barcodes[i]] = fh; if (pair_1_fhs[barcodes[i]]->fail()) { cerr << "Error opening output file '" << path << "'\n"; exit(1); } if (paired) { path = out_path + filename + suffix_2; fh = new ofstream(path.c_str(), ifstream::out); pair_2_fhs[barcodes[i]] = fh; if (pair_2_fhs[barcodes[i]]->fail()) { cerr << "Error opening output file '" << path << "'\n"; exit(1); } path = out_path + filename + ".rem" + suffix_1; fh = new ofstream(path.c_str(), ifstream::out); rem_1_fhs[barcodes[i]] = fh; if (rem_1_fhs[barcodes[i]]->fail()) { cerr << "Error opening remainder output file '" << path << "'\n"; exit(1); } path = out_path + filename + ".rem" + suffix_2; fh = new ofstream(path.c_str(), ifstream::out); rem_2_fhs[barcodes[i]] = fh; if (rem_2_fhs[barcodes[i]]->fail()) { cerr << "Error opening remainder output file '" << path << "'\n"; exit(1); } } } return 0; } int open_files(vector > &files, vector &barcodes, map &pair_1_fhs, map &pair_2_fhs, map &rem_1_fhs, map &rem_2_fhs, map > &counters) { string path, suffix_1, suffix_2, filepath, file; if (paired) { suffix_1 = ".1"; suffix_2 = ".2"; } if (out_file_type == FileT::gzfastq) { suffix_1 += ".fq.gz"; suffix_2 += ".fq.gz"; } else { suffix_1 += ".fa.gz"; suffix_2 += ".fa.gz"; } uint pos; gzFile *fh; BarcodePair bc; // // If the size of the barcodes vector is 0, then no barcodes // were submitted. In this case, we want to open output files // of the same name as input files, but in out_path. // if (barcodes.size() == 0 && merge == false) { struct stat sb_1, sb_2; for (uint i = 0; i < files.size(); i++) { bc.se = files[i].first; if (paired) bc.pe = files[i].second; path = out_path + files[i].first; // // Check that the file has a proper suffix for the output type. // pos = path.find_last_of("."); if (path.substr(pos) == ".bam") { path = path.substr(0, pos) + suffix_1; } else if (path.substr(pos) == ".gz") { path = path.substr(0, pos); pos = path.find_last_of("."); path = path.substr(0, pos) + suffix_1; } else { path = path.substr(0, pos) + suffix_1; } if (stat((in_path_1 + files[i].first).c_str(), &sb_1) == -1) { cerr << "Unable to stat input file '" << in_path_1 + files[i].first << "'\n"; exit(1); } if (stat(path.c_str(), &sb_2) == 0 && sb_2.st_dev == sb_1.st_dev && sb_2.st_ino == sb_1.st_ino) { cerr << "Input and output files ('" << path << "') are the same and will cause the input " << "file to be overwritten. Please specify a separate output directory using '-o'.\n"; help(); } fh = new gzFile; *fh = gzopen(path.c_str(), "wb"); pair_1_fhs[bc] = fh; if (!(*fh)) { cerr << "Error opening output file '" << path << "'\n"; exit(1); } if (paired) { file = interleaved ? files[i].first : files[i].second; path = out_path + file; pos = path.find_last_of("."); if (path.substr(pos) == ".bam") { path.replace(pos, 4, suffix_2); } else if (path.substr(pos) == ".gz") { path = path.substr(0, pos); pos = path.find_last_of("."); path = path.substr(0, pos) + suffix_2; } else { path = path.substr(0, pos) + suffix_2; } if (stat((in_path_2 + file).c_str(), &sb_1) == -1) { cerr << "Unable to stat input file '" << in_path_2 + file << "'\n"; exit(1); } if (stat(path.c_str(), &sb_2) == 0 && sb_2.st_dev == sb_1.st_dev && sb_2.st_ino == sb_1.st_ino) { cerr << "Input and output file names ('" << path << "') are the same and will cause the input " << "file to be overwritten. Please specify a separate output directory using '-o'.\n"; help(); } fh = new gzFile; *fh = gzopen(path.c_str(), "wb"); pair_2_fhs[bc] = fh; if (!(*fh)) { cerr << "Error opening output file '" << path << "'\n"; exit(1); } filepath = files[i].first; pos = filepath.find_last_of("."); if (filepath.substr(pos) == ".gz") { filepath = filepath.substr(0, pos); pos = filepath.find_last_of("."); filepath = filepath.substr(0, pos); } else if (filepath.substr(pos) == ".bam") { filepath = filepath.substr(0, pos); } path = out_path + filepath + ".rem" + suffix_1; fh = new gzFile; *fh = gzopen(path.c_str(), "wb"); rem_1_fhs[bc] = fh; if (!*(fh)) { cerr << "Error opening remainder output file '" << path << "'\n"; exit(1); } filepath = file; pos = filepath.find_last_of("."); if (filepath.substr(pos) == ".gz") { filepath = filepath.substr(0, pos); pos = filepath.find_last_of("."); filepath = filepath.substr(0, pos); } else if (filepath.substr(pos) == ".bam") { filepath = filepath.substr(0, pos); } path = out_path + filepath + ".rem" + suffix_2; fh = new gzFile; *fh = gzopen(path.c_str(), "wb"); rem_2_fhs[bc] = fh; if (!(*fh)) { cerr << "Error opening remainder output file '" << path << "'\n"; exit(1); } } } return 0; } else if (barcodes.size() == 0 && merge == true) { path = out_path + "sample_unbarcoded" + suffix_1; fh = new gzFile; *fh = gzopen(path.c_str(), "wb"); if (!(*fh)) { cerr << "Error opening output file '" << path << "'\n"; exit(1); } for (uint i = 0; i < files.size(); i++) { bc.se = files[i].first; if (paired) bc.pe = files[i].second; pair_1_fhs[bc] = fh; } if (paired) { path = out_path + "sample_unbarcoded" + suffix_2; fh = new gzFile; *fh = gzopen(path.c_str(), "wb"); if (!(*fh)) { cerr << "Error opening output file '" << path << "'\n"; exit(1); } for (uint i = 0; i < files.size(); i++) { bc.se = files[i].first; bc.pe = files[i].second; pair_2_fhs[bc] = fh; } path = out_path + "sample_unbarcoded.rem" + suffix_1; fh = new gzFile; *fh = gzopen(path.c_str(), "wb"); if (!(*fh)) { cerr << "Error opening remainder output file '" << path << "'\n"; exit(1); } for (uint i = 0; i < files.size(); i++) { bc.se = files[i].first; bc.pe = files[i].second; rem_1_fhs[bc] = fh; } path = out_path + "sample_unbarcoded.rem" + suffix_2; fh = new gzFile; *fh = gzopen(path.c_str(), "wb"); if (!(*fh)) { cerr << "Error opening remainder output file '" << path << "'\n"; exit(1); } for (uint i = 0; i < files.size(); i++) { bc.se = files[i].first; bc.pe = files[i].second; rem_2_fhs[bc] = fh; } } return 0; } string filename; for (uint i = 0; i < barcodes.size(); i++) { filename = barcodes[i].name_exists() ? barcodes[i].name : "sample_" + barcodes[i].str(); path = out_path + filename + suffix_1; fh = new gzFile; *fh = gzopen(path.c_str(), "wb"); pair_1_fhs[barcodes[i]] = fh; if (!(*pair_1_fhs[barcodes[i]])) { cerr << "Error opening output file '" << path << "'\n"; exit(1); } if (paired) { path = out_path + filename + suffix_2; fh = new gzFile; *fh = gzopen(path.c_str(), "wb"); pair_2_fhs[barcodes[i]] = fh; if (!(*pair_2_fhs[barcodes[i]])) { cerr << "Error opening output file '" << path << "'\n"; exit(1); } path = out_path + filename + ".rem" + suffix_1; fh = new gzFile; *fh = gzopen(path.c_str(), "wb"); rem_1_fhs[barcodes[i]] = fh; if (!(*rem_1_fhs[barcodes[i]])) { cerr << "Error opening remainder output file '" << path << "'\n"; exit(1); } path = out_path + filename + ".rem" + suffix_2; fh = new gzFile; *fh = gzopen(path.c_str(), "wb"); rem_2_fhs[barcodes[i]] = fh; if (!(*rem_2_fhs[barcodes[i]])) { cerr << "Error opening remainder output file '" << path << "'\n"; exit(1); } } } return 0; } int close_file_handles(map &fhs) { map::iterator i; set ptrs; set::iterator j; for (i = fhs.begin(); i != fhs.end(); i++) { i->second->close(); ptrs.insert(i->second); } for (j = ptrs.begin(); j != ptrs.end(); j++) { delete *j; } return 0; } int close_file_handles(map &fhs) { map::iterator i; set ptrs; set::iterator j; for (i = fhs.begin(); i != fhs.end(); i++) { gzclose(*(i->second)); ptrs.insert(i->second); } for (j = ptrs.begin(); j != ptrs.end(); j++) { delete *j; } return 0; } int load_barcodes(string barcode_file, vector &barcodes, set &se_bc, set &pe_bc, uint &min_se_len, uint &max_se_len, uint &min_pe_len, uint &max_pe_len) { switch(barcode_type) { case null_null: cerr << "Barcode type unspecified, assuming unbarcoded data.\n"; break; case null_index: cerr << "Searching for single, index barcode.\n"; break; case index_null: cerr << "Searching for single-end, indexed barcodes.\n"; break; case inline_null: cerr << "Searching for single-end, inlined barcodes.\n"; break; case index_index: cerr << "Searching for single and paired-end, indexed barcodes.\n"; break; case inline_inline: cerr << "Searching for single and paired-end, inlined barcodes.\n"; break; case inline_index: if (paired) cerr << "Searching for single-end, inlined and paired-end, indexed barcodes.\n"; else cerr << "Searching for single-end inlined and indexed barcodes.\n"; break; case index_inline: if (paired) cerr << "Searching for single-end, indexed and paired-end, inlined barcodes.\n"; else cerr << "Searching for single-end, indexed and inlined barcodes.\n"; break; } if (barcode_file.length() == 0) return 0; char line[id_len]; ifstream fh(barcode_file.c_str(), ifstream::in); if (fh.fail()) { cerr << "Error opening barcode file '" << barcode_file << "'\n"; exit(1); } char *p, *q, *r, *s; uint cols, line_num = 0; while (fh.good()) { memset(line, 0, id_len); fh.getline(line, id_len); line_num++; if (strlen(line) == 0) continue; // // Check that the proper number of columns exist. // cols = 1; for (p = line; *p != '\0'; p++) if (*p == '\t') cols++; if (cols > 2 && (barcode_type == inline_null || barcode_type == index_null)) { cerr << "Too many columns (" << cols << ") specified in '" << barcode_file << "' for single-end barcodes on line " << line_num << ".\n"; exit(1); } else if (cols > 3) { cerr << "Too many columns (" << cols << ") specified in '" << barcode_file << "' on line " << line_num << ".\n"; exit(1); } // // Identify the first barcode and check that it's legitimate. // p = line; q = p; while (*q != '\0') { switch (*q) { case 'A': case 'C': case 'G': case 'T': break; case 'a': *q = 'A'; break; case 'c': *q = 'C'; break; case 'g': *q = 'G'; break; case 't': *q = 'T'; break; case '\r': case '\t': *q = '\0'; break; default: cerr << "Invalid barcode on line " << line_num << ": '" << p << "'\n"; exit(1); } if (*q != '\0') q++; } // // If a second barcode was specified on the command line, identify it and check that it's legitimate. // r = NULL; if (barcode_type == inline_inline || barcode_type == inline_index || barcode_type == index_inline || barcode_type == index_index) { if (q - p < id_len) q++; r = q; while (*q != '\0') { switch (*q) { case 'A': case 'C': case 'G': case 'T': break; case 'a': *q = 'A'; break; case 'c': *q = 'C'; break; case 'g': *q = 'G'; break; case 't': *q = 'T'; break; case '\r': case '\t': *q = '\0'; break; default: cerr << "Invalid barcode on line " << line_num << ": '" << r << "'\n"; exit(1); } if (*q != '\0') q++; } } // // Check for the existence of a file name to associate with this barcode set. // if (q - p < id_len) q++; s = q; while (*q != '\0') { if (!isalnum(*q)) { switch (*q) { case '-': case '_': case '.': break; case '\r': case '\t': *q = '\0'; break; default: cerr << "Invalid filename on line " << line_num << ": '" << s << "' (filenames can consist of letters, numbers, '.', '-' and '_').\n"; exit(1); } } if (*q != '\0') q++; } barcodes.push_back(BarcodePair(p, r, s)); if (p != NULL && strlen(p) > 0) se_bc.insert(string(p)); if (r != NULL && strlen(r) > 0) pe_bc.insert(string(r)); } fh.close(); if (barcodes.size() == 0) { cerr << "Unable to load any barcodes from '" << barcode_file << "'\n"; help(); } // // Make sure barcodes are properly paired up. // int pe_cnt = 0; int se_cnt = 0; for (uint i = 0; i < barcodes.size(); i++) { se_cnt += (barcodes[i].se.length() > 0) ? 1 : 0; pe_cnt += (barcodes[i].pe.length() > 0) ? 1 : 0; } if (pe_cnt > 0 && se_cnt != pe_cnt) { cerr << "Single and paired-end barcodes must be properly paired.\n"; help(); } // // Determine the minimum and maximum barcode lengths for the single-end barcodes. // min_se_len = barcodes[0].se.length(); max_se_len = min_se_len; for (uint i = 1; i < barcodes.size(); i++) { if (barcodes[i].se.length() < min_se_len) min_se_len = barcodes[i].se.length(); else if (barcodes[i].se.length() > max_se_len) max_se_len = barcodes[i].se.length(); } // // Determine the minimum and maximum barcode lengths for the paired-end barcodes. // min_pe_len = barcodes[0].pe.length(); max_pe_len = min_pe_len; for (uint i = 0; i < barcodes.size(); i++) { if (barcodes[i].pe.length() < min_pe_len) min_pe_len = barcodes[i].pe.length(); else if (barcodes[i].pe.length() > max_pe_len) max_pe_len = barcodes[i].pe.length(); } // // If paired barcodes were supplied check that a paired barcode type was // specified and vice versa. // if (se_bc.size() > 0 && pe_bc.size() > 0) { if (barcode_type != inline_inline && barcode_type != index_index && barcode_type != inline_index && barcode_type != index_inline) { cerr << "You provided paried barcodes but did not specify a paired barcode type.\n"; help(); } } else { if (barcode_type != inline_null && barcode_type != index_null) { cerr << "You provided single-end barcodes but did not specify a single-end barcode type.\n"; help(); } } cerr << "Loaded " << barcodes.size() << " barcodes "; if (pe_bc.size() > 0) { if (min_se_len != max_se_len) cerr << "(" << min_se_len << "-" << max_se_len << "bp / "; else cerr << "(" << max_se_len << "bp / "; if (min_pe_len != max_pe_len) cerr << min_pe_len << "-" << max_pe_len << "bp).\n"; else cerr << max_pe_len << "bp).\n"; } else { if (min_se_len != max_se_len) cerr << "(" << min_se_len << "-" << max_se_len << "bp).\n"; else cerr << "(" << max_se_len << "bp).\n"; } return 0; } int build_file_list(vector > &files) { // // Scan a directory for a list of files. // if (in_path_1.length() > 0) { string file, paired_file; const char *p, *q, *end; struct dirent *direntry; DIR *dir = opendir(in_path_1.c_str()); if (dir == NULL) { cerr << "Unable to open directory '" << in_path_1 << "' for reading.\n"; exit(1); } while ((direntry = readdir(dir)) != NULL) { file = direntry->d_name; if (file.substr(0, 1) == ".") continue; // // Check the file suffix to make sure we should process it. // p = file.c_str(); q = p + file.length() + 1; end = q; while (q >= p && *q != '.') q--; if (strcmp(q, ".gz") == 0) { end = q; while (q >= p && *q != '.') q--; } if (strncmp(q, ".fq", end - q) != 0 && strncmp(q, ".fa", end - q) != 0 && strncmp(q, ".fastq", end - q) != 0 && strncmp(q, ".fasta", end - q) != 0 && strncmp(q, ".bam", end - q) != 0 && (file.length()<9 || file.substr(file.length()-9) != "_qseq.txt")) continue; // // If paired-end specified, parse file names to sort out which is which. // if (paired && interleaved == false) { int res; if ((res = parse_illumina_v1(file.c_str())) > 0 || (res = parse_illumina_v2(file.c_str())) > 0) { paired_file = file; paired_file.replace(res, 1, "2"); files.push_back(make_pair(file, paired_file)); } } else { files.push_back(make_pair(file, "")); } } if (files.size() == 0) { cerr << "Unable to locate any input files to process within '" << in_path_1 << "'\n"; } } else { // // Files specified directly: // Break off file path and store path and file name. // if (paired && interleaved == false) { int pos_1 = in_file_p1.find_last_of("/"); in_path_1 = in_file_p1.substr(0, pos_1 + 1); int pos_2 = in_file_p2.find_last_of("/"); in_path_2 = in_file_p2.substr(0, pos_2 + 1); files.push_back(make_pair(in_file_p1.substr(pos_1+1), in_file_p2.substr(pos_2+1))); } else if (paired && interleaved == true) { int pos = in_file.find_last_of("/"); in_path_1 = in_file.substr(0, pos + 1); in_path_2 = in_path_1; files.push_back(make_pair(in_file.substr(pos+1), "")); } else { int pos = in_file.find_last_of("/"); in_path_1 = in_file.substr(0, pos + 1); files.push_back(make_pair(in_file.substr(pos+1), "")); } } cerr << "Found " << files.size(); if (paired && interleaved) cerr << " interleaved, paired input file(s).\n"; else if (paired) cerr << " paired input file(s).\n"; else cerr << " input file(s).\n"; return 0; } string remove_suffix(FileT type, string file) { int pos = file.find_last_of("."); if ((type == FileT::gzfastq || type == FileT::gzfasta) && file.substr(pos) == ".gz") file = file.substr(0, pos); pos = file.find_last_of("."); if (type == FileT::gzfastq || type == FileT::fastq) { if (file.substr(pos) == ".fastq" || file.substr(pos) == ".fq") file = file.substr(0, pos); } else if (type == FileT::gzfasta || type == FileT::fasta) { if (file.substr(pos) == ".fasta" || file.substr(pos) == ".fa") file = file.substr(0, pos); } return file; } stacks-1.44/src/file_io.h000644 000765 000024 00000005710 12764047443 016066 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2012, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __FILE_IO_H__ #define __FILE_IO_H__ #include #include // Process command-line options #include // Open/Read contents of a directory #include #include #include #include #include using std::string; #include using std::map; #include using std::vector; #include using std::pair; #include #include using std::istream; using std::ofstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include #include #include "constants.h" #include "clean.h" #include "input.h" // // Command line options defined in process_radtags and process_shortreads. // extern FileT in_file_type; extern FileT out_file_type; extern barcodet barcode_type; extern bool paired; extern bool interleaved; extern bool merge; extern string out_path; extern string in_file; extern string in_file_p1; extern string in_file_p2; extern string in_path_1; extern string in_path_2; // // Defined externally in process_radtags and process_shortreads. // void help( void ); int build_file_list(vector > &); int load_barcodes(string, vector &, set &, set &, uint &, uint &, uint &, uint &); int open_files(vector > &, vector &, map &, map &, map &, map &, map > &); int open_files(vector > &, vector &, map &, map &, map &, map &, map > &); int close_file_handles(map &); int close_file_handles(map &); string remove_suffix(FileT, string); #endif // __FILE_IO_H__ stacks-1.44/src/GappedAln.h000644 000765 000024 00000037641 12775212561 016320 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __GAPPEDALN_H__ #define __GAPPEDALN_H__ enum dynprog {dynp_down, dynp_right, dynp_diag}; class AlignRes { public: string cigar; uint gap_cnt; uint contiguity; double pct_id; AlignRes() { this->gap_cnt = 0; this->contiguity = 0; this->pct_id = 0.0; } AlignRes(string cigar, uint gapcnt, uint contiguity, double pct_id) { this->cigar = cigar; this->gap_cnt = gapcnt; this->contiguity = contiguity; this->pct_id = pct_id; } }; class AlignPath { public: bool diag; bool left; bool up; AlignPath() { diag = false; left = false; up = false; } int count() { int cnt; cnt = this->up ? 1 : 0; cnt += this->left ? 1 : 0; cnt += this->diag ? 1 : 0; return cnt; } }; // // Needleman-Wunsch Alignment // const double gapopen_score = -10; const double gapext_score = -0.5; const double mismatch_score = -4; const double match_score = 5; class GappedAln { uint _m; uint _n; uint _m_size; uint _n_size; double **matrix; AlignPath **path; AlignRes _aln; inline int swap(double *, dynprog *, int, int); int trace_alignment(string, string); public: GappedAln(); GappedAln(int i) : GappedAln(i, i) {}; GappedAln(int, int); ~GappedAln(); int init(int, int); int align(string, string); AlignRes result(); int parse_cigar(vector > &); int dump_alignment(string, string); }; GappedAln::GappedAln() { this->_m = 0; this->_n = 0; this->_m_size = this->_m; this->_n_size = this->_n; this->matrix = NULL; this->path = NULL; } GappedAln::GappedAln(int len_1, int len_2) { this->_m = len_1 + 1; this->_n = len_2 + 1; this->_m_size = this->_m; this->_n_size = this->_n; this->matrix = new double * [this->_m]; for (uint i = 0; i < this->_m; i++) this->matrix[i] = new double [this->_n]; this->path = new AlignPath * [this->_m]; for (uint i = 0; i < this->_m; i++) this->path[i] = new AlignPath [this->_n]; } GappedAln::~GappedAln() { for (uint i = 0; i < this->_m_size; i++) { delete [] this->matrix[i]; delete [] this->path[i]; } delete [] this->matrix; delete [] this->path; } int GappedAln::init(int size_1, int size_2) { // // Resize the underlying matrix and path arrays, if necessary. // if ((size_1 + 1) > (int)this->_m_size || (size_2 + 1) > (int)this->_n_size) { for (uint i = 0; i < this->_m_size; i++) { delete [] this->matrix[i]; delete [] this->path[i]; } if (this->_m_size > 0) { delete [] this->matrix; delete [] this->path; } this->_m_size = size_1 + 1; this->_n_size = size_2 + 1; // // Resize the arrays to be 25% larger than the requested size. // int new_size = this->_m_size > this->_n_size ? this->_m_size : this->_n_size; new_size += int((double) new_size * 0.25); this->_m_size = new_size; this->_n_size = new_size; this->matrix = new double * [this->_m_size]; for (uint i = 0; i < this->_m_size; i++) this->matrix[i] = new double [this->_n_size]; this->path = new AlignPath * [this->_m_size]; for (uint i = 0; i < this->_m_size; i++) this->path[i] = new AlignPath [this->_n_size]; } // // Otherwise, set the dimensions of the matrix and path arrays to be the sequence lengths. // this->_m = size_1 + 1; this->_n = size_2 + 1; return 0; } int GappedAln::align(string tag_1, string tag_2) { // j----> // [0][1][2][3]...[n-1] // +-------------------- // i [0] | [i][j] // | [1] | // | [2] | // v [3] | // ... | // [m-1] | // // // Initialize the first column and row of the dynamic programming // matrix and the path array. // path[0][0].diag = false; path[0][0].up = false; path[0][0].left = false; matrix[0][0] = 0.0; for (uint i = 1; i < this->_m; i++) { this->matrix[i][0] = this->path[i - 1][0].up ? this->matrix[i - 1][0] + gapext_score : this->matrix[i - 1][0] + gapopen_score; this->path[i][0].diag = false; this->path[i][0].up = true; this->path[i][0].left = false; } for (uint j = 1; j < this->_n; j++) { this->matrix[0][j] = this->path[0][j - 1].left ? this->matrix[0][j - 1] + gapext_score : this->matrix[0][j - 1] + gapopen_score; this->path[0][j].diag = false; this->path[0][j].up = false; this->path[0][j].left = true; } double score_down, score_diag, score_right; double scores[3]; dynprog direction[3]; for (uint i = 1; i < this->_m; i++) { for (uint j = 1; j < this->_n; j++) { // Calculate the score: // 1) If we were to move down from the above cell. score_down = this->matrix[i - 1][j]; score_down += this->path[i - 1][j].up ? gapext_score : gapopen_score; // 2) If we were to move diagonally from the above and left cell. score_diag = this->matrix[i - 1][j - 1] + (tag_1[i - 1] == tag_2[j - 1] ? match_score : mismatch_score); // 3) If we were to move over from the cell left of us. score_right = this->matrix[i][j - 1]; score_right += this->path[i][j - 1].left ? gapext_score : gapopen_score; // // Sort the scores, highest to lowest. // scores[0] = score_down; direction[0] = dynp_down; scores[1] = score_diag; direction[1] = dynp_diag; scores[2] = score_right; direction[2] = dynp_right; if (scores[0] < scores[1]) this->swap(scores, direction, 0, 1); if (scores[1] < scores[2]) this->swap(scores, direction, 1, 2); if (scores[0] < scores[1]) this->swap(scores, direction, 0, 1); this->matrix[i][j] = scores[0]; if (scores[0] > scores[1]) { // // One path is best. // switch (direction[0]) { case dynp_diag: this->path[i][j].diag = true; this->path[i][j].up = false; this->path[i][j].left = false; break; case dynp_down: this->path[i][j].diag = false; this->path[i][j].up = true; this->path[i][j].left = false; break; case dynp_right: default: this->path[i][j].diag = false; this->path[i][j].up = false; this->path[i][j].left = true; } } else if (scores[0] == scores[1]) { // // Two of the paths are equivalent. // switch (direction[0]) { case dynp_diag: this->path[i][j].diag = true; switch (direction[1]) { case dynp_down: this->path[i][j].up = true; this->path[i][j].left = false; break; default: case dynp_right: this->path[i][j].up = false; this->path[i][j].left = true; break; } break; case dynp_down: this->path[i][j].up = true; switch (direction[1]) { case dynp_right: this->path[i][j].diag = false; this->path[i][j].left = true; break; default: case dynp_diag: this->path[i][j].diag = true; this->path[i][j].left = false; break; } break; default: case dynp_right: this->path[i][j].left = true; switch (direction[1]) { case dynp_diag: this->path[i][j].diag = true; this->path[i][j].up = false; break; default: case dynp_down: this->path[i][j].diag = false; this->path[i][j].up = true; break; } break; } } else { // // All paths equivalent. // this->path[i][j].diag = true; this->path[i][j].up = true; this->path[i][j].left = true; } } } // dump_alignment(tag_1, tag_2, matrix, path); if (this->trace_alignment(tag_1, tag_2)) return 1; return 0; } inline int GappedAln::swap(double *scores, dynprog *direction, int index_1, int index_2) { double swap = scores[index_1]; scores[index_1] = scores[index_2]; scores[index_2] = swap; dynprog swapdir = direction[index_1]; direction[index_1] = direction[index_2]; direction[index_2] = swapdir; return 0; } bool compare_alignres(AlignRes a, AlignRes b) { if (a.gap_cnt == b.gap_cnt) { if (a.pct_id == b.pct_id) return (a.contiguity > b.contiguity); else return (a.pct_id > b.pct_id); } else { return (a.gap_cnt < b.gap_cnt); } } int GappedAln::trace_alignment(string tag_1, string tag_2) { // j----> // [0][1][2][3]...[n-1] // +-------------------- // i [0] | [i][j] // | [1] | // | [2] | // v [3] | // ... | // [m-1] | // int i, j, cnt, len, gaps, contiguity; double ident; string cigar; char buf[id_len]; vector alns; bool more_paths = true; bool seq_break = false; do { more_paths = false; i = this->_m - 1; j = this->_n - 1; string aln_1, aln_2; while (i > 0 || j > 0) { cnt = this->path[i][j].count(); if (cnt > 1) more_paths = true; if (this->path[i][j].diag) { aln_1 += tag_1[i - 1]; aln_2 += tag_2[j - 1]; if (cnt > 1) this->path[i][j].diag = false; i--; j--; } else if (this->path[i][j].up) { aln_1 += tag_1[i - 1]; aln_2 += "-"; if (cnt > 1) this->path[i][j].up = false; i--; } else if (this->path[i][j].left) { aln_1 += "-"; aln_2 += tag_2[j - 1]; if (cnt > 1) this->path[i][j].left = false; j--; } } reverse(aln_1.begin(), aln_1.end()); reverse(aln_2.begin(), aln_2.end()); // // Convert to CIGAR strings. // cigar = ""; len = aln_1.length(); gaps = 0; contiguity = 0; seq_break = false; ident = 0.0; i = 0; while (i < len) { if (aln_1[i] != '-' && aln_2[i] != '-') { cnt = 0; do { if (aln_1[i] == aln_2[i]) ident++; cnt++; i++; if (seq_break == false) contiguity++; } while (i < len && aln_1[i] != '-' && aln_2[i] != '-'); sprintf(buf, "%dM", cnt); } else if (aln_1[i] == '-') { cnt = 0; do { cnt++; i++; } while (i < len && aln_1[i] == '-'); sprintf(buf, "%dD", cnt); gaps++; seq_break = true; } else { cnt = 0; do { cnt++; i++; } while (i < len && aln_2[i] == '-'); sprintf(buf, "%dI", cnt); gaps++; seq_break = true; } cigar += buf; } alns.push_back(AlignRes(cigar, gaps, contiguity, (ident / (double) len))); // cerr << aln_1 << " [" << cigar << ", contiguity: " << contiguity << ", gaps: " << gaps << "]\n" // << aln_2 << "\n"; } while (more_paths); sort(alns.begin(), alns.end(), compare_alignres); this->_aln = alns[0]; // cerr << "Final alignment: " << this->_aln.cigar << "; contiguity: " << contiguity << "; gaps: " << this->_aln.gap_cnt << "\n"; return 1; } AlignRes GappedAln::result() { return this->_aln; } int GappedAln::parse_cigar(vector > &cigar) { char buf[id_len]; int dist; const char *p, *q; p = this->_aln.cigar.c_str(); cigar.clear(); while (*p != '\0') { q = p + 1; while (*q != '\0' && isdigit(*q)) q++; strncpy(buf, p, q - p); buf[q-p] = '\0'; dist = atoi(buf); cigar.push_back(make_pair(*q, dist)); p = q + 1; } return 0; } int GappedAln::dump_alignment(string tag_1, string tag_2) { // j----> // [0][1][2][3]...[n-1] // +-------------------- // i [0] | [i][j] // | [1] | // | [2] | // v [3] | // ... | // [m-1] | // // // Output the score matrix. // cout << " "; for (uint j = 0; j < this->_n; j++) cout << " " << tag_2[j] << " |"; cout << "\n"; cout << " "; for (uint j = 0; j < this->_n; j++) printf("% 6.1f|", this->matrix[0][j]); cout << "\n"; for (uint i = 1; i < this->_m; i++) { cout << tag_1[i - 1] << " "; for (uint j = 0; j < this->_n; j++) printf("% 6.1f|", this->matrix[i][j]); cout << "\n"; } cout << "\n"; // // Output the path matrix. // cout << " "; for (uint j = 0; j < this->_n; j++) cout << " " << tag_2[j] << " |"; cout << "\n"; cout << " "; for (uint j = 0; j < this->_n; j++) { cout << " "; this->path[0][j].diag ? cout << "d" : cout << " "; this->path[0][j].up ? cout << "u" : cout << " "; this->path[0][j].left ? cout << "l" : cout << " "; cout << "|"; } cout << "\n"; for (uint i = 1; i < this->_m; i++) { cout << tag_1[i - 1] << " "; for (uint j = 0; j < this->_n; j++) { cout << " "; this->path[i][j].diag ? cout << "d" : cout << " "; this->path[i][j].up ? cout << "u" : cout << " "; this->path[i][j].left ? cout << "l" : cout << " "; cout << "|"; } cout << "\n"; } cout << "\n"; return 0; } #endif // __GAPPEDALN_H__ stacks-1.44/src/genotype_dictionaries.cc000644 000765 000024 00000036157 12750704270 021207 0ustar00catchenstaff000000 000000 #include "genotype_dictionaries.h" int encoded_gtypes[4][4] = { // A C G T {1, 2, 3, 4}, // A {2, 5, 6, 7}, // C {3, 6, 8, 9}, // G {4, 7, 9, 10} // T }; void initialize_dictionaries(map > &global_dictionary) { global_dictionary["ab/--"]["a"] = "aa"; global_dictionary["ab/--"]["b"] = "bb"; global_dictionary["--/ab"]["a"] = "aa"; global_dictionary["--/ab"]["b"] = "bb"; global_dictionary["aa/bb"]["a"] = "aa"; global_dictionary["aa/bb"]["ab"] = "ab"; global_dictionary["aa/bb"]["b"] = "bb"; global_dictionary["ab/ac"]["a"] = "aa"; global_dictionary["ab/ac"]["ab"] = "ab"; global_dictionary["ab/ac"]["b"] = "bb"; global_dictionary["ab/ac"]["ac"] = "ac"; global_dictionary["ab/ac"]["c"] = "cc"; global_dictionary["ab/ac"]["bc"] = "bc"; global_dictionary["ab/cd"]["a"] = "aa"; global_dictionary["ab/cd"]["ab"] = "ab"; global_dictionary["ab/cd"]["b"] = "bb"; global_dictionary["ab/cd"]["c"] = "cc"; global_dictionary["ab/cd"]["cd"] = "cd"; global_dictionary["ab/cd"]["d"] = "dd"; global_dictionary["ab/cd"]["ac"] = "ac"; global_dictionary["ab/cd"]["ad"] = "ad"; global_dictionary["ab/cd"]["bc"] = "bc"; global_dictionary["ab/cd"]["bd"] = "bd"; global_dictionary["ab/aa"]["a"] = "aa"; global_dictionary["ab/aa"]["ab"] = "ab"; global_dictionary["ab/aa"]["b"] = "bb"; global_dictionary["aa/ab"]["a"] = "aa"; global_dictionary["aa/ab"]["ab"] = "ab"; global_dictionary["aa/ab"]["b"] = "bb"; global_dictionary["ab/cc"]["a"] = "aa"; global_dictionary["ab/cc"]["ab"] = "ab"; global_dictionary["ab/cc"]["b"] = "bb"; global_dictionary["ab/cc"]["c"] = "cc"; global_dictionary["ab/cc"]["ac"] = "ac"; global_dictionary["ab/cc"]["bc"] = "bc"; global_dictionary["cc/ab"]["a"] = "aa"; global_dictionary["cc/ab"]["ab"] = "ab"; global_dictionary["cc/ab"]["b"] = "bb"; global_dictionary["cc/ab"]["c"] = "cc"; global_dictionary["cc/ab"]["ac"] = "ac"; global_dictionary["cc/ab"]["bc"] = "bc"; global_dictionary["ab/ab"]["a"] = "aa"; global_dictionary["ab/ab"]["b"] = "bb"; global_dictionary["ab/ab"]["ab"] = "ab"; } void load_cp_dictionary(map &types, map > &dictionary) { types["ab/--"] = "ab/--"; types["--/ab"] = "--/ab"; types["ab/aa"] = "ab/aa"; types["aa/ab"] = "aa/ab"; types["ab/a-"] = "ab/--"; types["-a/ab"] = "--/ab"; types["ab/c-"] = "ab/cd"; types["-c/ab"] = "ab/cd"; types["ab/cc"] = "ab/--"; types["cc/ab"] = "--/ab"; types["ab/ab"] = "ab/ab"; types["ab/ac"] = "ab/ac"; types["ab/cd"] = "ab/cd"; types["-a/bb"] = "ab/--"; types["aa/b-"] = "--/ab"; dictionary["ab/--"]["--"] = "--"; dictionary["ab/--"]["aa"] = "aa"; dictionary["ab/--"]["ab"] = "ab"; dictionary["ab/--"]["bb"] = "ab"; dictionary["ab/--"]["ac"] = "aa"; dictionary["ab/--"]["bc"] = "ab"; dictionary["--/ab"]["--"] = "--"; dictionary["--/ab"]["aa"] = "aa"; dictionary["--/ab"]["ab"] = "ab"; dictionary["--/ab"]["bb"] = "ab"; dictionary["--/ab"]["ac"] = "aa"; dictionary["--/ab"]["bc"] = "ab"; dictionary["ab/aa"]["--"] = "--"; dictionary["ab/aa"]["aa"] = "aa"; dictionary["ab/aa"]["ab"] = "ab"; dictionary["aa/ab"]["--"] = "--"; dictionary["aa/ab"]["aa"] = "aa"; dictionary["aa/ab"]["ab"] = "ab"; dictionary["ab/ab"]["--"] = "--"; dictionary["ab/ab"]["ab"] = "ab"; dictionary["ab/ab"]["aa"] = "aa"; dictionary["ab/ab"]["bb"] = "bb"; dictionary["ab/ac"]["--"] = "--"; dictionary["ab/ac"]["ab"] = "ab"; dictionary["ab/ac"]["ac"] = "ac"; dictionary["ab/ac"]["bc"] = "bc"; dictionary["ab/ac"]["aa"] = "aa"; dictionary["ab/cd"]["--"] = "--"; dictionary["ab/cd"]["ac"] = "ac"; dictionary["ab/cd"]["ad"] = "ad"; dictionary["ab/cd"]["bc"] = "bc"; dictionary["ab/cd"]["bd"] = "bd"; return; } void load_joinmap_cp_dictionary(map &types, map > &dictionary) { types["ab/--"] = "lmx--"; types["--/ab"] = "--xnp"; types["ab/aa"] = "lmxll"; types["aa/ab"] = "nnxnp"; types["ab/ab"] = "hkxhk"; types["ab/ac"] = "efxeg"; types["ab/cd"] = "abxcd"; dictionary["lmx--"]["--"] = "--"; dictionary["lmx--"]["aa"] = "ll"; dictionary["lmx--"]["ab"] = "lm"; dictionary["lmx--"]["bb"] = "lm"; dictionary["lmx--"]["ac"] = "ll"; dictionary["lmx--"]["bc"] = "lm"; dictionary["--xnp"]["--"] = "--"; dictionary["--xnp"]["aa"] = "nn"; dictionary["--xnp"]["ab"] = "np"; dictionary["--xnp"]["bb"] = "np"; dictionary["--xnp"]["ac"] = "nn"; dictionary["--xnp"]["bc"] = "np"; dictionary["lmxll"]["--"] = "--"; dictionary["lmxll"]["aa"] = "ll"; dictionary["lmxll"]["ab"] = "lm"; dictionary["nnxnp"]["--"] = "--"; dictionary["nnxnp"]["aa"] = "nn"; dictionary["nnxnp"]["ab"] = "np"; dictionary["hkxhk"]["--"] = "--"; dictionary["hkxhk"]["ab"] = "hk"; dictionary["hkxhk"]["aa"] = "hh"; dictionary["hkxhk"]["bb"] = "kk"; dictionary["efxeg"]["--"] = "--"; dictionary["efxeg"]["ab"] = "ef"; dictionary["efxeg"]["ac"] = "eg"; dictionary["efxeg"]["bc"] = "fg"; dictionary["efxeg"]["aa"] = "ee"; dictionary["abxcd"]["--"] = "--"; dictionary["abxcd"]["ac"] = "ac"; dictionary["abxcd"]["ad"] = "ad"; dictionary["abxcd"]["bc"] = "bc"; dictionary["abxcd"]["bd"] = "bd"; return; } void load_onemap_cp_dictionary(map &types, map > &dictionary) { types["ab/--"] = "abxoo"; types["--/ab"] = "ooxab"; types["ab/aa"] = "abxaa"; types["aa/ab"] = "aaxab"; types["ab/ab"] = "abxab"; types["ab/ac"] = "abxac"; types["ab/cd"] = "abxcd"; // D1.11 dictionary["abxoo"]["--"] = "-"; dictionary["abxoo"]["aa"] = "a"; dictionary["abxoo"]["bb"] = "b"; // D2.16 dictionary["ooxab"]["--"] = "-"; dictionary["ooxab"]["aa"] = "a"; dictionary["ooxab"]["bb"] = "b"; // D1.10 dictionary["abxaa"]["--"] = "-"; dictionary["abxaa"]["aa"] = "a"; dictionary["abxaa"]["ab"] = "ab"; // D2.15 dictionary["aaxab"]["--"] = "-"; dictionary["aaxab"]["aa"] = "a"; dictionary["aaxab"]["ab"] = "ab"; // B3.7 dictionary["abxab"]["--"] = "-"; dictionary["abxab"]["ab"] = "ab"; dictionary["abxab"]["aa"] = "a"; dictionary["abxab"]["bb"] = "b"; // A.2 dictionary["abxac"]["--"] = "-"; dictionary["abxac"]["ab"] = "ba"; dictionary["abxac"]["ac"] = "ac"; dictionary["abxac"]["bc"] = "bc"; dictionary["abxac"]["aa"] = "a"; // A.1 dictionary["abxcd"]["--"] = "-"; dictionary["abxcd"]["ac"] = "ac"; dictionary["abxcd"]["ad"] = "ad"; dictionary["abxcd"]["bc"] = "bc"; dictionary["abxcd"]["bd"] = "bd"; return; } void load_bc_dictionary(map &types, map > &dictionary) { types["aa/bb"] = "aa/bb"; types["bb/aa"] = "bb/aa"; types["ab/cc"] = "ab/cc"; types["cc/ab"] = "cc/ab"; dictionary["aa/bb"]["--"] = "--"; dictionary["aa/bb"]["aa"] = "aa"; dictionary["aa/bb"]["ab"] = "ab"; dictionary["aa/bb"]["bb"] = "bb"; dictionary["bb/aa"]["--"] = "--"; dictionary["bb/aa"]["aa"] = "aa"; dictionary["bb/aa"]["ab"] = "ab"; dictionary["bb/aa"]["bb"] = "bb"; dictionary["ab/cc"]["--"] = "--"; dictionary["ab/cc"]["ac"] = "ac"; dictionary["ab/cc"]["bc"] = "bc"; dictionary["ab/cc"]["ab"] = "ab"; dictionary["ab/cc"]["aa"] = "aa"; dictionary["ab/cc"]["bb"] = "bb"; dictionary["cc/ab"]["--"] = "--"; dictionary["cc/ab"]["ac"] = "ac"; dictionary["cc/ab"]["bc"] = "bc"; dictionary["cc/ab"]["ab"] = "ab"; dictionary["cc/ab"]["aa"] = "aa"; dictionary["cc/ab"]["bb"] = "bb"; } void load_f2_dictionary(map &types, map > &dictionary) { types["aa/bb"] = "aa/bb"; types["ab/cd"] = "ab/cd"; types["ab/aa"] = "ab/aa"; types["aa/ab"] = "aa/ab"; types["ab/cc"] = "ab/cc"; types["cc/ab"] = "cc/ab"; dictionary["aa/bb"]["aa"] = "aa"; dictionary["aa/bb"]["ab"] = "ab"; dictionary["aa/bb"]["bb"] = "bb"; dictionary["aa/bb"]["--"] = "--"; dictionary["ab/cd"]["aa"] = "aa"; dictionary["ab/cd"]["ab"] = "ab"; dictionary["ab/cd"]["bb"] = "bb"; dictionary["ab/cd"]["cc"] = "cc"; dictionary["ab/cd"]["cd"] = "cd"; dictionary["ab/cd"]["dd"] = "dd"; dictionary["ab/cd"]["ac"] = "ac"; dictionary["ab/cd"]["ad"] = "ad"; dictionary["ab/cd"]["bc"] = "bc"; dictionary["ab/cd"]["bd"] = "bd"; dictionary["ab/cd"]["--"] = "--"; dictionary["ab/aa"]["aa"] = "--"; dictionary["ab/aa"]["ab"] = "--"; dictionary["ab/aa"]["bb"] = "bb"; dictionary["ab/aa"]["--"] = "--"; dictionary["aa/ab"]["aa"] = "--"; dictionary["aa/ab"]["ab"] = "--"; dictionary["aa/ab"]["bb"] = "bb"; dictionary["aa/ab"]["--"] = "--"; dictionary["ab/cc"]["aa"] = "aa"; dictionary["ab/cc"]["ab"] = "ab"; dictionary["ab/cc"]["bb"] = "bb"; dictionary["ab/cc"]["cc"] = "cc"; dictionary["ab/cc"]["ac"] = "--"; dictionary["ab/cc"]["bc"] = "--"; dictionary["ab/cc"]["--"] = "--"; dictionary["cc/ab"]["aa"] = "aa"; dictionary["cc/ab"]["ab"] = "ab"; dictionary["cc/ab"]["bb"] = "bb"; dictionary["cc/ab"]["cc"] = "cc"; dictionary["cc/ab"]["ac"] = "--"; dictionary["cc/ab"]["bc"] = "--"; dictionary["cc/ab"]["--"] = "--"; } void load_mm_bc_dictionary(map &types, map > &dictionary) { types["aa/bb"] = "aaxbb"; types["bb/aa"] = "bbxaa"; types["ab/cc"] = "abxcc"; types["cc/ab"] = "ccxab"; dictionary["aaxbb"]["--"] = "-"; dictionary["aaxbb"]["aa"] = "b"; dictionary["aaxbb"]["ab"] = "h"; dictionary["aaxbb"]["bb"] = "h"; dictionary["bbxaa"]["--"] = "-"; dictionary["bbxaa"]["aa"] = "h"; dictionary["bbxaa"]["ab"] = "h"; dictionary["bbxaa"]["bb"] = "a"; dictionary["abxcc"]["--"] = "-"; dictionary["abxcc"]["ac"] = "h"; dictionary["abxcc"]["bc"] = "h"; dictionary["abxcc"]["ab"] = "b"; dictionary["abxcc"]["aa"] = "b"; dictionary["abxcc"]["bb"] = "b"; dictionary["ccxab"]["--"] = "-"; dictionary["ccxab"]["ac"] = "h"; dictionary["ccxab"]["bc"] = "h"; dictionary["ccxab"]["ab"] = "a"; dictionary["ccxab"]["aa"] = "a"; dictionary["ccxab"]["bb"] = "a"; } void load_mm_f2_dictionary(map &types, map > &dictionary) { types["aa/bb"] = "aaxbb"; types["ab/cd"] = "abxcd"; types["ab/aa"] = "abxaa"; types["aa/ab"] = "aaxab"; types["ab/cc"] = "abxcc"; types["cc/ab"] = "ccxab"; dictionary["aaxbb"]["aa"] = "a"; dictionary["aaxbb"]["ab"] = "h"; dictionary["aaxbb"]["bb"] = "b"; dictionary["aaxbb"]["--"] = "-"; dictionary["abxcd"]["aa"] = "a"; dictionary["abxcd"]["ab"] = "a"; dictionary["abxcd"]["bb"] = "a"; dictionary["abxcd"]["cc"] = "b"; dictionary["abxcd"]["cd"] = "b"; dictionary["abxcd"]["dd"] = "b"; dictionary["abxcd"]["ac"] = "h"; dictionary["abxcd"]["ad"] = "h"; dictionary["abxcd"]["bc"] = "h"; dictionary["abxcd"]["bd"] = "h"; dictionary["abxcd"]["--"] = "-"; dictionary["abxaa"]["aa"] = "-"; dictionary["abxaa"]["ab"] = "-"; dictionary["abxaa"]["bb"] = "a"; dictionary["abxaa"]["--"] = "-"; dictionary["aaxab"]["aa"] = "-"; dictionary["aaxab"]["ab"] = "-"; dictionary["aaxab"]["bb"] = "b"; dictionary["aaxab"]["--"] = "-"; dictionary["abxcc"]["aa"] = "a"; dictionary["abxcc"]["ab"] = "a"; dictionary["abxcc"]["bb"] = "a"; dictionary["abxcc"]["cc"] = "b"; dictionary["abxcc"]["ac"] = "-"; dictionary["abxcc"]["bc"] = "-"; dictionary["abxcc"]["--"] = "-"; dictionary["ccxab"]["aa"] = "b"; dictionary["ccxab"]["ab"] = "b"; dictionary["ccxab"]["bb"] = "b"; dictionary["ccxab"]["cc"] = "a"; dictionary["ccxab"]["ac"] = "-"; dictionary["ccxab"]["bc"] = "-"; dictionary["ccxab"]["--"] = "-"; } void load_dh_dictionary(map &types, map > &dictionary) { types["ab/--"] = "ab/--"; types["--/ab"] = "--/ab"; dictionary["ab/--"]["aa"] = "aa"; dictionary["ab/--"]["bb"] = "bb"; dictionary["ab/--"]["--"] = "--"; dictionary["--/ab"]["aa"] = "aa"; dictionary["--/ab"]["bb"] = "bb"; dictionary["--/ab"]["--"] = "--"; } void load_mm_dh_dictionary(map &types, map > &dictionary) { types["ab/--"] = "abx--"; types["--/ab"] = "--xab"; dictionary["abx--"]["aa"] = "a"; dictionary["abx--"]["bb"] = "b"; dictionary["abx--"]["--"] = "-"; dictionary["--xab"]["aa"] = "a"; dictionary["--xab"]["bb"] = "b"; dictionary["--xab"]["--"] = "-"; } void load_segregation_ratios(map_types type, map > &segregation_ratios) { switch(type) { case cp: segregation_ratios["ab/--"]["aa"] = 0.50; segregation_ratios["ab/--"]["ab"] = 0.50; segregation_ratios["--/ab"]["aa"] = 0.50; segregation_ratios["--/ab"]["ab"] = 0.50; segregation_ratios["ab/aa"]["aa"] = 0.50; segregation_ratios["ab/aa"]["ab"] = 0.50; segregation_ratios["aa/ab"]["aa"] = 0.50; segregation_ratios["aa/ab"]["ab"] = 0.50; segregation_ratios["ab/ab"]["ab"] = 0.50; segregation_ratios["ab/ab"]["aa"] = 0.25; segregation_ratios["ab/ab"]["bb"] = 0.25; segregation_ratios["ab/ac"]["ab"] = 0.25; segregation_ratios["ab/ac"]["ac"] = 0.25; segregation_ratios["ab/ac"]["bc"] = 0.25; segregation_ratios["ab/ac"]["aa"] = 0.25; segregation_ratios["ab/cd"]["ac"] = 0.25; segregation_ratios["ab/cd"]["ad"] = 0.25; segregation_ratios["ab/cd"]["bc"] = 0.25; segregation_ratios["ab/cd"]["bd"] = 0.25; break; case f2: segregation_ratios["aaxbb"]["a"] = 0.25; segregation_ratios["aaxbb"]["b"] = 0.25; segregation_ratios["aaxbb"]["h"] = 0.50; segregation_ratios["abxcd"]["a"] = 0.25; segregation_ratios["abxcd"]["b"] = 0.25; segregation_ratios["abxcd"]["h"] = 0.50; segregation_ratios["abxaa"]["a"] = 1.00; segregation_ratios["aaxab"]["b"] = 1.00; segregation_ratios["abxcc"]["a"] = 0.50; segregation_ratios["abxcc"]["b"] = 0.50; segregation_ratios["ccxab"]["b"] = 0.50; segregation_ratios["ccxab"]["a"] = 0.50; break; case bc1: segregation_ratios["aaxbb"]["h"] = 0.50; segregation_ratios["aaxbb"]["b"] = 0.50; segregation_ratios["bbxaa"]["h"] = 0.50; segregation_ratios["bbxaa"]["a"] = 0.50; segregation_ratios["abxcc"]["h"] = 0.50; segregation_ratios["abxcc"]["b"] = 0.50; segregation_ratios["ccxab"]["h"] = 0.50; segregation_ratios["ccxab"]["a"] = 0.50; break; case dh: segregation_ratios["ab/--"]["a"] = 0.50; segregation_ratios["ab/--"]["b"] = 0.50; segregation_ratios["--/ab"]["a"] = 0.50; segregation_ratios["--/ab"]["b"] = 0.50; break; case gen: case none: case unk: break; } return; } stacks-1.44/src/genotype_dictionaries.h000644 000765 000024 00000004513 12750704270 021040 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2011-2014, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __GENOTYPE_DICTIONARIES_H__ #define __GENOTYPE_DICTIONARIES_H__ #include #include using std::string; using std::map; enum map_types {unk, none, gen, dh, cp, bc1, f2}; enum out_types {rqtl, joinmap, onemap, genomic}; void initialize_dictionaries(map > &global_dictionary); void load_cp_dictionary(map &types, map > &dictionary); void load_joinmap_cp_dictionary(map &types, map > &dictionary); void load_onemap_cp_dictionary(map &types, map > &dictionary); void load_bc_dictionary(map &types, map > &dictionary); void load_f2_dictionary(map &types, map > &dictionary); void load_mm_bc_dictionary(map &types, map > &dictionary); void load_mm_f2_dictionary(map &types, map > &dictionary); void load_dh_dictionary(map &types, map > &dictionary); void load_mm_dh_dictionary(map &types, map > &dictionary); void load_segregation_ratios(map_types type, map > &segregation_ratios); inline int encode_gtype(char a) { switch (a) { case 'A': return 0; case 'C': return 1; case 'G': return 2; case 'T': return 3; } return -1; } #endif // __GENOTYPE_DICTIONARIES_H__ stacks-1.44/src/genotypes.cc000644 000765 000024 00000256444 12774470747 016657 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2011-2014, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // genotypes -- genotype a mapping cross. // // Julian Catchen // jcatchen@uoregon.edu // University of Oregon // #include "MetaPopInfo.h" #include "genotypes.h" extern int encoded_gtypes[4][4]; // Global variables to hold command-line options. int num_threads = 1; int batch_id = -1; map_types map_type = gen; out_types out_type = joinmap; string in_path; string out_path; string cor_path; string out_file; string bl_file; string wl_file; string enz; int progeny_limit = 1; bool man_corrections = false; bool corrections = false; bool expand_id = false; bool sql_out = false; bool filter_lnl = false; double lnl_limit = 0.0; double chisq_pval_limit = 0.05; int min_stack_depth = 0; int min_hom_seqs = 5; double min_het_seqs = 0.05; double max_het_seqs = 0.1; set whitelist, blacklist; // // Hold information about restriction enzymes // map renz; map renz_cnt; map renz_len; // // Dictionaries to hold legal genotypes for different map types. // map > global_dictionary; int main (int argc, char* argv[]) { initialize_renz(renz, renz_cnt, renz_len); parse_command_line(argc, argv); // // Set the number of OpenMP parallel threads to execute. // #ifdef _OPENMP omp_set_num_threads(num_threads); #endif initialize_dictionaries(global_dictionary); MetaPopInfo mpopi; mpopi.init_directory(in_path); if (mpopi.samples().empty()) { cerr << "Error: Failed to find sample files in directory '" << in_path << "'.\n"; return -1; } if (wl_file.length() > 0) { load_marker_list(wl_file, whitelist); cerr << "Loaded " << whitelist.size() << " whitelisted markers.\n"; } if (bl_file.length() > 0) { load_marker_list(bl_file, blacklist); cerr << "Loaded " << blacklist.size() << " blacklisted markers.\n"; } // // Load the catalog // stringstream catalog_file; map catalog; bool compressed = false; int res; catalog_file << in_path << "batch_" << batch_id << ".catalog"; if ((res = load_loci(catalog_file.str(), catalog, false, false, compressed)) == 0) { cerr << "Unable to load the catalog '" << catalog_file.str() << "'\n"; return 0; } // // Implement the black/white list // reduce_catalog(catalog, whitelist, blacklist); // // Load matches to the catalog // vector > catalog_matches; vector samples_to_remove; set seen_samples; for (size_t i=0; i()); vector& m = catalog_matches.back(); const MetaPopInfo::Sample& sample = mpopi.samples()[i]; load_catalog_matches(in_path + sample.name, m); if (m.size() == 0) { cerr << "Warning: Absent or malformed matches file '" << sample.name << ".matches.tsv(.gz)" <<"', excluding this sample from population analysis.\n"; samples_to_remove.push_back(i); catalog_matches.pop_back(); // n.b. Index shift will be resolved by the call to MetaPopInfo::delete_samples(). continue; } size_t sample_id = m[0]->sample_id; if (seen_samples.count(sample_id) > 0) { cerr << "Error: sample ID " << sample_id << " occurs twice in this data set, likely the pipeline was run incorrectly.\n"; return -1; } seen_samples.insert(sample_id); mpopi.set_sample_id(i, sample_id); } mpopi.delete_samples(samples_to_remove); if (mpopi.samples().size() == 0) { cerr << "Error: Couln't find any matches files.\n"; return -1; } // [mpopi] is definitive. cerr << "Working on " << mpopi.samples().size() << " samples.\n"; // Set the deprecated [samples] variable to the value it should have map samples; // map of {sample_id, sample_name} mpopi.fill_samples(samples); // Retrieve the IDs of the parents vector sample_ids; for (auto& sample : mpopi.samples()) sample_ids.push_back(sample.id); sort(sample_ids.begin(), sample_ids.end()); set parent_ids; identify_parental_ids(catalog, sample_ids, parent_ids); // // Create the population map // cerr << "Populating observed haplotypes for " << mpopi.samples().size() << " samples, " << catalog.size() << " loci.\n"; PopMap *pmap = new PopMap(mpopi, catalog.size()); pmap->populate(catalog, catalog_matches); apply_locus_constraints(catalog, pmap); // // Identify mappable markers in the parents // find_markers(catalog, pmap, parent_ids); // // Calculate F, inbreeding coefficient // calculate_f(catalog, pmap, parent_ids); // // Create genotypes maps, calculate mean log likelihood for each locus. // map::iterator it; Datum **d; CSLocus *loc; double mean, cnt; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; d = pmap->locus(loc->id); if (loc->marker.length() > 0) { create_genotype_map(loc, pmap, parent_ids); call_population_genotypes(loc, pmap, global_dictionary); } mean = 0.0; cnt = 0.0; for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] == NULL) continue; mean += d[i]->lnl; cnt++; } loc->lnl = mean / cnt; } // // Make automated corrections // if (corrections) automated_corrections(samples, parent_ids, catalog, catalog_matches, pmap); // // Check markers for potentially missing alleles. // switch(map_type) { case cp: correct_cp_markers_missing_alleles(parent_ids, catalog, pmap); break; case dh: case bc1: case f2: case gen: case none: case unk: break; } // // Reassign genotypes according to specific map type, record any // marker corrections made by detecting missing alleles. // if (map_type != gen) map_specific_genotypes(catalog, pmap, parent_ids); // // Incorporate manual corrections exported from a Stacks SQL database. // if (man_corrections) manual_corrections(cor_path, pmap); // // Calculate segregation distortion using chi-squared test. // map > seg_ratios; if (map_type != gen) { load_segregation_ratios(map_type, seg_ratios); calc_segregation_distortion(seg_ratios, catalog, pmap, parent_ids); } // // If a mapping type was specified, output it. // switch(map_type) { case dh: export_dh_map(catalog, pmap, parent_ids, samples); break; case cp: export_cp_map(catalog, pmap, parent_ids, samples); break; case bc1: export_bc1_map(catalog, pmap, parent_ids, samples); break; case f2: export_f2_map(catalog, pmap, parent_ids, samples); break; case gen: export_gen_map(catalog, pmap, parent_ids, samples); break; case none: case unk: break; } if (sql_out) write_sql(catalog, pmap, parent_ids); // // Output the observed haplotypes. // write_generic(catalog, pmap, samples, parent_ids, false); if (out_type == genomic) write_genomic(catalog, pmap); return 0; } int apply_locus_constraints(map &catalog, PopMap *pmap) { CSLocus *loc; Datum **d; uint below_stack_dep = 0; uint below_lnl_thresh = 0; if (min_stack_depth == 0) return 0; map::iterator it; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; d = pmap->locus(loc->id); for (int i = 0; i < pmap->sample_cnt(); i++) { // // Check that each sample is over the minimum stack depth for this locus. // if (d[i] != NULL && d[i]->tot_depth < min_stack_depth) { below_stack_dep++; delete d[i]; d[i] = NULL; } // // Check that each sample is over the log likelihood threshold. // if (d[i] != NULL && filter_lnl && d[i]->lnl < lnl_limit) { below_lnl_thresh++; delete d[i]; d[i] = NULL; } } } if (min_stack_depth > 0) cerr << "Removed " << below_stack_dep << " samples from loci that are below the minimum stack depth of " << min_stack_depth << "x\n"; if (filter_lnl) cerr << "Removed " << below_lnl_thresh << " samples from loci that are below the log likelihood threshold of " << lnl_limit << "\n"; return 0; } int identify_parental_ids(map &catalog, vector &sample_ids, set &parents) { set catalog_parents; map::iterator it; CSLocus *loc; int sample_id; // // We assume the catalog was constructed from one or more parents of one // or more crosses. These are listed in the catalog.tags.tsv file, column 8. // for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; for (uint i = 0; i < loc->comp.size(); i++) { sample_id = (int) strtol(loc->comp[i], NULL, 10); catalog_parents.insert(sample_id); } } // // Now we want to iterate over those individuals genotypes found when // searching the Stacks directory and crosscheck those found in the catalog. // for (uint i = 0; i < sample_ids.size(); i++) { if (catalog_parents.count(sample_ids[i]) > 0) parents.insert(sample_ids[i]); } set::iterator sit; cerr << "Identified parent IDs: "; for (sit = parents.begin(); sit != parents.end(); sit++) cerr << *sit << " "; cerr << "\n"; return 0; } int find_markers(map &catalog, PopMap *pmap, set &parent_ids) { map::iterator it; vector::iterator hit; set::iterator p, q; int pid_1, pid_2, parent_count, allele_cnt_1, allele_cnt_2; Datum *d_1, *d_2; CSLocus *loc; if (parent_ids.size() > 2) return 0; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; // // Count the number of parental tags matching this catalog tag. A proper marker should // contain a single representative from each parent; multiple alleles must be called from // a single tag from a single parent. // if (parent_ids.size() == 1) { p = parent_ids.begin(); pid_1 = *p; pid_2 = -1; d_1 = pmap->blacklisted(loc->id, pid_1) ? NULL : pmap->datum(loc->id, pid_1); d_2 = NULL; } else { p = parent_ids.begin(); q = p++; pid_1 = *p < *q ? *p : *q; pid_2 = *p < *q ? *q : *p; if (pmap->blacklisted(loc->id, pid_1) || pmap->blacklisted(loc->id, pid_2)) { d_1 = NULL; d_2 = NULL; } else { d_1 = pmap->datum(loc->id, pid_1); d_2 = pmap->datum(loc->id, pid_2); } } parent_count = 0; if (d_1 != NULL) parent_count++; if (d_2 != NULL) parent_count++; // // Locus is present in both parents. // if (parent_count == 2) { allele_cnt_1 = d_1->obshap.size(); allele_cnt_2 = d_2->obshap.size(); // // Determine the number of unique alleles // set unique_alleles; for (hit = d_1->obshap.begin(); hit != d_1->obshap.end(); hit++) unique_alleles.insert(*hit); for (hit = d_2->obshap.begin(); hit != d_2->obshap.end(); hit++) unique_alleles.insert(*hit); int num_unique_alleles = unique_alleles.size(); // // Locus is heterozygous in both parents. However, the number of alleles present distinguishes // what type of marker it is. Four unique alleles requries an ab/cd marker, while four // alleles that are the same in both parents requires an ab/ab marker. Finally, three unique // alleles requires either an ab/ac marker. // if (allele_cnt_1 == 2 && allele_cnt_2 == 2) { if (num_unique_alleles == 3) loc->marker = "ab/ac"; else if (num_unique_alleles == 2) loc->marker = "ab/ab"; else loc->marker = "ab/cd"; // // Locus is homozygous in one parent and heterozygous in the other. // } else if (allele_cnt_1 == 2 && allele_cnt_2 == 1) { if (num_unique_alleles == 3) loc->marker = "ab/cc"; else if (num_unique_alleles == 2) loc->marker = "ab/aa"; // // Locus is homozygous in one parent and heterozygous in the other. // } else if (allele_cnt_1 == 1 && allele_cnt_2 == 2) { if (num_unique_alleles == 3) loc->marker = "cc/ab"; else if (num_unique_alleles == 2) loc->marker = "aa/ab"; // // Locus is homozygous in both parents, but heterozygous between parents. // } else if (allele_cnt_1 == 1 && allele_cnt_2 == 1) { if (strcmp(d_1->obshap[0], d_2->obshap[0]) != 0) loc->marker = "aa/bb"; } // // Locus only exists in one parent. // } else if (parent_count == 1) { if (d_1 != NULL && d_1->obshap.size() == 2) loc->marker = "ab/--"; else if (d_2 != NULL && d_2->obshap.size() == 2) loc->marker = "--/ab"; } } return 0; } int calculate_f(map &catalog, PopMap *pmap, set &parent_ids) { map::iterator it; map::iterator j; Datum **d; CSLocus *loc; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; d = pmap->locus(loc->id); if (loc->snps.size() == 0) continue; double tot = 0.0; double hets = 0; double p, q, h, h0; map alle; for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] == NULL) continue; if (parent_ids.count(pmap->rev_sample_index(i))) continue; tot++; if (d[i]->obshap.size() > 1) hets++; // // We are measuring the first SNP in the haplotype // for (uint j = 0; j < d[i]->obshap.size(); j++) alle[d[i]->obshap[j][0]]++; } if (alle.size() > 2 || tot == 0) continue; j = alle.begin(); p = j->second; j++; q = j->second; h = hets / tot; // Observed frequency of heterozygotes in the population h0 = 2 * (p/tot) * (q/tot); // 2PQ, expected frequency of hets under Hardy-Weinberg if (h0 > 0) loc->f = (h0 - h) / h0; //cerr << "P: " << p << "; Q: " << q << "; Hets: " << hets << "; total: " << tot << "; f: " << loc->f << "\n"; } return 0; } int create_genotype_map(CSLocus *locus, PopMap *pmap, set &parent_ids) { // // Create a genotype map. For any set of alleles, this routine will // assign each allele to one of the constituent genotypes, e.g. given the // marker type 'aaxbb' and the alleles 'A' from the male, and 'G' // from the female, will assign 'G' == 'bb' and 'A'== 'aa'. It assumes that // recombination may have occurred as with an F2, F3 or later cross. // //cerr << "Creating genotype map for catalog ID " << locus->id << ", marker: " << locus->marker << ".\n"; // // Get the parent IDs ordered // set::iterator p = parent_ids.begin(); set::iterator q = p++; int pid_1 = *p < *q ? *p : *q; int pid_2 = *p < *q ? *q : *p; set p1_gtypes, p2_gtypes; set::iterator i; map legal_gtypes, com_gtypes; // // First, identify any alleles that are common between the two parents. // p1_gtypes.insert(locus->marker[0]); p1_gtypes.insert(locus->marker[1]); p2_gtypes.insert(locus->marker[3]); p2_gtypes.insert(locus->marker[4]); for (i = p1_gtypes.begin(); i != p1_gtypes.end(); i++) if (*i != '-') legal_gtypes[*i]++; for (i = p2_gtypes.begin(); i != p2_gtypes.end(); i++) if (*i != '-') legal_gtypes[*i]++; // // Find the common genotypes // vector types; map::iterator j; for (j = legal_gtypes.begin(); j != legal_gtypes.end(); j++) if (j->second > 1) types.push_back(j->first); sort(types.begin(), types.end()); Datum *d_1, *d_2; map haplotypes; map::iterator k; vector > sorted_haplotypes; d_1 = pmap->datum(locus->id, pid_1); d_2 = pmap->datum(locus->id, pid_2); if (d_1 != NULL) { for (uint n = 0; n < d_1->obshap.size(); n++) haplotypes[d_1->obshap[n]]++; } if (d_2 != NULL) { for (uint n = 0; n < d_2->obshap.size(); n++) haplotypes[d_2->obshap[n]]++; } // // Sort the haplotypes map by value // for (k = haplotypes.begin(); k != haplotypes.end(); k++) sorted_haplotypes.push_back(*k); sort(sorted_haplotypes.begin(), sorted_haplotypes.end(), hap_compare); for (uint n = 0, index = 0; n < sorted_haplotypes.size() && index < types.size(); n++, index++) { if (sorted_haplotypes[n].second > 1) { locus->gmap[sorted_haplotypes[n].first] = types[index]; com_gtypes[types[index]]++; // cerr << " Assigning common allele " << sorted_haplotypes[n].first << " to genotype '" << locus->gmap[sorted_haplotypes[n].first] << "'\n"; } } // // Now, examine the remaining first parent alleles. // if (d_1 != NULL) { legal_gtypes.clear(); for (i = p1_gtypes.begin(); i != p1_gtypes.end(); i++) if (*i != '-' && com_gtypes.count(*i) == 0) { // cerr << " Adding " << *i << " to first parent genotypes\n"; legal_gtypes[*i]++; } types.clear(); for (j = legal_gtypes.begin(); j != legal_gtypes.end(); j++) types.push_back(j->first); sort(types.begin(), types.end()); for (uint n = 0, index = 0; n < d_1->obshap.size() && index < types.size(); n++, index++) { if (locus->gmap.count(d_1->obshap[n])) { index--; continue; } locus->gmap[d_1->obshap[n]] = types[index]; // cerr << " Assinging '" << d_1->obshap[n] << "' to first parent genotype '" << locus->gmap[d_1->obshap[n]] << "'\n"; } } // // Finally, repeat in the second parent. // if (d_2 != NULL) { legal_gtypes.clear(); for (i = p2_gtypes.begin(); i != p2_gtypes.end(); i++) if (*i != '-' && com_gtypes.count(*i) == 0) { // cerr << " Adding " << *i << " to second genotypes\n"; legal_gtypes[*i]++; } types.clear(); for (j = legal_gtypes.begin(); j != legal_gtypes.end(); j++) types.push_back(j->first); sort(types.begin(), types.end()); for (uint n = 0, index = 0; n < d_2->obshap.size() && index < types.size(); n++, index++) { if (locus->gmap.count(d_2->obshap[n])) { index--; continue; } locus->gmap[d_2->obshap[n]] = types[index]; // cerr << " Assinging '" << d_2->obshap[n] << "' to second parent genotype '" << locus->gmap[d_2->obshap[n]] << "'\n"; } } return 0; } int call_population_genotypes(CSLocus *locus, PopMap *pmap, map > &dictionary) { // // Fetch the array of observed haplotypes from the population // Datum **d = pmap->locus(locus->id); for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] == NULL) continue; vector gtypes; string gtype; //cerr << "Sample Id: " << pmap->rev_sample_index(i) << "\n"; for (uint j = 0; j < d[i]->obshap.size(); j++) { // // Impossible allele encountered. // if (locus->gmap.count(d[i]->obshap[j]) == 0) { gtypes.clear(); gtypes.push_back("-"); goto impossible; } gtypes.push_back(locus->gmap[d[i]->obshap[j]]); //cerr << " Observed Haplotype: " << d[i]->obshap[j] << ", Genotype: " << locus->gmap[d[i]->obshap[j]] << "\n"; } impossible: sort(gtypes.begin(), gtypes.end()); for (uint j = 0; j < gtypes.size(); j++) { gtype += gtypes[j]; //cerr << " Adding genotype to string: " << gtypes[j] << "; " << gtype << "\n"; } string m = dictionary[locus->marker].count(gtype) ? dictionary[locus->marker][gtype] : "--"; if (d[i]->gtype != NULL) delete d[i]->gtype; d[i]->gtype = new char[m.length() + 1]; strcpy(d[i]->gtype, m.c_str()); if (m != "--") locus->gcnt++; //cerr << "Assigning datum, marker: " << locus->marker << ", string: " << m << ", haplotype: " << d[i]->obshap[0] << ", gtype: " << gtype << "\n"; } return 0; } int correct_cp_markers_missing_alleles(set &parent_ids, map &catalog, PopMap *pmap) { map::iterator it; CSLocus *loc; Datum **d; map > seg_ratios; // // The following segregation ratios will occur when one of the parents // in the cross is missing an allele. We will not see these ratios // in one of these markers with no missing alleles. // seg_ratios["ab/aa"]["aa"] = 0.50; seg_ratios["ab/aa"]["ab"] = 0.25; seg_ratios["ab/aa"]["bb"] = 0.25; seg_ratios["aa/ab"]["aa"] = 0.50; seg_ratios["aa/ab"]["ab"] = 0.25; seg_ratios["aa/ab"]["bb"] = 0.25; seg_ratios["ab/cc"]["ac"] = 0.25; seg_ratios["ab/cc"]["bc"] = 0.25; seg_ratios["ab/cc"]["aa"] = 0.25; seg_ratios["ab/cc"]["bb"] = 0.25; seg_ratios["cc/ab"]["ac"] = 0.25; seg_ratios["cc/ab"]["bc"] = 0.25; seg_ratios["cc/ab"]["aa"] = 0.25; seg_ratios["cc/ab"]["bb"] = 0.25; cerr << "Testing catalog loci for mapping parents with missing alleles..."; int corrected = 0; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; // // We only want to examine markers where one parent is homozygous. // if (loc->marker != "ab/aa" && loc->marker != "aa/ab" && loc->marker != "ab/cc" && loc->marker != "cc/ab") continue; map cnts; // // Calculate initial segregation distortion. // double n = tally_generic_gtypes(loc->id, pmap, parent_ids, cnts); double chisq_pval = chisq_test(seg_ratios, cnts, loc->marker, n); // // Check if our genotype ratios match the segregation ratios specified above. If so, // we have a dropped allele in one of the parents. // if (n == 0 || chisq_pval < chisq_pval_limit) continue; corrected++; if (loc->marker == "ab/aa") loc->marker = "ab/a-"; else if (loc->marker == "aa/ab") loc->marker = "-a/ab"; else if (loc->marker == "ab/cc") loc->marker = "ab/c-"; else if (loc->marker == "cc/ab") loc->marker = "-c/ab"; d = pmap->locus(loc->id); if (loc->marker == "ab/a-" || loc->marker == "-a/ab") { for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] == NULL) continue; if (parent_ids.count(pmap->rev_sample_index(i))) continue; if (strcmp(d[i]->gtype, "bb") == 0) strcpy(d[i]->gtype, "ab"); } } else if (loc->marker == "ab/c-") { for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] == NULL) continue; if (parent_ids.count(pmap->rev_sample_index(i))) continue; if (strcmp(d[i]->gtype, "bb") == 0) strcpy(d[i]->gtype, "bd"); else if (strcmp(d[i]->gtype, "aa") == 0) strcpy(d[i]->gtype, "ad"); } } else if (loc->marker == "-c/ab") { for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] == NULL) continue; if (parent_ids.count(pmap->rev_sample_index(i))) continue; if (strcmp(d[i]->gtype, "bb") == 0) strcpy(d[i]->gtype, "ad"); else if (strcmp(d[i]->gtype, "aa") == 0) strcpy(d[i]->gtype, "ac"); else if (strcmp(d[i]->gtype, "bc") == 0) strcpy(d[i]->gtype, "bd"); else if (strcmp(d[i]->gtype, "ac") == 0) strcpy(d[i]->gtype, "bc"); } } } // // Now we will deal with aa/bb markers separately, since there can be three possible // missing allele situations: // aa: 50%, ab: 50% - we have an aa/b- marker, which should be mapped as an --/ab // bb: 50%, ab: 50% - we have an -a/bb marker, which should be mapped as an ab/-- // aa: 33%, ab: 33%, bb: 33% - we have an -a/b- maker, which should be mapped as an ab/ab, but // we can't disambiguate the aa bb genotypes so it can't be mapped. // map > seg_ratio_1, seg_ratio_2; seg_ratio_1["aa/bb"]["aa"] = 0.50; seg_ratio_1["aa/bb"]["ab"] = 0.50; seg_ratio_2["aa/bb"]["bb"] = 0.50; seg_ratio_2["aa/bb"]["ab"] = 0.50; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (loc->marker != "aa/bb") continue; map cnts; double n = tally_generic_gtypes(loc->id, pmap, parent_ids, cnts); double chisq_pval = chisq_test(seg_ratio_1, cnts, loc->marker, n); if (n == 0) continue; if (chisq_pval >= chisq_pval_limit) { corrected++; loc->marker = "aa/b-"; d = pmap->locus(loc->id); for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] == NULL) continue; if (parent_ids.count(pmap->rev_sample_index(i))) continue; if (strcmp(d[i]->gtype, "ab") == 0) strcpy(d[i]->gtype, "bb"); } } else { chisq_pval = chisq_test(seg_ratio_2, cnts, loc->marker, n); if (chisq_pval >= chisq_pval_limit) { corrected++; loc->marker = "-a/bb"; d = pmap->locus(loc->id); for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] == NULL) continue; if (parent_ids.count(pmap->rev_sample_index(i))) continue; if (strcmp(d[i]->gtype, "ab") == 0) strcpy(d[i]->gtype, "aa"); } } } } cerr << "corrected " << corrected << " catalog loci.\n"; return 0; } int automated_corrections(map &samples, set &parent_ids, map &catalog, vector > &matches, PopMap *pmap) { int sample_id, catalog_id, tag_id; Datum *d; Locus *s; string file; cerr << "Performing automated corrections...\n"; for (uint i = 0; i < matches.size(); i++) { sample_id = matches[i][0]->sample_id; file = samples[sample_id]; //if (sample_id != 29) continue; if (parent_ids.count(sample_id)) continue; map stacks; bool compressed = false; int res; if ((res = load_loci(in_path + file, stacks, true, false, compressed)) == 0) { cerr << "Unable to load sample file '" << file << "'\n"; return 0; } set > processed; for (uint j = 0; j < matches[i].size(); j++) { catalog_id = matches[i][j]->cat_id; sample_id = matches[i][j]->sample_id; tag_id = matches[i][j]->tag_id; if (catalog.count(catalog_id) == 0) continue; // // There are multiple matches per stack, but we only need to process // each stack once to make corrections. // if (processed.count(make_pair(catalog_id, tag_id)) == 0 && catalog[catalog_id]->marker.length() > 0) { d = pmap->datum(catalog_id, sample_id); //cerr << "Accessing catalog ID " << catalog_id << "; sample: " << sample_id << "; marker: " << catalog[catalog_id]->marker << ": d: " << d << "; gtype: " << d->gtype << "\n"; if (d != NULL && strcmp(d->gtype, "--") != 0) { s = stacks[tag_id]; check_uncalled_snps(catalog[catalog_id], s, d); } processed.insert(make_pair(catalog_id, tag_id)); } } // // Free up memory // map::iterator it; for (it = stacks.begin(); it != stacks.end(); it++) delete it->second; } // // Summarize correction results // long pot_gen = 0; long tot_gen = 0; long tot_cor = 0; long het_cor = 0; long rem_cor = 0; int markers = 0; map::iterator it; for (it = catalog.begin(); it != catalog.end(); it++) { if (it->second->marker.length() == 0) continue; markers++; for (int j = 0; j < pmap->sample_cnt(); j++) { sample_id = pmap->rev_sample_index(j); if (parent_ids.count(sample_id)) continue; d = pmap->datum(it->first, sample_id); pot_gen++; if (d == NULL) continue; tot_gen++; if (d->corrected == true) { tot_cor++; if (strcmp(d->gtype, "--") == 0) rem_cor++; else het_cor++; } } } cerr << pot_gen << " potential genotypes in " << markers << " markers, " << tot_gen << " populated; " << tot_cor << " corrected, " << het_cor << " converted to heterozygotes, " << rem_cor << " unsupported homozygotes removed.\n"; return 0; } int check_uncalled_snps(CSLocus *clocus, Locus *stack, Datum *d) { // // If this locus is already known to be multi-allele, return, we only want // to verify uncalled SNPs. // if (strlen(d->gtype) > 1 && d->gtype[0] != d->gtype[1]) return 0; //cerr << "Catalog locus: " << clocus->id << ", marker: " << clocus->marker << ", tag_id: " << stack->id << "; Starting gtype: " << d->gtype << "\n"; vector verified_snps; string status = "false"; string homozygous; for (uint i = 0; i < clocus->snps.size(); i++) { check_homozygosity(stack->reads, clocus->snps[i]->col, clocus->snps[i]->rank_1, clocus->snps[i]->rank_2, homozygous); if (homozygous == "unknown") status = "true"; else if (homozygous == "false") verified_snps.push_back(clocus->snps[i]); } if (status == "true") { d->corrected = true; delete [] d->gtype; d->gtype = new char[2]; strcpy(d->gtype, "-"); return 0; } else if (verified_snps.size() < clocus->snps.size()) { return 0; } // // Detect the alleles present from the verified SNPs // vector haplotypes; call_alleles(verified_snps, stack->reads, haplotypes); vector types; for (uint i = 0; i < haplotypes.size(); i++) { if (clocus->gmap.count(haplotypes[i])) { //cerr << " Adding haplotype '" << haplotypes[i] << "', which maps to '" << clocus->gmap[haplotypes[i]] << "' to the genotype\n"; types.push_back(clocus->gmap[haplotypes[i]]); } else { //cerr << " Adding haplotype '-' for " << haplotypes[i] << "\n"; types.push_back("-"); } } sort(types.begin(), types.end()); string genotype; for (uint i = 0; i < types.size(); i++) genotype += types[i]; //cerr << "Final genotype: " << genotype << "\n"; genotype = global_dictionary[clocus->marker].count(genotype) ? global_dictionary[clocus->marker][genotype] : "--"; //cerr << "Final translated genotype: " << genotype << "\n"; if (strcmp(genotype.c_str(), d->gtype) != 0) { d->corrected = true; delete [] d->gtype; d->gtype = new char[genotype.length() + 1]; strcpy(d->gtype, genotype.c_str()); } //cerr << " Catalog: " << clocus->id << ", stack: " << stack->id << ", Ending Genotype: " << d->gtype << "\n\n"; return 0; } int call_alleles(vector &snps, vector &reads, vector &haplotypes) { map a; int height = reads.size(); char base; for (int i = 0; i < height; i++) { string haplotype; for (uint j = 0; j < snps.size(); j++) { base = reads[i][snps[j]->col]; // // Check to make sure the nucleotide at the location of this SNP is // of one of the two possible states the multinomial model called. // if (base == snps[j]->rank_1 || base == snps[j]->rank_2) haplotype += base; else break; } if (haplotype.length() == snps.size()) a[haplotype]++; } map::iterator it; for (it = a.begin(); it != a.end(); it++) { //cerr << " Calling haplotype: " << it->first << "\n"; haplotypes.push_back(it->first); } return 0; } int check_homozygosity(vector &reads, int col, char rank_1, char rank_2, string &homozygous) { //cerr << " Examining col " << col << ", rank 1: " << rank_1 << "; rank 2: " << rank_2 << "\n"; int height = reads.size(); homozygous = "true"; if (height < min_hom_seqs) { homozygous = "unknown"; return 0; } map nuc; vector > sorted_nuc; nuc['A'] = 0; nuc['C'] = 0; nuc['G'] = 0; nuc['T'] = 0; for (int j = 0; j < height; j++) nuc[reads[j][col]]++; map::iterator i; for (i = nuc.begin(); i != nuc.end(); i++) sorted_nuc.push_back(make_pair(i->first, i->second)); sort(sorted_nuc.begin(), sorted_nuc.end(), compare_pair); // // Check if more than a single nucleotide occurs in this column. Only // count nucleotides that are part of the called SNP, do not count // error-generated nucleotides. Also, check that the sorting was successful // by ensuring that sorted_nuc[0] > sorted_nuc[1] > sorted_nuc[2]. // if (sorted_nuc[2].second > 0 && sorted_nuc[1].second <= sorted_nuc[2].second) { homozygous = "unknown"; return 0; } // cerr << "Sorted_nuc[0], '" << sorted_nuc[0].first << "', count: " << sorted_nuc[0].second // << "; Sorted_nuc[1], '" << sorted_nuc[1].first << "', count: " << sorted_nuc[1].second // << "; Sorted_nuc[2], '" << sorted_nuc[2].first << "', count: " << sorted_nuc[2].second << "\n"; if ((sorted_nuc[0].second > 0) && (sorted_nuc[0].first == rank_1 || sorted_nuc[0].first == rank_2) && (sorted_nuc[1].second > 0) && (sorted_nuc[1].first == rank_1 || sorted_nuc[1].first == rank_2)) { homozygous = "false"; } // // If we find a second nucleotide present, check its prevelence. If it is // less than 1/20 of the total reads, don't count a heterozygote. If it is // less than 1/10 report that we can't tell if its homozygous or not. Otherwise, // report this tag as a heterozygote. // double frac = (double) sorted_nuc[1].second / (double) height; //cerr << " Frac: " << frac << "; Second-most Prominent Nuc count: " << sorted_nuc[1].second << "; Depth: " << height << "\n"; if (homozygous == "false" && frac < min_het_seqs) homozygous = "true"; else if (homozygous == "false" && frac < max_het_seqs) homozygous = "unknown"; //cerr << " Homozygous: " << homozygous << "\n"; return 0; } int manual_corrections(string cor_path, PopMap *pmap) { // // Load manual corrections from a tab-seprated file, as exported from a Stacks SQL // dataabse. Has the format: // idbatch_idcatalog_idsample_idgenotype // char line[max_len]; ifstream fh(cor_path.c_str(), ifstream::in); if (fh.fail()) { cerr << "Error opening manual corrections file '" << cor_path << "'\n"; exit(1); } vector parts; int catalog_id, sample_id, len; char gtype[id_len]; char *e; int line_num = 0; int total = 0; int skipped = 0; int success = 0; int i; while (fh.good()) { fh.getline(line, id_len); line_num++; len = strlen(line); if (len == 0) continue; // // Check that there is no carraige return in the buffer. // if (line[len - 1] == '\r') line[len - 1] = '\0'; // // Ignore comments // if (line[0] == '#') continue; parse_tsv(line, parts); if (parts.size() != 5) { cerr << "Error parsing '" << line << "' at line: " << line_num << ". (" << parts.size() << " fields).\n"; return 0; } catalog_id = (int) strtol(parts[2].c_str(), &e, 10); if (*e != '\0') { cerr << "Error parsing '" << parts[2].c_str() << "' at line: " << line_num << ".\n"; return 0; } sample_id = (int) strtol(parts[3].c_str(), &e, 10); if (*e != '\0') { cerr << "Error parsing '" << parts[3].c_str() << "' at line: " << line_num << ".\n"; return 0; } strcpy(gtype, parts[4].c_str()); // // Overwrite the genotype in the PopMap. // Datum **d = pmap->locus(catalog_id); total++; if (d == NULL) { skipped++; continue; } if ((i = pmap->sample_index(sample_id)) < 0) { skipped++; continue; } if (d[i] == NULL) { skipped++; continue; } for (uint k = 0; k < strlen(gtype); k++) gtype[k] = tolower(gtype[k]); if (strcmp(gtype, "--") == 0) strcpy(gtype, "-"); if (d[i]->gtype != NULL) delete [] d[i]->gtype; d[i]->gtype = new char[strlen(gtype) + 1]; strcpy(d[i]->gtype, gtype); d[i]->corrected = true; success++; } fh.close(); cerr << "Successfully imported " << success << " manually corrected genotypes. Skipped " << skipped << " genotypes due to invalid catalog or sample IDs, " << total << " genotypes read from file.\n"; return 0; } int export_gen_map(map &catalog, PopMap *pmap, set &parent_ids, map &samples) { // // We wish to export, a set of generic genotypes, not specific to any mapping type. // // // Mark those genotypes that have been corrected in uppercase letters. // map::iterator it; CSLocus *loc; uint len; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; Datum **d = pmap->locus(loc->id); for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] == NULL) continue; if (parent_ids.count(pmap->rev_sample_index(i))) continue; if (d[i]->corrected) { len = strlen(d[i]->gtype); for (uint k = 0; k < len; k++) d[i]->gtype[k] = toupper(d[i]->gtype[k]); } } } // // Output the results // write_generic(catalog, pmap, samples, parent_ids, true); return 0; } int export_f2_map(map &catalog, PopMap *pmap, set &parent_ids, map &samples) { // // We wish to export, according to the JoinMap manual, a locus genotype file (loc-file), // which contains the information of all the loci for a single segregating population. // // We are exporting an F2 population type: // The result of selfing the F1 of a cross between two fully homozygous diploid parents. // // Genotype codes for an F2 population, depending on the locus segregation type. // // Seg. type Possible genotypes // --------- ------------------ // a, b, h, – // a, b, h, – // a, – // b, – // a, b, – // b, – // map types; map > dictionary; load_mm_f2_dictionary(types, dictionary); // // Translate the genotypes for this particular map type. // translate_genotypes(types, dictionary, catalog, pmap, samples, parent_ids); // // Output the results // switch(out_type) { case joinmap: write_joinmap(catalog, pmap, types, samples, parent_ids); break; case rqtl: write_rqtl(catalog, pmap, types, samples, parent_ids); break; case onemap: write_onemap_mapmaker(catalog, pmap, types, samples, parent_ids); break; default: break; } return 0; } int export_dh_map(map &catalog, PopMap *pmap, set &parent_ids, map &samples) { // // We wish to export, according to the JoinMap manual, a locus genotype file (loc-file), // which contains the information of all the loci for a single segregating population. // // We are exporting a DH population type: // a doubled haploid population: the result of doubling the gametes of a single heterozygous // diploid individual. // // Segregation type codes for population type DH, from Joinmap manual: // // Code Description // ------- ----------- // locus homozygous in both parents, heterozygous between the parents // // Genotype codes for a CP population, depending on the locus segregation type. // // Seg. type Possible genotypes // --------- ------------------ // a the one genotype // b the other genotype // map types; map > dictionary; load_mm_dh_dictionary(types, dictionary); // // Translate the genotypes for this particular map type. // translate_genotypes(types, dictionary, catalog, pmap, samples, parent_ids); // // Output the results // switch(out_type) { case joinmap: write_joinmap(catalog, pmap, types, samples, parent_ids); break; case rqtl: write_rqtl(catalog, pmap, types, samples, parent_ids); break; default: break; } return 0; } int export_bc1_map(map &catalog, PopMap *pmap, set &parent_ids, map &samples) { // // We wish to export, according to the JoinMap manual, a locus genotype file (loc-file), // which contains the information of all the loci for a single segregating population. // // We are exporting a BC1 population type: // a first generation backcross population: the result of crossing the F1 of a cross between // two fully homozygous diploid parents to one of the parents. // // Segregation type codes for population type BC1, from Joinmap manual: // // Code Description // ------- ----------- // locus homozygous in both parents, heterozygous between the parents // // Genotype codes for a BC1 population, depending on the locus segregation type. // // Seg. type Possible genotypes // --------- ------------------ // a homozygote or haploid as the first parent // b homozygote or haploid as the second parent // h heterozygote (as the F1) // map types; map > dictionary; load_mm_bc_dictionary(types, dictionary); // // Translate the genotypes for this particular map type. // translate_genotypes(types, dictionary, catalog, pmap, samples, parent_ids); // // Output the results // switch(out_type) { case joinmap: write_joinmap(catalog, pmap, types, samples, parent_ids); break; case rqtl: write_rqtl(catalog, pmap, types, samples, parent_ids); break; case onemap: write_onemap_mapmaker(catalog, pmap, types, samples, parent_ids); break; default: break; } return 0; } int export_cp_map(map &catalog, PopMap *pmap, set &parent_ids, map &samples) { // // We wish to export, according to the JoinMap manual, a locus genotype file (loc-file), // which contains the information of all the loci for a single segregating population. // // We are exporting a CP population type: // a population resulting from a cross between two heterogeneously // heterozygous and homozygous diploid parents, linkage phases originally // (possibly) unknown. // // Segregation type codes for population type CP, from Joinmap manual: // // Code Description // ------- ----------- // locus heterozygous in both parents, four alleles // locus heterozygous in both parents, three alleles // locus heterozygous in both parents, two alleles // locus heterozygous in the first parent // locus heterozygous in the second parent // // Genotype codes for a CP population, depending on the locus segregation type. // // Seg. type Possible genotypes // --------- ------------------ // ac, ad, bc, bd, –– // ee, ef, eg, fg, –– // hh, hk, kk, h-, k-, –– // ll, lm, –– // nn, np, –– // map types; map > dictionary; switch(out_type) { case joinmap: load_joinmap_cp_dictionary(types, dictionary); break; case onemap: load_onemap_cp_dictionary(types, dictionary); break; default: break; } // // Translate the genotypes for this particular map type. // translate_genotypes(types, dictionary, catalog, pmap, samples, parent_ids); // // Output the results // switch(out_type) { case joinmap: write_joinmap(catalog, pmap, types, samples, parent_ids); break; case onemap: write_onemap(catalog, pmap, types, samples, parent_ids); break; default: break; } return 0; } int calc_segregation_distortion(map > &seg_ratios, map &catalog, PopMap *pmap, set &parent_ids) { map types; map > dictionary; switch(map_type) { case dh: load_dh_dictionary(types, dictionary); break; case cp: load_cp_dictionary(types, dictionary); break; case bc1: load_mm_bc_dictionary(types, dictionary); break; case f2: load_mm_f2_dictionary(types, dictionary); break; case gen: case none: case unk: break; } map::iterator it; CSLocus *loc; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (seg_ratios.count(loc->marker) == 0) continue; map cnts; double n = tally_translated_gtypes(loc->id, pmap, parent_ids, dictionary[loc->marker], cnts); if (n == 0) continue; // cerr << "ID: " << loc->id << "; marker: " << loc->marker << "\n"; loc->chisq = chisq_test(seg_ratios, cnts, loc->marker, n); } return 0; } double tally_translated_gtypes(int loc_id, PopMap *pmap, set &parent_ids, map &dictionary, map &cnts) { Datum **d = pmap->locus(loc_id); double n = 0.0; for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] == NULL) continue; if (parent_ids.count(pmap->rev_sample_index(i))) continue; if (strcmp(d[i]->gtype, "--") == 0) continue; n++; if (cnts.count(dictionary[d[i]->gtype]) > 0) cnts[dictionary[d[i]->gtype]]++; else cnts[dictionary[d[i]->gtype]] = 1; } return n; } double tally_generic_gtypes(int loc_id, PopMap *pmap, set &parent_ids, map &cnts) { Datum **d = pmap->locus(loc_id); double n = 0.0; for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] == NULL) continue; if (parent_ids.count(pmap->rev_sample_index(i))) continue; if (strcmp(d[i]->gtype, "--") == 0) continue; n++; if (cnts.count(d[i]->gtype) > 0) cnts[d[i]->gtype]++; else cnts[d[i]->gtype] = 1; } return n; } double chisq_test(map > &seg_ratios, map &cnts, string marker, double n) { // // Calculate chi-square value. // sit->second * n == the expected value for this genotype // double chisq = 0.0; double exp = 0.0; double obs = 0.0; double df = seg_ratios[marker].size() - 1; map::iterator sit; for (sit = seg_ratios[marker].begin(); sit != seg_ratios[marker].end(); sit++) { obs = cnts.count(sit->first) == 0 ? 0 : cnts[sit->first]; exp = sit->second * n; // cerr << " category: " << sit->first << "; obs: " << obs << "; exp: " << exp << "\n"; chisq += ((obs - exp) * (obs - exp)) / exp; } // cerr << " df: " << df << "; Chisq value: " << chisq << "; pvalue: " << chisq_pvalue(df, chisq) << "\n"; // // Determine p-value // return chisq_pvalue(df, chisq); } double chisq_pvalue(int df, double chisq) { int i = 0; while (chisq > chisq_crit_values[df][i] && i < chisq_crit_values_size) { i++; } if (i == chisq_crit_values_size) return chisq_crit_values[0][chisq_crit_values_size - 1]; return chisq_crit_values[0][i]; } int map_specific_genotypes(map &catalog, PopMap *pmap, set &parent_ids) { map types; map > dictionary; switch(map_type) { case dh: load_dh_dictionary(types, dictionary); break; case cp: load_cp_dictionary(types, dictionary); break; case bc1: load_bc_dictionary(types, dictionary); break; case f2: load_f2_dictionary(types, dictionary); break; case gen: case none: case unk: break; } map::iterator it; string marker, m; Datum **d; CSLocus *loc; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; loc->gcnt = 0; if (loc->marker.length() == 0) continue; if (types.count(loc->marker)) { loc->uncor_marker = loc->marker; loc->marker = types[loc->marker]; marker = loc->marker; } else { marker = ""; } d = pmap->locus(loc->id); for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] == NULL) continue; if (parent_ids.count(pmap->rev_sample_index(i))) continue; if (marker.length() == 0) { m = "--"; } else { m = dictionary[marker].count(d[i]->gtype) ? dictionary[marker][d[i]->gtype] : dictionary[marker]["--"]; } strcpy(d[i]->gtype, m.c_str()); if (m != dictionary[marker]["--"]) loc->gcnt++; } } return 0; } int translate_genotypes(map &types, map > &dictionary, map &catalog, PopMap *pmap, map &samples, set &parent_ids) { map::iterator it; CSLocus *loc; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; string marker = types.count(loc->marker) ? types[loc->marker] : ""; Datum **d = pmap->locus(loc->id); for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] == NULL) continue; if (parent_ids.count(pmap->rev_sample_index(i))) continue; //cerr << "Examining progeny " << samples[pmap->rev_sample_index(i)] << "; marker: " << loc->marker << "\n"; string m; if (marker.length() == 0) { m = dictionary[marker]["--"]; } else { m = dictionary[marker].count(d[i]->gtype) ? dictionary[marker][d[i]->gtype] : dictionary[marker]["--"]; } d[i]->trans_gtype = new char[m.length() + 1]; // // If the genotype was corrected, output it in uppercase letters. // if (d[i]->corrected) { for (uint k = 0; k < m.length(); k++) d[i]->trans_gtype[k] = toupper(m[k]); d[i]->trans_gtype[m.length()] = '\0'; } else { strcpy(d[i]->trans_gtype, m.c_str()); } if (m != dictionary[marker]["--"]) loc->trans_gcnt++; //cerr << " allele: " << d[i]->trans_gtype << "; trans_gcnt: " << loc->trans_gcnt << "\n"; } } return 0; } int tally_progeny_haplotypes(CSLocus *locus, PopMap *pmap, set &parent_ids, int &total, double &max, string &freq_str) { char gtype[id_len]; map freq; Datum **d = pmap->locus(locus->id); total = 0; max = 0; //cerr << "Examining marker: " << locus->id << "\n"; for (int i = 0; i < pmap->sample_cnt(); i++) { if (parent_ids.count(pmap->rev_sample_index(i))) continue; if (d[i] == NULL) continue; //cerr << " Sample: " << i << "; Haplotype: " << d[i]->obshap[0] << "; Genotype: " << d[i]->gtype << "\n"; if (strcmp(d[i]->gtype, "--") != 0) { // // Automated corrections will uppercase genotypes, convert them back to lowercase // in order to tally them properly. // int j = 0; while (d[i]->gtype[j] != '\0') { gtype[j] = tolower(d[i]->gtype[j]); j++; } gtype[j] = '\0'; freq[gtype]++; total++; } } if (total == 0) return 0; double frac; stringstream s; char f[id_len]; map::iterator it; for (it = freq.begin(); it != freq.end(); it++) { frac = (double) it->second / (double) total * 100; if (frac > max) max = frac; sprintf(f, "(%0.1f%%);", frac); s << it->first << ":" << it->second << f; } freq_str = s.str().substr(0, s.str().length() - 1); return 0; } int write_sql(map &catalog, PopMap *pmap, set &parent_ids) { if (map_type == none) return 0; stringstream pop_name; pop_name << "batch_" << batch_id << ".markers.tsv"; string file = in_path + pop_name.str(); cerr << "Writing SQL markers file to '" << file << "'\n"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening markers SQL file '" << file << "'\n"; exit(1); } fh << "# SQL ID" << "\t" << "Batch ID" << "\t" << "Catalog Locus ID" << "\t" << "Marker Type" << "\t" << "Total Genotypes" << "\t" << "Max" << "\t" << "Genotype Freqs" << "\t" << "Segregation Distortion" << "\t" << "Mean Log Likelihood" << "\t" << "Genotype Map" << "\t" << "Uncorrected Marker" << "\n"; map::iterator it; map::iterator j; CSLocus *loc; char max_str[id_len]; stringstream gtype_map; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (loc->marker.length() == 0) continue; double max = 0.0; int total = 0; string freq, map; tally_progeny_haplotypes(loc, pmap, parent_ids, total, max, freq); sprintf(max_str, "%0.2f", max); // // Record the haplotype to genotype map. // gtype_map.str(""); for (j = loc->gmap.begin(); j != loc->gmap.end(); j++) gtype_map << j->first << ":" << j->second << ";"; map = gtype_map.str().substr(0, gtype_map.str().length() - 1); fh << 0 << "\t" << batch_id << "\t" << loc->id << "\t" << loc->marker << "\t" << total << "\t" << max_str << "\t" << freq << "\t" << loc->chisq << "\t" << loc->lnl << "\t" << map << "\t" << (loc->uncor_marker.length() == 0 ? loc->marker : loc->uncor_marker) << "\n"; } fh.close(); pop_name.str(""); pop_name << "batch_" << batch_id << ".genotypes_" << progeny_limit << ".txt"; file = in_path + pop_name.str(); cerr << "Writing SQL genotypes file to '" << file << "'\n"; fh.open(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening genotypes SQL file '" << file << "'\n"; exit(1); } fh << "# SQL ID" << "\t" << "Batch ID" << "\t" << "Catalog Locus ID" << "\t" << "Sample ID" << "\t" << "Genotype" << "\n"; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (loc->gcnt < progeny_limit) continue; Datum **d = pmap->locus(loc->id); for (int i = 0; i < pmap->sample_cnt(); i++) { if (parent_ids.count(pmap->rev_sample_index(i))) continue; fh << 0 << "\t" << batch_id << "\t" << loc->id << "\t" << pmap->rev_sample_index(i) << "\t"; if (d[i] == NULL) map_type == cp ? fh << "--\n" : fh << "-\n"; else fh << d[i]->gtype << "\n"; } } fh.close(); return 0; } int write_genomic(map &catalog, PopMap *pmap) { stringstream pop_name; pop_name << "batch_" << batch_id << ".genomic_" << progeny_limit << ".tsv"; string file = in_path + pop_name.str(); ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening genomic output file '" << file << "'\n"; exit(1); } // // Count the number of markers that have enough samples to output. // map::iterator cit; CSLocus *loc; int num_loci = 0; for (cit = catalog.begin(); cit != catalog.end(); cit++) { loc = cit->second; if (loc->hcnt < progeny_limit) continue; num_loci += loc->len - renz_len[enz]; } cerr << "Writing " << num_loci << " nucleotide positions to genomic file, '" << file << "'\n"; // // Write the header // fh << num_loci << "\t" << pmap->sample_cnt() << "\n"; // // Output each locus. // map >::iterator it; int a, b; uint rcnt = renz_cnt[enz]; uint rlen = renz_len[enz]; char *p; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { for (uint i = 0; i < it->second.size(); i++) { loc = it->second[i]; if (loc->hcnt < progeny_limit) continue; Datum **d = pmap->locus(loc->id); set snp_locs; string obshap; for (uint i = 0; i < loc->snps.size(); i++) snp_locs.insert(loc->snps[i]->col); uint start = 0; uint end = loc->len; // // Check for the existence of the restriction enzyme cut site, mask off // its output. // for (uint n = 0; n < rcnt; n++) if (strncmp(loc->con, renz[enz][n], rlen) == 0) start += renz_len[enz]; if (start == 0) { p = loc->con + (loc->len - rlen); for (uint n = rcnt; n < rcnt + rcnt; n++) if (strncmp(p, renz[enz][n], rlen) == 0) end -= renz_len[enz]; } uint k = 0; for (uint n = start; n < end; n++) { fh << loc->id << "\t" << loc->loc.chr << "\t" << loc->loc.bp + n; if (snp_locs.count(n) == 0) { for (int j = 0; j < pmap->sample_cnt(); j++) { a = encode_gtype(loc->con[n]); fh << "\t" << encoded_gtypes[a][a]; } } else { for (int j = 0; j < pmap->sample_cnt(); j++) { fh << "\t"; if (d[j] == NULL) fh << "0"; else switch (d[j]->obshap.size()) { case 1: a = encode_gtype(d[j]->obshap[0][k]); fh << encoded_gtypes[a][a]; break; case 2: a = encode_gtype(d[j]->obshap[0][k]); b = encode_gtype(d[j]->obshap[1][k]); fh << encoded_gtypes[a][b]; break; default: fh << "0"; break; } } k++; } fh << "\n"; } } } fh.close(); return 0; } int write_generic(map &catalog, PopMap *pmap, map &samples, set &parent_ids, bool write_gtypes) { stringstream pop_name; pop_name << "batch_" << batch_id; if (write_gtypes) pop_name << ".genotypes_" << progeny_limit << ".tsv"; else pop_name << ".haplotypes_" << progeny_limit << ".tsv"; string file = in_path + pop_name.str(); ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening generic output file '" << file << "'\n"; exit(1); } // // Count the number of markers that have enough samples to output. // map::iterator it; CSLocus *loc; int num_loci = 0; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (write_gtypes == false && loc->hcnt < progeny_limit) continue; if (write_gtypes == true && loc->gcnt < progeny_limit) continue; num_loci++; } cerr << "Writing " << num_loci << " loci to " << (write_gtypes ? "genotype" : "observed haplotype") << " file, '" << file << "'\n"; // // Write the header // fh << "# Catalog ID\t"; if (expand_id) fh << "\t"; if (write_gtypes) fh << "Marker\t"; fh << "Cnt\t" << "Seg Dist\t"; map::iterator s; for (int i = 0; i < pmap->sample_cnt(); i++) { if (write_gtypes && parent_ids.count(pmap->rev_sample_index(i))) continue; fh << samples[pmap->rev_sample_index(i)]; if (i < pmap->sample_cnt() - 1) fh << "\t"; } fh << "\n"; // // Output each locus. // for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (write_gtypes == false && loc->hcnt < progeny_limit) continue; if (write_gtypes == true && loc->gcnt < progeny_limit) continue; stringstream id; loc->annotation.length() > 0 ? id << loc->id << "|" << loc->annotation : id << loc->id; fh << id.str(); if (expand_id) { if (loc->annotation.length() > 0) id << "\t" << loc->id << "\t" << loc->annotation; else if (strlen(loc->loc.chr) > 0) id << "\t" << loc->id << "\t" << loc->loc.chr << "_" << loc->loc.bp; else id << "\t" << loc->id << "\t"; } if (write_gtypes) fh << "\t" << loc->marker; write_gtypes ? fh << "\t" << loc->gcnt : fh << "\t" << loc->hcnt; fh << "\t" << loc->chisq; Datum **d = pmap->locus(loc->id); string obshap; for (int i = 0; i < pmap->sample_cnt(); i++) { if (write_gtypes && parent_ids.count(pmap->rev_sample_index(i))) continue; fh << "\t"; if (d[i] == NULL) fh << "-"; else if (write_gtypes) { fh << d[i]->gtype; } else { obshap = ""; for (uint j = 0; j < d[i]->obshap.size(); j++) obshap += string(d[i]->obshap[j]) + "/"; obshap = obshap.substr(0, obshap.length()-1); fh << obshap; } } fh << "\n"; } fh.close(); return 0; } int write_joinmap(map &catalog, PopMap *pmap, map &types, map &samples, set &parent_ids) { stringstream pop_name; pop_name << "batch_" << batch_id << ".genotypes_" << progeny_limit; string file = in_path + pop_name.str() + ".loc"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening joinmap output file '" << file << "'\n"; exit(1); } // // Count the number of mappable progeny // map::iterator it; CSLocus *loc; int num_loci = 0; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (loc->trans_gcnt < progeny_limit) continue; num_loci++; } cerr << "Writing " << num_loci << " loci to JoinMap file, '" << file << "'\n"; map map_types; map_types[cp] = "CP"; map_types[dh] = "DH"; map_types[bc1] = "BC1"; map_types[f2] = "F2"; // // Output the header of the file // fh << "name = " << pop_name.str() << "\n" << "popt = " << map_types[map_type] << "\n" << "nloc = " << num_loci << "\n" << "nind = " << pmap->sample_cnt() - parent_ids.size() << "\n\n"; // // Output each locus. // for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (loc->trans_gcnt < progeny_limit) continue; stringstream id; loc->annotation.length() > 0 ? id << loc->id << "|" << loc->annotation : id << loc->id; fh << id.str() << "\t"; if (expand_id) { id.str(""); if (loc->annotation.length() > 0) id << loc->id << "\t" << loc->annotation; else if (strlen(loc->loc.chr) > 0) id << loc->id << "\t" << loc->loc.chr << "_" << loc->loc.bp; else id << loc->id << "\t"; fh << id.str() << "\t"; } if (types[loc->marker] == "lmx--") fh << ""; else if (types[loc->marker] == "--xnp") fh << ""; else fh << "<" << types[loc->marker] << ">"; Datum **d = pmap->locus(loc->id); for (int i = 0; i < pmap->sample_cnt(); i++) { if (parent_ids.count(pmap->rev_sample_index(i))) continue; fh << "\t"; if (d[i] == NULL) map_type == cp ? fh << "--" : fh << "-"; else fh << d[i]->trans_gtype; } fh << "\n"; } fh << "\nindividual names:\n"; map::iterator s; for (s = samples.begin(); s != samples.end(); s++) { if (parent_ids.count(s->first)) continue; fh << s->second << "\n"; } fh.close(); return 0; } int write_onemap_mapmaker(map &catalog, PopMap *pmap, map &types, map &samples, set &parent_ids) { stringstream pop_name; pop_name << "batch_" << batch_id << ".genotypes_" << progeny_limit; string file = in_path + pop_name.str() + ".onemap.txt"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening joinmap output file '" << file << "'\n"; exit(1); } // // Count the number of mappable progeny // map::iterator it; CSLocus *loc; int num_loci = 0; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (loc->trans_gcnt < progeny_limit) continue; num_loci++; } cerr << "Writing " << num_loci << " loci to OneMap file, '" << file << "'\n"; // // Output map type. // if (map_type == f2 ) fh << "data type f2 intercross\n"; else if (map_type == bc1) fh << "data type f2 backcross\n"; // // Output the header: number of individuals, number of markers, number of // quantitative traits (none). // fh << pmap->sample_cnt() - parent_ids.size() << " " << num_loci << " " << "0\n\n"; // // Output each locus. // for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (loc->trans_gcnt < progeny_limit) continue; fh << "*" << loc->id; Datum **d = pmap->locus(loc->id); for (int i = 0; i < pmap->sample_cnt(); i++) { if (parent_ids.count(pmap->rev_sample_index(i))) continue; fh << " "; if (d[i] == NULL) fh << "-"; else fh << d[i]->trans_gtype; } fh << "\n"; } fh.close(); return 0; } int write_onemap(map &catalog, PopMap *pmap, map &types, map &samples, set &parent_ids) { stringstream pop_name; pop_name << "batch_" << batch_id << ".genotypes_" << progeny_limit; string file = in_path + pop_name.str() + "onemap.tsv"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening joinmap output file '" << file << "'\n"; exit(1); } // // Count the number of mappable progeny // map::iterator it; CSLocus *loc; int num_loci = 0; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (loc->trans_gcnt < progeny_limit) continue; num_loci++; } cerr << "Writing " << num_loci << " loci to OneMap file, '" << file << "'\n"; map marker_types; marker_types["abxoo"] = "D1.11"; marker_types["ooxab"] = "D2.16"; marker_types["abxaa"] = "D1.10"; marker_types["aaxab"] = "D2.15"; marker_types["abxab"] = "B3.7"; marker_types["abxac"] = "A.2"; marker_types["abxcd"] = "A.1"; // // Output the header: number of individuals followed by number of markers. // fh << pmap->sample_cnt() - parent_ids.size() << "\t" << num_loci << "\n"; // // Output each locus. // for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (loc->trans_gcnt < progeny_limit) continue; fh << "*" << loc->id << " " << marker_types[types[loc->marker]] << "\t"; Datum **d = pmap->locus(loc->id); for (int i = 0; i < pmap->sample_cnt(); i++) { if (parent_ids.count(pmap->rev_sample_index(i))) continue; if (d[i] == NULL) fh << "-"; else fh << d[i]->trans_gtype; if (i < pmap->sample_cnt() - 1) fh << ","; } fh << "\n"; } fh.close(); return 0; } int write_rqtl(map &catalog, PopMap *pmap, map &types, map &samples, set &parent_ids) { stringstream pop_name; pop_name << "batch_" << batch_id << ".genotypes_" << progeny_limit; string file = in_path + pop_name.str() + ".rqtl.tsv"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening R/QTL output file '" << file << "'\n"; exit(1); } // // Count the number of mappable progeny // map::iterator it; CSLocus *loc; int num_loci = 0; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (loc->trans_gcnt < progeny_limit) continue; num_loci++; } cerr << "Writing " << num_loci << " loci to R/QTL file, '" << file << "'\n"; map map_types; map_types[cp] = "CP"; map_types[dh] = "DH"; map_types[bc1] = "BC1"; map_types[f2] = "F2"; // // Output the header of the file, followed by the list of markers, one per column // fh << "# Exported: " << pop_name.str() << "\n" << "# Map Type: " << map_types[map_type] << "\n" << "# Num Loci: " << num_loci << "\n" << "# Num Samples " << pmap->sample_cnt() - parent_ids.size() << "\n"; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (loc->gcnt < progeny_limit) continue; fh << ","; stringstream id; loc->annotation.length() > 0 ? id << loc->id << "|" << loc->annotation : id << loc->id; fh << id.str(); } fh << "\n"; // // Output the chromosome (if available) for each marker and then the location // for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (loc->gcnt < progeny_limit) continue; fh << ","; string chr; chr = strlen(loc->loc.chr) > 0 ? loc->loc.chr : "1"; fh << chr; } fh << "\n"; int i = 1; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (loc->gcnt < progeny_limit) continue; fh << ","; int bp = loc->loc.bp > 0 ? loc->loc.bp : i; fh << bp; i++; } fh << "\n"; // // For each sample, print out the genotypes for each marker // Datum *d; for (int i = 0; i < pmap->sample_cnt(); i++) { if (parent_ids.count(pmap->rev_sample_index(i))) continue; fh << samples[pmap->rev_sample_index(i)]; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; //if (loc->gcnt < progeny_limit) continue; d = pmap->datum(loc->id, pmap->rev_sample_index(i)); fh << ","; if (d == NULL) map_type == cp ? fh << "--" : fh << "-"; else fh << d->trans_gtype; } fh << "\n"; } fh.close(); return 0; } // sub create_imputed_genotype_map { // my ($order, $marker, $tag_id, $parents, $progeny, $map) = @_; // my (@keys, $key, $m, $type, $allele, $uall); // my (%gtypes, %allcnt, %uniqall); // // // // Count up the number of each type of observed allele in the progeny, // // record whether those genotypes are heterozygous or homozygous. // // // foreach $key (keys %{$progeny}) { // my $alleles; // print STDERR "Examining progeny $key\n" if ($debug); // // // // Discard progeny with more than one locus matched to this catalog tag. // // // @keys = keys %{$progeny->{$key}}; // next if (scalar(@keys) > 1); // $alleles = join("|", sort @{$progeny->{$key}->{$keys[0]}}); // if (!defined($allcnt{$alleles})) { // $allcnt{$alleles} = scalar(@{$progeny->{$key}->{$keys[0]}}); // } // //print STDERR "Adding genotype $alleles\n"; // $gtypes{$alleles}++; // foreach $allele (@{$progeny->{$key}->{$keys[0]}}) { // $uniqall{$allele}++; // } // } // // // // Examine the first parent alleles (the only alleles we have, since // // we are imputing the second parent. // // // my @parents = keys %{$parents}; // my %legal_genotypes = (); // $key = $order->{$parents[0]} eq "first" ? $parents[0] : $parents[1]; // $m = substr($marker, 0, 2); // foreach $type (split(//, $m)) { // //print STDERR " Adding $type to genotypes\n" if ($debug); // $legal_genotypes{$type}++; // } // my @types = sort keys %legal_genotypes; // if ($marker eq "lmxll") { // @keys = sort {$gtypes{$b} <=> $gtypes{$a}} keys %gtypes; // // // // Discard heterozygous alleles and find the highest frequency homozygote, // // this is the "l" in the "lmxll" marker. // // // while ($allcnt{$keys[0]} == 2) { // shift @keys; // } // $map->{$keys[0]} = shift @types; // print STDERR " Assinging '$keys[0]' to first parent genotype '", $map->{$keys[0]}, "'\n" if ($debug); // foreach $uall (sort {$uniqall{$b} <=> $uniqall{$a}} keys %uniqall) { // if ($uall ne $keys[0]) { // $allele = $uall; // last; // } // } // $map->{$allele} = shift @types; // print STDERR " Assinging '$allele' to first parent genotype '", $map->{$allele}, "'\n" if ($debug); // } // } int load_marker_list(string path, set &list) { char line[id_len]; ifstream fh(path.c_str(), ifstream::in); if (fh.fail()) { cerr << "Error opening white/black list file '" << path << "'\n"; exit(1); } int marker; char *e; while (fh.good()) { fh.getline(line, id_len); if (strlen(line) == 0) continue; marker = (int) strtol(line, &e, 10); if (*e == '\0') list.insert(marker); } fh.close(); if (list.size() == 0) { cerr << "Unable to load any markers from '" << path << "'\n"; help(); } return 0; } bool hap_compare(pair a, pair b) { return (a.second > b.second); } int parse_command_line(int argc, char* argv[]) { int c; while (1) { static struct option long_options[] = { {"help", no_argument, NULL, 'h'}, {"version", no_argument, NULL, 'v'}, {"corr", no_argument, NULL, 'c'}, {"sql", no_argument, NULL, 's'}, {"num_threads", required_argument, NULL, 'p'}, {"batch_id", required_argument, NULL, 'b'}, {"in_path", required_argument, NULL, 'P'}, {"map_type", required_argument, NULL, 't'}, {"out_type", required_argument, NULL, 'o'}, {"progeny", required_argument, NULL, 'r'}, {"min_depth", required_argument, NULL, 'm'}, {"min_hom_seqs", required_argument, NULL, 'H'}, {"min_het_seqs", required_argument, NULL, 'N'}, {"max_het_seqs", required_argument, NULL, 'X'}, {"renz", required_argument, NULL, 'e'}, {"whitelist", required_argument, NULL, 'W'}, {"blacklist", required_argument, NULL, 'B'}, {"man_corr", required_argument, NULL, 'C'}, {"lnl_lim", required_argument, NULL, 'L'}, {0, 0, 0, 0} }; // getopt_long stores the option index here. int option_index = 0; c = getopt_long(argc, argv, "hvcsib:p:t:o:r:P:m:e:H:N:X:W:B:C:L:", long_options, &option_index); // Detect the end of the options. if (c == -1) break; switch (c) { case 'h': help(); break; case 'P': in_path = optarg; break; case 'b': batch_id = is_integer(optarg); if (batch_id < 0) { cerr << "Batch ID (-b) must be an integer, e.g. 1, 2, 3\n"; help(); } break; case 't': if (strcasecmp(optarg, "cp") == 0) map_type = cp; else if (strcasecmp(optarg, "bc1") == 0) map_type = bc1; else if (strcasecmp(optarg, "f2") == 0) map_type = f2; else if (strcasecmp(optarg, "dh") == 0) map_type = dh; else if (strcasecmp(optarg, "gen") == 0) map_type = gen; else map_type = unk; break; case 'o': if (strcasecmp(optarg, "joinmap") == 0) out_type = joinmap; else if (strcasecmp(optarg, "rqtl") == 0) out_type = rqtl; else if (strcasecmp(optarg, "onemap") == 0) out_type = onemap; else if (strcasecmp(optarg, "genomic") == 0) out_type = genomic; break; case 'r': progeny_limit = atoi(optarg); break; case 'c': corrections = true; break; case 'L': lnl_limit = is_double(optarg); filter_lnl = true; break; case 'i': expand_id = true; break; case 's': sql_out = true; break; case 'W': wl_file = optarg; break; case 'B': bl_file = optarg; break; case 'C': man_corrections = true; cor_path = optarg; break; case 'm': min_stack_depth = is_integer(optarg); break; case 'H': min_hom_seqs = is_integer(optarg); break; case 'N': min_het_seqs = is_double(optarg); break; case 'X': max_het_seqs = is_double(optarg); break; case 'e': enz = optarg; break; case 'v': version(); break; case '?': // getopt_long already printed an error message. help(); break; default: help(); abort(); } } if (in_path.length() == 0) { cerr << "You must specify a path to the directory containing Stacks output files.\n"; help(); } if (in_path.at(in_path.length() - 1) != '/') in_path += "/"; if (batch_id < 0) { cerr << "You must specify a batch ID.\n"; help(); } if (map_type != cp && map_type != dh && map_type != bc1 && map_type != f2 && map_type != gen && map_type != none) { cerr << "You must specify a valid map type. 'CP', 'DH', 'F2', 'BC1' and 'GEN' are the currently supported map types.\n"; help(); } if (map_type != none && min_stack_depth > 0) cerr << "Warning: using a minimum stack depth when building genetic markers is not recommended.\n"; if (out_type == genomic && enz.length() == 0) { cerr << "You must specify the restriction enzyme used with 'genomic' output.\n"; help(); } if (out_type == genomic && renz.count(enz) == 0) { cerr << "Unrecognized restriction enzyme specified: '" << enz.c_str() << "'.\n"; help(); } return 0; } void version() { std::cerr << "genotypes " << VERSION << "\n\n"; exit(0); } void help() { std::cerr << "genotypes " << VERSION << "\n" << "genotypes -b batch_id -P path [-r min] [-m min] [-t map_type -o type] [-B blacklist] [-W whitelist] [-c] [-s] [-e renz] [-v] [-h]" << "\n" << " b: Batch ID to examine when exporting from the catalog.\n" << " r: minimum number of progeny required to print a marker.\n" << " c: make automated corrections to the data.\n" << " P: path to the Stacks output files.\n" << " t: map type to write. 'CP', 'DH', 'F2', 'BC1' and 'GEN' are the currently supported map types.\n" << " o: output file type to write, 'joinmap', 'onemap', 'rqtl', and 'genomic' are currently supported.\n" << " m: specify a minimum stack depth required before exporting a locus in a particular individual.\n" << " s: output a file to import results into an SQL database.\n" << " B: specify a file containing Blacklisted markers to be excluded from the export.\n" << " W: specify a file containign Whitelisted markers to include in the export.\n" << " e: restriction enzyme, required if generating 'genomic' output.\n" << " v: print program version." << "\n" << " h: display this help messsage." << "\n" << " Filtering options:\n" << " --lnl_lim [num]: filter loci with log likelihood values below this threshold.\n" << " Automated corrections options:\n" << " --min_hom_seqs: minimum number of reads required at a stack to call a homozygous genotype (default 5).\n" << " --min_het_seqs: below this minor allele frequency a stack is called a homozygote, above it (but below --max_het_seqs) it is called unknown (default 0.05).\n" << " --max_het_seqs: minimum frequency of minor allele to call a heterozygote (default 0.1).\n" << " Manual corrections options:\n" << " --cor_path : path to file containing manual genotype corrections from a Stacks SQL database to incorporate into output.\n"; exit(0); } stacks-1.44/src/genotypes.h000644 000765 000024 00000012474 12764047443 016502 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __GENOTYPES_H__ #define __GENOTYPES_H__ #ifdef _OPENMP #include // OpenMP library #endif #include // Process command-line options #include // Open/Read contents of a directory #include #include #include #include using std::pair; using std::make_pair; #include using std::string; #include #include using std::ifstream; using std::ofstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include using std::stringstream; #include using std::vector; #include using std::map; #include using std::set; #include "constants.h" #include "input.h" #include "stacks.h" #include "locus.h" #include "renz.h" #include "PopMap.h" #include "sql_utilities.h" #include "catalog_utils.h" #include "utils.h" #include "genotype_dictionaries.h" // // Chi-squared distribution critical values // [0] => p-values // [1] => one degree of freedom // [2] => two degrees of freedom // [3] => three degrees of freedom // const int chisq_crit_values_size = 10; const double chisq_crit_values[4][10] = { {0.50, 0.25, 0.20, 0.15, 0.10, 0.05, 0.01, 0.005, 0.001, 0.0005}, {0.46, 1.32, 1.64, 2.07, 2.71, 3.84, 6.63, 7.880, 10.830, 12.1200}, {1.39, 2.77, 3.22, 3.79, 4.61, 5.99, 9.21, 10.600, 13.820, 15.2000}, {2.37, 4.11, 4.64, 5.32, 6.25, 7.81, 11.34, 12.840, 16.270, 17.7300} }; void help( void ); void version( void ); int parse_command_line(int, char**); int build_file_list(vector &); int load_marker_list(string, set &); int identify_parental_ids(map &, vector &, set &); int find_markers(map &, PopMap *, set &); int calculate_f(map &, PopMap *, set &); int create_genotype_map(CSLocus *, PopMap *, set &); int apply_locus_constraints(map &, PopMap *); int call_population_genotypes(CSLocus *, PopMap *, map > &); int tally_progeny_haplotypes(CSLocus *, PopMap *, set &, int &, double &, string &); int map_specific_genotypes(map &, PopMap *, set &); int translate_genotypes(map &, map > &, map &, PopMap *, map &, set &); int automated_corrections(map &, set &, map &, vector > &, PopMap *); int check_uncalled_snps(CSLocus *, Locus *, Datum *); int call_alleles(vector &, vector &, vector &); int check_homozygosity(vector &, int, char, char, string &); int manual_corrections(string, PopMap *); int correct_cp_markers_missing_alleles(set &, map &, PopMap *); int calc_segregation_distortion(map > &, map &, PopMap *, set &); double tally_generic_gtypes(int, PopMap *, set &, map &); double tally_translated_gtypes(int, PopMap *, set &, map &, map &); double chisq_test(map > &, map &, string, double); double chisq_pvalue(int, double); int export_gen_map(map &, PopMap *, set &, map &); int export_cp_map(map &, PopMap *, set &, map &); int export_bc1_map(map &, PopMap *, set &, map &); int export_dh_map(map &, PopMap *, set &, map &); int export_f2_map(map &, PopMap *, set &, map &); int write_generic(map &, PopMap *, map &, set &, bool); int write_sql(map &, PopMap *, set &); int write_joinmap(map &, PopMap *, map &, map &, set &); int write_onemap(map &, PopMap *, map &, map &, set &); int write_onemap_mapmaker(map &, PopMap *, map &, map &, set &); int write_rqtl(map &, PopMap *, map &, map &, set &); int write_genomic(map &, PopMap *); bool hap_compare(pair, pair); #endif // __GENOTYPES_H__ stacks-1.44/src/gzFasta.h000644 000765 000024 00000013010 12764047443 016047 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2013, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __GZFASTA_H__ #define __GZFASTA_H__ #ifdef HAVE_LIBZ #include #include #include "input.h" class GzFasta: public Input { gzFile gz_fh; string buf; public: GzFasta(const char *path) : Input() { this->gz_fh = gzopen(path, "rb"); if (!this->gz_fh) { cerr << "Failed to open gzipped file '" << path << "': " << strerror(errno) << ".\n"; exit(EXIT_FAILURE); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(this->gz_fh, libz_buffer_size); #endif }; GzFasta(string path) : Input() { this->gz_fh = gzopen(path.c_str(), "rb"); if (!this->gz_fh) { cerr << "Failed to open gzipped file '" << path << "': " << strerror(errno) << ".\n"; exit(EXIT_FAILURE); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(this->gz_fh, libz_buffer_size); #endif }; ~GzFasta() { gzclose(this->gz_fh); }; Seq *next_seq(); int next_seq(Seq &); }; Seq *GzFasta::next_seq() { // // Check the contents of the line buffer. When we finish reading a FASTA record // the buffer will either contain whitespace or the header of the next FAST // record. // while (this->line[0] != '>' && !gzeof(this->gz_fh)) { gzgets(this->gz_fh, this->line, max_len); } if (gzeof(this->gz_fh)) { return NULL; } // // Check if there is a carraige return in the buffer // uint len = strlen(this->line); if (len > 0 && this->line[len - 1] == '\n') this->line[len - 1] = '\0'; if (len > 0 && this->line[len - 2] == '\r') this->line[len - 2] = '\0'; // // Initialize the Seq structure and store the FASTA ID // Seq *s = new Seq; s->id = new char[len + 1]; strcpy(s->id, this->line + 1); // // Read the sequence from the file -- keep reading lines until we reach the next // record or the end of file. // gzgets(this->gz_fh, this->line, max_len); while (this->line[0] != '>' && !gzeof(this->gz_fh)) { len = strlen(this->line); if (len > 0 && this->line[len - 1] == '\n') this->line[len - 1] = '\0'; if (len > 0 && this->line[len - 2] == '\r') this->line[len - 2] = '\0'; this->buf += this->line; this->line[0] = '\0'; gzgets(this->gz_fh, this->line, max_len); } if (gzeof(this->gz_fh)) { len = strlen(this->line); if (len > 0 && this->line[len - 1] == '\n') this->line[len - 1] = '\0'; if (len > 0 && this->line[len - 2] == '\r') this->line[len - 2] = '\0'; this->buf += this->line; this->line[0] = '\0'; } s->seq = new char[this->buf.length() + 1]; strcpy(s->seq, this->buf.c_str()); this->buf.clear(); return s; } int GzFasta::next_seq(Seq &s) { // // Check the contents of the line buffer. When we finish reading a FASTA record // the buffer will either contain whitespace or the header of the next FAST // record. // while (this->line[0] != '>' && !gzeof(this->gz_fh)) { gzgets(this->gz_fh, this->line, max_len); } if (gzeof(this->gz_fh)) { return 0; } // // Check if there is a carraige return in the buffer // uint len = strlen(this->line); if (len > 0 && this->line[len - 1] == '\n') this->line[len - 1] = '\0'; if (len > 0 && this->line[len - 2] == '\r') this->line[len - 2] = '\0'; // // Store the FASTA ID // strcpy(s.id, this->line + 1); // // Read the sequence from the file -- keep reading lines until we reach the next // record or the end of file. // gzgets(this->gz_fh, this->line, max_len); while (this->line[0] != '>' && !gzeof(this->gz_fh)) { len = strlen(this->line); if (len > 0 && this->line[len - 1] == '\n') this->line[len - 1] = '\0'; if (len > 0 && this->line[len - 2] == '\r') this->line[len - 2] = '\0'; this->buf += this->line; this->line[0] = '\0'; gzgets(this->gz_fh, this->line, max_len); } if (gzeof(this->gz_fh)) { len = strlen(this->line); if (len > 0 && this->line[len - 1] == '\n') this->line[len - 1] = '\0'; if (len > 0 && this->line[len - 2] == '\r') this->line[len - 2] = '\0'; this->buf += this->line; this->line[0] = '\0'; } strcpy(s.seq, this->buf.c_str()); this->buf.clear(); return 1; } #else // If HAVE_LIBZ is undefined and zlib library is not present. #include "input.h" class GzFasta: public Input { public: GzFasta(const char *path) : Input() { cerr << "Gzip support was not enabled when Stacks was compiled.\n"; }; ~GzFasta() {}; Seq *next_seq() { return NULL; }; int next_seq(Seq &) { return 0; }; }; #endif // HAVE_LIBZ #endif // __GZFASTA_H__ stacks-1.44/src/gzFastq.h000644 000765 000024 00000014520 12764047443 016076 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2013, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __GZFASTQ_H__ #define __GZFASTQ_H__ #ifdef HAVE_LIBZ #include #include #include "input.h" class GzFastq: public Input { gzFile gz_fh; public: GzFastq(string path) : Input() { this->gz_fh = gzopen(path.c_str(), "rb"); if (!this->gz_fh) { cerr << "Failed to open gzipped file '" << path << "': " << strerror(errno) << ".\n"; exit(EXIT_FAILURE); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(this->gz_fh, libz_buffer_size); #endif }; GzFastq(const char *path) : Input() { this->gz_fh = gzopen(path, "rb"); if (!this->gz_fh) { cerr << "Failed to open gzipped file '" << path << "': " << strerror(errno) << ".\n"; exit(EXIT_FAILURE); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(this->gz_fh, libz_buffer_size); #endif }; ~GzFastq() { gzclose(this->gz_fh); }; Seq *next_seq(); int next_seq(Seq &s); }; Seq *GzFastq::next_seq() { char *res = NULL; // // Check the contents of the line buffer. When we finish reading a FASTQ record // the buffer will either contain whitespace or the header of the next FASTQ // record. // this->line[0] = '\0'; do { res = gzgets(this->gz_fh, this->line, max_len); } while (this->line[0] != '@' && res != NULL); if (res == NULL) { return NULL; } // // Check if there is a carraige return in the buffer // uint len = strlen(this->line); if (len > 0 && this->line[len - 1] == '\n') this->line[len - 1] = '\0'; if (len > 0 && this->line[len - 2] == '\r') this->line[len - 2] = '\0'; // // Initialize the Seq structure and store the FASTQ ID // Seq *s = new Seq; s->id = new char[strlen(this->line) + 1]; strcpy(s->id, this->line + 1); // // Read the sequence from the file // gzgets(this->gz_fh, this->line, max_len); if (gzeof(this->gz_fh)) { return NULL; } len = strlen(this->line); if (len > 0 && this->line[len - 1] == '\n') this->line[len - 1] = '\0'; if (len > 0 && this->line[len - 2] == '\r') this->line[len - 2] = '\0'; s->seq = new char[len + 1]; strcpy(s->seq, this->line); // // Read the repeat of the ID // this->line[0] = '\0'; res = gzgets(this->gz_fh, this->line, max_len); if (this->line[0] != '+' || res == NULL) { return NULL; } // // Read the quality score from the file // this->line[0] = '\0'; res = gzgets(this->gz_fh, this->line, max_len); if (res == NULL && strlen(this->line) == 0) { return NULL; } len = strlen(this->line); if (len > 0 && this->line[len - 1] == '\n') this->line[len - 1] = '\0'; if (len > 0 && this->line[len - 2] == '\r') this->line[len - 2] = '\0'; s->qual = new char[len + 1]; strcpy(s->qual, this->line); // // Clear the line buffer so it is set up for the next record. If a '@' // appears in the quality scores read, it will break parsing next time // it is called. // this->line[0] = '\0'; return s; } int GzFastq::next_seq(Seq &s) { char *res = NULL; // // Check the contents of the line buffer. When we finish reading a FASTQ record // the buffer will either contain whitespace or the header of the next FASTQ // record. // this->line[0] = '\0'; do { res = gzgets(this->gz_fh, this->line, max_len); } while (this->line[0] != '@' && res != NULL); if (res == NULL) { return 0; } // // Check if there is a carraige return in the buffer // uint len = strlen(this->line); if (len > 0 && this->line[len - 1] == '\n') this->line[len - 1] = '\0'; if (len > 0 && this->line[len - 2] == '\r') this->line[len - 2] = '\0'; // // Store the FASTQ ID // strcpy(s.id, this->line + 1); // // Read the sequence from the file // this->line[0] = '\0'; res = gzgets(this->gz_fh, this->line, max_len); if (res == NULL) { return 0; } len = strlen(this->line); if (len > 0 && this->line[len - 1] == '\n') this->line[len - 1] = '\0'; if (len > 0 && this->line[len - 2] == '\r') this->line[len - 2] = '\0'; strcpy(s.seq, this->line); // // Read the repeat of the ID // this->line[0] = '\0'; res = gzgets(this->gz_fh, this->line, max_len); if (this->line[0] != '+' || res == NULL) { return 0; } // // Read the quality score from the file // this->line[0] = '\0'; res = gzgets(this->gz_fh, this->line, max_len); if (res == NULL && strlen(this->line) == 0) { return 0; } len = strlen(this->line); if (len > 0 && this->line[len - 1] == '\n') this->line[len - 1] = '\0'; if (len > 0 && this->line[len - 2] == '\r') this->line[len - 2] = '\0'; strcpy(s.qual, this->line); // // Clear the line buffer so it is set up for the next record. If a '@' // appears in the quality scores read, it will break parsing next time // it is called. // this->line[0] = '\0'; return 1; } #else // If HAVE_LIBZ is undefined and zlib library is not present. #include "input.h" class GzFastq: public Input { public: GzFastq(const char *path) : Input() { cerr << "Gzip support was not enabled when Stacks was compiled.\n"; }; GzFastq(string path) : Input() { cerr << "Gzip support was not enabled when Stacks was compiled.\n"; }; ~GzFastq() {}; Seq *next_seq() { return NULL; }; int next_seq(Seq &) { return 0; }; }; #endif // HAVE_LIBZ #endif // __GZFASTQ_H__ stacks-1.44/src/hstacks.cc000644 000765 000024 00000066337 12764047443 016272 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // hstacks -- find homologous stacks among a set of samples // // Match stacks between samples to identify homologous loci. Stacks // may contain masked sites (N's), resulting from using the fixed-model in // the ustacks program, or an explicit number of mismatches per tag may be // specified, allowing non-exact matching homologus sites to be identified // across a set of samples. // // Julian Catchen // jcatchen@uoregon.edu // University of Oregon // // $Id$ // #include "hstacks.h" // Global variables to hold command-line options. string in_path; string out_path; int batch_id = 0; int num_threads = 1; int stack_depth_min = 1; int stack_dist = 0; int n_limit = 4; int main (int argc, char* argv[]) { parse_command_line(argc, argv); // // Set the number of OpenMP parallel threads to execute. // #ifdef _OPENMP omp_set_num_threads(num_threads); #endif vector input_files; vector::iterator in_file; map samples; build_file_list(in_path, input_files); int id = 0; for (in_file = input_files.begin(); in_file != input_files.end(); in_file++) { map sample; map::iterator it; size_t pos_1 = (*in_file).find_last_of("/"); size_t pos_2 = (*in_file).find_last_of("."); string sample_id = (*in_file).substr(pos_1 + 1, (pos_2 - pos_1 - 1)); bool compressed = false; load_loci(*in_file, sample, false, false, compressed); // // Give each locus a unique ID among all samples // for (it = sample.begin(); it != sample.end(); it++) { it->second->uniq_id = id; samples[id] = (*it).second; id++; } } // // Calculate distance between tags in different samples. // cerr << "Calculating distance between stacks...\n"; calc_kmer_distance(samples, stack_dist); // // Write out tags matched between samples. // write_homologous_loci(samples); return 0; } int calc_kmer_distance(map &loci, int stack_dist) { // // Calculate the distance (number of mismatches) between each pair // of Radtags. We expect all radtags to be the same length; // CatKmerHashMap kmer_map; map::iterator it; vector >::iterator allele; HLocus *tag_1, *tag_2; int i, j; // // Calculate the number of k-mers we will generate. If kmer_len == 0, // determine the optimal length for k-mers. // int con_len = strlen(loci.begin()->second->con); int kmer_len = determine_kmer_length(con_len, stack_dist); int num_kmers = con_len - kmer_len + 1; // // Calculate the minimum number of matching k-mers required for a possible sequence match. // int min_hits = calc_min_kmer_matches(kmer_len, stack_dist, con_len, false); // // If more mismatches are allowed than can be handled by the k-mer algorithm, revert // to the simple, slow matching algorithm. // if (min_hits <= 0) { cerr << " Unable to use k-mer matching due to the number of allowed mismatches. Switching to slower algorithm...\n"; calc_distance(loci, stack_dist); return 0; } cerr << " Number of kmers per sequence: " << num_kmers << "\n"; populate_kmer_hash(loci, kmer_map, kmer_len); cerr << " " << loci.size() << " loci, " << kmer_map.size() << " elements in the kmer hash.\n"; // // OpenMP can't parallelize random access iterators, so we convert // our map to a vector of integer keys. // vector keys; for (it = loci.begin(); it != loci.end(); it++) keys.push_back(it->first); #pragma omp parallel private(i, j, tag_1, tag_2, allele) { #pragma omp for schedule(dynamic) for (i = 0; i < (int) keys.size(); i++) { tag_1 = loci[keys[i]]; for (allele = tag_1->strings.begin(); allele != tag_1->strings.end(); allele++) { vector kmers; generate_kmers(allele->second.c_str(), kmer_len, num_kmers, kmers); map > hits; vector >::iterator map_it; int d; // // Lookup the occurances of each k-mer in the kmer_map // for (j = 0; j < num_kmers; j++) { if (kmer_map.count(kmers[j]) > 0) for (map_it = kmer_map[kmers[j]].begin(); map_it != kmer_map[kmers[j]].end(); map_it++) hits[map_it->second].push_back(map_it->first); } // // Free the allocated k-mers. // for (j = 0; j < num_kmers; j++) delete [] kmers[j]; kmers.clear(); //cerr << " Tag " << tag_1->id << " hit " << hits.size() << " kmers.\n"; map >::iterator hit_it; vector::iterator all_it; // // Iterate through the list of hits. For each hit, total up the hits to the various alleles. // Any allele that has more than min_hits check its full length to verify a match. // for (hit_it = hits.begin(); hit_it != hits.end(); hit_it++) { //cerr << " Tag " << hit_it->first << " has " << hit_it->second << " hits (min hits: " << min_hits << ")\n"; map allele_cnts; map::iterator cnt_it; for (all_it = hit_it->second.begin(); all_it != hit_it->second.end(); all_it++) allele_cnts[*all_it]++; for (cnt_it = allele_cnts.begin(); cnt_it != allele_cnts.end(); cnt_it++) { //cerr << " allele " << cnt_it->first << " has " << cnt_it->second << " hits\n"; if (cnt_it->second < min_hits) continue; //cerr << " Match found, checking full-length match\n"; tag_2 = loci[hit_it->first]; d = dist(allele->second.c_str(), tag_2, cnt_it->first); if (d < 0) cerr << "Unknown error calculating distance between " << tag_1->id << " and " << tag_2->id << "; query allele: " << allele->first << "\n"; //cerr << " Distance: " << d << " CTAG_DIST: " << ctag_dist << "\n"; // // Add a match to the query sequence: catalog ID, catalog allele, query allele, distance // if (d <= stack_dist) { if (tag_1->depth < stack_depth_min || tag_2->depth < stack_depth_min) continue; tag_1->add_match(tag_2->uniq_id, d); } } } } // Sort the vector of distances. sort(tag_1->matches.begin(), tag_1->matches.end(), compare_mdist); } } return 0; } int populate_kmer_hash(map &loci, CatKmerHashMap &kmer_map, int kmer_len) { map::iterator it; vector >::iterator allele; vector kmers; HLocus *tag; char *hash_key; bool exists; int j; // // Break each stack down into k-mers and create a hash map of those k-mers // recording in which sequences they occur. // int num_kmers = strlen(loci.begin()->second->con) - kmer_len + 1; for (it = loci.begin(); it != loci.end(); it++) { tag = it->second; // // Iterate through the possible Loci alleles // for (allele = tag->strings.begin(); allele != tag->strings.end(); allele++) { // // Generate and Hash the kmers for this allele string // generate_kmers(allele->second.c_str(), kmer_len, num_kmers, kmers); for (j = 0; j < num_kmers; j++) { exists = kmer_map.count(kmers[j]) == 0 ? false : true; if (exists) { hash_key = kmers[j]; } else { hash_key = new char [strlen(kmers[j]) + 1]; strcpy(hash_key, kmers[j]); } kmer_map[hash_key].push_back(make_pair(allele->first, tag->uniq_id)); } for (j = 0; j < num_kmers; j++) delete [] kmers[j]; kmers.clear(); } } //dump_kmer_map(kmer_map); return 0; } int calc_distance(map &loci, int utag_dist) { // // Calculate the distance (number of mismatches) between each pair // of Radtags. We expect all radtags to be the same length; // map::iterator it; HLocus *tag_1, *tag_2; int i, j; cerr << "Calculating distance between stacks...\n"; // OpenMP can't parallelize random access iterators, so we convert // our map to a vector of integer keys. vector keys; for (it = loci.begin(); it != loci.end(); it++) keys.push_back(it->first); #pragma omp parallel private(i, j, tag_1, tag_2) { #pragma omp for schedule(dynamic) for (i = 0; i < (int) keys.size(); i++) { tag_1 = loci[keys[i]]; int d; for (j = 0; j < (int) keys.size(); j++) { tag_2 = loci[keys[j]]; // Don't compare tag_1 against itself. if (tag_1 == tag_2) continue; d = dist(tag_1, tag_2); // // Store the distance between these two sequences if it is // below the maximum distance. // if (d == utag_dist) { if (tag_1->depth < stack_depth_min || tag_2->depth < stack_depth_min) continue; tag_1->add_match(tag_2->uniq_id, d); } } // Sort the vector of distances. sort(tag_1->matches.begin(), tag_1->matches.end(), compare_mdist); } } return 0; } int dist(HLocus *tag_1, HLocus *tag_2) { int dist = 0; char *p = tag_1->con; char *q = tag_2->con; char *end = p + strlen(p); // Count the number of characters that are different // between the two sequences. Don't count wildcard 'N' // nucleotides. while (p < end) { dist += ((*p == *q) || (*q == 'N' || *p == 'N')) ? 0 : 1; p++; q++; } return dist; } bool compare_mdist(Match *a, Match *b) { return (a->dist < b->dist); } int call_consensus(map &loci, set &merge_list, string &consensus, vector &snps, vector &alleles) { // // Create a two-dimensional array, each row containing one read. For // each unique tag that has been merged together, add the sequence for // that tag into our array as many times as it originally occurred. // HLocus *tag; set::iterator j; vector reads; for (j = merge_list.begin(); j != merge_list.end(); j++) { tag = loci[*j]; reads.push_back(tag->con); } // // Iterate over each column of the array and call the consensus base. // int row, col; int length = strlen(reads[0]); int height = reads.size(); char *base; for (col = 0; col < length; col++) { vector > nuc; nuc.push_back(make_pair('A', 0)); nuc.push_back(make_pair('C', 0)); nuc.push_back(make_pair('G', 0)); nuc.push_back(make_pair('T', 0)); for (row = 0; row < height; row++) { base = reads[row]; base = base + col; //cerr << " Row: " << row << " Col: " << col << " Base: " << *base << "\n"; switch(*base) { case 'A': nuc[0].second++; break; case 'C': nuc[1].second++; break; case 'G': nuc[2].second++; break; case 'T': nuc[3].second++; break; default: break; } } //cerr << "A: " << nuc[0].second << " C: " << nuc[1].second << " G: " << nuc[2].second << " T: " << nuc[3].second << "\n"; // // Find the base with a plurality of occurances and call it. // sort(nuc.begin(), nuc.end(), compare_pair); consensus += nuc[0].second > 0 ? nuc[0].first : 'N'; // // If the nucleotides are not fixed record a SNP. // if (nuc[0].second > 0 && nuc[1].second > 0) { SNP *s = new SNP; s->col = col; s->lratio = 0; s->rank_1 = nuc[0].first; s->rank_2 = nuc[1].first; snps.push_back(s); } } if (!call_alleles(reads, snps, alleles)) { cerr << "Error calling alleles.\n"; exit(0); } return 0; } int call_alleles(vector &reads, vector &snps, vector &alleles) { int row; int height = reads.size(); string allele; char *base; vector::iterator snp; if (snps.size() == 0) return 1; for (row = 0; row < height; row++) { allele.clear(); for (snp = snps.begin(); snp != snps.end(); snp++) { base = reads[row]; base = base + (*snp)->col; allele += *base; } if (allele.size() == snps.size()) alleles.push_back(allele); else return 0; } return 1; } int write_homologous_loci(map &samples) { map::iterator i; vector::iterator k; set write_map; HLocus *tag_1, *tag_2; cerr << "Writing homologous stacks...\n"; // // Parse the input file name to create the output files // stringstream prefix; string out_file; prefix << out_path << "batch_" << batch_id; // Open the output files for writing. out_file = prefix.str() + ".homologous.nucs.tsv"; ofstream nuc_file(out_file.c_str()); out_file = prefix.str() + ".homologous.snps.tsv"; ofstream snp_file(out_file.c_str()); out_file = prefix.str() + ".homologous.alleles.tsv"; ofstream all_file(out_file.c_str()); out_file = prefix.str() + ".homologous.matches.tsv"; ofstream mat_file(out_file.c_str()); out_file = prefix.str() + ".homologous.tags.tsv"; ofstream tag_file(out_file.c_str()); int id = 1; for (i = samples.begin(); i != samples.end(); i++) { tag_1 = i->second; // // This tag may already have been merged by an earlier operation. // if (write_map.find(tag_1->uniq_id) != write_map.end()) continue; set unique_merge_list; set unique_alleles; set::iterator it; trace_stack_graph(tag_1, samples, unique_merge_list); // // Call the consensus for this locus and identify SNPs and associated alleles. // string consensus; vector snps; vector alleles; call_consensus(samples, unique_merge_list, consensus, snps, alleles); // // Output the consensus tag for a locus in this sample. // tag_file << "0" << "\t" << batch_id << "\t" << id << "\t" << tag_1->loc.chr << "\t" << tag_1->loc.bp << "\t" << "consensus" << "\t" << 0 << "\t" << "" << "\t" << consensus << "\t" << 0 << "\t" << // These flags are unused in hstacks, but important in ustacks 0 << "\t" << 0 << "\n"; // // Output the SNPs and alleles // string allele; vector::iterator s; set::iterator u; for (s = snps.begin(); s != snps.end(); s++) snp_file << "0" << "\t" << batch_id << "\t" << id << "\t" << (*s)->col << "\t" << (*s)->lratio << "\t" << (*s)->rank_1 << "\t" << (*s)->rank_2 << "\n"; for (uint a = 0; a < alleles.size(); a++) unique_alleles.insert(alleles[a]); for (u = unique_alleles.begin(); u != unique_alleles.end(); u++) all_file << "0" << "\t" << batch_id << "\t" << id << "\t" << *u << "\t" << 0 << "\t" << 0 << "\n"; unique_alleles.clear(); int sub_id = 0; int a = 0; for (it = unique_merge_list.begin(); it != unique_merge_list.end(); it++) { tag_2 = samples[(*it)]; // Record the nodes that have been merged in this round. write_map.insert(tag_2->uniq_id); // // For each tag we are outputting, output the depth of coverage for each // nucleotide in that stack separately (in order to calculate correlations // between depth of coverage and fixed/non-fixed nucleotides. // if (unique_merge_list.size() > 1) { char *p, *end; end = tag_2->con + strlen(tag_2->con); for (p = tag_2->con; p < end; p++) nuc_file << tag_2->sample_id << "_" << tag_2->id << "\t" << *p << "\t" << tag_2->depth << "\n"; } // // Output the consensus sequenes for all homologous loci. // tag_file << "0" << "\t" << batch_id << "\t" << id << "\t" << tag_2->loc.chr << "\t" << tag_2->loc.bp << "\t" << "primary" << "\t" << sub_id << "\t" << tag_2->sample_id << "_" << tag_2->id << "\t" << tag_2->con << "\t" << "" << "\t" << // These flags are unused in hstacks, but important in ustacks "" << "\t" << "" << "\n"; allele = (alleles.size() == 0) ? "consensus" : alleles[a]; mat_file << "0" << "\t" << batch_id << "\t" << id << "\t" << tag_2->sample_id << "\t" << tag_2->uniq_id << "\t" << allele << "\n"; sub_id++; a++; } id++; } tag_file.close(); nuc_file.close(); snp_file.close(); all_file.close(); mat_file.close(); cerr << " Wrote " << id - 1 << " tags to " << out_file << "\n"; return 0; } int trace_stack_graph(HLocus *tag_1, map &loci, set &unique_merge_list) { queue merge_list; pair::iterator,bool> ret; vector::iterator k; HLocus *tag_2; unique_merge_list.insert(tag_1->uniq_id); merge_list.push(tag_1->uniq_id); while (!merge_list.empty()) { tag_2 = loci[merge_list.front()]; merge_list.pop(); for (k = tag_2->matches.begin(); k != tag_2->matches.end(); k++) { ret = unique_merge_list.insert((*k)->cat_id); // // If this Tag has not already been added to the merge list (i.e. we were able // to insert it in to our unique_merge_list, which is a set), add it for consideration // later in the loop. // if (ret.second == true) merge_list.push((*k)->cat_id); } } return 0; } int build_file_list(string in_path, vector &sql_files) { struct dirent *dirp; string d; DIR *dp = opendir(in_path.c_str()); if (dp == NULL) { cerr << "Error (" << errno << ") opening " << in_path << "\n"; return errno; } while ((dirp = readdir(dp)) != NULL) { d = string(dirp->d_name); if (d.find("tags.tsv") != string::npos && d.find("batch") == string::npos) { size_t pos = d.find(".tags.tsv"); d = in_path + d.substr(0, pos); sql_files.push_back(d); } } closedir(dp); cerr << "Identified " << sql_files.size() << " samples.\n"; return 0; } HLocus::~HLocus() { vector::iterator it; for (it = this->matches.begin(); it != this->matches.end(); it++) delete *it; } int HLocus::add_match(int id, int distance) { Match *m = new Match; m->cat_id = id; m->dist = distance; this->matches.push_back(m); return 0; } int HLocus::populate_alleles() { this->strings.clear(); string s; int j; vector::iterator s_it; map::iterator a; if (this->strings.size() == 0) this->strings.push_back(make_pair("consensus", this->con)); else for (a = this->alleles.begin(); a != this->alleles.end(); a++) { s = this->con; j = 0; for (s_it = this->snps.begin(); s_it != this->snps.end(); s_it++) { s.replace((*s_it)->col, 1, 1, a->first[j]); j++; } this->strings.push_back(make_pair(a->first, s)); } // // Count the number of N's // vector col; char *p = this->con; int i = 0; while (*p != '\0') { if (*p == 'N') col.push_back(i); i++; p++; } int n_cnt = col.size(); if (n_cnt == 0) return 0; // // If there are too many Ns in this stack, do not include it in the // search. // if (n_cnt > n_limit) { this->strings.clear(); return 0; } // // Generate all permutations of strings for n_cnt N's // if (pstrings.count(n_cnt) == 0) generate_permutations(pstrings, n_cnt); vector > new_strings; vector >::iterator k; vector::iterator c; char *q, **r; int n = (int) pow(4, n_cnt); for (k = this->strings.begin(); k != this->strings.end(); k++) { for (i = 0; i < n; i++) { r = pstrings[n_cnt]; q = r[i]; s = k->second; j = 0; for (c = col.begin(); c != col.end(); c++) { //cerr << "Str: " << s << "; rep str: " << q << "; replacing col: " << *c << " with '" << q[j] << "'\n"; s.replace(*c, 1, 1, q[j]); j++; } new_strings.push_back(make_pair(k->first, s)); } } this->strings.clear(); for (k = new_strings.begin(); k != new_strings.end(); k++) this->strings.push_back(*k); return 0; } int parse_command_line(int argc, char* argv[]) { int c; while (1) { static struct option long_options[] = { {"help", no_argument, NULL, 'h'}, {"version", no_argument, NULL, 'v'}, {"stack_dist", required_argument, NULL, 'n'}, {"depth_min", required_argument, NULL, 'm'}, {"inpath", required_argument, NULL, 'p'}, {"outpath", required_argument, NULL, 'o'}, {"n_limit", required_argument, NULL, 'N'}, {"batch_id", required_argument, NULL, 'b'}, {0, 0, 0, 0} }; // getopt_long stores the option index here. int option_index = 0; c = getopt_long(argc, argv, "hvi:p:o:b:e:m:n:N:", long_options, &option_index); // Detect the end of the options. if (c == -1) break; switch (c) { case 'h': help(); break; case 'v': version(); break; case 'i': in_path = optarg; break; case 'o': out_path = optarg; break; case 'b': batch_id = atoi(optarg); break; case 'N': n_limit = atoi(optarg); break; case 'm': stack_depth_min = atoi(optarg); break; case 'n': stack_dist = atoi(optarg); break; case 'p': num_threads = atoi(optarg); break; case '?': // getopt_long already printed an error message. help(); break; default: cerr << "Unknown command line option '" << (char) c << "'\n"; help(); abort(); } } if (in_path.length() == 0) { cerr << "You must specify a path to a set of input files.\n"; help(); } if (in_path.at(in_path.length() - 1) != '/') in_path += "/"; if (out_path.length() == 0) out_path = "."; if (out_path.at(out_path.length() - 1) != '/') out_path += "/"; return 0; } void version() { std::cerr << "hstacks " << VERSION << "\n\n"; exit(0); } void help() { std::cerr << "hstacks " << VERSION << "\n" << "hstacks -i path [-o path] [-b batch_id] [-n mismatches] [-m min] [-p min_threads] [-N limit] [-h]" << "\n" << " i: path to the set of SQL files from which to load loci." << "\n" << " o: output path to write results." << "\n" << " b: SQL Batch ID to insert into the output to identify a group of samples." << "\n" << " m: minimum stack depth required for a locus to be included in the search." << "\n" << " n: number of mismatches to allow between stacks." << "\n" << " N: number of 'N' characters to allow in a stack (default: 4)." << "\n" << " p: enable parallel execution with num_threads threads.\n" << " h: display this help messsage." << "\n\n"; exit(0); } stacks-1.44/src/hstacks.h000644 000765 000024 00000005326 12335173442 016114 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __HSTACKS_H__ #define __HSTACKS_H__ #ifdef _OPENMP #include // OpenMP library #endif #include // Process command-line options #include #include #include #include #include #include using std::pair; using std::make_pair; #include using std::string; #include #include #include using std::ofstream; using std::stringstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include using std::vector; #include using std::map; #include using std::queue; #include using std::set; #include #include "constants.h" #include "stacks.h" #include "locus.h" #include "kmers.h" #include "models.h" #include "sql_utilities.h" // // A map holding k-mer permutation strings. For use when generating fuzzy k-mers. // map pstrings; // // Homologous Locus Class // class HLocus : public Locus { public: int uniq_id; // An ID that is unique among all samples in this analysis. vector matches; // Matching tags found for the catalog. HLocus(): Locus() {} ~HLocus(); int populate_alleles(); int add_match(int, int); }; void help( void ); void version( void ); int parse_command_line(int, char**); int build_file_list(string, vector &); int calc_kmer_distance(map &, int); int calc_distance(map &, int); int dist(HLocus *, HLocus *); int write_homologous_loci(map &); int trace_stack_graph(HLocus *, map &, set &); int call_consensus(map &, set &, string &, vector &, vector &); int call_alleles(vector &, vector &, vector &); int populate_kmer_hash(map &, CatKmerHashMap &, int); bool compare_mdist(Match *, Match *); bool compare_pair(pair, pair); #endif // __HSTACKS_H__ stacks-1.44/src/input.cc000644 000765 000024 00000016723 12775212561 015760 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // input.cc -- routines to read various formats of data into the XXX data structure. // #include "input.h" Seq::Seq() { this->id = NULL; this->seq = NULL; this->qual = NULL; this->loc_str = NULL; this->aln_type = pri_aln; this->pct_aln = 1.0; } Seq::Seq(const Seq& other) : loc(other.loc) { if (other.id != NULL) { id = new char[strlen(other.id)+1]; strcpy(id, other.id); } else { id = NULL; } if (other.seq != NULL) { seq = new char[strlen(other.seq)+1]; strcpy(seq, other.seq); } else { seq = NULL; } if (other.qual != NULL) { qual = new char[strlen(other.qual)+1]; strcpy(qual, other.qual); } else { qual = NULL; } if (other.loc_str != NULL) { loc_str = new char[strlen(other.loc_str)+1]; strcpy(loc_str, other.loc_str); } else { loc_str = NULL; } pct_aln = other.pct_aln; aln_type = other.aln_type; } Seq::Seq(const char *id, const char *seq) { this->id = new char[strlen(id) + 1]; this->seq = new char[strlen(seq) + 1]; this->qual = NULL; this->loc_str = NULL; strcpy(this->id, id); strcpy(this->seq, seq); this->aln_type = pri_aln; this->pct_aln = 1.0; } Seq::Seq(const char *id, const char *seq, const char *qual) { this->id = new char[strlen(id) + 1]; this->seq = new char[strlen(seq) + 1]; this->qual = new char[strlen(qual) + 1]; this->loc_str = NULL; strcpy(this->id, id); strcpy(this->seq, seq); strcpy(this->qual, qual); this->aln_type = pri_aln; this->pct_aln = 1.0; } Seq::Seq(const char *id, const char *seq, const char *qual, const char *chr, uint bp, strand_type strand) { this->id = new char[strlen(id) + 1]; this->qual = new char[strlen(qual) + 1]; this->loc_str = new char[strlen(chr) + 15]; strcpy(this->id, id); strcpy(this->qual, qual); this->loc.set(chr, bp, strand); sprintf(this->loc_str, "%s|%d|%c", chr, bp, strand == strand_plus ? '+' : '-'); // // Reverse complement sequences from the negative strand // if (strand == strand_plus) { this->seq = new char[strlen(seq) + 1]; strcpy(this->seq, seq); } else { this->seq = rev_comp(seq); } this->aln_type = pri_aln; this->pct_aln = 1.0; } Seq::Seq(const char *id, const char *seq, const char *qual, const char *chr, uint bp, strand_type strand, alnt aln_type, double pct_aln) { this->id = new char[strlen(id) + 1]; this->qual = new char[strlen(qual) + 1]; this->loc_str = new char[strlen(chr) + 15]; strcpy(this->id, id); strcpy(this->qual, qual); this->loc.set(chr, bp, strand); sprintf(this->loc_str, "%s|%d|%c", chr, bp, strand == strand_plus ? '+' : '-'); // // Reverse complement sequences from the negative strand // if (strand == strand_plus) { this->seq = new char[strlen(seq) + 1]; strcpy(this->seq, seq); } else { this->seq = rev_comp(seq); } this->aln_type = aln_type; this->pct_aln = pct_aln; } void swap(Seq& s1, Seq& s2) { char *ptr; alnt a; double p; ptr = s1.id; s1.id = s2.id; s2.id = ptr; ptr = s1.seq; s1.seq = s2.seq; s2.seq = ptr; ptr = s1.qual; s1.qual = s2.qual; s2.qual = ptr; ptr = s1.loc_str; s1.loc_str = s2.loc_str; s2.loc_str = ptr; a = s1.aln_type; s1.aln_type = s2.aln_type; s2.aln_type = a; p = s1.pct_aln; s1.pct_aln = s2.pct_aln; s2.pct_aln = p; swap(s1.loc, s2.loc); } Input::Input() { memset(this->line, '\0', max_len); } Input::Input(const char *path) { memset(this->line, '\0', max_len); this->path = string(path); // // Open the file for reading // this->fh.open(path, ifstream::in); if (this->fh.fail()) cerr << "Error opening input file '" << path << "'\n"; } Input::~Input() { // Close the file this->fh.close(); } int parse_tsv(const char *line, vector &parts) { const char *p, *q; string part; parts.clear(); p = line; do { for (q = p; *q != '\t' && *q != '\0'; q++); if (q - p == 0) part = ""; else part.assign(p, (q - p)); parts.push_back(part); p = q + 1; } while (*q != '\0'); //for (size_t i = 0; i < parts.size(); i++) // cerr << "Parts[" << i << "]: " << parts[i].c_str() << "\n"; //cerr << "\n"; return 0; } int parse_ssv(const char *line, vector &parts) { const char *p, *q; string part; parts.clear(); p = line; do { for (q = p; *q != ' ' && *q != '\0'; q++); if (q - p == 0) part = ""; else part.assign(p, (q - p)); parts.push_back(string(part)); p = q + 1; } while (*q != '\0'); return 0; } int read_line(ifstream &fh, char **line, int *size) { char buf[max_len]; int blen, llen; memset(*line, 0, *size); llen = 0; // // Make sure we read the entire line. // do { fh.clear(); fh.getline(buf, max_len); blen = strlen(buf); if (blen + llen <= (*size) - 1) { strcat(*line, buf); llen += blen; } else { *size *= 2; llen += blen; *line = (char *) realloc(*line, *size); strcat(*line, buf); } } while (fh.fail() && !fh.bad() && !fh.eof()); if (fh.eof() || fh.bad()) return 0; return 1; } int read_gzip_line(gzFile &fh, char **line, int *size) { char buf[max_len]; int blen, llen; bool eol; memset(*line, 0, *size); llen = 0; eol = false; // // Make sure we read the entire line. // do { if (gzgets(fh, buf, max_len) == NULL) break; blen = strlen(buf); if (blen > 0 && buf[blen - 1] == '\n') { eol = true; buf[blen - 1] = '\0'; } if (blen + llen <= (*size) - 1) { strcat(*line, buf); llen += blen; } else { *size *= 2; llen += blen; *line = (char *) realloc(*line, *size); strcat(*line, buf); } } while (!gzeof(fh) && !eol); if (gzeof(fh)) return 0; return 1; } bool is_comment(const char *line) { const char *p = line; while (*p != '\0') switch(*p) { case '#': return true; break; case ' ': case '\t': p++; break; default: return false; break; } return false; } stacks-1.44/src/input.h000644 000765 000024 00000005155 12775212561 015617 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __INPUT_H__ #define __INPUT_H__ #include #include #include #include #include #include #include #include using std::ifstream; using std::string; using std::vector; using std::cin; using std::cout; using std::cerr; using std::endl; #include "constants.h" #include "utils.h" #include "stacks.h" enum alnt {pri_aln, sec_aln, sup_aln}; class Seq { public: char *id; char *seq; char *qual; // // Information for an aligned sequence. // alnt aln_type; double pct_aln; char *loc_str; PhyLoc loc; Seq(); Seq(const Seq& other); Seq(const char *, const char *); Seq(const char *, const char *, const char *); Seq(const char *, const char *, const char *, const char *, uint, strand_type); Seq(const char *, const char *, const char *, const char *, uint, strand_type, alnt, double); ~Seq( void ) { if (id != NULL) delete[] id; if (seq != NULL) delete[] seq; if (qual != NULL) delete[] qual; if (loc_str != NULL) delete[] loc_str; } friend void swap(Seq&, Seq&); Seq& operator=(Seq&& other) {swap(*this, other); return *this;} Seq& operator=(const Seq& other) = delete; }; // // The base class for all of our Input classes, such as Tsv, Fastq, Fasta, etc. // class Input { public: string path; ifstream fh; char line[max_len]; Input(); Input(const char *path); virtual ~Input(); virtual Seq *next_seq() = 0; virtual int next_seq(Seq &) = 0; }; int parse_tsv(const char *, vector &); int parse_ssv(const char *, vector &); int read_line(ifstream &, char **, int *); int read_gzip_line(gzFile &, char **, int *); bool is_comment(const char *); #endif // __INPUT_H__ stacks-1.44/src/kmer_filter.cc000644 000765 000024 00000163060 12764047443 017124 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2011-2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // kmer_filter -- // #include "kmer_filter.h" // // Global variables to hold command-line options. // FileT in_file_type = FileT::unknown; FileT out_file_type = FileT::fastq; vector in_files; vector in_pair_files; string in_path; string out_path; string k_freq_path; bool filter_rare_k = false; bool filter_abundant_k = false; bool normalize = false; bool discards = false; bool write_k_freq = false; bool read_k_freq = false; bool kmer_distr = false; bool ill_barcode = false; uint truncate_seq = 0; int transition_lim = 3; int normalize_lim = 0; int kmer_len = 15; int max_k_freq = 20000; double max_k_pct = 0.80; double min_k_pct = 0.15; int min_lim = 0; int max_lim = 0; int num_threads = 1; int barcode_size = 0; int main (int argc, char* argv[]) { parse_command_line(argc, argv); if (min_lim == 0) min_lim = (int) round((double) kmer_len * 0.80); cerr << "Using a kmer size of " << kmer_len << "\n"; if (filter_rare_k) { cerr << "Filtering out reads by identifying rare kmers: On.\n" << " A kmer is considered rare when its coverage is at " << min_k_pct * 100 << "% or below the median kmer coverage for the read.\n" << " A read is dropped when it contains " << min_lim << " or more rare kmers in a row.\n"; } else cerr << "Filtering out reads by identifying rare kmers: Off.\n"; if (filter_abundant_k) { cerr << "Filtering out reads by identifying abundant kmers: On.\n" << " Kmer is considered abundant when it occurs " << max_k_freq << " or more times.\n"; if (max_lim == 0) cerr << " A read is dropped when it contains " << max_k_pct * 100 << "% or more abundant kmers.\n"; else cerr << " A read is dropped when it contains " << max_lim << " or more abundant kmers.\n"; } else cerr << "Filtering out reads by identifying abundant kmers: Off.\n"; if (normalize) cerr << "Normalizing read depth: On.\n" << " Read depth limit: " << normalize_lim << "x\n"; else cerr << "Normalizing read depth: Off.\n"; vector > files, pair_files; map > counters; SeqKmerHash kmers; vector kmers_keys; build_file_list(in_files, files); cerr << "Found " << files.size() << " input file(s).\n"; build_file_list(in_pair_files, pair_files); cerr << "Found " << pair_files.size() << " paired input file(s).\n"; if (filter_rare_k || filter_abundant_k || kmer_distr || write_k_freq) { cerr << "Generating kmer distribution...\n"; if (read_k_freq) read_kmer_freq(k_freq_path, kmers, kmers_keys); else populate_kmers(pair_files, files, kmers, kmers_keys); // double kmer_med, kmer_mad; // calc_kmer_median(kmers, kmer_med, kmer_mad); // cerr << "Median kmer frequency: " << kmer_med << "; median absolute deviation: " << kmer_mad << "\n"; if (kmer_distr) { generate_kmer_dist(kmers); if (write_k_freq == false) exit(0); } if (write_k_freq) { write_kmer_freq(k_freq_path, kmers); exit(0); } cerr << "Filtering reads by kmer frequency...\n"; for (uint i = 0; i < pair_files.size(); i += 2) { cerr << "Processing paired file " << i+1 << " of " << (pair_files.size() / 2) << " [" << pair_files[i].second << "]\n"; counters[pair_files[i].second]["total"] = 0; counters[pair_files[i].second]["retained"] = 0; counters[pair_files[i].second]["rare_k"] = 0; counters[pair_files[i].second]["abundant_k"] = 0; process_paired_reads(pair_files[i].first, pair_files[i].second, pair_files[i+1].first, pair_files[i+1].second, kmers, counters[pair_files[i].second]); cerr << " " << counters[pair_files[i].second]["total"] << " total reads; " << "-" << counters[pair_files[i].second]["rare_k"] << " rare k-mer reads; " << "-" << counters[pair_files[i].second]["abundant_k"] << " abundant k-mer reads; " << counters[pair_files[i].second]["retained"] << " retained reads.\n"; } for (uint i = 0; i < files.size(); i++) { cerr << "Processing file " << i+1 << " of " << files.size() << " [" << files[i].second << "]\n"; counters[files[i].second]["total"] = 0; counters[files[i].second]["retained"] = 0; counters[files[i].second]["rare_k"] = 0; counters[files[i].second]["abundant_k"] = 0; process_reads(files[i].first, files[i].second, kmers, counters[files[i].second]); cerr << " " << counters[files[i].second]["total"] << " total reads; " << "-" << counters[files[i].second]["rare_k"] << " rare k-mer reads; " << "-" << counters[files[i].second]["abundant_k"] << " abundant k-mer reads; " << counters[files[i].second]["retained"] << " retained reads.\n"; } free_kmer_hash(kmers, kmers_keys); print_results(counters); } if (normalize) { cerr << "Normalizing read depth...\n"; // // Add the remainder files from the previous step to the queue. // if (filter_rare_k || filter_abundant_k) { string file; int pos; for (uint i = 0; i < pair_files.size(); i += 2) { file = pair_files[i].second; pos = file.find_last_of("."); if (file.substr(pos - 2, 2) == ".1") pos -= 2; file = file.substr(0, pos) + ".rem.fil"; file += out_file_type == FileT::fastq ? ".fq" : ".fa"; cerr << "Adding remainder file generated in previous step to queue, '" << file << "\n"; files.push_back(make_pair(pair_files[i].first, file)); } } for (uint i = 0; i < pair_files.size(); i += 2) { cerr << "Processing paired files " << i+1 << " of " << (pair_files.size() / 2) << " [" << pair_files[i].second << " / " << pair_files[i+1].second << "]\n"; counters[pair_files[i].second]["total"] = 0; counters[pair_files[i].second]["retained"] = 0; counters[pair_files[i].second]["overep"] = 0; normalize_paired_reads(pair_files[i].first, pair_files[i].second, pair_files[i+1].first, pair_files[i+1].second, kmers, kmers_keys, counters[pair_files[i].second]); cerr << " " << counters[pair_files[i].second]["total"] << " total reads; " << "-" << counters[pair_files[i].second]["overep"] << " over-represented reads; " << counters[pair_files[i].second]["retained"] << " retained reads.\n"; } for (uint i = 0; i < files.size(); i++) { cerr << "Processing file " << i+1 << " of " << files.size() << " [" << files[i].second << "]\n"; counters[files[i].second]["total"] = 0; counters[files[i].second]["retained"] = 0; counters[files[i].second]["overep"] = 0; normalize_reads(files[i].first, files[i].second, kmers, kmers_keys, counters[files[i].second]); cerr << " " << counters[files[i].second]["total"] << " total reads; " << "-" << counters[files[i].second]["overep"] << " over-represented reads; " << counters[files[i].second]["retained"] << " retained reads.\n"; } free_kmer_hash(kmers, kmers_keys); } return 0; } int process_paired_reads(string in_path_1, string in_file_1, string in_path_2, string in_file_2, SeqKmerHash &kmers, map &counter) { Input *fh_1, *fh_2; ofstream *discard_fh_1, *discard_fh_2; int pos; string path; string path_1 = in_path_1 + in_file_1; string path_2 = in_path_2 + in_file_2; if (in_file_type == FileT::fastq) { fh_1 = new Fastq(path_1); fh_2 = new Fastq(path_2); } else if (in_file_type == FileT::fasta) { fh_1 = new Fasta(path_1); fh_2 = new Fasta(path_2); } else if (in_file_type == FileT::gzfasta) { fh_1 = new GzFasta(path_1 + ".gz"); fh_2 = new GzFasta(path_2 + ".gz"); } else if (in_file_type == FileT::gzfastq) { fh_1 = new GzFastq(path_1 + ".gz"); fh_2 = new GzFastq(path_2 + ".gz"); } else if (in_file_type == FileT::bustard) { fh_1 = new Bustard(path_1); fh_2 = new Bustard(path_2); } // // Open the output files. // pos = in_file_1.find_last_of("."); path = out_path + in_file_1.substr(0, pos) + ".fil" + in_file_1.substr(pos); ofstream *ofh_1 = new ofstream(path.c_str(), ifstream::out); if (ofh_1->fail()) { cerr << "Error opening filtered output file '" << path << "'\n"; exit(1); } pos = in_file_2.find_last_of("."); path = out_path + in_file_2.substr(0, pos) + ".fil" + in_file_2.substr(pos); ofstream *ofh_2 = new ofstream(path.c_str(), ifstream::out); if (ofh_2->fail()) { cerr << "Error opening filtered paired output file '" << path << "'\n"; exit(1); } pos = in_file_2.find_last_of("."); // // Pull the ".2" suffix off the paired file name to make the remainder file name. // if (in_file_2.substr(pos - 2, 2) == ".2") pos -= 2; path = out_path + in_file_2.substr(0, pos) + ".rem.fil"; path += out_file_type == FileT::fastq ? ".fq" : ".fa"; ofstream *rem_fh = new ofstream(path.c_str(), ifstream::out); if (rem_fh->fail()) { cerr << "Error opening filtered remainder output file '" << path << "'\n"; exit(1); } // // Open a file for recording discarded reads // if (discards) { pos = in_file_1.find_last_of("."); path = out_path + in_file_1.substr(0, pos) + ".discards" + in_file_1.substr(pos); discard_fh_1 = new ofstream(path.c_str(), ifstream::out); if (discard_fh_1->fail()) { cerr << "Error opening discard output file '" << path << "'\n"; exit(1); } pos = in_file_2.find_last_of("."); path = out_path + in_file_2.substr(0, pos) + ".discards" + in_file_2.substr(pos); discard_fh_2 = new ofstream(path.c_str(), ifstream::out); if (discard_fh_2->fail()) { cerr << "Error opening discard output file '" << path << "'\n"; exit(1); } } // // Read in the first record, initializing the Seq object s. Then // initialize the Read object r, then loop, using the same objects. // Seq *s_1 = fh_1->next_seq(); Seq *s_2 = fh_2->next_seq(); if (s_1 == NULL || s_2 == NULL) { cerr << "Unable to allocate Seq object.\n"; exit(1); } int rare_k, abundant_k, num_kmers, max_kmer_lim; bool retain_1, retain_2; char *kmer = new char[kmer_len + 1]; long i = 1; do { if (i % 10000 == 0) cerr << " Processing short read pair " << i << " \r"; counter["total"] += 2; stringstream msg_1, msg_2; retain_1 = true; retain_2 = true; num_kmers = strlen(s_1->seq) - kmer_len + 1; max_kmer_lim = max_lim == 0 ? (int) round((double) num_kmers * max_k_pct) : max_lim; // // Drop the first sequence if it has too many rare or abundant kmers. // kmer_lookup(kmers, s_1->seq, kmer, num_kmers, rare_k, abundant_k); if (filter_rare_k && rare_k > 0) { counter["rare_k"]++; retain_1 = false; msg_1 << "rare_k_" << rare_k; } if (retain_1 && filter_abundant_k && abundant_k > max_kmer_lim) { counter["abundant_k"]++; retain_1 = false; msg_1 << "abundant_k_" << abundant_k; } rare_k = 0; abundant_k = 0; num_kmers = strlen(s_2->seq) - kmer_len + 1; max_kmer_lim = max_lim == 0 ? (int) round((double) num_kmers * max_k_pct) : max_lim; // // Drop the second sequence if it has too many rare or abundant kmers. // kmer_lookup(kmers, s_2->seq, kmer, num_kmers, rare_k, abundant_k); if (filter_rare_k && rare_k > 0) { counter["rare_k"]++; retain_2 = false; msg_2 << "rare_k_" << rare_k; } if (retain_2 && filter_abundant_k && abundant_k > max_kmer_lim) { counter["abundant_k"]++; retain_2 = false; msg_2 << "abundant_k_" << abundant_k; } if (retain_1 && retain_2) { counter["retained"] += 2; out_file_type == FileT::fastq ? write_fastq(ofh_1, s_1) : write_fasta(ofh_1, s_1); out_file_type == FileT::fastq ? write_fastq(ofh_2, s_2) : write_fasta(ofh_2, s_2); } if (retain_1 && !retain_2) { counter["retained"]++; out_file_type == FileT::fastq ? write_fastq(rem_fh, s_1) : write_fasta(rem_fh, s_1); } if (!retain_1 && retain_2) { counter["retained"]++; out_file_type == FileT::fastq ? write_fastq(rem_fh, s_2) : write_fasta(rem_fh, s_2); } if (discards && !retain_1) out_file_type == FileT::fastq ? write_fastq(discard_fh_1, s_1, msg_1.str()) : write_fasta(discard_fh_1, s_1, msg_1.str()); if (discards && !retain_2) out_file_type == FileT::fastq ? write_fastq(discard_fh_2, s_2, msg_2.str()) : write_fasta(discard_fh_2, s_2, msg_2.str()); delete s_1; delete s_2; i++; } while ((s_1 = fh_1->next_seq()) != NULL && (s_2 = fh_2->next_seq()) != NULL); delete [] kmer; if (discards) { delete discard_fh_1; delete discard_fh_2; } // // Close the file and delete the Input object. // delete fh_1; delete fh_2; delete ofh_1; delete ofh_2; delete rem_fh; return 0; } int process_reads(string in_path, string in_file, SeqKmerHash &kmers, map &counter) { Input *fh; ofstream *discard_fh; int pos; string path = in_path + in_file; if (in_file_type == FileT::fastq) fh = new Fastq(path); else if (in_file_type == FileT::fasta) fh = new Fasta(path); else if (in_file_type == FileT::gzfastq) fh = new GzFastq(path + ".gz"); else if (in_file_type == FileT::gzfasta) fh = new GzFasta(path + ".gz"); else if (in_file_type == FileT::bustard) fh = new Bustard(path); // // Open the output file. // pos = in_file.find_last_of("."); path = out_path + in_file.substr(0, pos) + ".fil" + in_file.substr(pos); ofstream *out_fh = new ofstream(path.c_str(), ifstream::out); if (out_fh->fail()) { cerr << "Error opening output file '" << path << "'\n"; exit(1); } // // Open a file for recording discarded reads // if (discards) { pos = in_file.find_last_of("."); path = out_path + in_file.substr(0, pos) + ".discards" + in_file.substr(pos); discard_fh = new ofstream(path.c_str(), ifstream::out); if (discard_fh->fail()) { cerr << "Error opening discard output file '" << path << "'\n"; exit(1); } } // // Read in the first record, initializing the Seq object s. Then // initialize the Read object r, then loop, using the same objects. // Seq *s = fh->next_seq(); if (s == NULL) { cerr << "Unable to allocate Seq object.\n"; exit(1); } int rare_k, abundant_k, num_kmers, max_kmer_lim; bool retain; char *kmer = new char[kmer_len + 1]; long i = 1; do { if (i % 10000 == 0) cerr << " Processing short read " << i << " \r"; counter["total"]++; stringstream msg; // // Drop this sequence if it has too many rare or abundant kmers. // retain = true; num_kmers = strlen(s->seq) - kmer_len + 1; max_kmer_lim = max_lim == 0 ? (int) round((double) num_kmers * max_k_pct) : max_lim; kmer_lookup(kmers, s->seq, kmer, num_kmers, rare_k, abundant_k); if (filter_rare_k && rare_k > 0) { counter["rare_k"]++; retain = false; msg << "rare_k_" << rare_k; } if (retain && filter_abundant_k && abundant_k > max_kmer_lim) { counter["abundant_k"]++; retain = false; msg << "abundant_k_" << abundant_k; } if (retain) { counter["retained"]++; out_file_type == FileT::fastq ? write_fastq(out_fh, s) : write_fasta(out_fh, s); } if (discards && !retain) out_file_type == FileT::fastq ? write_fastq(discard_fh, s, msg.str()) : write_fasta(discard_fh, s, msg.str()); delete s; i++; } while ((s = fh->next_seq()) != NULL); delete [] kmer; if (discards) delete discard_fh; // // Close the file and delete the Input object. // delete fh; delete out_fh; return 0; } int normalize_paired_reads(string in_path_1, string in_file_1, string in_path_2, string in_file_2, SeqKmerHash &kmers, vector &kmer_keys, map &counter) { Input *fh_1, *fh_2; ofstream *discard_fh_1, *discard_fh_2; ofstream *ofh_1, *ofh_2, *rem_fh; string path_1, path_2; int pos; if (filter_abundant_k || filter_rare_k) { // // If we already filtered the data, open the files we created in the output // directory to normalize. // pos = in_file_1.find_last_of("."); path_1 = out_path + in_file_1.substr(0, pos) + ".fil" + in_file_1.substr(pos); pos = in_file_2.find_last_of("."); path_2 = out_path + in_file_2.substr(0, pos) + ".fil" + in_file_2.substr(pos); if (in_file_type == FileT::fastq) { fh_1 = new Fastq(path_1); fh_2 = new Fastq(path_2); } else if (in_file_type == FileT::gzfastq) { fh_1 = new Fastq(path_1); fh_2 = new Fastq(path_2); } else if (in_file_type == FileT::fasta) { fh_1 = new Fasta(path_1); fh_2 = new Fasta(path_2); } else if (in_file_type == FileT::gzfasta) { fh_1 = new Fasta(path_1); fh_2 = new Fasta(path_2); } } else { // // Otherwise, open unmodified files. // path_1 = in_path_1 + in_file_1; path_2 = in_path_2 + in_file_2; if (in_file_type == FileT::fastq) { fh_1 = new Fastq(path_1); fh_2 = new Fastq(path_2); } else if (in_file_type == FileT::gzfastq) { fh_1 = new GzFastq(path_1 + ".gz"); fh_2 = new GzFastq(path_2 + ".gz"); } else if (in_file_type == FileT::fasta) { fh_1 = new Fasta(path_1); fh_2 = new Fasta(path_2); } else if (in_file_type == FileT::gzfasta) { fh_1 = new GzFasta(path_1 + ".gz"); fh_2 = new GzFasta(path_2 + ".gz"); } else if (in_file_type == FileT::bustard) { fh_1 = new Bustard(path_1); fh_2 = new Bustard(path_2); } } // // Open the output files. // if (filter_abundant_k || filter_rare_k) { pos = in_file_1.find_last_of("."); path_1 = out_path + in_file_1.substr(0, pos) + ".fil.norm" + in_file_1.substr(pos); ofh_1 = new ofstream(path_1.c_str(), ifstream::out); if (ofh_1->fail()) { cerr << "Error opening normalized output file '" << path_1 << "'\n"; exit(1); } pos = in_file_2.find_last_of("."); path_2 = out_path + in_file_2.substr(0, pos) + ".fil.norm" + in_file_2.substr(pos); ofh_2 = new ofstream(path_2.c_str(), ifstream::out); if (ofh_2->fail()) { cerr << "Error opening normalized paired output file '" << path_2 << "'\n"; exit(1); } if (in_file_2.substr(pos - 2, 2) == ".2") pos -= 2; path_2 = out_path + in_file_2.substr(0, pos) + ".fil.norm.rem"; path_2 += out_file_type == FileT::fastq ? ".fq" : ".fa"; rem_fh = new ofstream(path_2.c_str(), ifstream::out); if (rem_fh->fail()) { cerr << "Error opening normalized remainder output file '" << path_2 << "'\n"; exit(1); } } else { pos = in_file_1.find_last_of("."); path_1 = out_path + in_file_1.substr(0, pos) + ".norm" + in_file_1.substr(pos); ofh_1 = new ofstream(path_1.c_str(), ifstream::out); if (ofh_1->fail()) { cerr << "Error opening normalized output file '" << path_1 << "'\n"; exit(1); } pos = in_file_2.find_last_of("."); path_2 = out_path + in_file_2.substr(0, pos) + ".norm" + in_file_2.substr(pos); ofh_2 = new ofstream(path_2.c_str(), ifstream::out); if (ofh_2->fail()) { cerr << "Error opening normalized paired output file '" << path_2 << "'\n"; exit(1); } if (in_file_2.substr(pos - 2, 2) == ".2") pos -= 2; path_2 = out_path + in_file_2.substr(0, pos) + ".norm.rem"; path_2 += out_file_type == FileT::fastq ? ".fq" : ".fa"; rem_fh = new ofstream(path_2.c_str(), ifstream::out); if (rem_fh->fail()) { cerr << "Error opening normalized remainder output file '" << path_2 << "'\n"; exit(1); } } // // Open a file for recording discarded reads // if (discards) { pos = in_file_1.find_last_of("."); if (filter_abundant_k || filter_rare_k) path_1 = out_path + in_file_1.substr(0, pos) + ".fil.discards" + in_file_1.substr(pos); else path_1 = out_path + in_file_1.substr(0, pos) + ".discards" + in_file_1.substr(pos); discard_fh_1 = new ofstream(path_1.c_str(), ifstream::out); if (discard_fh_1->fail()) { cerr << "Error opening discard output file '" << path_1 << "'\n"; exit(1); } pos = in_file_2.find_last_of("."); if (filter_abundant_k || filter_rare_k) path_2 = out_path + in_file_2.substr(0, pos) + ".fil.discards" + in_file_2.substr(pos); else path_2 = out_path + in_file_2.substr(0, pos) + ".discards" + in_file_2.substr(pos); discard_fh_2 = new ofstream(path_2.c_str(), ifstream::out); if (discard_fh_2->fail()) { cerr << "Error opening discard output file '" << path_1 << "'\n"; exit(1); } } // // Read in the first record, initializing the Seq object s. Then // initialize the Read object r, then loop, using the same objects. // Seq *s_1 = fh_1->next_seq(); Seq *s_2 = fh_2->next_seq(); if (s_1 == NULL || s_2 == NULL) { cerr << "Unable to allocate Seq object.\n"; exit(1); } int num_kmers; bool retain_1, retain_2; char *kmer = new char[kmer_len + 1]; long i = 1; do { if (i % 10000 == 0) cerr << " Processing short read pair " << i << " \r"; counter["total"] += 2; retain_1 = true; retain_2 = true; num_kmers = strlen(s_1->seq) - kmer_len + 1; // // Drop the first sequence if it has too many rare or abundant kmers. // retain_1 = normalize_kmer_lookup(kmers, s_1->seq, kmer, num_kmers, kmer_keys); num_kmers = strlen(s_2->seq) - kmer_len + 1; // // Drop the second sequence if it has too many rare or abundant kmers. // retain_2 = normalize_kmer_lookup(kmers, s_2->seq, kmer, num_kmers, kmer_keys); if (retain_1 && retain_2) { counter["retained"] += 2; out_file_type == FileT::fastq ? write_fastq(ofh_1, s_1) : write_fasta(ofh_1, s_1); out_file_type == FileT::fastq ? write_fastq(ofh_2, s_2) : write_fasta(ofh_2, s_2); } else { counter["overep"] +=2; } if (retain_1 && !retain_2) { counter["retained"]++; counter["overep"]++; out_file_type == FileT::fastq ? write_fastq(rem_fh, s_1) : write_fasta(rem_fh, s_1); } if (!retain_1 && retain_2) { counter["retained"]++; counter["overep"]++; out_file_type == FileT::fastq ? write_fastq(rem_fh, s_2) : write_fasta(rem_fh, s_2); } if (discards && !retain_1) out_file_type == FileT::fastq ? write_fastq(discard_fh_1, s_1) : write_fasta(discard_fh_1, s_1); if (discards && !retain_2) out_file_type == FileT::fastq ? write_fastq(discard_fh_2, s_2) : write_fasta(discard_fh_2, s_2); delete s_1; delete s_2; i++; } while ((s_1 = fh_1->next_seq()) != NULL && (s_2 = fh_2->next_seq()) != NULL); delete [] kmer; if (discards) { delete discard_fh_1; delete discard_fh_2; } // // Close the file and delete the Input object. // delete fh_1; delete fh_2; delete ofh_1; delete ofh_2; delete rem_fh; return 0; } int normalize_reads(string in_path, string in_file, SeqKmerHash &kmers, vector &kmer_keys, map &counter) { Input *fh; ofstream *discard_fh; string path; int pos = in_file.find_last_of("."); if (filter_abundant_k || filter_rare_k) { if (in_file.substr(pos - 4, 4) == ".fil") path = out_path + in_file; else path = out_path + in_file.substr(0, pos) + ".fil" + in_file.substr(pos); if (in_file_type == FileT::fastq) fh = new Fastq(path); else if (in_file_type == FileT::gzfastq) fh = new Fastq(path); else if (in_file_type == FileT::fasta) fh = new Fasta(path); else if (in_file_type == FileT::gzfasta) fh = new Fasta(path); else if (in_file_type == FileT::bustard) fh = new Bustard(path); } else { path = in_path + in_file; if (in_file_type == FileT::fastq) fh = new Fastq(path); else if (in_file_type == FileT::gzfastq) fh = new GzFastq(path + ".gz"); else if (in_file_type == FileT::fasta) fh = new Fasta(path); else if (in_file_type == FileT::gzfasta) fh = new GzFasta(path + ".gz"); else if (in_file_type == FileT::bustard) fh = new Bustard(path); } // // Open the output file. // // if (filter_abundant_k || filter_rare_k) { // path = out_path + in_file.substr(0, pos) + ".norm" + in_file.substr(pos); // } else { // path = out_path + in_file.substr(0, pos) + ".norm" + in_file.substr(pos); // } path = out_path + in_file.substr(0, pos) + ".norm" + in_file.substr(pos); ofstream *out_fh = new ofstream(path.c_str(), ifstream::out); if (out_fh->fail()) { cerr << "Error opening normalized output file '" << path << "'\n"; exit(1); } // // Open a file for recording discarded reads // if (discards) { if (filter_abundant_k || filter_rare_k) path = out_path + in_file.substr(0, pos) + ".fil.discards" + in_file.substr(pos); else path = out_path + in_file.substr(0, pos) + ".discards" + in_file.substr(pos); discard_fh = new ofstream(path.c_str(), ifstream::out); if (discard_fh->fail()) { cerr << "Error opening discard output file '" << path << "'\n"; exit(1); } } // // Read in the first record, initializing the Seq object s. Then // initialize the Read object r, then loop, using the same objects. // Seq *s = fh->next_seq(); if (s == NULL) { cerr << "Unable to allocate Seq object.\n"; exit(1); } int num_kmers; bool retain; char *kmer = new char[kmer_len + 1]; long i = 1; do { if (i % 10000 == 0) cerr << " Processing short read " << i << " \r"; counter["total"]++; // // Drop this sequence if it has too many rare or abundant kmers. // retain = true; num_kmers = strlen(s->seq) - kmer_len + 1; retain = normalize_kmer_lookup(kmers, s->seq, kmer, num_kmers, kmer_keys); if (retain) { counter["retained"]++; out_file_type == FileT::fastq ? write_fastq(out_fh, s) : write_fasta(out_fh, s); } else { counter["overep"]++; } if (discards && !retain) out_file_type == FileT::fastq ? write_fastq(discard_fh, s) : write_fasta(discard_fh, s); delete s; i++; } while ((s = fh->next_seq()) != NULL); delete [] kmer; if (discards) delete discard_fh; // // Close the file and delete the Input object. // delete fh; delete out_fh; return 0; } int populate_kmers(vector > &pair_files, vector > &files, SeqKmerHash &kmers, vector &kmers_keys) { // // Break each read down into k-mers and create a hash map of those k-mers // recording in which sequences they occur. // uint j = 1; uint cnt = files.size() + pair_files.size(); for (uint i = 0; i < files.size(); i++) { cerr << "Generating kmers from file " << j << " of " << cnt << " [" << files[i].second << "]\n"; process_file_kmers(files[i].first + files[i].second, kmers, kmers_keys); j++; } for (uint i = 0; i < pair_files.size(); i++) { cerr << "Generating kmers from file " << j << " of " << cnt << " [" << pair_files[i].second << "]\n"; process_file_kmers(pair_files[i].first + pair_files[i].second, kmers, kmers_keys); j++; } cerr << kmers.size() << " unique k-mers recorded.\n"; return 0; } int read_kmer_freq(string in_path, SeqKmerHash &kmer_map, vector &kmer_map_keys) { cerr << "Reading kmer frequencies from '" << in_path.c_str() << "'...\n"; ifstream fh(in_path.c_str(), ifstream::in); if (fh.fail()) { cerr << "Error opening rare kmer frequency input file '" << in_path << "'\n"; exit(1); } char *hash_key; bool exists; int len, cnt; char kmer[id_len]; char line[max_len]; vector parts; long i = 1; while (fh.good()) { if (i % 10000 == 0) cerr << " Processing kmer " << i << " \r"; fh.getline(line, max_len); len = strlen(line); if (len == 0) continue; // // Check that there is no carraige return in the buffer. // if (line[len - 1] == '\r') line[len - 1] = '\0'; // // Ignore comments // if (line[0] == '#') continue; // // Parse the kmer and the number of times it occurred // // parse_tsv(line, parts); if (parts.size() != 2) { cerr << "kmer frequencies are not formated correctly: expecting two, tab separated columns, found " << parts.size() << ".\n"; exit(0); } strcpy(kmer, parts[1].c_str()); cnt = is_integer(kmer); if (cnt < 0) { cerr << "Non integer found in second column.\n"; exit(0); } strcpy(kmer, parts[0].c_str()); exists = kmer_map.count(kmer) == 0 ? false : true; if (exists) { cerr << "Warning: kmer '" << kmer << "' already exists in the kmer hash map.\n"; hash_key = kmer; kmer_map[hash_key] += cnt; } else { hash_key = new char [strlen(kmer) + 1]; strcpy(hash_key, kmer); kmer_map_keys.push_back(hash_key); kmer_map[hash_key] = cnt; } i++; } fh.close(); cerr << kmer_map.size() << " unique k-mers read.\n"; kmer_len = strlen(kmer_map.begin()->first); cerr << "Setting kmer length to " << kmer_len << "bp.\n"; return 0; } int write_kmer_freq(string path, SeqKmerHash &kmer_map) { cerr << "Writing kmer frequencies to '" << path.c_str() << "'..."; ofstream out_fh(path.c_str(), ifstream::out); if (out_fh.fail()) { cerr << "Error opening rare kmer output file '" << path << "'\n"; exit(1); } SeqKmerHash::iterator i; out_fh << "# Kmer\tCount\n"; for (i = kmer_map.begin(); i != kmer_map.end(); i++) { out_fh << i->first << "\t" << i->second << "\n"; } out_fh.close(); cerr << "done.\n"; return 0; } int process_file_kmers(string path, SeqKmerHash &kmer_map, vector &kmer_map_keys) { vector kmers; char *hash_key; bool exists; int j; Input *fh = NULL; if (in_file_type == FileT::fastq) fh = new Fastq(path); else if (in_file_type == FileT::gzfastq) fh = new GzFastq(path + ".gz"); else if (in_file_type == FileT::fasta) fh = new Fasta(path); else if (in_file_type == FileT::gzfasta) fh = new GzFasta(path + ".gz"); else if (in_file_type == FileT::bustard) fh = new Bustard(path.c_str()); // // Read in the first record, initializing the Seq object s. // Seq *s = fh->next_seq(); if (s == NULL) { cerr << "Unable to allocate Seq object.\n"; exit(1); } int num_kmers; char *kmer = new char [kmer_len + 1]; long i = 1; do { if (i % 10000 == 0) cerr << " Processing short read " << i << " \r"; num_kmers = strlen(s->seq) - kmer_len + 1; // // Generate and hash the kmers for this raw read // kmer[kmer_len] = '\0'; for (j = 0; j < num_kmers; j++) { strncpy(kmer, s->seq + j, kmer_len); exists = kmer_map.count(kmer) == 0 ? false : true; if (exists) { hash_key = kmer; } else { hash_key = new char [kmer_len + 1]; strcpy(hash_key, kmer); kmer_map_keys.push_back(hash_key); } kmer_map[hash_key]++; } delete s; i++; } while ((s = fh->next_seq()) != NULL); delete [] kmer; // // Close the file and delete the Input object. // delete fh; return 0; } int generate_kmer_dist(SeqKmerHash &kmer_map) { SeqKmerHash::iterator i; map bins; cerr << "Generating kmer distribution...\n"; for (i = kmer_map.begin(); i != kmer_map.end(); i++) bins[i->second]++; map::iterator j; vector > sorted_kmers; for (j = bins.begin(); j != bins.end(); j++) sorted_kmers.push_back(make_pair(j->first, j->second)); cout << "KmerFrequency\tCount\n"; for (unsigned long k = 0; k < sorted_kmers.size(); k++) cout << sorted_kmers[k].first << "\t" << sorted_kmers[k].second << "\n"; return 0; } int calc_kmer_median(SeqKmerHash &kmers, double &kmer_med, double &kmer_mad) { kmer_med = 0.0; kmer_mad = 0.0; int num_kmers = kmers.size(); vector freqs, residuals; freqs.reserve(num_kmers); SeqKmerHash::iterator i; for (i = kmers.begin(); i != kmers.end(); i++) freqs.push_back(i->second); sort(freqs.begin(), freqs.end()); kmer_med = num_kmers % 2 == 0 ? (double) (freqs[num_kmers / 2 - 1] + freqs[num_kmers / 2]) / 2.0 : (double) freqs[num_kmers / 2 - 1]; // // Calculate the median absolute deviation. // residuals.reserve(num_kmers); for (int j = 0; j < num_kmers; j++) residuals.push_back(abs(freqs[j] - (int) kmer_med)); sort(residuals.begin(), residuals.end()); kmer_mad = num_kmers % 2 == 0 ? (double) (residuals[num_kmers / 2 - 1] + residuals[num_kmers / 2]) / 2.0 : (double) residuals[num_kmers / 2 - 1]; return 0; } int kmer_map_cmp(pair a, pair b) { return (a.second < b.second); } inline bool normalize_kmer_lookup(SeqKmerHash &kmer_map, char *read, char *kmer, int num_kmers, vector &kmer_keys) { kmer[kmer_len] = '\0'; int cnt = 0; bool retain = true; // // Generate kmers from this read, increment kmer frequency in dataset. // vector sorted_cnts; sorted_cnts.reserve(num_kmers); // cout << "# " << read << "\n"; for (int j = 0; j < num_kmers; j++) { strncpy(kmer, read + j, kmer_len); cnt = kmer_map.count(kmer) > 0 ? kmer_map[kmer] : 0; sorted_cnts.push_back(cnt); // cout << kmer << "\t" << j << "\t" << cnt << "\n"; } // // Calculate the median kmer frequency along the read. // sort(sorted_cnts.begin(), sorted_cnts.end()); double median = num_kmers % 2 == 0 ? (double) (sorted_cnts[num_kmers / 2 - 1] + sorted_cnts[num_kmers / 2]) / 2.0 : (double) sorted_cnts[num_kmers / 2 - 1]; // cout << "# median: " << median << "\n"; if (median > normalize_lim) retain = false; // // Generate and hash the kmers for this raw read // bool exists; char *hash_key; kmer[kmer_len] = '\0'; for (int j = 0; j < num_kmers; j++) { strncpy(kmer, read + j, kmer_len); exists = kmer_map.count(kmer) == 0 ? false : true; if (exists) { hash_key = kmer; } else { hash_key = new char [kmer_len + 1]; strcpy(hash_key, kmer); kmer_keys.push_back(hash_key); } kmer_map[hash_key]++; } return retain; } inline int kmer_lookup(SeqKmerHash &kmer_map, char *read, char *kmer, int num_kmers, int &rare_k, int &abundant_k) { // // Generate kmers from this read, lookup kmer frequency in dataset. // rare_k = 0; abundant_k = 0; kmer[kmer_len] = '\0'; int cnt = 0; vector cnts, sorted_cnts; cnts.reserve(num_kmers); sorted_cnts.reserve(num_kmers); // cout << "# " << read << "\n"; for (int j = 0; j < num_kmers; j++) { strncpy(kmer, read + j, kmer_len); cnt = kmer_map[kmer]; cnts.push_back(cnt); sorted_cnts.push_back(cnt); // cout << kmer << "\t" << j << "\t" << cnt << "\n"; if (cnt >= max_k_freq) abundant_k++; } // // Calculate the median kmer frequency along the read. // sort(sorted_cnts.begin(), sorted_cnts.end()); double median = num_kmers % 2 == 0 ? (double) (sorted_cnts[num_kmers / 2 - 1] + sorted_cnts[num_kmers / 2]) / 2.0 : (double) sorted_cnts[num_kmers / 2 - 1]; // cout << "# median: " << median << "\n"; double bound = round(median * min_k_pct); // cout << "# kmer cov bound: " << bound << "\n"; // // Look for runs of rare kmers. // // We will slide a window across the read, f represents the front of the window, b // represents the back. Each time a kmer is below the bound we will increment run_cnt, // which represents the number of kmers in the window below the bound. If 2/3 of the // kmers in the window go below the bound, assume a sequencing error has occurred. // int run_cnt = 0; int b = 0; for (int f = 0; f < num_kmers; f++) { if (f >= kmer_len) { b++; if (cnts[b] <= bound) run_cnt--; } if (cnts[f] <= bound) { run_cnt++; if (run_cnt >= min_lim) { rare_k++; // cout << "# Rejecting read, position: " << f << "; run_cnt: " << run_cnt << "\n"; return 0; } } // cout << "# b: " << b << "; f: " << f << "; run_cnt: " << run_cnt << "; counts[front]: " << cnts[f] << "; bound: " << bound << "\n"; } // cout << "\n\n"; return 0; } // inline int // kmer_lookup(SeqKmerHash &kmer_map, // char *read, char *kmer, // int num_kmers, // int &rare_k, int &abundant_k, bool &complex) // { // // // // Generate kmers from this read, lookup kmer frequency in dataset. // // // rare_k = 0; // abundant_k = 0; // complex = false; // kmer[kmer_len] = '\0'; // int cnt = 0; // int rare_k_lim = (int) round((double) kmer_len * (1.0/2.0)); // vector cnts; // cnts.reserve(num_kmers); // // cout << "# " << read << "\n"; // for (int j = 0; j < num_kmers; j++) { // strncpy(kmer, read + j, kmer_len); // cnt = kmer_map[kmer]; // if (cnt >= 100000) // cnts.push_back(100000); // else if (cnt >= 10000) // cnts.push_back(10000); // else if (cnt >= 1000) // cnts.push_back(1000); // else if (cnt >= 100) // cnts.push_back(100); // else if (cnt >= 10) // cnts.push_back(10); // else // cnts.push_back(1); // // cout << kmer << "\t" << j << "\t" << cnt << "\n"; // if (cnt >= max_k_freq) abundant_k++; // } // // // // // // Detect the number of kmer coverage transitions. // // // // // int t = 0; // // int cov = cnts[0]; // // cout << "\nDetermining transitions:\n" << kmer << "\t" << "0" << "\t" << cnts[0] << "\n"; // // for (int j = 1; j < num_kmers; j++) // // if (cnts[j] != cov) { // // cov = cnts[j]; // // t++; // // cout << kmer << "\t" << j << "\t" << cnts[j] << ": Transition." << "\n"; // // } else { // // cout << kmer << "\t" << j << "\t" << cnts[j] << "\n"; // // } // // cerr << t << " total cnts.\n"; // // // // Look for runs of kmers at various orders of magnitude. // // // // We will slide a window across the read, f represents the front of the window, b // // represents the back. Each time a kmer is below the bound we will increment run_cnt, // // which represents the number of kmers in the window below the bound. If 2/3 of the // // kmers in the window go below the bound, assume a sequencing error has occurred. // // Run counters: // // 1 10 100 1k 10k 100k // // runs[0] runs[1] runs[2] runs[3] runs[4] runs[5] // int runs[6] = {0}; // int prev_cnt, run_cnt, tot_trans; // int f = 0; // while (f < num_kmers) { // tot_trans = 0; // run_cnt = 1; // prev_cnt = cnts[f]; // f++; // while (f < num_kmers && cnts[f] == prev_cnt) { // // cout << "# window front: " << f << "; run_cnt: " << run_cnt << "; prev_cnt: " << prev_cnt << "\n"; // f++; // run_cnt++; // } // if (run_cnt >= rare_k_lim) { // // cout << "# found transition run, position: " << f-1 << "; run_cnt: " << run_cnt << "\n"; // switch(prev_cnt) { // case 1: // runs[0]++; // break; // case 10: // runs[1]++; // break; // case 100: // runs[2]++; // break; // case 1000: // runs[3]++; // break; // case 10000: // runs[4]++; // break; // case 100000: // runs[5]++; // break; // } // } // for (int j = 0; j < 6; j++) // if (runs[j] > 0) tot_trans++; // // cout << "# Total transitions: " << tot_trans << "\n"; // if (tot_trans >= transition_lim) { // // cout << "# Rejecting read.\n"; // rare_k++; // return 0; // } // } // // cout << "\n\n"; // return 0; // } int free_kmer_hash(SeqKmerHash &kmer_map, vector &kmer_map_keys) { for (uint i = 0; i < kmer_map_keys.size(); i++) { delete [] kmer_map_keys[i]; } kmer_map_keys.clear(); kmer_map.clear(); return 0; } int print_results(map > &counters) { map >::iterator it; string log_path = out_path + "kmer_filter.log"; ofstream log(log_path.c_str()); if (log.fail()) { cerr << "Unable to open log file '" << log_path << "'\n"; return 0; } cerr << "Outputing details to log: '" << log_path << "'\n\n"; log << "File\t" << "Retained Reads\t" << "Rare K\t" << "Abundant K\t" << "Total\n"; for (it = counters.begin(); it != counters.end(); it++) { log << it->first << "\t" << it->second["retained"] << "\t" << it->second["rare_k"] << "\t" << it->second["abundant_k"] << "\t" << it->second["total"] << "\n"; } map c; c["total"] = 0; // // Total up the individual counters // for (it = counters.begin(); it != counters.end(); it++) { c["total"] += it->second["total"]; c["retained"] += it->second["retained"]; c["rare_k"] += it->second["rare_k"]; c["abundant_k"] += it->second["abundant_k"]; } cerr << c["total"] << " total sequences;\n" << " " << c["rare_k"] << " rare k-mer reads;\n" << " " << c["abundant_k"] << " abundant k-mer reads;\n" << c["retained"] << " retained reads.\n"; log << "Total Sequences\t" << c["total"] << "\n" << "Retained Reads\t" << c["retained"] << "\n"; log.close(); return 0; } int build_file_list(vector &in_files, vector > &files) { string file, suffix; int pos; // // Scan a directory for a list of files. // if (in_path.length() > 0) { struct dirent *direntry; DIR *dir = opendir(in_path.c_str()); if (dir == NULL) { cerr << "Unable to open directory '" << in_path << "' for reading.\n"; exit(1); } while ((direntry = readdir(dir)) != NULL) { file = direntry->d_name; if (file.substr(0, 1) == ".") continue; // // If the file is gzip'ed, remove the '.gz' suffix. // pos = file.find_last_of("."); if ((in_file_type == FileT::gzfastq || in_file_type == FileT::gzfasta) && file.substr(pos) == ".gz") { file = file.substr(0, pos); pos = file.find_last_of("."); } // // Check that the remaining file name has the right suffix. // suffix = file.substr(pos + 1); if (in_file_type == FileT::fastq && (suffix.substr(0, 2) == "fq" || suffix.substr(0, 5) == "fastq")) files.push_back(make_pair(in_path, file)); else if (in_file_type == FileT::fasta && (suffix.substr(0, 2) == "fa" || suffix.substr(0, 5) == "fasta")) files.push_back(make_pair(in_path, file)); } if (files.size() == 0) cerr << "Unable to locate any input files to process within '" << in_path << "'\n"; } else { string path; for (uint i = 0; i < in_files.size(); i++) { // // Files specified directly: // Break off file path and store path and file name. // Check if this is a gzip'ed file and if so, remove 'gz' suffix. // file = in_files[i]; pos = file.find_last_of("."); if ((in_file_type == FileT::gzfastq || in_file_type == FileT::gzfasta) && file.substr(pos) == ".gz") { file = file.substr(0, pos); pos = file.find_last_of("."); } pos = file.find_last_of("/"); path = file.substr(0, pos + 1); files.push_back(make_pair(path, file.substr(pos+1))); } } return 0; } int parse_command_line(int argc, char* argv[]) { string pair_1, pair_2; int c; while (1) { static struct option long_options[] = { {"help", no_argument, NULL, 'h'}, {"version", no_argument, NULL, 'v'}, {"discards", no_argument, NULL, 'D'}, {"pair_1", required_argument, NULL, '1'}, {"pair_2", required_argument, NULL, '2'}, {"infile_type", required_argument, NULL, 'i'}, {"outfile_type", required_argument, NULL, 'y'}, {"file", required_argument, NULL, 'f'}, {"path", required_argument, NULL, 'p'}, {"outpath", required_argument, NULL, 'o'}, {"k_dist", no_argument, NULL, 'I'}, {"rare", no_argument, NULL, 'R'}, {"abundant", no_argument, NULL, 'A'}, {"normalize", required_argument, NULL, 'N'}, {"k_len", required_argument, NULL, 'K'}, {"max_k_freq", required_argument, NULL, 'M'}, {"min_lim", required_argument, NULL, 'F'}, {"max_lim", required_argument, NULL, 'G'}, {"min_k_pct", required_argument, NULL, 'P'}, {"read_k_freq", required_argument, NULL, 'r'}, {"write_k_freq", required_argument, NULL, 'w'}, {0, 0, 0, 0} }; // getopt_long stores the option index here. int option_index = 0; c = getopt_long(argc, argv, "hvRADkP:N:I:w:r:K:F:G:M:m:i:y:f:o:t:p:1:2:", long_options, &option_index); // Detect the end of the options. if (c == -1) break; switch (c) { case 'h': help(); break; case 'i': if (strcasecmp(optarg, "fasta") == 0) in_file_type = FileT::fasta; else if (strcasecmp(optarg, "gzfasta") == 0) in_file_type = FileT::gzfasta; else if (strcasecmp(optarg, "gzfastq") == 0) in_file_type = FileT::gzfastq; else in_file_type = FileT::fastq; break; case 'y': if (strcasecmp(optarg, "fasta") == 0) out_file_type = FileT::fasta; else out_file_type = FileT::fastq; break; case 'f': in_files.push_back(optarg); break; case '1': pair_1 = optarg; break; case '2': pair_2 = optarg; if (pair_1.length() == 0) help(); in_pair_files.push_back(pair_1); in_pair_files.push_back(pair_2); pair_1 = ""; pair_2 = ""; break; case 'p': in_path = optarg; break; case 'o': out_path = optarg; break; case 'D': discards = true; break; case 'I': kmer_distr = true; break; case 'R': filter_rare_k = true; break; case 'A': filter_abundant_k = true; break; case 'N': normalize = true; normalize_lim = is_integer(optarg); break; case 'K': kmer_len = is_integer(optarg); break; case 'M': max_k_freq = is_integer(optarg); break; case 'F': min_lim = is_integer(optarg); break; case 'G': max_lim = is_integer(optarg); break; case 'P': min_k_pct = is_double(optarg); break; case 'r': read_k_freq = true; k_freq_path = optarg; break; case 'w': write_k_freq = true; k_freq_path = optarg; break; case 'v': version(); break; case '?': // getopt_long already printed an error message. help(); break; default: cerr << "Unknown command line option '" << (char) c << "'\n"; help(); abort(); } } if (in_files.size() == 0 && in_pair_files.size() == 0 && in_path.length() == 0) { cerr << "You must specify an input file of a directory path to a set of input files.\n"; help(); } if (in_files.size() > 0 && in_path.length() > 0) { cerr << "You must specify either a single input file (-f) or a directory path (-p), not both.\n"; help(); } if (in_path.length() > 0 && in_path.at(in_path.length() - 1) != '/') in_path += "/"; if (out_path.length() == 0) out_path = "."; if (out_path.at(out_path.length() - 1) != '/') out_path += "/"; if (in_file_type == FileT::unknown) in_file_type = FileT::fastq; if (read_k_freq && write_k_freq) { cerr << "You may either read a set of kmer frequencies, or write kmer frequencies, not both.\n"; help(); } if (min_k_pct < 0.0 || min_k_pct > 1.0) { cerr << "Percentage to consider a kmer rare must be between 0 and 1.0.\n"; help(); } // // Check that the output path exists. // struct stat info; if (stat(out_path.c_str(), &info) != 0) { cerr << "Unable to locate the specified output path, '" << out_path << "'\n"; exit(1); } return 0; } void version() { std::cerr << "kmer_filter " << VERSION << "\n\n"; exit(0); } void help() { std::cerr << "kmer_filter " << VERSION << "\n" << "kmer_filter [-f in_file_1 [-f in_file_2...] | -p in_dir] [-1 pair_1 -2 pair_2 [-1 pair_1...]] -o out_dir [-i type] [-y type] [-D] [-h]\n" << " f: path to the input file if processing single-end seqeunces.\n" << " i: input file type, either 'bustard' for the Illumina BUSTARD output files, 'fasta', 'fastq', 'gzfasta', or 'gzfastq' (default 'fastq').\n" << " p: path to a directory of files (for single-end files only).\n" << " 1: specify the first in a pair of files to be processed together.\n" << " 2: specify the second in a pair of files to be processed together.\n" << " o: path to output the processed files.\n" << " y: output type, either 'fastq' or 'fasta' (default fastq).\n" << " D: capture discarded reads to a file.\n" << " h: display this help messsage.\n\n" << " Filtering options:\n" << " --rare: turn on filtering based on rare k-mers.\n" << " --abundant: turn on filtering based on abundant k-mers.\n" << " --k_len : specify k-mer size (default 15).\n\n" << " Advanced filtering options:\n" << " --max_k_freq : specify the number of times a kmer must occur to be considered abundant (default 20,000).\n" << " --min_lim : specify number of rare kmers occuring in a row required to discard a read (default 80% of the k-mer length).\n" << " --max_lim : specify number of abundant kmers required to discard a read (default 80% of the k-mers in a read).\n\n" << " Normalize data:\n" << " --normalize : normalize read depth according to k-mer coverage.\n\n" << " Characterizing K-mers:\n" << " --write_k_freq: write kmers along with their frequency of occurrence and exit.\n" << " --k_dist: print k-mer frequency distribution and exit.\n\n" << " Advanced input options:\n" << " --read_k_freq : read a set of kmers along with their frequencies of occurrence instead of reading raw input files.\n" << "\n"; exit(0); } stacks-1.44/src/kmer_filter.h000644 000765 000024 00000006741 12764047443 016770 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2011-2014, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __KMER_FILTER_H__ #define __KMER_FILTER_H__ #include "constants.h" #include #include // Process command-line options #include // Open/Read contents of a directory #include #include #include #include #include #include #include #include using std::stringstream; using std::istream; using std::ofstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include using std::vector; #include using std::map; #include using std::set; #include using std::pair; #include using std::unordered_map; #ifdef HAVE_SPARSEHASH #include using google::sparse_hash_map; #endif #include "clean.h" #include "utils.h" #include "kmers.h" #include "write.h" #include "BustardI.h" // Reading input files in Tab-separated Bustard format #include "FastaI.h" // Reading input files in FASTA format #include "FastqI.h" // Reading input files in FASTQ format #include "gzFasta.h" // Reading gzipped input files in FASTA format #include "gzFastq.h" // Reading gzipped input files in FASTQ format #ifdef HAVE_SPARSEHASH typedef sparse_hash_map SeqKmerHash; #else typedef unordered_map SeqKmerHash; #endif void help( void ); void version( void ); int parse_command_line(int, char**); int build_file_list(vector &, vector > &); int process_reads(string, string, SeqKmerHash &, map &); int process_paired_reads(string, string, string, string, SeqKmerHash &, map &); int print_results(map > &); // // Functions to normalize read depth // int normalize_reads(string, string, SeqKmerHash &, vector &, map &); int normalize_paired_reads(string, string, string, string, SeqKmerHash &, vector &, map &); bool normalize_kmer_lookup(SeqKmerHash &, char *, char *, int, vector &); // // Functions for finding and removing reads with rare kmers // int populate_kmers(vector > &, vector > &, SeqKmerHash &, vector &); int process_file_kmers(string, SeqKmerHash &, vector &); int generate_kmer_dist(SeqKmerHash &); int calc_kmer_median(SeqKmerHash &, double &, double &); int kmer_map_cmp(pair, pair); int kmer_lookup(SeqKmerHash &, char *, char *, int, int &, int &); int free_kmer_hash(SeqKmerHash &, vector &); int read_kmer_freq(string, SeqKmerHash &, vector &); int write_kmer_freq(string, SeqKmerHash &); #endif // __KMER_FILTER_H__ stacks-1.44/src/kmers.cc000644 000765 000024 00000037777 12764047443 015761 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // kmers.cc -- routines to generate and hash K-mers // #include "kmers.h" int determine_kmer_length(int read_len, int dist) { int kmer_len, span, min_matches; // // If distance allowed between sequences is 0, then k-mer length equals read length. // if (dist == 0) return read_len; // // Longer k-mer lengths will provide a smaller hash, with better key placement. // Increase the kmer_len until we start to miss hits at the given distance. Then // back the kmer_len off one unit to get the final value. // for (kmer_len = 5; kmer_len < read_len; kmer_len += 2) { span = (kmer_len * (dist + 1)) - 1; min_matches = read_len - span; if (min_matches <= 0) break; } if (kmer_len >= read_len) { cerr << "Unable to find a suitable k-mer length for matching.\n"; exit(1); } kmer_len -= 2; return kmer_len; } int calc_min_kmer_matches(int kmer_len, int dist, int read_len, bool exit_err) { int span, min_matches; span = (kmer_len * (dist + 1)) - 1; min_matches = read_len - span; if (min_matches <= 0) { cerr << "Warning: combination of k-mer length (" << kmer_len << ") and edit distance (" << dist << ") allows for " << "sequences to be missed by the matching algorithm.\n"; } if (min_matches <= 0 && exit_err) exit(1); else if (min_matches <= 0) min_matches = 1; return min_matches; } int initialize_kmers(int kmer_len, int num_kmers, vector &kmers) { char *kmer; for (int i = 0; i < num_kmers; i++) { kmer = new char[kmer_len + 1]; kmers.push_back(kmer); } return 0; } int generate_kmers_lazily(const char *seq, uint kmer_len, uint num_kmers, vector &kmers) { char *kmer; const char *k = seq; if (num_kmers > kmers.size()) { int new_kmers = num_kmers - kmers.size(); for (int i = 0; i < new_kmers; i++) { kmer = new char[kmer_len + 1]; kmers.push_back(kmer); } } for (uint i = 0; i < num_kmers; i++) { kmer = kmers.at(i); strncpy(kmer, k, kmer_len); kmer[kmer_len] = '\0'; k++; } return 0; } int generate_kmers(const char *seq, int kmer_len, int num_kmers, vector &kmers) { char *kmer; const char *k = seq; for (int i = 0; i < num_kmers; i++) { kmer = new char[kmer_len + 1]; strncpy(kmer, k, kmer_len); kmer[kmer_len] = '\0'; kmers.push_back(kmer); k++; } return 0; } int generate_permutations(map &pstrings, int width) { int i, j, rem, div, num; char *p; // // Given a k-mer that allows wildcards -- 'N' characters, we need to generate all // possible k-mers. To do so, we will generate a range of numbers that we convert to // base 4, assuming that 0 = 'A', 1 = 'C', 2 = 'G', 3 = 'T'. // const int base = 4; int range = (int) pow(4, width); // // Create an array of strings to hold the permuted nucleotides. // char **strings = new char * [range]; for (i = 0; i < range; i++) strings[i] = new char[width + 1]; for (i = 0; i < range; i++) { for (j = 0; j < width; j++) strings[i][j] = 'A'; strings[i][width] = '\0'; } for (i = 0; i < range; i++) { // // Convert this number to base 4 // p = strings[i]; p += width - 1; num = i; do { div = (int) floor(num / base); rem = num % base; switch(rem) { case 0: *p = 'A'; break; case 1: *p = 'C'; break; case 2: *p = 'G'; break; case 3: *p = 'T'; break; } num = div; p--; } while (div > 0); } pstrings[width] = strings; return 0; } int populate_kmer_hash(map &merged, KmerHashMap &kmer_map, vector &kmer_map_keys, int kmer_len) { map::iterator it; MergedStack *tag; vector kmers; bool exists; // // Break each stack down into k-mers and create a hash map of those k-mers // recording in which sequences they occur. // int num_kmers = strlen(merged.begin()->second->con) - kmer_len + 1; for (it = merged.begin(); it != merged.end(); it++) { tag = it->second; // Don't compute distances for masked tags if (tag->masked) continue; generate_kmers(tag->con, kmer_len, num_kmers, kmers); // Hash the kmers for (int j = 0; j < num_kmers; j++) { exists = kmer_map.count(kmers[j]) == 0 ? false : true; kmer_map[kmers[j]].push_back(tag->id); if (exists) delete [] kmers[j]; else kmer_map_keys.push_back(kmers[j]); } kmers.clear(); } //dump_kmer_map(kmer_map); return 0; } int populate_kmer_hash(map &catalog, CatKmerHashMap &kmer_map, vector &kmer_map_keys, int kmer_len) { map::iterator it; vector >::iterator allele; vector kmers; Locus *tag; char *hash_key; bool exists; int num_kmers; // // Break each stack down into k-mers and create a hash map of those k-mers // recording in which sequences they occur. // for (it = catalog.begin(); it != catalog.end(); it++) { tag = it->second; num_kmers = strlen(tag->con) - kmer_len + 1; // // Iterate through the possible Catalog alleles // for (allele = tag->strings.begin(); allele != tag->strings.end(); allele++) { // // Generate and hash the kmers for this allele string // generate_kmers(allele->second.c_str(), kmer_len, num_kmers, kmers); for (int j = 0; j < num_kmers; j++) { hash_key = kmers[j]; exists = kmer_map.count(hash_key) == 0 ? false : true; kmer_map[hash_key].push_back(make_pair(allele->first, tag->id)); if (exists) delete [] kmers[j]; else kmer_map_keys.push_back(hash_key); } kmers.clear(); } } //dump_kmer_map(kmer_map); return 0; } int populate_kmer_hash(map &catalog, KmerHashMap &kmer_map, vector &kmer_map_keys, map > &allele_map, int kmer_len) { map::iterator it; KmerHashMap::iterator map_it; vector >::iterator allele; map >::iterator allele_it; vector kmers; Locus *tag; char *hash_key; // // Break each stack down into k-mers and create a hash map of those k-mers // recording in which sequences they occur. // int num_kmers; int allele_index = 0; allele_it = allele_map.begin(); for (it = catalog.begin(); it != catalog.end(); it++) { tag = it->second; num_kmers = strlen(tag->con) - kmer_len + 1; // // Iterate through the possible Catalog alleles // for (allele = tag->strings.begin(); allele != tag->strings.end(); allele++) { // // Generate and hash the kmers for this allele string // generate_kmers(allele->second.c_str(), kmer_len, num_kmers, kmers); allele_it = allele_map.insert(allele_it, make_pair(allele_index, make_pair(allele->first, tag->id))); for (int j = 0; j < num_kmers; j++) { hash_key = kmers[j]; map_it = kmer_map.find(hash_key); if (map_it != kmer_map.end()) { map_it->second.push_back(allele_index); delete [] kmers[j]; } else { kmer_map[hash_key].push_back(allele_index); kmer_map_keys.push_back(hash_key); } } kmers.clear(); allele_index++; } } //dump_kmer_map(kmer_map); return 0; } int populate_kmer_hash(map &catalog, KmerHashMap &kmer_map, vector &kmer_map_keys, map > &allele_map, int kmer_len) { map::iterator it; KmerHashMap::iterator map_it; vector >::iterator allele; map >::iterator allele_it; vector kmers; Locus *tag; char *hash_key; // // Break each stack down into k-mers and create a hash map of those k-mers // recording in which sequences they occur. // int num_kmers; int allele_index = 0; allele_it = allele_map.begin(); for (it = catalog.begin(); it != catalog.end(); it++) { tag = it->second; num_kmers = strlen(tag->con) - kmer_len + 1; // // Iterate through the possible Catalog alleles // for (allele = tag->strings.begin(); allele != tag->strings.end(); allele++) { // // Generate and hash the kmers for this allele string // generate_kmers(allele->second.c_str(), kmer_len, num_kmers, kmers); allele_it = allele_map.insert(allele_it, make_pair(allele_index, make_pair(allele->first, tag->id))); for (int j = 0; j < num_kmers; j++) { hash_key = kmers[j]; map_it = kmer_map.find(hash_key); if (map_it != kmer_map.end()) { map_it->second.push_back(allele_index); delete [] kmers[j]; } else { kmer_map[hash_key].push_back(allele_index); kmer_map_keys.push_back(hash_key); } } kmers.clear(); allele_index++; } } //dump_kmer_map(kmer_map); return 0; } int free_kmer_hash(CatKmerHashMap &kmer_map, vector &kmer_map_keys) { for (uint i = 0; i < kmer_map_keys.size(); i++) { kmer_map[kmer_map_keys[i]].clear(); } kmer_map.clear(); for (uint i = 0; i < kmer_map_keys.size(); i++) { delete [] kmer_map_keys[i]; } kmer_map_keys.clear(); return 0; } int free_kmer_hash(KmerHashMap &kmer_map, vector &kmer_map_keys) { for (uint i = 0; i < kmer_map_keys.size(); i++) { kmer_map[kmer_map_keys[i]].clear(); } kmer_map.clear(); for (uint i = 0; i < kmer_map_keys.size(); i++) { delete [] kmer_map_keys[i]; } kmer_map_keys.clear(); return 0; } int dist(const char *tag_1, Locus *tag_2, allele_type allele) { int dist = 0; const char *p = tag_1; const char *p_end = p + strlen(p); const char *q = NULL; // // Identify which matching string has the proper allele // vector >::iterator it; for (it = tag_2->strings.begin(); it != tag_2->strings.end(); it++) if (it->first == allele) q = it->second.c_str(); if (q == NULL) return -1; const char *q_end = q + strlen(q); // Count the number of characters that are different // between the two sequences. while (p < p_end && q < q_end) { dist += (*p == *q) ? 0 : 1; p++; q++; } return dist; } int dist(const char *tag_1, const char *tag_2, vector > &cigar) { uint size = cigar.size(); char op; uint dist, len, pos_1, pos_2, stop; int mismatches = 0; len = strlen(tag_1); pos_1 = 0; pos_2 = 0; for (uint i = 0; i < size; i++) { op = cigar[i].first; dist = cigar[i].second; switch(op) { case 'D': // // A deletion has occured in tag_1 relative to tag_2. // pos_2 += dist; break; case 'I': // // An insertion has occured in tag_1 relative to tag_2. // pos_1 += dist; break; case 'M': stop = pos_1 + dist; while (pos_1 < stop && pos_1 < len && pos_2 < len) { if (tag_1[pos_1] != 'N' && tag_2[pos_2] != 'N' && tag_1[pos_1] != tag_2[pos_2]) mismatches++; pos_1++; pos_2++; } break; default: break; } } return mismatches; } int dist(Locus *tag_1, Locus *tag_2) { int dist = 0; char *p = tag_1->con; char *q = tag_2->con; char *p_end = p + tag_1->len; char *q_end = q + tag_2->len; if (tag_1->len != tag_2->len) { if (tag_1->len < tag_2->len) dist += tag_2->len - tag_1->len; else if (tag_1->len > tag_2->len) dist += tag_1->len - tag_2->len; } // // Count the number of characters that are different // between the two sequences. // while (p < p_end && q < q_end) { dist += (*p == *q) ? 0 : 1; p++; q++; } return dist; } int dist(MergedStack *tag_1, MergedStack *tag_2) { int dist = 0; char *p = tag_1->con; char *q = tag_2->con; char *p_end = p + tag_1->len; char *q_end = q + tag_2->len; // // If the sequences are of different lengths, count the missing // nucleotides as mismatches. // if (tag_1->len != tag_2->len) { if (tag_1->len < tag_2->len) dist += tag_2->len - tag_1->len; else if (tag_1->len > tag_2->len) dist += tag_1->len - tag_2->len; } // // Count the number of characters that are different // between the two sequences. // while (p < p_end && q < q_end) { dist += (*p == *q) ? 0 : 1; p++; q++; } return dist; } int dist(MergedStack *tag_1, char *seq) { int dist = 0; char *p = tag_1->con; char *q = seq; uint q_len = strlen(q); char *p_end = p + tag_1->len; char *q_end = q + q_len; // // If the sequences are of different lengths, count the missing // nucleotides as mismatches. // if (tag_1->len != q_len) { if (tag_1->len < q_len) dist += q_len - tag_1->len; else if (tag_1->len > q_len) dist += tag_1->len - q_len; } // // Count the number of characters that are different // between the two sequences. // while (p < p_end && q < q_end) { dist += (*p == *q) ? 0 : 1; p++; q++; } return dist; } bool compare_dist(pair a, pair b) { return (a.second < b.second); } int dump_kmer_map(KmerHashMap &kmer_map) { KmerHashMap::iterator kit; vector::iterator vit; cerr << kmer_map.size() << " keys in the map.\n"; int i = 1; for (kit = kmer_map.begin(); kit != kmer_map.end(); kit++) { cerr << "Key #" << i << " " << kit->first << ": "; for (vit = (kit->second).begin(); vit != (kit->second).end(); vit++) cerr << " " << *vit; cerr << "\n"; i++; if (i > 1000) break; } return 0; } stacks-1.44/src/kmers.h000644 000765 000024 00000006763 12702770100 015573 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __KMERS_H__ #define __KMERS_H__ #include "constants.h" #include #include #include using std::string; #include using std::vector; #include using std::map; #include using std::pair; using std::make_pair; #include using std::ifstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include using std::unordered_map; #ifdef HAVE_SPARSEHASH #include using google::sparse_hash_map; #endif #include "stacks.h" #include "locus.h" #include "mstack.h" #include "input.h" struct hash_charptr { size_t operator()(const char *__s) const { size_t __result = static_cast(14695981039346656037ULL); unsigned int __len = strlen(__s); for (unsigned int i = 0; i < __len; i++) { __result ^= static_cast(__s[i]); __result *= static_cast(1099511628211ULL); } return __result; } }; struct eqstr { bool operator()(const char* s1, const char* s2) const { return strcmp(s1, s2) == 0; } }; #ifdef HAVE_SPARSEHASH typedef sparse_hash_map, hash_charptr, eqstr> KmerHashMap; typedef sparse_hash_map >, hash_charptr, eqstr> CatKmerHashMap; #else typedef unordered_map, hash_charptr, eqstr> KmerHashMap; typedef unordered_map >, hash_charptr, eqstr> CatKmerHashMap; #endif int determine_kmer_length(int, int); int calc_min_kmer_matches(int, int, int, bool); int initialize_kmers(int, int, vector &); int generate_kmers(const char *, int, int, vector &); int generate_kmers_lazily(const char *, uint, uint, vector &); int populate_kmer_hash(map &, KmerHashMap &, vector &, int); int populate_kmer_hash(map &, CatKmerHashMap &, vector &, int); int populate_kmer_hash(map &, KmerHashMap &, vector &, map > &, int); int populate_kmer_hash(map &, KmerHashMap &, vector &, map > &, int); int free_kmer_hash(KmerHashMap &, vector &); int free_kmer_hash(CatKmerHashMap &, vector &); int generate_permutations(map &, int); // // Utilities // int dist(const char *, const char *, vector > &); int dist(const char *, Locus *, allele_type); int dist(Locus *, Locus *); int dist(MergedStack *, MergedStack *); int dist(MergedStack *, char *); // // For sorting functions. // bool compare_dist(pair, pair); // // Debugging // int dump_kmer_map(KmerHashMap &); #endif // __KMERS_H__ stacks-1.44/src/locus.cc000644 000765 000024 00000007636 12774470747 015764 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2013-2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // locus.cc -- routines for the Locus class and its derivatives. // #include "locus.h" uint Locus::sort_bp(uint k) { if (this->loc.strand == strand_plus) return this->loc.bp + k; else return (k == 0 ? this->loc.bp - this->len + 1 : this->loc.bp - k); } int Locus::snp_index(uint col) { for (uint i = 0; i < this->snps.size(); i++) if (this->snps[i]->col == col) return i; return -1; } int Locus::add_consensus(const char *seq) { if (this->con != NULL) delete [] this->con; this->len = strlen(seq); this->con = new char[this->len + 1]; strcpy(this->con, seq); return 0; } int Locus::add_model(const char *seq) { if (this->model != NULL) delete [] this->model; this->model = new char[this->len + 1]; strncpy(this->model, seq, this->len); this->model[this->len] = '\0'; return 0; } int Locus::populate_alleles() { vector::iterator i; map::iterator j; string s; int k; if (this->len > strlen(this->con)) cerr << "Recorded locus->len: " << this->len << "; consensus length: " << strlen(this->con) << "\n"; // // Is this effective? // for (uint n = 0; n < this->strings.size(); n++) { this->strings[n].first.clear(); this->strings[n].second.clear(); } this->strings.clear(); if (this->snps.size() == 0) { this->strings.push_back(make_pair("consensus", this->con)); return 0; } for (j = this->alleles.begin(); j != this->alleles.end(); j++) { s = this->con; k = 0; for (i = this->snps.begin(); i != this->snps.end(); i++) { if ((*i)->col < this->len) s.replace((*i)->col, 1, 1, j->first[k]); k++; } this->strings.push_back(make_pair(j->first, s)); } return 0; } bool bp_compare(Locus *a, Locus *b) { return (a->sort_bp() < b->sort_bp()); } QLocus::~QLocus() { vector::iterator it; for (it = this->matches.begin(); it != this->matches.end(); it++) delete *it; } int QLocus::add_match(int catalog_id, allele_type cat_type, allele_type query_type, int distance) { Match *m = new Match; m->cat_id = catalog_id; m->cat_type = cat_type; m->query_type = query_type; m->dist = distance; this->matches.push_back(m); return 0; } int QLocus::add_match(int catalog_id, allele_type cat_type, allele_type query_type, int distance, string cigar) { Match *m = new Match; m->cat_id = catalog_id; m->cat_type = cat_type; m->query_type = query_type; m->dist = distance; m->cigar = cigar; this->matches.push_back(m); return 0; } int QLocus::add_match(int catalog_id, allele_type cat_type) { Match *m = new Match; m->cat_id = catalog_id; m->cat_type = cat_type; m->query_type = ""; m->dist = 0; this->matches.push_back(m); return 0; } int QLocus::clear_matches() { vector::iterator it; for (it = this->matches.begin(); it != this->matches.end(); it++) delete *it; this->matches.clear(); return 0; } stacks-1.44/src/locus.h000644 000765 000024 00000012716 12764047443 015611 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2013, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __LOCUS_H__ #define __LOCUS_H__ #include #include using std::string; #include using std::vector; #include using std::map; #include #include using std::pair; using std::make_pair; #include "constants.h" #include "stacks.h" typedef struct match { uint cat_id; allele_type cat_type; allele_type query_type; string cigar; uint dist; } Match; class Locus { public: int id; // Locus ID int sample_id; // Sample ID int depth; // Stack depth char *con; // Consensus sequence char *model; // Model calls for each nucleotide uint len; // Sequence length double lnl; // Log likelihood of this locus // // Flags // bool blacklisted; bool deleveraged; bool lumberjackstack; vector comp; // Raw components in this stack. vector reads; // Sequence reads contributing to this locus. vector comp_cnt; // Counter for internal stacks merged into this locus. vector comp_type; // Read types for reads contributing to this locus. PhyLoc loc; // Physical genome location of this stack. vector snps; // Single Nucleotide Polymorphisms in this stack. map alleles; // Map of the allelic configuration of SNPs in this stack along with the count of each vector > strings; // Strings for matching (representing the various allele combinations) Locus() { id = 0; sample_id = 0; depth = 0; model = NULL; con = NULL; len = 0; lnl = 0.0; blacklisted = false; deleveraged = false; lumberjackstack = false; } virtual ~Locus() { delete [] con; delete [] model; for (uint i = 0; i < snps.size(); i++) delete snps[i]; for (uint i = 0; i < comp.size(); i++) delete [] comp[i]; for (uint i = 0; i < reads.size(); i++) delete [] reads[i]; } uint sort_bp(uint k = 0); int snp_index(uint); int add_consensus(const char *); int add_model(const char *); virtual int populate_alleles(); }; // // Query Locus Class // class QLocus : public Locus { public: vector matches; // Matching tags found for the catalog. QLocus(): Locus() {} ~QLocus(); int add_match(int, allele_type, allele_type, int, string); int add_match(int, allele_type, allele_type, int); int add_match(int, allele_type); int clear_matches(); }; // // Catalog Locus Class, for use in cstacks, contains catalog loci and records the // constiuent tags this locus was built from. // class CLocus : public Locus { public: vector > sources; // Sample/ID pairs for the sources contributing to this catalog entry uint match_cnt; CLocus() : Locus() { this->match_cnt = 0; }; int merge_snps(QLocus *); int reduce_alleles(set &); }; // // Catalog Summary Locus Class; used in genotypes and populations, records a catalog // locus with summary information derived from individuals in the population. // class CSLocus : public Locus { public: CSLocus() : Locus() { this->f = 0.0; this->cnt = 0; this->hcnt = 0; this->gcnt = 0; this->trans_gcnt = 0; this->chisq = 1.0; this->confounded_cnt = 0; }; string annotation; string marker; string uncor_marker; map hap_cnts; // Counts of each observed haplotype for this locus in the population. double f; // Inbreeder's coefficient map gmap; // Observed haplotype to genotype map for this locus. int confounded_cnt; // Number of samples/progeny containing confounded loci (more than one // locus from an individual sample matches this catalog locus). int hcnt; // Number of samples/progeny containing a haplotype for this locus. int cnt; // Number of samples/progeny containing data for this locus. int gcnt; // Number of progeny containing a valid genotype. int trans_gcnt; // Number of progeny containing a valid // genotype, translated for a particular map type. double chisq; // Chi squared p-value testing the null hypothesis of no segregation distortion. }; bool bp_compare(Locus *, Locus *); #endif // __LOCUS_H__ stacks-1.44/src/log_utils.cc000644 000765 000024 00000002510 12764047443 016612 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2013-2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #include "log_utils.h" int init_log(ofstream &fh, int argc, char **argv) { // // Obtain the current date. // time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%F %T", timeinfo); // // Write the command line that was executed. // for (int i = 0; i < argc; i++) { fh << argv[i]; if (i < argc - 1) fh << " "; } fh << "\n" << argv[0] << " version " << VERSION << " executed " << date << "\n\n"; return 0; } stacks-1.44/src/log_utils.h000644 000765 000024 00000002161 12533677757 016471 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2013-2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __LOG_UTILS_H__ #define __LOG_UTILS_H__ #include #include #include using std::ifstream; using std::ofstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include using std::stringstream; #include "constants.h" int init_log(ofstream &, int, char **); #endif // __LOG_UTILS_H__ stacks-1.44/src/MetaPopInfo.cc000644 000765 000024 00000025511 12750704270 016771 0ustar00catchenstaff000000 000000 #include #include #include #include #include #include #include "config.h" #include "constants.h" #include "input.h" #include "MetaPopInfo.h" using std::ifstream; using std::cerr; using std::exception; const string MetaPopInfo::Pop::default_name = "defaultpop"; const string MetaPopInfo::Group::default_name = "defaultgrp"; void MetaPopInfo::reset_sample_map() { sample_indexes_.clear(); for (size_t i = 0; i < samples_.size(); ++i) sample_indexes_.insert( {samples_[i].name, i} ); } void MetaPopInfo::reset_pop_map() { pop_indexes_.clear(); for (size_t i = 0; i < pops_.size(); ++i) pop_indexes_.insert( {pops_[i].name, i} ); } void MetaPopInfo::reset_group_map() { group_indexes_.clear(); for (size_t i = 0; i < groups_.size(); ++i) group_indexes_.insert( {groups_[i].name, i} ); } bool MetaPopInfo::init_popmap(const string& pmap_path) { ifstream fh(pmap_path.c_str(), ifstream::in); if (fh.fail()) return false; size_t p = 0; // pop index counter size_t g = 0; // group index counter char line[max_len]; memset(line, '\0', max_len); vector parts; while (fh.getline(line, max_len)) { size_t len = strlen(line); // Skip empty lines and comments if (len == 0 || line[0] == '#') continue; // Check for Windows line endings if (line[len - 1] == '\r') { line[len - 1] = '\0'; len -= 1; } // // Parse the contents, we expect: // tab [tab ] // parse_tsv(line, parts); if (parts.size() < 2 || parts.size() > 3 || parts[0].empty() || parts[1].empty()) { cerr << "Error: Malformed population map -- expected 'SAMPLE\\tPOP[\\tGROUP]'. In file '" << pmap_path << "', line :\n" << line << "\n"; throw exception(); } // // Process the sample name. // samples_.push_back(Sample(parts[0])); // // Process the population field. // pair::iterator, bool> pop_ins = pop_indexes_.insert( {parts[1], p} ); size_t pop_index = pop_ins.first->second; samples_.back().pop = pop_index; // Set the sample's population index. if (pop_ins.second) { // Unknown pop pops_.push_back(Pop(parts[1])); ++p; } // // Process the group field, if any // if (parts.size() == 3 && ! parts[2].empty()) { // Get the index of this group -- create it if necessary. pair::iterator, bool> grp_ins = group_indexes_.insert( {parts[2], g} ); if (grp_ins.second) { groups_.push_back(Group(parts[2])); ++g; } size_t grp_index = grp_ins.first->second; if (pops_[pop_index].group != size_t(-1)) { // The current pop already has a group, check that it is the same. if (pops_[pop_index].group != grp_index) { cerr << "Error: In population map file '" << pmap_path << "': population '" << pops_[pop_index].name << "' belongs to two groups, '" << groups_[pops_[pop_index].group].name << "' and '" << groups_[grp_index].name << "'\n."; throw exception(); } } else { pops_[pop_index].group = grp_index; groups_[grp_index].pops.push_back(pop_index); } } } if (samples_.empty()) // Empty population map. return false; // // Check that all the populations are in a group. Put // populations that do not have a group in a default // one. // bool missing_group = false; if (not groups_.empty()) { for (vector::iterator p = pops_.begin(); p != pops_.end(); ++p) { if (p->group == size_t(-1)) { cerr << "Warning: Population '" << p->name << "' did not have a group, adding it to '" << Group::default_name << "'.\n"; missing_group = true; } } } else { missing_group = true; } if (missing_group) { groups_.push_back(Group(Group::default_name)); g = groups_.size()-1; group_indexes_.insert( {Group::default_name, g} ); for (size_t p = 0; p < pops_.size(); ++p) { if (pops_[p].group == size_t(-1)) { pops_[p].group = g; groups_[g].pops.push_back(p); } } } // // Sort the samples. Determine the first/last indexes for each population. // sort(samples_.begin(), samples_.end()); reset_sample_map(); size_t curr_pop = 0; pops_[curr_pop].first_sample = 0; for (size_t s = 1; s < samples_.size(); ++s) { if (samples_[s].pop != curr_pop) { pops_[curr_pop].last_sample = s-1; ++curr_pop; pops_[curr_pop].first_sample = s; } } pops_[curr_pop].last_sample = samples_.size()-1; return true; } bool MetaPopInfo::init_names(const vector& sample_names) { if(sample_names.empty()) return false; // Create the samples for (vector::const_iterator s=sample_names.begin(); s!= sample_names.end(); ++s) { samples_.push_back(Sample(*s)); samples_.back().pop = 0; } sort(samples_.begin(), samples_.end()); // Create a default population pops_.push_back(Pop(Pop::default_name)); pops_[0].first_sample = 0; pops_[0].last_sample = samples_.size()-1; pops_[0].group = 0; // Create a default group groups_.push_back(Group(Group::default_name)); groups_[0].pops.push_back(0); // Set the support members. reset_sample_map(); reset_pop_map(); reset_group_map(); return true; } bool MetaPopInfo::init_directory(const string& dir_path) { // // Find all sample names. // vector sample_names; DIR* dir = opendir(dir_path.c_str()); if (dir == NULL) { cerr << "Unable to open directory '" << dir_path << "' for reading.\n"; throw exception(); } dirent *direntry; while ((direntry = readdir(dir)) != NULL) { string filename = direntry->d_name; if (filename == "." || filename == ".." || filename.substr(0, 6) == "batch_") continue; size_t pos = filename.rfind(".tags.tsv"); if (pos == string::npos) pos = filename.rfind(".tags.tsv.gz"); if (pos != string::npos) sample_names.push_back(filename.substr(0, pos)); } closedir(dir); // // Initialize the MetaPopInfo // return init_names(sample_names); } void MetaPopInfo::delete_samples(const vector& rm_samples) { // Remove these samples from [samples_]. for (vector::const_iterator s = rm_samples.begin(); s != rm_samples.end(); ++s) { samples_.at(*s).name.clear(); // Mark the sample for removal. } samples_.erase( remove_if(samples_.begin(), samples_.end(), [] (Sample& s) {return s.name.empty();} ), samples_.end()); // Update the indexes of the populations. for (vector::iterator p = pops_.begin(); p != pops_.end(); ++p) { for (vector::const_reverse_iterator rm_sample = rm_samples.rbegin(); rm_sample != rm_samples.rend(); ++rm_sample) { if (p->first_sample > *rm_sample) // n.b. ">" --p->first_sample; if (p->last_sample >= *rm_sample) // n.b. ">=". Thus if the population becomes // empty, [first_sample] will be past [last_sample]. // n.b. If removing the first pop, last_sample=size_t(-1). --p->last_sample; } } auto pop_is_empty = [] (Pop& p) {return (p.first_sample > p.last_sample || p.last_sample == size_t(-1));}; // Remove the empty populations from [groups_]. for(vector::iterator group = groups_.begin(); group != groups_.end(); ++group) group->pops.erase( remove_if(group->pops.begin(), group->pops.end(), [this,&pop_is_empty] (size_t p) {return pop_is_empty(pops_[p]);}), group->pops.end()); // Remove the empty populations from [pops_]. pops_.erase( remove_if(pops_.begin(), pops_.end(), pop_is_empty), pops_.end()); // Remove empty groups from [groups_]. groups_.erase( remove_if(groups_.begin(), groups_.end(), [] (Group& g) {return g.pops.empty();}), groups_.end()); // Update the support members. reset_sample_map(); reset_pop_map(); reset_group_map(); reset_sample_id_map(); } void MetaPopInfo::reset_sample_id_map() { sample_indexes_by_id_.clear(); for (size_t i = 0; i < samples_.size(); ++i) sample_indexes_by_id_[samples_[i].id] = i; } void MetaPopInfo::fill_files(vector >& files) const { files.clear(); for (vector::const_iterator sample = samples_.begin(); sample != samples_.end(); ++sample) files.push_back( {sample->pop, sample->name} ); } void MetaPopInfo::fill_sample_ids(vector& sample_ids) const { sample_ids.clear(); for (vector::const_iterator sample = samples_.begin(); sample != samples_.end(); ++sample) sample_ids.push_back(sample->id); } void MetaPopInfo::fill_samples(map& samples) const { samples.clear(); for (vector::const_iterator sample = samples_.begin(); sample != samples_.end(); ++sample) samples.insert({sample->id, sample->name}); } void MetaPopInfo::fill_pop_key(map& pop_key) const { pop_key.clear(); for (size_t i = 0; i < pops_.size(); ++i) pop_key.insert( {i, pops_[i].name} ); } void MetaPopInfo::fill_pop_indexes(map >& pop_indexes) const { pop_indexes.clear(); for (size_t i = 0; i < pops_.size(); ++i) pop_indexes.insert( {i, {pops_[i].first_sample, pops_[i].last_sample}} ); } void MetaPopInfo::fill_grp_key(map& grp_key) const { grp_key.clear(); for (size_t i = 0; i < groups_.size(); ++i) grp_key.insert({i, groups_[i].name}); } void MetaPopInfo::fill_grp_members(map >& grp_members) const { grp_members.clear(); for (size_t i = 0; i < groups_.size(); ++i) { vector& pop_ids = grp_members.insert( {i, vector()} ).first->second; for(vector::const_iterator p = groups_[i].pops.begin(); p != groups_[i].pops.end(); ++p) pop_ids.push_back(*p); } } stacks-1.44/src/MetaPopInfo.h000644 000765 000024 00000007021 12750704270 016627 0ustar00catchenstaff000000 000000 #ifndef METAPOPINFO_H #define METAPOPINFO_H #include #include #include using std::size_t; using std::pair; using std::vector; using std::string; using std::map; /* * MetaPopInfo * Class for reprensenting a metapopulation : its samples, populations, * groups of populations, and associated information. */ class MetaPopInfo { public: struct Sample { string name; size_t pop; size_t id; // optional, deprecated Sample(const string& n) : name(n), pop(-1), id(-1) {} inline bool operator<(const Sample& other) const; }; struct Pop { string name; size_t first_sample; size_t last_sample; size_t group; Pop(const string& n) : name(n), first_sample(-1), last_sample(-1), group(-1) {} static const string default_name; }; struct Group { string name; vector pops; Group(const string& n) : name(n), pops() {} static const string default_name; }; private: vector samples_; //n.b. Samples are sorted primarily by population index, and secondarily by name. vector pops_; vector groups_; map sample_indexes_; // Links a name with an index in [samples_]. map pop_indexes_; map group_indexes_; void reset_sample_map(); // Resets [sample_indexes_]. void reset_pop_map(); void reset_group_map(); map sample_indexes_by_id_; // Links a sample ID with an index in [samples_]. void reset_sample_id_map(); public: // Create the representation : // -- from a population map file. // -- from just a vector of sample names. // -- or by looking for "*.tags.tsv(.gz)" files in a directory. bool init_popmap(const string& popmap_path); bool init_names(const vector& sample_names); bool init_directory(const string& dir_path); // Delete samples from the metapopulation. // (As samples, populations or groups may be deleted, the indexes of // the remaining ones change, but the order in which they appear // is preserved.) void delete_samples(const vector& rm_samples); // Retrieve information. const vector& samples() const {return samples_;} const vector& pops() const {return pops_;} const vector& groups() const {return groups_;} size_t get_sample_index(const string& name) const {return sample_indexes_.at(name);} size_t get_pop_index(const string& name) const {return pop_indexes_.at(name);} size_t get_group_index(const string& name) const {return group_indexes_.at(name);} // Work with sample IDs. (IDs unicity is not enforced.) void set_sample_id(size_t index, size_t id) {samples_.at(index).id = id; sample_indexes_by_id_[id] = index;} size_t get_sample_index(const size_t& id) const {return sample_indexes_by_id_.at(id);} /* * Methods for backwards compatibility */ // Fill former globals. void fill_files(vector >&) const; void fill_sample_ids(vector&) const; void fill_samples(map&) const; void fill_pop_key(map&) const; void fill_pop_indexes(map >&) const; void fill_grp_key(map&) const; void fill_grp_members(map >&) const; }; inline bool MetaPopInfo::Sample::operator<(const Sample& other) const { if (pop == other.pop) return name < other.name; else return pop < other.pop; } #endif // METAPOPINFO_H stacks-1.44/src/models.cc000644 000765 000024 00000045645 12764047443 016114 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010 - 2012, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // models.cc -- routines to detect polymorphism (snp) and detect a lack of polymorphism (fixed). // // Julian Catchen // jcatchen@uoregon.edu // University of Oregon // // $Id$ // #include "models.h" snp_type call_multinomial_snp(MergedStack *tag, int col, map &n, bool record_snps) { vector > nuc; map::iterator i; int total = 0; for (i = n.begin(); i != n.end(); i++) { if (i->first != 'N') { total += i->second; nuc.push_back(make_pair(i->first, i->second)); } } sort(nuc.begin(), nuc.end(), compare_pair); // // If this column was simply uncalled Ns, return. // if (nuc[0].second == 0) { if (record_snps) { SNP *snp = new SNP; snp->type = snp_type_unk; snp->col = col; snp->lratio = 0; snp->rank_1 = 'N'; snp->rank_2 = '-'; tag->snps.push_back(snp); } return snp_type_unk; } // // Method of Paul Hohenlohe , personal communication. // // For a diploid individual, there are ten possible genotypes // (four homozygous and six heterozygous genotypes). We calculate // the likelihood of each possible genotype by using a multinomial // sampling distribution, which gives the probability of observing // a set of read counts (n1,n2,n3,n4) given a particular genotype. // double nuc_1 = nuc[0].second; double nuc_2 = nuc[1].second; double nuc_3 = nuc[2].second; double nuc_4 = nuc[3].second; double l_ratio = 0; l_ratio = (nuc_1 * log(nuc_1 / total)); if (total - nuc_1 > 0) l_ratio += ((total - nuc_1) * log((total - nuc_1) / (3 * total))); if (nuc_1 + nuc_2 > 0) l_ratio -= ((nuc_1 + nuc_2) * log((nuc_1 + nuc_2) / (2 * total))); if (nuc_3 + nuc_4 > 0) l_ratio -= ((nuc_3 + nuc_4) * log((nuc_3 + nuc_4) / (2 * total))); l_ratio *= 2; snp_type res; if (l_ratio <= heterozygote_limit) { // // This locus is a heterozygote. // if (record_snps) { SNP *snp = new SNP; snp->type = snp_type_het; snp->col = col; snp->lratio = l_ratio; snp->rank_1 = nuc[0].first; snp->rank_2 = nuc[1].first; tag->snps.push_back(snp); } res = snp_type_het; } else if (l_ratio >= homozygote_limit) { // // This locus is a homozygote. // if (record_snps) { SNP *snp = new SNP; snp->type = snp_type_hom; snp->col = col; snp->lratio = l_ratio; snp->rank_1 = nuc[0].first; snp->rank_2 = '-'; tag->snps.push_back(snp); } res = snp_type_hom; } else { // // Unknown whether this is a heterozygote or homozygote. // if (record_snps) { SNP *snp = new SNP; snp->type = snp_type_unk; snp->col = col; snp->lratio = l_ratio; snp->rank_1 = nuc[0].first; snp->rank_2 = nuc[1].second > 0 ? nuc[1].first : '-'; tag->snps.push_back(snp); } res = snp_type_unk; } return res; } snp_type call_multinomial_snp(Locus *tag, int col, map &n) { vector > nuc; map::iterator i; int total = 0; for (i = n.begin(); i != n.end(); i++) { if (i->first != 'N') { total += i->second; nuc.push_back(make_pair(i->first, i->second)); } } sort(nuc.begin(), nuc.end(), compare_pair); // // If this column was simply uncalled Ns, return. // if (nuc[0].second == 0) { tag->snps[col]->type = snp_type_unk; tag->snps[col]->col = col; tag->snps[col]->lratio = 0; tag->snps[col]->rank_1 = 'N'; tag->snps[col]->rank_2 = '-'; return snp_type_unk; } // // Method of Paul Hohenlohe , personal communication. // // For a diploid individual, there are ten possible genotypes // (four homozygous and six heterozygous genotypes). We calculate // the likelihood of each possible genotype by using a multinomial // sampling distribution, which gives the probability of observing // a set of read counts (n1,n2,n3,n4) given a particular genotype. // double nuc_1 = nuc[0].second; double nuc_2 = nuc[1].second; double nuc_3 = nuc[2].second; double nuc_4 = nuc[3].second; double l_ratio = 0; l_ratio = (nuc_1 * log(nuc_1 / total)); if (total - nuc_1 > 0) l_ratio += ((total - nuc_1) * log((total - nuc_1) / (3 * total))); if (nuc_1 + nuc_2 > 0) l_ratio -= ((nuc_1 + nuc_2) * log((nuc_1 + nuc_2) / (2 * total))); if (nuc_3 + nuc_4 > 0) l_ratio -= ((nuc_3 + nuc_4) * log((nuc_3 + nuc_4) / (2 * total))); l_ratio *= 2; snp_type res; if (l_ratio <= heterozygote_limit) { // // This locus is a heterozygote. // tag->snps[col]->type = snp_type_het; tag->snps[col]->col = col; tag->snps[col]->lratio = l_ratio; tag->snps[col]->rank_1 = nuc[0].first; tag->snps[col]->rank_2 = nuc[1].first; res = snp_type_het; } else if (l_ratio >= homozygote_limit) { // // This locus is a homozygote. // tag->snps[col]->type = snp_type_hom; tag->snps[col]->col = col; tag->snps[col]->lratio = l_ratio; tag->snps[col]->rank_1 = nuc[0].first; tag->snps[col]->rank_2 = '-'; res = snp_type_hom; } else { // // Unknown whether this is a heterozygote or homozygote. // tag->snps[col]->type = snp_type_unk; tag->snps[col]->col = col; tag->snps[col]->lratio = l_ratio; tag->snps[col]->rank_1 = nuc[0].first; tag->snps[col]->rank_2 = nuc[1].second > 0 ? nuc[1].first : '-'; res = snp_type_unk; } return res; } snp_type call_bounded_multinomial_snp(MergedStack *tag, int col, map &n, bool record_snps) { vector > nuc; map::iterator i; double total = 0.0; for (i = n.begin(); i != n.end(); i++) { if (i->first != 'N') { total += i->second; nuc.push_back(make_pair(i->first, i->second)); } } sort(nuc.begin(), nuc.end(), compare_pair); // // If this column was simply uncalled Ns, return. // if (nuc[0].second == 0) { if (record_snps) { SNP *snp = new SNP; snp->type = snp_type_unk; snp->col = col; snp->lratio = 0; snp->rank_1 = 'N'; snp->rank_2 = '-'; tag->snps.push_back(snp); } return snp_type_unk; } double nuc_1 = nuc[0].second; double nuc_2 = nuc[1].second; double nuc_3 = nuc[2].second; double nuc_4 = nuc[3].second; // // Method of Paul Hohenlohe , personal communication. // // // Calculate the site specific error rate for homozygous and heterozygous genotypes. // double epsilon_hom = (4.0 / 3.0) * ((total - nuc_1) / total); double epsilon_het = 2.0 * ((nuc_3 + nuc_4) / total); // cerr << "Epsilon_hom: " << epsilon_hom << "; epsilon_het: " << epsilon_het << "\n"; // // Check if the error rate is above or below the specified bound. // if (epsilon_hom < bound_low) epsilon_hom = bound_low; else if (epsilon_hom > bound_high) epsilon_hom = bound_high; if (epsilon_het < bound_low) epsilon_het = bound_low; else if (epsilon_het > bound_high) epsilon_het = bound_high; // // Calculate the log likelihood for the homozygous and heterozygous genotypes. // double ln_L_hom = nuc_1 * log(1 - ((3.0/4.0) * epsilon_hom)); ln_L_hom += epsilon_hom > 0 ? ((nuc_2 + nuc_3 + nuc_4) * log(epsilon_hom / 4.0)) : 0; double ln_L_het = (nuc_1 + nuc_2) * log(0.5 - (epsilon_het / 4.0)); ln_L_het += epsilon_het > 0 ? ((nuc_3 + nuc_4) * log(epsilon_het / 4.0)) : 0; // // Calculate the likelihood ratio. // double l_ratio = 2 * (ln_L_hom - ln_L_het); // cerr << " Nuc_1: " << nuc_1 << " Nuc_2: " << nuc_2 << " Nuc_3: " << nuc_3 << " Nuc_4: " << nuc_4 // << " epsilon homozygote: " << epsilon_hom // << " epsilon heterozygote: " << epsilon_het // << " Log likelihood hom: " << ln_L_hom // << " Log likelihood het: " << ln_L_het // << " Likelihood ratio: " << l_ratio << "\n"; snp_type res; if (l_ratio <= heterozygote_limit) { // // This locus is a heterozygote. // if (record_snps) { SNP *snp = new SNP; snp->type = snp_type_het; snp->col = col; snp->lratio = l_ratio; snp->rank_1 = nuc[0].first; snp->rank_2 = nuc[1].first; tag->snps.push_back(snp); } res = snp_type_het; } else if (l_ratio >= homozygote_limit) { // // This locus is a homozygote. // if (record_snps) { SNP *snp = new SNP; snp->type = snp_type_hom; snp->col = col; snp->lratio = l_ratio; snp->rank_1 = nuc[0].first; snp->rank_2 = '-'; tag->snps.push_back(snp); } res = snp_type_hom; } else { // // Unknown whether this is a heterozygote or homozygote. // if (record_snps) { SNP *snp = new SNP; snp->type = snp_type_unk; snp->col = col; snp->lratio = l_ratio; snp->rank_1 = nuc[0].first; snp->rank_2 = nuc[1].first; tag->snps.push_back(snp); } res = snp_type_unk; } return res; } snp_type call_bounded_multinomial_snp(Locus *tag, int col, map &n) { vector > nuc; map::iterator i; double total = 0.0; for (i = n.begin(); i != n.end(); i++) { if (i->first != 'N') { total += i->second; nuc.push_back(make_pair(i->first, i->second)); } } sort(nuc.begin(), nuc.end(), compare_pair); // // If this column was simply uncalled Ns, return. // if (nuc[0].second == 0) { tag->snps[col]->type = snp_type_unk; tag->snps[col]->col = col; tag->snps[col]->lratio = 0; tag->snps[col]->rank_1 = 'N'; tag->snps[col]->rank_2 = '-'; return snp_type_unk; } double nuc_1 = nuc[0].second; double nuc_2 = nuc[1].second; double nuc_3 = nuc[2].second; double nuc_4 = nuc[3].second; // // Method of Paul Hohenlohe , personal communication. // // // Calculate the site specific error rate for homozygous and heterozygous genotypes. // double epsilon_hom = (4.0 / 3.0) * ((total - nuc_1) / total); double epsilon_het = 2.0 * ((nuc_3 + nuc_4) / total); // cerr << "Epsilon_hom: " << epsilon_hom << "; epsilon_het: " << epsilon_het << "\n"; // // Check if the error rate is above or below the specified bound. // if (epsilon_hom < bound_low) epsilon_hom = bound_low; else if (epsilon_hom > bound_high) epsilon_hom = bound_high; if (epsilon_het < bound_low) epsilon_het = bound_low; else if (epsilon_het > bound_high) epsilon_het = bound_high; // // Calculate the log likelihood for the homozygous and heterozygous genotypes. // double ln_L_hom = nuc_1 * log(1 - ((3.0/4.0) * epsilon_hom)); ln_L_hom += epsilon_hom > 0 ? ((nuc_2 + nuc_3 + nuc_4) * log(epsilon_hom / 4.0)) : 0; double ln_L_het = (nuc_1 + nuc_2) * log(0.5 - (epsilon_het / 4.0)); ln_L_het += epsilon_het > 0 ? ((nuc_3 + nuc_4) * log(epsilon_het / 4.0)) : 0; // // Calculate the likelihood ratio. // double l_ratio = 2 * (ln_L_hom - ln_L_het); // cerr << " Nuc_1: " << nuc_1 << " Nuc_2: " << nuc_2 << " Nuc_3: " << nuc_3 << " Nuc_4: " << nuc_4 // << " epsilon homozygote: " << epsilon_hom // << " epsilon heterozygote: " << epsilon_het // << " Log likelihood hom: " << ln_L_hom // << " Log likelihood het: " << ln_L_het // << " Likelihood ratio: " << l_ratio << "\n"; snp_type res; if (l_ratio <= heterozygote_limit) { // // This locus is a heterozygote. // tag->snps[col]->type = snp_type_het; tag->snps[col]->col = col; tag->snps[col]->lratio = l_ratio; tag->snps[col]->rank_1 = nuc[0].first; tag->snps[col]->rank_2 = nuc[1].first; res = snp_type_het; } else if (l_ratio >= homozygote_limit) { // // This locus is a homozygote. // tag->snps[col]->type = snp_type_hom; tag->snps[col]->col = col; tag->snps[col]->lratio = l_ratio; tag->snps[col]->rank_1 = nuc[0].first; tag->snps[col]->rank_2 = '-'; res = snp_type_hom; } else { // // Unknown whether this is a heterozygote or homozygote. // tag->snps[col]->type = snp_type_unk; tag->snps[col]->col = col; tag->snps[col]->lratio = l_ratio; tag->snps[col]->rank_1 = nuc[0].first; tag->snps[col]->rank_2 = nuc[1].first; res = snp_type_unk; } return res; } int call_multinomial_fixed (MergedStack *tag, int col, map &n) { const double nucleotide_fixed_limit = 1.92; vector > nuc; map::iterator i; int total = 0; for (i = n.begin(); i != n.end(); i++) { if (i->first != 'N') { total += i->second; nuc.push_back(make_pair(i->first, i->second)); } } sort(nuc.begin(), nuc.end(), compare_pair); if (nuc[0].second == 0) { SNP *snp = new SNP; snp->type = snp_type_unk; snp->col = col; snp->lratio = 0; snp->rank_1 = 'N'; snp->rank_2 = '-'; tag->snps.push_back(snp); return snp_type_unk; } // // Method of Paul Hohenlohe , personal communication. // // Each population sample contains DNA from 6 individuals, so a // sample of 12 alleles from the population. We want to assign a // nucleotide (A,C,G,T) to each position where the population is // fixed or nearly so, and N to each position that is either // polymorphic within the population or has insufficient coverage // depth to make a call. We can do this with a likelihood ratio // test of the read counts, testing whether the allele frequency // of the dominant allele is significantly larger than some // threshold p) , stepping through each nucleotide position across // RAD tags. // double nuc_1 = nuc[0].second; double nuc_2 = nuc[1].second; double n_ratio = 0.0; double l_ratio = 0.0; double epsilon = -1 * (log(1 - barcode_err_freq) / barcode_size); n_ratio = nuc_1 / (nuc_1 + nuc_2); l_ratio = nuc_1 * log( ((4 * nuc_1 * (1 - epsilon)) + ((nuc_1 + nuc_2) * epsilon)) / ((4 * p_freq * (nuc_1 + nuc_2) * (1 - epsilon)) + ((nuc_1 + nuc_2) * epsilon)) ); l_ratio += nuc_2 * log( ((4 * nuc_2 * (1 - epsilon)) + ((nuc_1 + nuc_2) * epsilon)) / ((4 * (1 - p_freq) * (nuc_1 + nuc_2) * (1 - epsilon)) + ((nuc_1 + nuc_2) * epsilon)) ); //cerr << "Nuc_1: " << nuc_1 << " Nuc_2: " << nuc_2 << " Likelihood ratio: " << l_ratio << "\n"; if (n_ratio < p_freq || l_ratio < nucleotide_fixed_limit) { // // This position is likely a SNP, record it's homozygosity as 'unknown'. // SNP *snp = new SNP; snp->type = snp_type_unk; snp->col = col; snp->lratio = l_ratio; snp->rank_1 = nuc[0].first; snp->rank_2 = nuc[1].first; tag->snps.push_back(snp); } else { // // Otherwise, this position is homozygous. // SNP *snp = new SNP; snp->type = snp_type_hom; snp->col = col; snp->lratio = l_ratio; snp->rank_1 = nuc[0].first; snp->rank_2 = nuc[1].first; tag->snps.push_back(snp); } return 0; } // // ln L(1/2) = ln(n! / n_1!n_2!n_3!n_4!) + // (n_1 + n_2) * ln(n_1 + n_2 / 2n) + // (n_3 + n_4) * ln(n_3 + n_4 / 2n) // double heterozygous_likelihood(int col, map &nuc) { vector > cnts; map::iterator i; double n = 0; for (i = nuc.begin(); i != nuc.end(); i++) { n += i->second; cnts.push_back(make_pair(i->first, i->second)); } sort(cnts.begin(), cnts.end(), compare_pair); double n_1 = cnts[0].second; double n_2 = cnts[1].second; double n_3 = cnts[2].second; double n_4 = cnts[3].second; double term_1 = reduced_log_factorial(n, n_1) - (log_factorial(n_2) + log_factorial(n_3) + log_factorial(n_4)); double term_3 = (n_3 + n_4 > 0) ? log((n_3 + n_4) / (2 * n)) : 0; double lnl = term_1 + ((n_1 + n_2) * log((n_1 + n_2) / (2 * n))) + ((n_3 + n_4) * term_3); return lnl; } // // ln L(1/1) = ln(n! / n_1!n_2!n_3!n_4!) + // n_1 * ln(n_1 / n) + // (n - n_1) * ln(n - n_1 / 3n) // double homozygous_likelihood(int col, map &nuc) { vector > cnts; map::iterator i; double n = 0; for (i = nuc.begin(); i != nuc.end(); i++) { n += i->second; cnts.push_back(make_pair(i->first, i->second)); } sort(cnts.begin(), cnts.end(), compare_pair); double n_1 = cnts[0].second; double n_2 = cnts[1].second; double n_3 = cnts[2].second; double n_4 = cnts[3].second; double term_1 = reduced_log_factorial(n, n_1) - (log_factorial(n_2) + log_factorial(n_3) + log_factorial(n_4)); double term_3 = n - n_1 > 0 ? log((n - n_1) / (3 * n)) : 0; double lnl = term_1 + (n_1 * log(n_1 / n)) + ((n - n_1) * term_3); return lnl; } stacks-1.44/src/models.h000644 000765 000024 00000004153 12441417455 015737 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2012, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __MODELS_H__ #define __MODELS_H__ #include #include #include using std::string; #include using std::vector; #include using std::map; #include using std::pair; using std::make_pair; #include #include using std::ifstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include "constants.h" #include "utils.h" #include "mstack.h" #include "locus.h" // // Possible models for calling nucleotide positions as fixed or variable // enum modelt {fixed, snp, bounded}; // // For use with the multinomial model to call fixed nucleotides. // extern const int barcode_size; extern modelt model_type; extern double p_freq; extern double barcode_err_freq; extern double bound_low; extern double bound_high; extern double heterozygote_limit; extern double homozygote_limit; snp_type call_bounded_multinomial_snp(MergedStack *, int, map &, bool); snp_type call_bounded_multinomial_snp(Locus *, int, map &); snp_type call_multinomial_snp(MergedStack *, int, map &, bool); snp_type call_multinomial_snp(Locus *, int, map &); int call_multinomial_fixed(MergedStack *, int, map &); double heterozygous_likelihood(int, map &); double homozygous_likelihood(int, map &); #endif // __MODELS_H__ stacks-1.44/src/mst.cc000644 000765 000024 00000016517 12764047443 015430 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // mst.cc -- routines to implement the Minimum Spanning Tree Class:. // // Julian Catchen // jcatchen@uoregon.edu // University of Oregon // // $Id$ // #include "mst.h" Edge *Node::add_edge(Node *n, int dist) { Edge *e = new Edge; e->child = n; e->dist = dist; this->edges.push_back(e); return e; } bool min_span_tree_cmp(const Node *lhs, const Node *rhs) { return (lhs->min_dist > rhs->min_dist); } Node *MinSpanTree::add_node(int id) { Node *n = new Node(id); this->nodes[id] = n; return n; } Node *MinSpanTree::add_node(string label) { // // Obtain an ID for this node. // uint id = this->id_cnt; Node *n = new Node(id); n->label = label; this->nodes[id] = n; this->node_key[label] = id; this->id_cnt++; return n; } Node *MinSpanTree::node(int id) { return this->nodes[id]; } Node *MinSpanTree::node(string label) { uint id = this->node_key[label]; return this->nodes[id]; } Node *MinSpanTree::head() { return this->nodes.begin()->second; } int MinSpanTree::node_count() { return this->nodes.size(); } bool MinSpanTree::connected(int *ids, int size) { set valid, visited; queue q; if (size == 1) return true; for (int i = 0; i < size; i++) valid.insert(ids[i]); // // Take the first ID and begin traversing the tree. If we hit // a node not in the ids set, stop traversing this branch. Check that // all nodes are directly connected. // int valid_cnt = 0; Node *n = this->node(ids[0]); q.push(n); while (!q.empty() && valid_cnt < size) { n = q.front(); q.pop(); visited.insert(n->id); if (valid.count(n->id)) { valid_cnt++; for (uint i = 0; i < n->min_adj_list.size(); i++) if (visited.count(n->min_adj_list[i]->id) == false) q.push(n->min_adj_list[i]); } } if (valid_cnt == size) return true; else return false; } // // Build a minimum spanning tree using Prim's alogorithm. Assume all necessary // nodes have been added using the add_node function. // int MinSpanTree::build_tree() { // // Vector, which we treat as a binary heap to access nodes that are of minimal distance // vector q; // // Select an initial node to process and initialize its minimum distance. // Node *n = this->nodes.begin()->second; n->min_dist = 0; // // Add all of the nodes to the binary heap; process them in order of min_dist // map::iterator it; for (it = this->nodes.begin(); it != this->nodes.end(); it++) q.push_back((*it).second); make_heap(q.begin(), q.end(), min_span_tree_cmp); while (q.size() > 0) { n = q.front(); pop_heap(q.begin(), q.end()); q.pop_back(); n->update = false; //cerr << "Examining node: " << n->id << " (" << n->min_dist << ")\n"; // // Record the minimum connection between parent and n. // if (n->parent != NULL) { n->parent->min_adj_list.push_back(n); n->min_adj_list.push_back(n->parent); } // // Iterate through all of the edges of n and update the // minimum distance to the proper nodes. // Edge *e; for (uint i = 0; i < n->edges.size(); i++) { e = n->edges[i]; if (e->child->update == true && e->dist < e->child->min_dist) { e->child->parent = n; e->child->min_dist = e->dist; //cerr << " Updating node: " << e->child->id << " to have distance: " << e->child->min_dist << "\n"; } } // // Resort the heap after possibly changing many min_dist values // make_heap(q.begin(), q.end(), min_span_tree_cmp); } return 0; } string MinSpanTree::vis(bool overlay) { uint j; double d, scale, scaled_d; char label[32]; int scale_factor = 20; // // Output a specification to visualize the minimum spanning tree using graphviz: // http://www.graphviz.org/ // stringstream data; data << "graph stacks_" << this->nodes.size() << " {\n" << "rankdir=LR\n" << "size=\"" << scale_factor << "!\"\n" << "overlap=false\n" << "node [shape=circle style=filled fillcolor=\"#3875d7\" fontname=\"Arial\"];\n" << "edge [fontsize=8.0 fontname=\"Arial\" color=\"#aaaaaa\"];\n"; map::iterator i; set visited; queue q; // // If overlay==true, write the minimum spanning tree on top of the full tree as a subgraph. // data << "subgraph mst {\n" << " edge [penwidth=5 fontsize=12.0 fontcolor=\"black\" color=\"black\"]\n" << " node [fillcolor=\"red\" fontcolor=\"white\"]\n"; Node *n = this->head(); q.push(n); while (!q.empty()) { n = q.front(); q.pop(); visited.insert(n->id); for (uint i = 0; i < n->min_adj_list.size(); i++) { data << " "; n->label.length() > 0 ? data << n->label : data << n->id; data << "--"; n->min_adj_list[i]->label.length() > 0 ? (data << n->min_adj_list[i]->label) : (data << n->min_adj_list[i]->id); data << "\n"; if (visited.count(n->min_adj_list[i]->id) == 0) q.push(n->min_adj_list[i]); } } data << "}\n"; // // Scale the graph to display on a scale_factor inch canvas. Find the largest edge weight // and scale the edge lengths to fit the canvas. // for (i = this->nodes.begin(); i != this->nodes.end(); i++) { n = i->second; for (j = 0; j < n->edges.size(); j++) scale = n->edges[j]->dist > scale ? n->edges[j]->dist : scale; } scale = scale / scale_factor; // // Write out edges. // for (i = this->nodes.begin(); i != this->nodes.end(); i++) { n = i->second; for (j = 0; j < n->edges.size(); j++) { d = n->edges[j]->dist; scaled_d = d / scale; scaled_d = scaled_d < 0.75 ? 0.75 : scaled_d; sprintf(label, "%.1f", d); n->label.length() > 0 ? (data << n->label) : (data << n->id); data << " -- "; n->edges[j]->child->label.length() > 0 ? (data << n->edges[j]->child->label) : (data << n->edges[j]->child->id); data << " [len=" << scaled_d << ", label=" << label << "];\n"; } } data << "}\n"; return data.str(); } stacks-1.44/src/mst.h000644 000765 000024 00000004537 12764047443 015271 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __MST_H__ #define __MST_H__ #include #include using std::string; #include using std::map; #include using std::vector; #include using std::set; #include using std::queue; #include using std::stringstream; #include using std::cerr; #include "constants.h" typedef unsigned int uint; class Node; class Edge { public: uint dist; // Distance or weight Node *child; }; class Node { public: uint id; string label; vector edges; Node *parent; bool update; uint min_dist; // // List of adjacent nodes that are connected by minimal distance // vector min_adj_list; Node(uint id) { this->id = id; this->parent = NULL; this->update = true; this->min_dist = 1000000; } ~Node() { for (uint i = 0; i < this->edges.size(); i++) delete this->edges[i]; } Edge *add_edge(Node *, int); }; class MinSpanTree { map nodes; map node_key; uint id_cnt; public: MinSpanTree() { id_cnt = 0; } ~MinSpanTree() { for (uint i = 0; i < this->nodes.size(); i++) delete this->nodes[i]; } Node *add_node(int id); Node *add_node(string label); int build_tree(); int node_count(); Node *node(int id); Node *node(string label); Node *head(); bool connected(int *, int); string vis(bool); }; bool min_span_tree_cmp(const Node *, const Node *); #endif // __MST_H__ stacks-1.44/src/mstack.cc000644 000765 000024 00000015644 12774470747 016117 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // mstack.cc -- implementation of the MergedStack Class // // Julian Catchen // jcatchen@uoregon.edu // University of Oregon // // $Id: mstack.cc 1987 2010-11-01 05:43:50Z catchen $ // #include "mstack.h" #include "models.h" MergedStack::MergedStack() { this->id = 0; this->count = 0; this->len = 0; this->con = NULL; this->matrix = NULL; this->lnl = 0.0; this->cohort_id = -1; this->deleveraged = false; this->masked = false; this->blacklisted = false; this->lumberjackstack = false; this->gappedlumberjack = false; } MergedStack::~MergedStack() { delete [] this->con; for (uint i = 0; i < snps.size(); i++) delete this->snps[i]; delete [] this->matrix; } int MergedStack::add_consensus(const char *seq) { if (this->con != NULL) delete [] this->con; this->len = strlen(seq); this->con = new char[len + 1]; strncpy(this->con, seq, len); this->con[len] = '\0'; return 0; } int MergedStack::add_consensus(DNASeq *seq) { if (this->con != NULL) delete [] this->con; this->len = seq->size; this->con = new char[this->len + 1]; this->con = seq->seq(this->con); return 0; } int MergedStack::add_consensus(DNANSeq *seq) { if (this->con != NULL) delete [] this->con; this->len = seq->size(); this->con = new char[this->len + 1]; this->con = seq->seq(this->con); return 0; } int MergedStack::add_dist(const int id, const int dist) { // // Store the ID and distance as a pair, ID in the first position, // dist in the second. // pair p(id, dist); this->dist.push_back(p); return 0; } DNANSeq ** MergedStack::gen_matrix(map &unique, map &rem) { Stack *tag; // // Create a two-dimensional array, each row containing one read. For // each unique tag that has been merged together, add the sequence for // that tag into our array as many times as it originally occurred. // // We do not allocate memory for the second dimension of the array, we simply // reuse the existing char arrays in the unique and rem maps // uint cnt = this->count + this->remtags.size(); if (this->matrix != NULL) delete [] this->matrix; this->matrix = new DNANSeq * [cnt]; vector::iterator j; int i = 0; for (j = this->utags.begin(); j != this->utags.end(); j++) { tag = unique[*j]; for (uint k = 0; k < tag->count(); k++) { this->matrix[i] = tag->seq; i++; } } // For each remainder tag that has been merged into this Stack, add the sequence. for (j = this->remtags.begin(); j != this->remtags.end(); j++) { this->matrix[i] = rem[*j]->seq; i++; } return this->matrix; } DNANSeq ** MergedStack::gen_matrix(map &unique) { PStack *tag; // // Create a two-dimensional array, each row containing one read. For // each unique tag that has been merged together, add the sequence for // that tag into our array as many times as it originally occurred. // // We do not allocate memory for the second dimension of the array, we simply // reuse the existing char arrays in the unique and rem maps // uint cnt = this->count; if (this->matrix != NULL) delete [] this->matrix; this->matrix = new DNANSeq * [cnt]; vector::iterator j; int i = 0; for (j = this->utags.begin(); j != this->utags.end(); j++) { tag = unique[*j]; for (uint k = 0; k < tag->count; k++) { this->matrix[i] = tag->seq; i++; } } return this->matrix; } double MergedStack::calc_likelihood() { if (this->matrix == NULL || this->snps.size() == 0) return 0; // // Iterate over each column of the array and call the consensus base. // int row, col, tot; int length = this->matrix[0]->size(); int height = this->count + this->remtags.size(); map nuc; map::iterator max, n; DNANSeq *d; uint cur_gap = this->gaps.size() > 0 ? 0 : 1; this->lnl = 0; for (col = 0; col < length; col++) { // // Don't invoke the model within gaps. // if (cur_gap < gaps.size() && col == (int)gaps[cur_gap].start) { do { col++; } while (col < (int)gaps[cur_gap].end && col < length); col--; continue; } nuc['A'] = 0; nuc['G'] = 0; nuc['C'] = 0; nuc['T'] = 0; // // Count the nucleotide type at each position in the column. // for (row = 0; row < height; row++) { d = this->matrix[row]; nuc[(*d)[col]]++; } // // Find the base with a plurality of occurances and call it. // max = nuc.end(); tot = 0; for (n = nuc.begin(); n != nuc.end(); n++) { tot += n->second; if (max == nuc.end() || n->second > max->second) max = n; } // // For nucleotide positions with potential polymorphism (i.e. two or more alleles at // the locus that differ at that position), first find the ML genotype (call_multinomial_snp). // If it returns 'het' calculate the heterozygous_likelihood(), otherwise calculate homozygous // likelihood. // snp_type res = this->snps[col]->type; if (res == snp_type_het) this->lnl += heterozygous_likelihood(col, nuc); else if (res == snp_type_hom) this->lnl += homozygous_likelihood(col, nuc); else { double homlnl = homozygous_likelihood(col, nuc); double hetlnl = heterozygous_likelihood(col, nuc); this->lnl += hetlnl > homlnl ? hetlnl : homlnl; } } return this->lnl; } string MergedStack::write_cmb() { stringstream s; uint size = this->utags.size(); s << "{"; for (uint i = 0; i < size; i++) { s << this->utags[i]; if (i < size - 1) s << ", "; } s << "}"; return s.str(); } stacks-1.44/src/mstack.h000644 000765 000024 00000005464 12701475247 015746 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2012, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __MSTACK_H__ #define __MSTACK_H__ #include using std::string; #include using std::vector; #include using std::map; #include using std::pair; using std::make_pair; #include using std::cerr; #include "stacks.h" class MergedStack { public: int id; // Identifier for the merged stack. char *con; // Consensus sequence uint len; // Sequence length // // Stack component parts // int count; // Number of merged stacks vector utags; // Stack IDs that have been merged into this MergedStack vector remtags; // Remainder tag IDs that have been merged into this Stack DNANSeq **matrix; // Two-dimensional array for iterating over the combined stack (stacks and remainders). vector > dist; // Vector describing the distance between this stack and other stacks. vector alns; // Vector describing gapped alignments between this stack and other stacks. int cohort_id; // Group ID of all stacks that were originally part of the same subgraph double lnl; // Log likelihood of this stack // // Mapping components // PhyLoc loc; // Physical genome location of this Stack. vector snps; // Single Nucleotide Polymorphisms found in this Stack map alleles; // Set of alleles defined by the SNPs found in this Stack vector gaps; // // Flags // bool deleveraged; bool masked; bool blacklisted; bool gappedlumberjack; bool lumberjackstack; MergedStack(); ~MergedStack(); int add_consensus(const char *); int add_consensus(DNASeq *); int add_consensus(DNANSeq *); int add_dist(const int id, const int dist); DNANSeq **gen_matrix(map &, map &); DNANSeq **gen_matrix(map &); double calc_likelihood(); string write_cmb(); }; #endif // __MSTACK_H__ stacks-1.44/src/ordered.h000644 000765 000024 00000030701 12764047443 016102 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2014, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __ORDERED_H__ #define __ORDERED_H__ #include #include using std::ifstream; using std::ofstream; using std::cin; using std::cout; using std::cerr; #include using std::vector; #include using std::map; #include using std::set; #include "MetaPopInfo.h" #include "PopSum.h" extern MetaPopInfo mpopi; enum loc_type {haplotype, snp}; template class Ordered { public: ofstream *log_fh; PopSum *psum; int incompatible_loci; int multiple_loci; Ordered() { } virtual ~Ordered() { } int init_sites(vector &, map &, vector &); int init_sites(vector &, map &, vector &, int); int init_sites(vector &, map &, vector &, int, int); int init_haplotypes(vector &, map &, vector &); }; template int Ordered::init_sites(vector &sites, map &sites_key, vector &sorted_loci) { CSLocus *loc; LocTally *ltally; int len; set bps; // // We need to create an array to store all the SNPs for exporting. We must // account for positions in the genome that are covered by more than one RAD tag. // for (uint pos = 0; pos < sorted_loci.size(); pos++) { loc = sorted_loci[pos]; len = strlen(loc->con); ltally = this->psum->locus_tally(loc->id); for (int k = 0; k < len; k++) { if (ltally->nucs[k].allele_cnt == 2) bps.insert(ltally->nucs[k].bp); } } sites.resize(bps.size(), NULL); // // Create a key describing where in the sites array to find each basepair coordinate. // set::iterator it; int i = 0; for (it = bps.begin(); it != bps.end(); it++) { sites_key[*it] = i; i++; } return 0; } template int Ordered::init_sites(vector &sites, map &sites_key, vector &sorted_loci, int pop_id) { CSLocus *loc; LocSum *lsum; int len; set bps; // // We need to create an array to store all the summary statistics for smoothing. We must // account for positions in the genome that are covered by more than one RAD tag. // for (uint pos = 0; pos < sorted_loci.size(); pos++) { loc = sorted_loci[pos]; len = strlen(loc->con); lsum = this->psum->pop(loc->id, pop_id); for (int k = 0; k < len; k++) { if (lsum->nucs[k].num_indv > 0) bps.insert(lsum->nucs[k].bp); } } sites.resize(bps.size(), NULL); // // Create a key describing where in the sites array to find each basepair coordinate. // set::iterator it; int i = 0; for (it = bps.begin(); it != bps.end(); it++) { sites_key[*it] = i; i++; } return 0; } template int Ordered::init_sites(vector &sites, map &sites_key, vector &sorted_loci, int pop_id_1, int pop_id_2) { CSLocus *loc; LocSum *lsum_1, *lsum_2; int len; set bps; // // We need to create an array to store all the pair values for computing smoothed Fst. We must // account for positions in the genome that are covered by more than one RAD tag. // for (uint pos = 0; pos < sorted_loci.size(); pos++) { loc = sorted_loci[pos]; len = strlen(loc->con); lsum_1 = this->psum->pop(loc->id, pop_id_1); lsum_2 = this->psum->pop(loc->id, pop_id_2); for (int k = 0; k < len; k++) { if (lsum_1->nucs[k].num_indv > 0 && lsum_2->nucs[k].num_indv > 0) bps.insert(lsum_1->nucs[k].bp); // slow } } sites.resize(bps.size(), NULL); // // Create a key describing where in the sites array to find each basepair coordinate. // set::iterator it; int i = 0; for (it = bps.begin(); it != bps.end(); it++) { sites_key[*it] = i; // slow i++; } return 0; } template int Ordered::init_haplotypes(vector &sites, map &sites_key, vector &sorted_loci) { CSLocus *loc; int bp; set bps; for (uint pos = 0; pos < sorted_loci.size(); pos++) { loc = sorted_loci[pos]; bp = loc->sort_bp(); bps.insert(bp); } sites.resize(bps.size(), NULL); // // Create a key describing where in the sites array to find each basepair coordinate. // set::iterator it; int i = 0; for (it = bps.begin(); it != bps.end(); it++) { sites_key[*it] = i; i++; } return 0; } template class OHaplotypes: public Ordered { public: OHaplotypes(): Ordered() { } int order(vector &, map &, vector &); }; template int OHaplotypes::order(vector &sites, map &sites_key, vector &sorted_loci) { this->init_haplotypes(sites, sites_key, sorted_loci); return 0; }; template class OPopPair: public Ordered { public: OPopPair(PopSum *psum, ofstream &log_fh): Ordered() { this->log_fh = &log_fh; this->psum = psum; } int order(vector &, map &, vector &, int, int); }; template int OPopPair::order(vector &sites, map &sites_key, vector &sorted_loci, int pop_1, int pop_2) { CSLocus *loc; StatT *pair; int len; this->incompatible_loci = 0; this->multiple_loci = 0; this->init_sites(sites, sites_key, sorted_loci, pop_1, pop_2); for (uint pos = 0; pos < sorted_loci.size(); pos++) { loc = sorted_loci[pos]; len = strlen(loc->con); for (int k = 0; k < len; k++) { pair = this->psum->Fst(loc->id, pop_1, pop_2, k); // // Locus is incompatible, log this position. // if (pair == NULL) { this->incompatible_loci++; *(this->log_fh) << "between_population\t" << "incompatible_locus\t" << loc->id << "\t" << loc->loc.chr << "\t" << loc->sort_bp(k) +1 << "\t" << k << "\t" << mpopi.pops()[pop_1].name << "\t" << mpopi.pops()[pop_2].name << "\n"; delete pair; continue; } pair->loc_id = loc->id; pair->bp = loc->sort_bp(k); pair->col = k; // // Locus is fixed in both populations, or was only found in one population. // if (pair->pi == 0) { delete pair; continue; } // // Check if this basepair position is already covered by a RAD site. // if (sites[sites_key[pair->bp]] != NULL) { this->multiple_loci++; *(this->log_fh) << "between_population\t" << "multiple_locus\t" << loc->id << "\t" << loc->loc.chr << "\t" << pair->bp +1 << "\t" << k << "\t" << mpopi.pops()[pop_1].name << "\t" << mpopi.pops()[pop_2].name << "\n"; delete pair; continue; } sites[sites_key[pair->bp]] = pair; } } return 0; }; template class OSumStat: public Ordered { public: OSumStat(PopSum *psum, ofstream &log_fh): Ordered() { this->log_fh = &log_fh; this->psum = psum; } int order(vector &, vector &, int); }; template int OSumStat::order(vector &sites, vector &sorted_loci, int pop_id) { this->incompatible_loci = 0; this->multiple_loci = 0; map sites_key; this->init_sites(sites, sites_key, sorted_loci, pop_id); CSLocus *loc; LocSum *lsum; int len; // // Assign nucleotides to their proper, ordered location in the genome, // checking that a site hasn't already been covered by another RAD locus. // for (uint pos = 0; pos < sorted_loci.size(); pos++) { loc = sorted_loci[pos]; len = strlen(loc->con); lsum = this->psum->pop(loc->id, pop_id); for (int k = 0; k < len; k++) { if (lsum->nucs[k].num_indv == 0) continue; if (sites_key.count(lsum->nucs[k].bp) == 0) { cerr << "Error: locus " << lsum->nucs[k].loc_id << " at " << lsum->nucs[k].bp +1 << "bp is not defined in the sites map.\n"; } else if (sites[sites_key[lsum->nucs[k].bp]] == NULL) { sites[sites_key[lsum->nucs[k].bp]] = &(lsum->nucs[k]); } else { this->multiple_loci++; *(this->log_fh) << "within_population\t" << "multiple_locus\t" << loc->id << "\t" << loc->loc.chr << "\t" << lsum->nucs[k].bp +1 << "\t" << k << "\t" << mpopi.pops()[pop_id].name << "\t" << "conflicts with locus " << sites[sites_key[lsum->nucs[k].bp]]->loc_id << "\n"; } } } return 0; }; template class OLocTally: public Ordered { public: OLocTally(PopSum *psum, ofstream &log_fh): Ordered() { this->log_fh = &log_fh; this->psum = psum; } int order(vector &, vector &); }; template int OLocTally::order(vector &sites, vector &sorted_loci) { this->incompatible_loci = 0; this->multiple_loci = 0; map sites_key; this->init_sites(sites, sites_key, sorted_loci); CSLocus *loc; LocTally *ltally; int len; // // Assign nucleotides to their proper, ordered location in the genome, // checking that a site hasn't already been covered by another RAD locus. // for (uint pos = 0; pos < sorted_loci.size(); pos++) { loc = sorted_loci[pos]; len = strlen(loc->con); ltally = this->psum->locus_tally(loc->id); for (int k = 0; k < len; k++) { if (ltally->nucs[k].allele_cnt != 2) continue; if (sites_key.count(ltally->nucs[k].bp) == 0) { cerr << "Error: locus " << ltally->nucs[k].loc_id << " at " << ltally->nucs[k].bp +1 << "bp is not defined in the sites map.\n"; } else if (sites[sites_key[ltally->nucs[k].bp]] == NULL) { sites[sites_key[ltally->nucs[k].bp]] = &(ltally->nucs[k]); } else { this->multiple_loci++; *(this->log_fh) << "within_population\t" << "multiple_locus\t" << loc->id << "\t" << loc->loc.chr << "\t" << ltally->nucs[k].bp +1 << "\t" << k << "\t" << "conflicts with locus " << sites[sites_key[ltally->nucs[k].bp]]->loc_id << "\n"; } } } return 0; }; #endif // __ORDERED_H__ stacks-1.44/src/phasedstacks.cc000644 000765 000024 00000156774 12764047443 017314 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2013-2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // phasedstacks -- analyse phased data, descended from a Stacks analysis. // #include "phasedstacks.h" // Global variables to hold command-line options. FileT in_file_type = FileT::unknown; int num_threads = 1; int batch_id = 0; string cat_path; string in_path; string out_path; string out_file; string pmap_path; bool haplotypes = false; bool write_zeros = true; double p_value_cutoff = 0.05; double chi_sq_limit = 3.84; double minor_freq_lim = 0.1; double min_inform_pairs = 0.90; uint max_pair_dist = 1000000; uint bucket_dist = 5000; double dprime_threshold = false; double dprime_threshold_level = 0.0; set whitelist, blacklist; map pop_map; map pop_cnts; int main (int argc, char* argv[]) { parse_command_line(argc, argv); if (p_value_cutoff == 0.1) { chi_sq_limit = 2.71; } else if (p_value_cutoff == 0.05) { chi_sq_limit = 3.84; } else if (p_value_cutoff == 0.01) { chi_sq_limit = 6.64; } else if (p_value_cutoff == 0.001) { chi_sq_limit = 10.83; } cerr << "Minor allele frequency cutoff: " << minor_freq_lim << "\n" << "Looking for "; switch(in_file_type) { case FileT::beagle: cerr << "Beagle"; break; case FileT::phase: cerr << "PHASE"; break; case FileT::fastphase: default: cerr << "fastPhase"; break; } cerr << " input files.\n" << "Size of buckets for binning D' values at a particular distance: " << bucket_dist / 1000 << "kb.\n"; if (dprime_threshold) cerr << "D' Threshold set at " << dprime_threshold_level << ". D' values above this limit will be set to 1.0, values below will be set to 0.0.\n"; // // Parse the population map. // parse_population_map(pmap_path, pop_map, pop_cnts); // // Set the number of OpenMP parallel threads to execute. // #ifdef _OPENMP omp_set_num_threads(num_threads); #endif vector > files; if (!build_file_list(files)) exit(1); cerr << "Identified " << files.size() << " files.\n"; // // Open the log file. // stringstream log; log << "phasedstacks.log"; string log_path = in_path + log.str(); ofstream log_fh(log_path.c_str(), ofstream::out); if (log_fh.fail()) { cerr << "Error opening log file '" << log_path << "'\n"; exit(1); } init_log(log_fh, argc, argv); // // Load the catalog // cerr << "Parsing the catalog...\n"; stringstream catalog_file; map catalog; bool compressed = false; int res; catalog_file << cat_path << "batch_" << batch_id << ".catalog"; if ((res = load_loci(catalog_file.str(), catalog, false, false, compressed)) == 0) { cerr << "Unable to load the catalog '" << catalog_file.str() << "'\n"; return 0; } cerr << "done.\n"; // // Implement the black/white list // reduce_catalog(catalog, whitelist, blacklist); map fgt_block_lens, fgt_snp_cnts; map dp_block_lens, dp_snp_cnts; // // Vectors to store D' measures of SNPs at bucketed distances. // vector dprime_buckets, dprime_bucket_cnts; for (uint i = 0; i < files.size(); i++) { // if (files[i].second != "batch_1.groupV.phase") continue; PhasedSummary *psum = NULL; if (in_file_type == FileT::fastphase) { if ((psum = parse_fastphase(in_path + files[i].second)) == NULL) { cerr << "Unable to parse fastPhase input files.\n"; exit(1); } } else if (in_file_type == FileT::beagle && haplotypes) { if ((psum = parse_beagle_haplotypes(catalog, in_path + files[i].second)) == NULL) { cerr << "Unable to parse Beagle input files.\n"; exit(1); } } else if (in_file_type == FileT::beagle) { if ((psum = parse_beagle(catalog, in_path + files[i].second)) == NULL) { cerr << "Unable to parse Beagle input files.\n"; exit(1); } } // // Summarize the genotypes in the populations. // summarize_phased_genotypes(psum); // for (uint j = 0; j < psum->size; j++) { // cerr << "BP: " << psum->nucs[j].bp << "\t" // << "A: " << std::setw(3) << psum->nucs[j].nuc[0] << " " // << "C: " << std::setw(3) << psum->nucs[j].nuc[1] << " " // << "G: " << std::setw(3) << psum->nucs[j].nuc[2] << " " // << "T: " << std::setw(3) << psum->nucs[j].nuc[3] << "\n"; // } // // Calculate D' // cerr << "Calculating D'..."; calc_dprime(psum); cerr << "done.\n"; write_dprime(in_path + files[i].second, psum); // // Generate haplotype blocks based on D'. // dprime_blocks(in_path + files[i].second, pop_map, psum, dp_block_lens, dp_snp_cnts); // // Generate haplotype blocks using the four gamete test. // four_gamete_test(in_path + files[i].second, pop_map, psum, fgt_block_lens, fgt_snp_cnts); // // Bucket the D' measures by distance between SNPs. // bucket_dprime(dprime_buckets, dprime_bucket_cnts, psum); // // Free the Samples objects // delete psum; } // // Write average D' values bucketed according to their distance in the genome. // write_buckets(in_path, dprime_buckets, dprime_bucket_cnts); // // Write the FGT bucketed distances. // log_fh << "# Distribution of FGT haplotype block lengths.\n"; map::iterator buck_it; for (buck_it = fgt_block_lens.begin(); buck_it != fgt_block_lens.end(); buck_it++) log_fh << buck_it->first << "\t" << buck_it->second << "\n"; // // Write the FGT bucketed SNP counts. // log_fh << "\n\n" << "# Distribution of FGT SNP counts per haplotype block.\n"; for (buck_it = fgt_snp_cnts.begin(); buck_it != fgt_snp_cnts.end(); buck_it++) log_fh << buck_it->first << "\t" << buck_it->second << "\n"; // // Write the D' haplotype block bucketed distances. // log_fh << "\n\n" << "# Distribution of D' haplotype block lengths.\n"; for (buck_it = dp_block_lens.begin(); buck_it != dp_block_lens.end(); buck_it++) log_fh << buck_it->first << "\t" << buck_it->second << "\n"; // // Write the D' bucketed SNP counts. // log_fh << "\n\n" << "# Distribution of D' SNP counts per haplotype block.\n"; for (buck_it = dp_snp_cnts.begin(); buck_it != dp_snp_cnts.end(); buck_it++) log_fh << buck_it->first << "\t" << buck_it->second << "\n"; log_fh.close(); return 0; } int bucket_dprime(vector &dprime_buckets, vector &dprime_bucket_cnts, PhasedSummary *psum) { uint bucket, dist, max_bucket; uint max_dist = 0; // // Check that we have enough buckets in our vectors. Find the maximum distance between // SNPs on this chromosome and add buckets as necessary. // for (uint i = 0; i < psum->size; i++) { for (uint j = i+1; j < psum->size; j++) { if (psum->nucs[i].freq < minor_freq_lim || psum->nucs[j].freq < minor_freq_lim) continue; if (write_zeros == false && psum->dprime[i][j].chisq_p == false) continue; dist = psum->nucs[j].bp - psum->nucs[i].bp; max_dist = dist > max_dist ? dist : max_dist; } } max_bucket = max_dist / bucket_dist; if (dprime_buckets.size() < max_bucket) { uint cnt = max_bucket + 1 - dprime_buckets.size(); for (uint i = 0; i < cnt; i++) { dprime_buckets.push_back(0.0); dprime_bucket_cnts.push_back(0.0); } } // // Populate buckets // for (uint i = 0; i < psum->size; i++) { for (uint j = i+1; j < psum->size; j++) { if (psum->nucs[i].freq < minor_freq_lim || psum->nucs[j].freq < minor_freq_lim) continue; if (write_zeros == false && psum->dprime[i][j].chisq_p == false) continue; bucket = ((psum->nucs[j].bp - psum->nucs[i].bp) / bucket_dist); dprime_buckets[bucket] += (psum->dprime[i][j].chisq_p ? psum->dprime[i][j].dprime : 0.0); dprime_bucket_cnts[bucket]++; } } return 0; } int write_buckets(string path, vector &dprime_buckets, vector &dprime_bucket_cnts) { // // Write the bucketed D' data for plotting. // stringstream file; file << path << "Dprime_dist_buckets" << bucket_dist/1000 << "kb.tsv"; cerr << "Writing bucketed D' data to '" << file.str() << "'..."; ofstream fh(file.str().c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening D' file '" << file.str() << "'\n"; exit(1); } fh << "# Distance (Kb)\tD' Average\n"; for (uint i = 0; i < dprime_buckets.size(); i++) fh << (i * bucket_dist) << "\t" << std::setprecision(3) << (dprime_buckets[i] / dprime_bucket_cnts[i]) << "\n"; fh.close(); cerr << "done\n"; return 0; } int four_gamete_test(string path, map &pop_map, PhasedSummary *psum, map &len_buckets, map &snp_buckets) { // // Write haplotypes as found by the four gamete test: // Wang, et al., Am. J. Hum. Genet. 71:1227–1234, 2002 // string file = path + ".fgt.tsv"; cerr << "Determining four gamete test haplotypes blocks, writing to:\n '" << file << "'...\n"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening FGT file '" << file << "'\n"; exit(1); } fh << "# ID\tStart\tEnd\tLen\tSNP Count\tHaplotype Count\tHaplotype\tPopulations\tHapPopCnt\n"; uint id = 1; uint start, end, cnt, dist; bool bound; map buckets, snps; for (uint i = 0; i < psum->size; i++) { if (psum->nucs[i].freq < minor_freq_lim) continue; // // Start a new block. // start = i; bound = false; cnt = 0; uint j = i; do { if (psum->nucs[j].freq < minor_freq_lim) { j++; continue; } for (int k = j; k >= (int) start; k--) { if (psum->nucs[k].freq < minor_freq_lim) continue; if (psum->recomb[k][j] == true) { bound = true; end = j; } } j++; cnt++; } while (bound == false && j < psum->size); if (j == psum->size) end = j - 1; fh << id << "\t" << psum->nucs[start].bp << "\t" << psum->nucs[end].bp << "\t" << psum->nucs[end].bp - psum->nucs[start].bp + 1 << "\t" << cnt << "\t"; // // Bucket the SNP counts for plotting. // snps[cnt]++; // // Bucket the haplotype block lengths for plotting. // dist = (psum->nucs[end].bp - psum->nucs[start].bp + 1) / 10000 * 10000; buckets[dist]++; enumerate_haplotypes(fh, pop_map, psum, start, end); i = end; id++; } // // Write the bucketed distances. // fh << "\n\n" << "# Distribution of FGT haplotype block lengths.\n"; map::iterator it; for (it = buckets.begin(); it != buckets.end(); it++) { fh << it->first << "\t" << it->second << "\n"; len_buckets[it->first] += it->second; } // // Write the bucketed SNP counts. // fh << "\n\n" << "# Distribution of SNP counts per FGT haplotype block.\n"; for (it = snps.begin(); it != snps.end(); it++) { fh << it->first << "\t" << it->second << "\n"; snp_buckets[it->first] += it->second; } fh.close(); cerr << "done.\n"; return 0; } int dprime_blocks(string path, map &pop_map, PhasedSummary *psum, map &len_buckets, map &snp_buckets) { // // Generate haplotype blocks according to strength of linkage disequilibrium measured using D'. // Stacey B. Gabriel et al. (2002). The Structure of Haplotype Blocks in the Human Genome. Science 296:2225-2229 // string file = path + ".dpblocks.tsv"; cerr << "Determining D' haplotypes blocks, writing to:\n '" << file << "'...\n"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening D' blocks file '" << file << "'\n"; exit(1); } fh << "# ID\tStart\tEnd\tLen\tSNP Count\tHaplotype Count\tHaplotype\tPopulations\tHapPopCnt\n"; uint dist; set loci; vector > ld_pairs; map > ld_map; map buckets, snps; uint tot_pairs = 0; uint recomb_pairs = 0; for (uint i = 0; i < psum->size; i++) { if (psum->nucs[i].freq < minor_freq_lim) continue; for (uint j = i+1; j < psum->size; j++) { if (psum->nucs[j].freq < minor_freq_lim) continue; tot_pairs++; dist = psum->nucs[j].bp - psum->nucs[i].bp + 1; // // Does this pair of markers show a strong measure of LD? // if (psum->dprime[i][j].ci_high > 0.98 && psum->dprime[i][j].ci_low > 0.7 && dist <= max_pair_dist) { psum->dprime[i][j].type = strong_ld; ld_pairs.push_back(make_pair(i, j)); ld_map[i].push_back(j); loci.insert(i); loci.insert(j); } // // Does this pair of markers show a strong measure of historical recombination? // if (psum->dprime[i][j].ci_high < 0.9) { psum->dprime[i][j].type = recomb; recomb_pairs++; } } } // map >::iterator it; // for (it = ld_map.begin(); it != ld_map.end(); it++) { // cerr << " " << it->first << " ->\n"; // for (uint i = 0; i < it->second.size(); i++) // cerr << " " << it->second[i] << " dist: " << (psum->nucs[it->second[i]].bp - psum->nucs[it->first].bp + 1) << "bp\n"; // } cerr << " Total pairs examined: " << tot_pairs << "; Strong LD pairs: " << ld_pairs.size() << "; Recombination pairs: " << recomb_pairs << "; Informative markers: " << std::setprecision(3) << ((double) (ld_pairs.size() + recomb_pairs) / (double) tot_pairs) * 100 << "%\n"; // // Convert our list of loci into an ordered, linked list, where each node // represents a haplotype block. // dPrimeBlocks blocks; blocks.initialize(loci); // // Merge nodes together where D' is strong enough to maintain the block. // HBlock *cur; cur = blocks.head(); do { // // Can we merge these two nodes together? // if (check_adjacent_blocks(psum, cur)) { // cerr << " Merging blocks: "; // for (uint i = 0; i < cur->loci.size(); i++) // cerr << cur->loci[i] << ", "; // cerr << " and "; // for (uint i = 0; i < cur->next->loci.size(); i++) // cerr << cur->next->loci[i] << ", "; // cerr << "\n"; blocks.merge_adjacent(cur); } else { cur = cur->next; } } while (cur->next != NULL); // blocks.print(); // // Write the blocks. // uint id = 1; uint start, end; cur = blocks.head(); do { start = *(cur->loci.begin()); end = *(cur->loci.rbegin()); fh << id << "\t" << psum->nucs[start].bp << "\t" << psum->nucs[end].bp << "\t" << psum->nucs[end].bp - psum->nucs[start].bp + 1 << "\t" << cur->loci.size() << "\t"; // // Bucket the SNP counts for plotting. // snps[cur->loci.size()]++; // // Bucket the haplotype block lengths for plotting. // dist = (psum->nucs[end].bp - psum->nucs[start].bp + 1) / 10000 * 10000; buckets[dist]++; enumerate_haplotypes(fh, pop_map, psum, start, end); id++; cur = cur->next; } while (cur != NULL); // // Write the bucketed distances. // fh << "\n\n" << "# Distribution of D' haplotype block lengths.\n"; map::iterator it; for (it = buckets.begin(); it != buckets.end(); it++) { fh << it->first << "\t" << it->second << "\n"; len_buckets[it->first] += it->second; } // // Write the bucketed SNP counts. // fh << "\n\n" << "# Distribution of SNP counts per D' haplotype block.\n"; for (it = snps.begin(); it != snps.end(); it++) { fh << it->first << "\t" << it->second << "\n"; snp_buckets[it->first] += it->second; } fh.close(); cerr << "done.\n"; return 0; } bool check_adjacent_blocks(PhasedSummary *psum, HBlock *block) { // // Create a list of all loci contained in the two blocks. // uint start = *(block->loci.begin()); uint end = *(block->next->loci.rbegin()); // // Check the D' measure between each pair in the proposed combined block. // double tot = 0.0; double strong_ld = 0.0; for (uint i = start; i <= end; i++) { if (psum->nucs[i].freq < minor_freq_lim) continue; for (uint j = i + 1; j <= end; j++) { if (psum->dprime[i][j].type == uninformative || psum->nucs[j].freq < minor_freq_lim) continue; tot++; if (psum->dprime[i][j].type == strong_ld) strong_ld++; } } // cerr << "Comparing range " << start << " to " << end // << "; total pairs: " << tot << "; strong LD: " << strong_ld // << "; proportion: " << std::setprecision(3) << strong_ld / tot << "\n"; if (strong_ld / tot >= min_inform_pairs) return true; return false; } HBlock * dPrimeBlocks::merge_adjacent(HBlock *a) { // // Merge two adjacent nodes. // HBlock *b = a->next; for (uint i = 0; i < b->loci.size(); i++) a->loci.push_back(b->loci[i]); a->next = b->next; delete b; return a; } HBlock * dPrimeBlocks::initialize(set &loci) { set::iterator it, prev_it; HBlock *cur, *next; this->_head = new HBlock; it = loci.begin(); this->_head->loci.push_back(*it); it++; // // // // Create a node from each locus and add to it all immediately adjacent loci. // // // do { // this->_head->loci.push_back(*it); // prev_it = it; // it++; // } while (it != loci.end() && (*prev_it) + 1 == *it); next = this->_head; // if (it == loci.end()) return this->_head; do { cur = new HBlock; cur->loci.push_back(*it); it++; // do { // cur->loci.push_back(*it); // prev_it = it; // it++; // } while (it != loci.end() && // (*prev_it) + 1 == *it); next->next = cur; next = next->next; } while (it != loci.end()); return this->_head; } int dPrimeBlocks::print() { HBlock *cur = this->_head; while (cur != NULL) { for (uint i = 0; i < cur->loci.size(); i++) { if (i > 0) cerr << ", "; cerr << cur->loci[i]; } cerr << "\n"; cur = cur->next; } return 0; } int enumerate_haplotypes(ofstream &fh, map &pop_map, PhasedSummary *psum, uint start, uint end) { map >::iterator it; map > haplotypes; map::iterator sit; string haplotype; set pops; // // Enumerate all haplotypes occurring in this block. // for (uint k = 0; k < psum->sample_cnt; k++) { for (uint n = start; n <= end; n++) if (psum->nucs[n].freq >= minor_freq_lim) haplotype += psum->samples[k].nucs_1[n]; pops.insert(pop_map[psum->samples[k].name]); if (haplotypes.count(haplotype) == 0) haplotypes[haplotype][pop_map[psum->samples[k].name]] = 1; else haplotypes[haplotype][pop_map[psum->samples[k].name]]++; haplotype.clear(); } for (uint k = 0; k < psum->sample_cnt; k++) { for (uint n = start; n <= end; n++) if (psum->nucs[n].freq >= minor_freq_lim) haplotype += psum->samples[k].nucs_2[n]; pops.insert(pop_map[psum->samples[k].name]); if (haplotypes.count(haplotype) == 0) haplotypes[haplotype][pop_map[psum->samples[k].name]] = 1; else haplotypes[haplotype][pop_map[psum->samples[k].name]]++; haplotype.clear(); } // // Write the haplotypes. // float tot = 0.0; fh << haplotypes.size() << "\t"; for (it = haplotypes.begin(); it != haplotypes.end(); it++) { // // Haplotypes are stored per population; sum them up here. // for (sit = it->second.begin(); sit != it->second.end(); sit++) tot += sit->second; if (it != haplotypes.begin()) fh << ","; fh << it->first << "|" << std::setprecision(3) << tot / ((float) psum->sample_cnt * 2.0); } fh << "\t"; set::iterator pit; stringstream pops_str; // // Write which populations this haplotype block occurs in. // if (pops.size() == 0) fh << "-1\t"; else for (pit = pops.begin(); pit != pops.end(); pit++) pops_str << *pit << ","; fh << pops_str.str().substr(0, pops_str.str().length()-1); pops_str.str(""); // // Write the frequency of occurence of each haplotype in each population. // for (it = haplotypes.begin(); it != haplotypes.end(); it++) { pops_str << "\t"; for (pit = pops.begin(); pit != pops.end(); pit++) pops_str << (it->second)[*pit] << "|" << std::setprecision(3) << (float) (it->second)[*pit] / (float) (pop_cnts[*pit] * 2.0) << ","; fh << pops_str.str().substr(0, pops_str.str().length()-1); pops_str.str(""); } fh << "\n"; return 0; } int calc_dprime(PhasedSummary *psum) { #pragma omp parallel { char allele_A, allele_a, allele_B, allele_b; double freq_A, freq_a, freq_B, freq_b; double freq_AB, freq_Ab, freq_aB, freq_ab; double D, min, var, chisq; double tot = psum->sample_cnt * 2.0; uint hap_cnt; #pragma omp for schedule(dynamic, 1) for (uint i = 0; i < psum->size; i++) { // // Assign nucleotides to allele A, and a. // assign_alleles(psum->nucs[i], allele_A, allele_a, freq_A, freq_a); for (uint j = i+1; j < psum->size; j++) { // // Assign nucleotides to allele B, and b. // assign_alleles(psum->nucs[j], allele_B, allele_b, freq_B, freq_b); freq_AB = 0.0; freq_Ab = 0.0; freq_aB = 0.0; freq_ab = 0.0; hap_cnt = 0; D = 0.0; // // Tally up haplotype frequencies. // for (uint k = 0; k < psum->sample_cnt; k++) { if (psum->samples[k].nucs_1[i] == allele_A && psum->samples[k].nucs_1[j] == allele_B) freq_AB++; else if (psum->samples[k].nucs_1[i] == allele_A && psum->samples[k].nucs_1[j] == allele_b) freq_Ab++; else if (psum->samples[k].nucs_1[i] == allele_a && psum->samples[k].nucs_1[j] == allele_B) freq_aB++; else if (psum->samples[k].nucs_1[i] == allele_a && psum->samples[k].nucs_1[j] == allele_b) freq_ab++; if (psum->samples[k].nucs_2[i] == allele_A && psum->samples[k].nucs_2[j] == allele_B) freq_AB++; else if (psum->samples[k].nucs_2[i] == allele_A && psum->samples[k].nucs_2[j] == allele_b) freq_Ab++; else if (psum->samples[k].nucs_2[i] == allele_a && psum->samples[k].nucs_2[j] == allele_B) freq_aB++; else if (psum->samples[k].nucs_2[i] == allele_a && psum->samples[k].nucs_2[j] == allele_b) freq_ab++; } freq_AB = freq_AB / tot; freq_Ab = freq_Ab / tot; freq_aB = freq_aB / tot; freq_ab = freq_ab / tot; // // Using the four-gamete test, check whether recombination has occurred // between these two loci. // Four-gamete test: if no recombination has occurred between any two loci (SNPs) there will // be three haplotypes present, if recombination has occurred there will be four haplotypes. // hap_cnt += freq_AB > 0 ? 1 : 0; hap_cnt += freq_Ab > 0 ? 1 : 0; hap_cnt += freq_aB > 0 ? 1 : 0; hap_cnt += freq_ab > 0 ? 1 : 0; if (hap_cnt == 3) psum->recomb[i][j] = false; else psum->recomb[i][j] = true; D = freq_AB - (freq_A * freq_B); // cerr << "D_AB: " << D << "; "; // D = freq_Ab - (freq_A * freq_b); // cerr << "D_Ab: " << D << "; "; // D = freq_aB - (freq_a * freq_B); // cerr << "D_aB: " << D << "; "; // D = freq_ab - (freq_a * freq_b); // cerr << "D_ab: " << D << "\n"; // cerr << " freq_AB: " << freq_AB << "; freq_Ab: " << freq_Ab << "; freq_aB: " << freq_aB << "; freq_ab: " << freq_ab << "\n"; if (D > 0) { min = (freq_A * freq_b) < (freq_a * freq_B) ? (freq_A * freq_b) : (freq_a * freq_B); psum->dprime[i][j].dprime = min == 0 ? 0.0 : D / min; } else { min = (freq_A * freq_B) < (freq_a * freq_b) ? (freq_A * freq_B) : (freq_a * freq_b); psum->dprime[i][j].dprime = min == 0 ? 0.0 :(-1 * D) / min; } // // Test D against a chi square distribution with 1 degree of freedom to show // whether these two loci have a D that is statistically significantly different from 0. // chisq = (tot * (D * D)) / (freq_A * freq_a * freq_B * freq_b); if (chisq >= chi_sq_limit) psum->dprime[i][j].chisq_p = true; // // Calculate variance and confidence limits. // if (psum->dprime[i][j].chisq_p) { var = (1.0 / tot) * ((freq_A * freq_a * freq_B * freq_b) + ((1 - (2 * freq_A)) * (1 - (2 * freq_B)) * D) - (D * D)); psum->dprime[i][j].var = var; psum->dprime[i][j].ci_high = psum->dprime[i][j].dprime + (1.96 * sqrt(var)); psum->dprime[i][j].ci_low = psum->dprime[i][j].dprime - (1.96 * sqrt(var)); } else { psum->dprime[i][j].ci_high = 0.0; psum->dprime[i][j].ci_low = 0.0; } } } } return 0; } int assign_alleles(NucSum nsum, char &p_allele, char &q_allele, double &p_freq, double &q_freq) { p_allele = 0; q_allele = 0; uint i = 0; float tot = 0; while (p_allele == 0 && i < 4) { if (nsum.nuc[i] > 0) { tot += nsum.nuc[i]; switch(i) { case 0: p_allele = 'A'; p_freq = nsum.nuc[0]; break; case 1: p_allele = 'C'; p_freq = nsum.nuc[1]; break; case 2: p_allele = 'G'; p_freq = nsum.nuc[2]; break; case 3: p_allele = 'T'; p_freq = nsum.nuc[3]; break; } } i++; } while (q_allele == 0 && i < 4) { if (nsum.nuc[i] > 0) { tot += nsum.nuc[i]; switch(i) { case 1: q_allele = 'C'; q_freq = nsum.nuc[1]; break; case 2: q_allele = 'G'; q_freq = nsum.nuc[2]; break; case 3: q_allele = 'T'; q_freq = nsum.nuc[3]; break; } } i++; } p_freq = p_freq / tot; q_freq = 1 - p_freq; return 0; } int write_dprime(string path, PhasedSummary *psum) { // // Write the D' data for plotting as a heatmap. // string file = path + ".dprime.tsv"; cerr << "Writing D' data to '" << file << "'..."; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening D' file '" << file << "'\n"; exit(1); } fh << "# Basepair 1\tBasepair 2\tD'\tCorrected D'\tVariance\tCI Low\tCI High\n"; double dprime = 0.0; for (uint i = 0; i < psum->size; i++) { for (uint j = i+1; j < psum->size; j++) { if (psum->nucs[i].freq < minor_freq_lim || psum->nucs[j].freq < minor_freq_lim) continue; dprime = psum->dprime[i][j].dprime; if (dprime_threshold) dprime = dprime >= dprime_threshold_level ? 1.0 : 0.0; if (write_zeros == false && (dprime == 0.0 || psum->dprime[i][j].chisq_p == false)) continue; fh << psum->nucs[i].bp << "\t" << psum->nucs[j].bp << "\t" << std::setprecision(3) << dprime << "\t" << std::setprecision(3) << (psum->dprime[i][j].chisq_p ? dprime : 0.0) << "\t" << psum->dprime[i][j].var << "\t" << psum->dprime[i][j].ci_low << "\t" << psum->dprime[i][j].ci_high << "\n"; } } fh.close(); cerr << "done.\n"; return 0; } int summarize_phased_genotypes(PhasedSummary *psum) { // // Construct a two dimensional array out of all the nucleotide arrays in the samples. // char **gtypes = new char *[psum->sample_cnt]; for (uint i = 0; i < psum->sample_cnt; i++) { gtypes[i] = psum->samples[i].nucs_1; } // // Sum up the occurences of each nucleotide. // for (uint i = 0; i < psum->size; i++) { for (uint j = 0; j < psum->sample_cnt; j++) { switch(gtypes[j][i]) { case 'A': psum->nucs[i].nuc[0]++; break; case 'C': psum->nucs[i].nuc[1]++; break; case 'G': psum->nucs[i].nuc[2]++; break; case 'T': psum->nucs[i].nuc[3]++; break; case 'N': default: break; } } } // // Repeat for the second set of phased genotypes. // for (uint i = 0; i < psum->sample_cnt; i++) { gtypes[i] = psum->samples[i].nucs_2; } // // Sum up the occurences of each nucleotide. // for (uint i = 0; i < psum->size; i++) { for (uint j = 0; j < psum->sample_cnt; j++) { switch(gtypes[j][i]) { case 'A': psum->nucs[i].nuc[0]++; break; case 'C': psum->nucs[i].nuc[1]++; break; case 'G': psum->nucs[i].nuc[2]++; break; case 'T': psum->nucs[i].nuc[3]++; break; case 'N': default: break; } } // // Calculate minor allele frequency. // float tot = (float) psum->sample_cnt * 2.0; float freq = 0.0; for (uint j = 0; j < 4; j++) { if (psum->nucs[i].nuc[j] > 0) { freq = (float) psum->nucs[i].nuc[j] / tot; psum->nucs[i].freq = freq < psum->nucs[i].freq ? freq : psum->nucs[i].freq; } } } delete [] gtypes; return 0; } // // Code to parse fastPhase format. // PhasedSummary * parse_fastphase(string path) { ifstream fh; char line[max_len]; string buf, filepath; const char *p, *q, *end; int i, sindex, pos; memset(line, '\0', max_len); // // Read in the original fastPhase export from Stacks to obtain the original base pair positions. // // // Open the file for reading // filepath = path + ".inp"; fh.open(filepath.c_str(), ifstream::in); if (fh.fail()) { cerr << "Error opening input file '" << path << "'\n"; return NULL; } cerr << "Parsing " << filepath << "...\n"; int num_samples, num_genotypes; char bp[id_len]; // // Get the number of samples in the dataset. // fh.getline(line, max_len); num_samples = is_integer(line); if (num_samples < 0) { cerr << "Unable to find the number of samples, should be the first line.\n"; return NULL; } // // Get the number of genotypes in the dataset. // fh.getline(line, max_len); num_genotypes = is_integer(line); if (num_genotypes < 0) { cerr << "Unable to find the number of genotypes, should be the second line.\n"; return NULL; } PhasedSummary *psum = new PhasedSummary(num_samples, num_genotypes); // // Get the set of base pair positions. // buf.clear(); do { fh.clear(); fh.getline(line, max_len); buf += line; } while (fh.fail() && !fh.bad() && !fh.eof()); i = 0; p = buf.c_str(); end = p + buf.length(); if (*p != 'P') { cerr << "Unable to locate line of basepair positions, should be the third line.\n"; delete psum; return NULL; } for (p += 2, q = p; p < end; p++, q++) { while (*q != ' ' && q < end) { q++; } strncpy(bp, p, q - p); bp[q - p] = '\0'; pos = is_integer(bp); if (pos < 0) { cerr << "Unable to parse base pair positions.\n"; delete psum; return NULL; } else { psum->nucs[i].bp = (uint) pos; } i++; p = q; } fh.close(); // // Open the file for reading // filepath = path + "_hapguess_switch.out"; fh.open(filepath.c_str(), ifstream::in); if (fh.fail()) { cerr << "Error opening input file '" << path << "'\n"; return NULL; } cerr << "Parsing " << filepath << "...\n"; // // Read from the "*_hapguess_switch.out" file until we hit the genotypes section // marked by the string "BEGIN GENOTYPES". // do { fh.getline(line, max_len); if (!fh.good()) { cerr << "Unable to find file section entitled 'BEGIN GENOTYPES'\n"; delete psum; return NULL; } } while (strcmp(line, "BEGIN GENOTYPES") != 0); // // Now read lines from the file in groups of three: // 1. Sample label // 2. Phased genotypes from chromosome 1 // 3. Phased genotypes from chromosome 2 // Stop reading individuals when we encounter the string, "END GENOTYPES". // fh.getline(line, max_len); do { // // Create a new Sample object and store the sample label. // sindex = psum->add_sample(line); // // Get the first set of phased genotypes. // buf.clear(); do { fh.clear(); fh.getline(line, max_len); buf += line; } while (fh.fail() && !fh.bad() && !fh.eof()); // // Count the number of genotypes on this line (they should be space deliniated). // i = 0; for (p = buf.c_str(); *p != '\0'; p++) if (*p != ' ') psum->samples[sindex].size++; // // Store the genotypes into our internal buffer. // psum->samples[sindex].nucs_1 = new char[psum->samples[sindex].size]; for (p = buf.c_str(); *p != '\0'; p++) { if (*p == ' ') continue; psum->samples[sindex].nucs_1[i] = *p; i++; } // len = strlen(line); // if (line[len - 1] == '\r') line[len - 1] = '\0'; // // Get the second set of phased genotypes. // buf.clear(); do { fh.clear(); fh.getline(line, max_len); buf += line; } while (fh.fail() && !fh.bad() && !fh.eof()); i = 0; psum->samples[sindex].nucs_2 = new char[psum->samples[sindex].size]; for (p = buf.c_str(); *p != '\0'; p++) { if (*p == ' ') continue; psum->samples[sindex].nucs_2[i] = *p; i++; } // // Get the sample label of the next record. // fh.getline(line, max_len); } while (strcmp(line, "END GENOTYPES") != 0 && fh.good()); fh.close(); return psum; } // // Code to parse Beagle format. // PhasedSummary * parse_beagle(map &catalog, string path) { gzFile gz_fh; char *line; string buf, filepath; const char *p, *q; uint len, line_len, i, sindex; bool eol; line_len = max_len; line = new char[line_len]; memset(line, '\0', line_len); // // Open the Beagle file for reading // filepath = path + ".phased.gz"; gz_fh = gzopen(filepath.c_str(), "rb"); if (!gz_fh) { cerr << "Failed to open gzipped file '" << filepath << "': " << strerror(errno) << ".\n"; return NULL; } cerr << "Parsing " << filepath << "...\n"; vector parts; uint num_samples = 0; uint num_genotypes = 0; char cat_loc_str[id_len], col_str[id_len]; // // Parse the file twice. On the first round: // 1. Determine the number of samples in the dataset (column count) // 2. Determine the number of markers (row count). // On the second round, parse the SNP genotypes. // // // Read each line in the file. If it starts with: // '#' it is a comment, skip the line. // 'I' it is the list of samples, parse them. // 'S' is the population ID for each SNP, skip this line. // 'M' is a marker, count the number of markers. // do { eol = false; buf.clear(); do { gzgets(gz_fh, line, line_len); buf += line; len = strlen(line); if (len > 0 && line[len - 1] == '\n') { eol = true; line[len - 1] = '\0'; } } while (!gzeof(gz_fh) && !eol); if (line_len < buf.length()) { // cerr << "Resizing line buffer from " << line_len << " to " << buf.length() << "\n"; delete [] line; line = new char[buf.length() + 1]; line_len = buf.length() + 1; memset(line, '\0', line_len); } if (buf[0] == 'M') { num_genotypes++; } else if (buf[0] == 'I') { // // Count the number of samples. // parse_ssv(buf.c_str(), parts); num_samples = (parts.size() - 2) / 2; } } while (!gzeof(gz_fh)); PhasedSummary *psum = new PhasedSummary(num_samples, num_genotypes); for (uint j = 2; j < parts.size(); j++) { if (j % 2 == 0) { sindex = psum->add_sample(parts[j]); psum->samples[sindex].size = num_genotypes; psum->samples[sindex].nucs_1 = new char[psum->samples[sindex].size]; psum->samples[sindex].nucs_2 = new char[psum->samples[sindex].size]; } } cerr << " Found " << num_samples << " samples; " << num_genotypes << " genotypes.\n"; gzrewind(gz_fh); uint marker_num = 0; memset(line, '\0', line_len); do { do { gzgets(gz_fh, line, line_len); } while (!gzeof(gz_fh) && line[0] != 'M'); len = strlen(line); if (len == 0) break; if (len > 0 && line[len - 1] == '\n') line[len - 1] = '\0'; parse_ssv(line, parts); // // Parse the catalog locus ID and the column number of the SNP: // e.g. LocId_column or 10329_37 // p = parts[1].c_str(); for (q = p + 1; *q != '_' && *q != '\0'; q++); strncpy(cat_loc_str, p, q - p); cat_loc_str[q-p] = '\0'; q++; strcpy(col_str, q); psum->nucs[marker_num].clocus = is_integer(cat_loc_str); psum->nucs[marker_num].col = is_integer(col_str); // // Store the genotypes into our internal buffer. // sindex = 0; i = 2; while (i < parts.size()) { p = parts[i].c_str(); psum->samples[sindex].nucs_1[marker_num] = *p; i++; p = parts[i].c_str(); psum->samples[sindex].nucs_2[marker_num] = *p; i++; sindex++; } marker_num++; } while (!gzeof(gz_fh)); gzclose(gz_fh); // // Use the catalog to look up the basepair positions for each catalog locus. // CSLocus *loc; for (i = 0; i < psum->size; i++) { loc = catalog[psum->nucs[i].clocus]; psum->nucs[i].bp = loc->sort_bp(psum->nucs[i].col); } return psum; } // // Code to parse Beagle format. // PhasedSummary * parse_beagle_haplotypes(map &catalog, string path) { gzFile gz_fh; char *line; string buf, filepath; const char *p; uint len, line_len, i, j, sindex; bool eol; line_len = max_len; line = new char[line_len]; memset(line, '\0', line_len); // // Open the Beagle file for reading // filepath = path + ".phased.gz"; gz_fh = gzopen(filepath.c_str(), "rb"); if (!gz_fh) { cerr << "Failed to open gzipped file '" << filepath << "': " << strerror(errno) << ".\n"; return NULL; } cerr << "Parsing " << filepath << "...\n"; vector parts, samples; uint num_samples = 0; uint num_genotypes = 0; uint cat_loc; // // Parse the file twice. On the first round: // 1. Determine the number of samples in the dataset (column count) // 2. Determine the number of markers (row count). // On the second round, parse the SNP genotypes. // // // Read each line in the file. If it starts with: // '#' it is a comment, skip the line. // 'I' it is the list of samples, parse them. // 'S' is the population ID for each SNP, skip this line. // 'M' is a marker, count the number of markers. // do { eol = false; buf.clear(); do { gzgets(gz_fh, line, line_len); buf += line; len = strlen(line); if (len > 0 && line[len - 1] == '\n') { eol = true; line[len - 1] = '\0'; } } while (!gzeof(gz_fh) && !eol); if (line_len < buf.length()) { // cerr << "Resizing line buffer from " << line_len << " to " << buf.length() << "\n"; delete [] line; line = new char[buf.length() + 1]; line_len = buf.length() + 1; memset(line, '\0', line_len); } if (buf[0] == 'M') { // // Count the number of genotypes by counting the number or nucleotides in each // haplotype for each marker. // parse_ssv(buf.c_str(), parts); num_genotypes += parts[2].length(); } else if (buf[0] == 'I') { // // Count the number of samples. // parse_ssv(buf.c_str(), samples); num_samples = (samples.size() - 2) / 2; } } while (!gzeof(gz_fh)); PhasedSummary *psum = new PhasedSummary(num_samples, num_genotypes); for (uint j = 2; j < samples.size(); j++) { if (j % 2 == 0) { sindex = psum->add_sample(samples[j]); psum->samples[sindex].size = num_genotypes; psum->samples[sindex].nucs_1 = new char[psum->samples[sindex].size]; psum->samples[sindex].nucs_2 = new char[psum->samples[sindex].size]; } } cerr << " Found " << num_samples << " samples; " << num_genotypes << " genotypes.\n"; gzrewind(gz_fh); CSLocus *loc; uint hap_len = 0; uint marker_num = 0; memset(line, '\0', line_len); do { do { gzgets(gz_fh, line, line_len); } while (!gzeof(gz_fh) && line[0] != 'M'); len = strlen(line); if (len == 0) break; if (len > 0 && line[len - 1] == '\n') line[len - 1] = '\0'; parse_ssv(line, parts); // // Use the catalog to look up the basepair positions for each catalog locus. // cat_loc = is_integer(parts[1].c_str()); loc = catalog[cat_loc]; hap_len = parts[2].length(); if (hap_len != loc->snps.size()) cerr << "Haplotypes don't match between catalog and beagle; Locus ID: " << loc->id << "; beagle hap len: " << hap_len << "; catalog hap len: " << loc->snps.size() << "\n"; for (j = 0, i = marker_num; i < marker_num + hap_len; i++, j++) { psum->nucs[i].clocus = cat_loc; psum->nucs[i].col = loc->snps[j]->col; psum->nucs[i].bp = loc->sort_bp(psum->nucs[i].col); } // // Store the genotypes into our internal buffer. // sindex = 0; i = 2; while (i < parts.size()) { p = parts[i].c_str(); for (j = marker_num; j < marker_num + hap_len; j++) { psum->samples[sindex].nucs_1[j] = *p; p++; } i++; p = parts[i].c_str(); for (j = marker_num; j < marker_num + hap_len; j++) { psum->samples[sindex].nucs_2[j] = *p; p++; } i++; sindex++; } marker_num += hap_len; } while (!gzeof(gz_fh)); gzclose(gz_fh); return psum; } int parse_population_map(string popmap_path, map &pop_map, map &pop_cnts) { char line[max_len]; char pop_id_str[id_len]; vector parts; uint len; if (pmap_path.length() == 0) return 0; cerr << "Parsing population map.\n"; ifstream fh(popmap_path.c_str(), ifstream::in); if (fh.fail()) { cerr << "Error opening population map '" << popmap_path << "'\n"; return 0; } while (fh.good()) { fh.getline(line, max_len); len = strlen(line); if (len == 0) continue; // // Check that there is no carraige return in the buffer. // if (line[len - 1] == '\r') line[len - 1] = '\0'; // // Ignore comments // if (line[0] == '#') continue; // // Parse the population map, we expect: // // parse_tsv(line, parts); if (parts.size() != 2) { cerr << "Population map is not formated correctly: expecting two, tab separated columns, found " << parts.size() << ".\n"; return 0; } strncpy(pop_id_str, parts[1].c_str(), id_len); for (int i = 0; i < id_len && pop_id_str[i] != '\0'; i++) if (!isdigit(pop_id_str[i])) { cerr << "Population map is not formated correctly: expecting numerical ID in second column, found '" << parts[1] << "'.\n"; return 0; } // // Add the sample name to population number mapping. // pop_map[parts[0]] = atoi(parts[1].c_str()); if (pop_cnts.count(atoi(parts[1].c_str())) == 0) pop_cnts[atoi(parts[1].c_str())] = 1; else pop_cnts[atoi(parts[1].c_str())]++; } fh.close(); return 0; } int build_file_list(vector > &files) { vector parts; string pattern; // // Read all the files from the Stacks directory. // uint pos; string file; struct dirent *direntry; DIR *dir = opendir(in_path.c_str()); if (dir == NULL) { cerr << "Unable to open directory '" << in_path << "' for reading.\n"; exit(1); } switch(in_file_type) { case FileT::beagle: pattern = ".phased.gz"; break; case FileT::fastphase: default: pattern = "_hapguess_switch.out"; break; } while ((direntry = readdir(dir)) != NULL) { file = direntry->d_name; if (file == "." || file == "..") continue; pos = file.rfind(pattern); if (pos < file.length()) files.push_back(make_pair(1, file.substr(0, pos))); } closedir(dir); if (files.size() == 0) { cerr << "Unable to locate any input files to process within '" << in_path << "'\n"; return 0; } return 1; } int parse_command_line(int argc, char* argv[]) { int c; while (1) { static struct option long_options[] = { {"help", no_argument, NULL, 'h'}, {"version", no_argument, NULL, 'v'}, {"haplotypes", no_argument, NULL, 'H'}, {"skip_zeros", no_argument, NULL, 'Z'}, {"infile_type", required_argument, NULL, 't'}, {"num_threads", required_argument, NULL, 'p'}, {"in_path", required_argument, NULL, 'P'}, {"cat_path", required_argument, NULL, 'S'}, {"pop_map", required_argument, NULL, 'M'}, {"batch_id", required_argument, NULL, 'b'}, {"dprime_bin_size", required_argument, NULL, 'B'}, {"minor_allele_freq", required_argument, NULL, 'a'}, {"min_inform_pairs", required_argument, NULL, 'm'}, {"dprime_threshold", required_argument, NULL, 'T'}, {0, 0, 0, 0} }; // getopt_long stores the option index here. int option_index = 0; c = getopt_long(argc, argv, "hvZHAb:M:t:P:S:p:a:B:T:", long_options, &option_index); // Detect the end of the options. if (c == -1) break; switch (c) { case 'h': help(); break; case 'b': batch_id = is_integer(optarg); if (batch_id < 0) { cerr << "Batch ID (-b) must be an integer, e.g. 1, 2, 3\n"; help(); } break; case 'p': num_threads = atoi(optarg); break; case 'a': minor_freq_lim = atof(optarg); break; case 'm': min_inform_pairs = atof(optarg); break; case 'P': in_path = optarg; break; case 'S': cat_path = optarg; break; case 't': if (strcasecmp(optarg, "phase") == 0) in_file_type = FileT::phase; else if (strcasecmp(optarg, "fastphase") == 0) in_file_type = FileT::fastphase; else if (strcasecmp(optarg, "beagle") == 0) in_file_type = FileT::beagle; else in_file_type = FileT::unknown; break; case 'M': pmap_path = optarg; break; case 'H': haplotypes = true; break; case 'Z': write_zeros = false; break; case 'B': bucket_dist = atoi(optarg); break; case 'T': dprime_threshold = true; dprime_threshold_level = atof(optarg); break; case 'v': version(); break; case '?': // getopt_long already printed an error message. help(); break; default: help(); abort(); } } if (in_path.length() == 0) { cerr << "You must specify a path to the directory containing Stacks output files.\n"; help(); } if (in_path.at(in_path.length() - 1) != '/') in_path += "/"; if (minor_freq_lim > 0) { if (minor_freq_lim > 1) minor_freq_lim = minor_freq_lim / 100; if (minor_freq_lim > 0.5) { cerr << "Unable to parse the minor allele frequency\n"; help(); } } if (min_inform_pairs > 0) { if (min_inform_pairs > 1) min_inform_pairs = min_inform_pairs / 100; } return 0; } void version() { std::cerr << "phasedstacks " << VERSION << "\n\n"; exit(0); } void help() { std::cerr << "phasedstacks " << VERSION << "\n" << "phasedstacks -b id -S path -P path -t file_type [-p threads] [-M popmap] [-v] [-h]" << "\n" << " b: Stacks batch ID.\n" << " P: path to the phased output files.\n" << " S: path to the Stacks output files.\n" << " t: input file type. Supported types: fastphase, and beagle.\n" << " p: number of processes to run in parallel sections of code.\n" << " M: path to the population map, a tab separated file describing which individuals belong in which population.\n" << " v: print program version." << "\n" << " h: display this help messsage." << "\n" << " --haplotypes: data were phased as RAD locus haplotypes.\n" << " --dprime_bin_size: size of buckets for binning SNPs at a particular distance to calculate the mean D' value.\n" << " --dprime_threshold : if D' values fall above , set the D' to 1, otherwise set D' to 0.\n\n" << " Filtering options:\n" << " --skip_zeros: do not include D' values of zero in the D' output.\n" << " --minor_allele_freq: specify a minimum minor allele frequency required to process a nucleotide site (0 < a < 0.5).\n" << " --min_inform_pairs: when building D' haplotype blocks, the minimum number of informative D' measures to combine two blocks (default 0.9).\n\n"; exit(0); } stacks-1.44/src/phasedstacks.h000644 000765 000024 00000013715 12755650665 017147 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2013, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __PHASEDSTACKS_H__ #define __PHASEDSTACKS_H__ #ifdef _OPENMP #include // OpenMP library #endif #include // Process command-line options #include // Open/Read contents of a directory #include #include #include #include #include using std::pair; using std::make_pair; #include using std::string; #include #include using std::ifstream; using std::ofstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include using std::stringstream; #include #include using std::vector; #include using std::map; #include using std::set; #include "constants.h" #include "utils.h" #include "log_utils.h" #include "catalog_utils.h" #include "input.h" #include "sql_utilities.h" #include "locus.h" #ifdef HAVE_LIBZ #include #include #endif enum loc_t {strong_ld, recomb, uninformative}; class Sample { public: string name; int id; int size; char *nucs_1; char *nucs_2; Sample() { this->id = 0; this->size = 0; this->nucs_1 = NULL; this->nucs_2 = NULL; } ~Sample() { if (this->nucs_1 != NULL) delete [] this->nucs_1; if (this->nucs_2 != NULL) delete [] this->nucs_2; } }; class NucSum { public: uint bp; uint col; uint clocus; float freq; uint nuc[4]; // nuc[0] == A // nuc[1] == C // nuc[2] == G // nuc[3] == T NucSum() { this->freq = 1.0; this->bp = 0; this->clocus = 0; for (uint i = 0; i < 4; i++) this->nuc[i] = 0; } }; class dPrime { public: double dprime; bool chisq_p; double var; double ci_high; double ci_low; bool informative; loc_t type; dPrime() { this->dprime = 0.0; this->chisq_p = false; this->var = 0.0; this->ci_high = 0.0; this->ci_low = 0.0; this->type = uninformative; } }; class PhasedSummary { map sample_map; public: uint size; uint sample_cnt; NucSum *nucs; Sample *samples; dPrime **dprime; bool **recomb; PhasedSummary(uint num_samples, uint num_genotypes) { this->sample_cnt = num_samples; this->samples = new Sample[this->sample_cnt]; this->size = num_genotypes; this->nucs = new NucSum[this->size]; this->dprime = new dPrime *[this->size]; for (uint i = 0; i < this->size; i++) this->dprime[i] = new dPrime[this->size]; this->recomb = new bool *[this->size]; for (uint i = 0; i < this->size; i++) { this->recomb[i] = new bool[this->size]; memset(this->recomb[i], 0, this->size); } } ~PhasedSummary() { if (this->nucs != NULL) delete [] this->nucs; if (this->dprime != NULL) { for (uint i = 0; i < this->size; i++) delete [] this->dprime[i]; delete [] this->dprime; } if (this->recomb != NULL) { for (uint i = 0; i < this->size; i++) delete [] this->recomb[i]; delete [] this->recomb; } if (this->samples != NULL) delete [] this->samples; } int add_sample(string name) { uint i = this->sample_map.size(); this->sample_map[name] = i; this->samples[i].name = name; return i; } }; class HBlock { public: vector loci; HBlock *next; HBlock() { this->next = NULL; } }; class dPrimeBlocks { HBlock *_head; public: dPrimeBlocks() { this->_head = NULL; } ~dPrimeBlocks() { HBlock *cur, *next; cur = this->_head; next = cur->next; while (next != NULL) { delete cur; cur = next; next = cur->next; } } HBlock *head() { return this->_head; } HBlock *initialize(set &); HBlock *merge_adjacent(HBlock *); int print(); }; void help( void ); void version( void ); int parse_command_line(int, char**); int build_file_list(vector > &); int parse_population_map(string, map &, map &); PhasedSummary *parse_fastphase(string); PhasedSummary *parse_beagle(map &, string); PhasedSummary *parse_beagle_haplotypes(map &, string); int summarize_phased_genotypes(PhasedSummary *); int calc_dprime(PhasedSummary *); int assign_alleles(NucSum, char &, char &, double &, double &); int write_dprime(string, PhasedSummary *); int four_gamete_test(string, map &, PhasedSummary *, map &, map &); int dprime_blocks(string, map &, PhasedSummary *, map &, map &); bool check_adjacent_blocks(PhasedSummary *, HBlock *); int enumerate_haplotypes(ofstream &, map &, PhasedSummary *, uint, uint); int bucket_dprime(vector &, vector &, PhasedSummary *); int write_buckets(string, vector &, vector &); #endif // __PHASEDSTACKS_H__ stacks-1.44/src/PopMap.h000644 000765 000024 00000044572 12764047443 015665 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2011-2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __POPMAP_H__ #define __POPMAP_H__ #include using std::exception; #include #include using std::string; #include using std::vector; #include using std::map; #include using std::set; #include using std::accumulate; #include #include using std::pair; using std::make_pair; #include "stacks.h" #include "locus.h" #include "aln_utils.h" #include "MetaPopInfo.h" #include "Vcf.h" class Datum { public: int id; // Stack ID int merge_partner; // Stack ID of merged datum, if this datum was merged/phased from two, overlapping datums. int len; // Length of locus int tot_depth; // Stack depth vector depth; // Stack depth of each matching allele bool corrected; // Has this genotype call been corrected char *model; // String representing SNP model output for each nucleotide at this locus. char *gtype; // Genotype char *trans_gtype; // Translated Genotype char *cigar; // CIGAR string describing how the datum aligns to the catalog locus. double lnl; // Log likelihood of this locus. vector obshap; // Observed Haplotypes vector snps; // All calls for this sample for this locus. size() is [len]. Datum() { this->id = -1; this->corrected = false; this->gtype = NULL; this->trans_gtype = NULL; this->model = NULL; this->cigar = NULL; this->tot_depth = 0; this->len = 0; this->lnl = 0.0; this->merge_partner = 0; } ~Datum() { for (uint i = 0; i < this->obshap.size(); i++) delete [] this->obshap[i]; for (uint i = 0; i < this->snps.size(); i++) delete this->snps[i]; delete [] this->gtype; delete [] this->trans_gtype; delete [] this->model; delete [] this->cigar; } int add_model(const char *model) { if (this->cigar == NULL) { this->len = strlen(model); this->model = new char[this->len + 1]; strcpy(this->model, model); } else { vector > c; this->len = parse_cigar(this->cigar, c); this->model = new char[this->len + 1]; apply_cigar_to_model_seq(this->model, this->len, model, c); // cerr << "Cigar: " << this->cigar << "\n" // << "Old model: " << model << "\n" // << "Gapped model: " << this->model << "\n"; } return 0; } }; template class PopMap { const MetaPopInfo& metapopinfo; int num_loci; set > blacklist; Datum ***data; map locus_order; // LocusID => ArrayIndex; map catalog IDs to their first dimension // position in the Datum array. map rev_locus_order; public: map > ordered_loci; // Loci ordered by genomic position PopMap(const MetaPopInfo& mpopi, int n_loci); ~PopMap(); // Populates the PopMap based on sstack matches files. // The catalog is modified (LocusT must be CSLocus, and // members [cnt, hcnt, confounded_cnt] are modified). int populate(map& catalog, const vector >& matches); // Populates the PopMap based on VCF (SNP) records. // The catalog is modified (LocusT must be CSLocus, and // members [cnt, hcnt] are modified). // N.B. The IDs of the loci in the catalog MUST be the same // as the indexes in the records vector. int populate(map& catalog, const vector& records, const VcfHeader& header); int order_loci(const map& catalog); int prune(set& loc_ids); int loci_cnt() const { return this->num_loci; } int locus_index(int id) const {return locus_order.at(id);} int rev_locus_index(int index) const {try {return rev_locus_order.at(index);} catch(exception&) {return -1;}} int sample_cnt() const { return metapopinfo.samples().size(); } int sample_index(int id) const {try {return metapopinfo.get_sample_index(id);} catch (exception&) {return -1;}} int rev_sample_index(int index) const {return metapopinfo.samples().at(index).id;} Datum **locus(int id); Datum *datum(int loc_id, int sample_id); bool blacklisted(int loc_id, int sample_id); }; template PopMap::PopMap(const MetaPopInfo& mpopi, int num_loci) : metapopinfo(mpopi) { this->data = new Datum **[num_loci]; for (int i = 0; i < num_loci; i++) { this->data[i] = new Datum *[metapopinfo.samples().size()]; for (size_t j = 0; j < metapopinfo.samples().size(); j++) this->data[i][j] = NULL; } this->num_loci = num_loci; } template PopMap::~PopMap() { for (int i = 0; i < this->num_loci; i++) { for (int j = 0; j < metapopinfo.samples().size(); j++) delete this->data[i][j]; delete [] this->data[i]; } delete [] this->data; } template int PopMap::populate(map &catalog, const vector > &matches) { // // Create an index showing what position each catalog locus is stored at in the datum // array. Create a second index allowing ordering of Loci by genomic position. // typename std::map::const_iterator it; uint i = 0; for (it = catalog.begin(); it != catalog.end(); it++) { this->locus_order[it->first] = i; this->rev_locus_order[i] = it->first; i++; } // // Sort the catalog loci on each chromosome according to base pair. // this->order_loci(catalog); // // Populate the datum array // Datum *d; int locus, sample; for (i = 0; i < matches.size(); i++) { for (uint j = 0; j < matches[i].size(); j++) { sample = metapopinfo.get_sample_index(matches[i][j]->sample_id); if (this->locus_order.count(matches[i][j]->cat_id) == 0) continue; locus = this->locus_order[matches[i][j]->cat_id]; // cerr << "Translating sample id: " << matches[i][j]->sample_id << " to index " << sample << "\n"; // cerr << "Translating locus id: " << matches[i][j]->cat_id << " to index " << locus << "\n"; if (this->data[locus][sample] == NULL) { if (this->blacklist.count(make_pair(matches[i][j]->sample_id, matches[i][j]->cat_id)) == 0) { // cerr << "Creating new datum for tag ID: " << matches[i][j]->tag_id << "\n"; d = new Datum; d->id = matches[i][j]->tag_id; char *h = new char[strlen(matches[i][j]->haplotype) + 1]; strcpy(h, matches[i][j]->haplotype); d->obshap.push_back(h); d->depth.push_back(matches[i][j]->depth); d->tot_depth += matches[i][j]->depth; d->lnl = matches[i][j]->lnl; if (matches[i][j]->cigar != NULL) { d->cigar = new char[strlen(matches[i][j]->cigar) + 1]; strcpy(d->cigar, matches[i][j]->cigar); } this->data[locus][sample] = d; catalog[matches[i][j]->cat_id]->hcnt++; catalog[matches[i][j]->cat_id]->cnt++; } } else { // cerr << " Adding haplotype to existing datum: sample: " << matches[i][j]->sample_id << ". tag: " << matches[i][j]->tag_id << "\n"; // // Check that the IDs of the two matches are the same. If not, then two tags // match this locus and the locus is invalid, set back to NULL. // if (matches[i][j]->tag_id == this->data[locus][sample]->id) { char *h = new char[strlen(matches[i][j]->haplotype) + 1]; strcpy(h, matches[i][j]->haplotype); if (matches[i][j]->cigar != NULL && strcmp(this->data[locus][sample]->cigar, matches[i][j]->cigar) != 0) cerr << "Warning: disparate CIGAR strings, catalog locus " << matches[i][j]->cat_id << "; sample ID: " << matches[i][j]->sample_id << "; sample locus ID: " << matches[i][j]->tag_id << "; datum cigar: " << this->data[locus][sample]->cigar << "; matches cigar: " << matches[i][j]->cigar << "\n"; this->data[locus][sample]->obshap.push_back(h); this->data[locus][sample]->depth.push_back(matches[i][j]->depth); this->data[locus][sample]->tot_depth += matches[i][j]->depth; this->data[locus][sample]->lnl = matches[i][j]->lnl; } else { //cerr << " Deleting sample, multiple tag matches\n"; delete this->data[locus][sample]; this->data[locus][sample] = NULL; this->blacklist.insert(make_pair(matches[i][j]->sample_id, matches[i][j]->cat_id)); catalog[matches[i][j]->cat_id]->hcnt--; catalog[matches[i][j]->cat_id]->confounded_cnt++; } } } } return 0; } template int PopMap::populate(map& catalog, const vector& records, const VcfHeader& header) { // Initalize [locus_order], [rev_locus_order]. size_t loc_index = 0; for (typename map::iterator l = catalog.begin(); l != catalog.end(); ++l) { locus_order[l->first] = loc_index; rev_locus_order[loc_index] = l->first; ++loc_index; } // Initialize [ordered_loci]. order_loci(catalog); /* * Fill the PopMap. * * We observe the following rules to create the Datums : * [id] is the locus id. * [len] is the locus length (expected to be one, for a SNP) * [model] "E" or "O" according to the SAMPLE/GT field, or * "U" if the GT field is absent. * [obshap] the nucleotide(s) observed for this SNP for this individual. * If one of a sample's VCF alleles is missing ('.') or has an index * corresponding to the special '*' allele, the Datum* is left to NULL. * * When no depth information is available, [tot_depth] and the [depths] * of all alleles are set to 0. * (n.b. the parsing of depth information in VCF is not implemented as * of Mar 21, 2016.) * * When no likelihood information is available, [lnl] is set to 0. * (n.b. the parsing of depth information in VCF is not implemented as * of Mar 21, 2016.) * * The following members are left unset, on the premise that * "populations" does not use them : * corrected, genotype, trans_genotype * * [merge_partner] is set later by [merge_datums()] (in populations.cc). * [snps] is only used by [write_vcf()] and [write_vcf_strict()], which * first call [populate_snp_calls()] then use the SNP data in the * datum. */ loc_index = 0; for (typename map::iterator l = catalog.begin(); l != catalog.end(); ++l) { LocusT* loc = l->second; const VcfRecord& rec = records[loc->id]; // n.b. assumes locus ID == record index. int ad_index; try { ad_index = rec.index_of_gt_subfield("AD"); } catch (exception& e) { ad_index = -1; } for (size_t s = 0; s < metapopinfo.samples().size(); ++s) { size_t vcf_index = header.sample_indexes().at(metapopinfo.samples()[s].name); const string& sample = rec.samples.at(vcf_index); pair gt = rec.parse_genotype(sample); if (gt.first < 0 || gt.second < 0 || rec.alleles[gt.first]=="*" || rec.alleles[gt.second]=="*") // Missing or incomplete genotype. continue; vector ad; if (ad_index != -1) { string ad_str = rec.parse_gt_subfield(sample, ad_index); size_t start = 0; size_t coma; try { while ((coma = ad_str.find(',', start)) != string::npos) { ad.push_back(std::stoi(ad_str.substr(start,coma))); if (ad.back() < 0) throw exception(); start=coma+1; } ad.push_back(std::stoi(ad_str.substr(start))); if (ad.size() != rec.alleles.size()) throw exception(); } catch (exception& e) { cerr << "Warning: Badly formatted AD string '" << ad_str << "' at VCF record '" << rec.chrom << ":" << rec.pos << "'.\n"; ad = vector(rec.alleles.size(), 0); } } Datum* d = new Datum(); data[loc_index][s] = d; ++loc->cnt; ++loc->hcnt; // id, len, lnl d->id = loc->id; d->len = loc->len; d->lnl = 0; // model, obshap, depth d->model = new char[2]; if (gt.first == gt.second) { strcpy(d->model, "O"); const string& allele = rec.alleles[gt.first]; d->obshap.push_back(new char[allele.size()+1]); strcpy(d->obshap[0], allele.c_str()); if (ad_index != -1) d->depth = { ad[gt.first] }; else d->depth = {0}; } else { strcpy(d->model, "E"); const string& allele1 = rec.alleles[gt.first]; const string& allele2 = rec.alleles[gt.second]; d->obshap.push_back(new char[allele1.size()+1]); d->obshap.push_back(new char[allele2.size()+1]); strcpy(d->obshap[0], allele1.c_str()); strcpy(d->obshap[1], allele2.c_str()); if (ad_index != -1) d->depth = {ad[gt.first], ad[gt.second]}; else d->depth = {0, 0}; } // tot_depth d->tot_depth = std::accumulate(d->depth.begin(), d->depth.end(), 0); } ++loc_index; } return 0; } template int PopMap::order_loci(const map &catalog) { this->ordered_loci.clear(); typename std::map::const_iterator it; for (it = catalog.begin(); it != catalog.end(); it++) { if (strlen(it->second->loc.chr) > 0) this->ordered_loci[it->second->loc.chr].push_back(it->second); } // // Sort the catalog loci on each chromosome according to base pair. // typename map >::iterator cit; for (cit = this->ordered_loci.begin(); cit != this->ordered_loci.end(); cit++) sort(cit->second.begin(), cit->second.end(), bp_compare); return 0; } template int PopMap::prune(set &remove_ids) { uint new_size = this->num_loci - remove_ids.size(); uint loc_id; map new_loc_order, new_rev_loc_order; Datum ***d = new Datum **[new_size]; int j = 0; for (int i = 0; i < this->num_loci; i++) { loc_id = this->rev_locus_order[i]; if (remove_ids.count(loc_id) == 0) { // Keep this locus. d[j] = this->data[i]; new_loc_order[loc_id] = j; new_rev_loc_order[j] = loc_id; j++; } else { // Remove this locus. for (size_t k = 0; k < metapopinfo.samples().size(); k++) delete this->data[i][k]; delete [] this->data[i]; } } delete [] this->data; this->data = d; this->locus_order.clear(); this->locus_order = new_loc_order; this->rev_locus_order.clear(); this->rev_locus_order = new_rev_loc_order; this->num_loci = new_size; // // Re-sort the catalog loci on each chromosome according to base pair. // map > new_ordered_loci; typename map >::iterator cit; for (cit = this->ordered_loci.begin(); cit != this->ordered_loci.end(); cit++) { for (uint k = 0; k < cit->second.size(); k++) { if (remove_ids.count(cit->second[k]->id) == 0) new_ordered_loci[cit->first].push_back(cit->second[k]); } } this->ordered_loci.clear(); this->ordered_loci = new_ordered_loci; for (cit = this->ordered_loci.begin(); cit != this->ordered_loci.end(); cit++) sort(cit->second.begin(), cit->second.end(), bp_compare); return new_size; } template Datum **PopMap::locus(int id) { return this->data[this->locus_order[id]]; } template Datum *PopMap::datum(int loc_id, int sample_id) { return this->data[this->locus_order[loc_id]][metapopinfo.get_sample_index(sample_id)]; } template bool PopMap::blacklisted(int locus, int sample) { if (this->blacklist.count(make_pair(sample, locus)) > 0) return true; else return false; } #endif // __POPMAP_H__ stacks-1.44/src/PopSum.h000644 000765 000024 00000116455 12764047443 015714 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2011-2012, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __POPSUM_H__ #define __POPSUM_H__ #include #include using std::string; #include using std::vector; #include using std::map; #include using std::set; #include #include using std::pair; using std::make_pair; #include #include #include "stacks.h" #include "locus.h" #include "PopMap.h" #include "MetaPopInfo.h" extern bool log_fst_comp; extern double minor_allele_freq; const uint PopStatSize = 5; class PopStat { public: int loc_id; int bp; bool fixed; double alleles; // Number of alleles sampled at this location. uint snp_cnt; // Number of SNPs in kernel-smoothed window centered on this SNP. double stat[PopStatSize]; double smoothed[PopStatSize]; double bs[PopStatSize]; PopStat() { this->loc_id = 0; this->bp = 0; this->fixed = false; this->alleles = 0.0; this->snp_cnt = 0; for (uint i = 0; i < PopStatSize; i++) { this->stat[i] = 0.0; this->smoothed[i] = 0.0; this->bs[i] = 0.0; } } virtual ~PopStat() { } }; class HapStat: public PopStat { // PopStat[0]: Phi_st // PopStat[1]: Phi_ct // PopStat[2]: Phi_sc // PopStat[3]: Fst' // PopStat[4]: D_est public: double *comp; uint popcnt; HapStat(): PopStat() { comp = NULL; popcnt = uint(-1); } ~HapStat() { if (this->comp != NULL) delete [] comp; } }; class LocStat: public PopStat { // PopStat[0]: gene diversity // PopStat[1]: haplotype diversity (Pi) public: uint hap_cnt; // Number of unique haplotypes at this locus. string hap_str; // Human-readable string of haplotype counts. LocStat(): PopStat() { this->hap_cnt = 0; } ~LocStat() {}; }; class PopPair: public PopStat { // PopStat[0]: corrected Fst, (by p-value or Bonferroni p-value). // PopStat[1]: corrected AMOVA Fst public: int col; double pi; double fst; double fet_p; // Fisher's Exact Test p-value. double fet_or; // Fisher's exact test odds ratio. double or_se; // Fisher's exact test odds ratio standard error. double lod; // base 10 logarithm of odds score. double ci_low; // Fisher's exact test lower confidence interval. double ci_high; // Fisher's exact test higher confidence interval. double amova_fst; // AMOVA Fst method, from Weir, Genetic Data Analysis II . double *comp; PopPair() { col = 0; pi = 0.0; fst = 0.0; fet_p = 0.0; fet_or = 0.0; or_se = 0.0; lod = 0.0; ci_low = 0.0; ci_high = 0.0; amova_fst = 0.0; comp = NULL; } ~PopPair() { if (this->comp != NULL) delete [] comp; } }; class SumStat: public PopStat { // PopStat[0]: pi // PopStat[1]: fis public: bool incompatible_site; bool filtered_site; double num_indv; char p_nuc; char q_nuc; double p; double obs_het; double obs_hom; double exp_het; double exp_hom; double π SumStat(): PopStat(), pi(this->stat[0]) { num_indv = 0.0; p = 0.0; p_nuc = 0; q_nuc = 0; obs_het = 0.0; obs_hom = 0.0; exp_het = 0.0; exp_hom = 0.0; snp_cnt = 0; incompatible_site = false; filtered_site = false; } }; class LocSum { public: SumStat *nucs; // Array containing summary statistics for // each nucleotide position at this locus. LocSum(int len) { this->nucs = new SumStat[len]; } ~LocSum() { delete [] this->nucs; } }; class NucTally { public: int loc_id; int bp; uint16_t col; uint16_t num_indv; uint16_t pop_cnt; uint16_t allele_cnt; char p_allele; char q_allele; double p_freq; double obs_het; bool fixed; int priv_allele; NucTally() { loc_id = 0; bp = 0; col = 0; num_indv = 0; pop_cnt = 0; allele_cnt = 0; p_allele = 0; q_allele = 0; p_freq = 0.0; obs_het = 0.0; priv_allele = -1; fixed = true; } }; class LocTally { public: NucTally *nucs; LocTally(int len) { this->nucs = new NucTally[len]; } ~LocTally() { delete [] this->nucs; } }; // // Population Summary class contains a two dimensional array storing the // summary statistics for each locus in each of a set of populations: // // Pop1 Pop2 Pop3 // Locus1 +-LocSum----+-LocSum----+-LocSum // | | | | // | +-SumStat +-SumStat +-SumStat (Nuc0) // | +-SumStat +-SumStat +-SumStat (Nuc1) // | ... // Locus2 +-LocSum----+-LocSum----+-LocSum // | | | | // | +-SumStat +-SumStat +-SumStat (Nuc0) // | +-SumStat +-SumStat +-SumStat (Nuc1) // | ... // Locus3 +-LocSum----+-LocSum----+-LocSum // | | | | // | +-SumStat +-SumStat +-SumStat (Nuc0) // | +-SumStat +-SumStat +-SumStat (Nuc1) // | ... // ... // template class PopSum { const PopMap& popmap; const MetaPopInfo& metapopinfo; LocSum ***data; LocTally **loc_tally; public: PopSum(const PopMap& pmap, const MetaPopInfo& mpopi); ~PopSum(); int initialize(PopMap *); int add_population(map &, PopMap *, size_t, bool, ofstream &); int tally(map &); int loci_cnt() const {return popmap.loci_cnt();} int rev_locus_index(int index) const {return popmap.rev_locus_index(index);} int pop_cnt() const {return metapopinfo.pops().size();} int pop_size(size_t pop_index) const { return metapopinfo.pops().at(pop_index).last_sample - metapopinfo.pops().at(pop_index).first_sample + 1; } LocSum **locus(int locus_id); LocSum *pop(int locus_id, size_t pop_index); LocTally *locus_tally(int locus_id); PopPair *Fst(int, int, int, int); int fishers_exact_test(PopPair *, double, double, double, double); private: int tally_heterozygous_pos(LocusT *, Datum **, LocSum *, int, int, uint, uint); int tally_fixed_pos(LocusT *, Datum **, LocSum *, int, uint, uint); int tally_ref_alleles(LocSum **, int, short unsigned int &, char &, char &, short unsigned int &, short unsigned int &); int tally_observed_haplotypes(vector &, int); double pi(double, double, double); double binomial_coeff(double, double); }; template PopSum::PopSum(const PopMap& pmap, const MetaPopInfo& mpopi) : popmap(pmap), metapopinfo(mpopi) { this->loc_tally = new LocTally*[popmap.loci_cnt()]; this->data = new LocSum**[popmap.loci_cnt()]; for (int i = 0; i < popmap.loci_cnt(); i++) { this->data[i] = new LocSum *[metapopinfo.pops().size()]; for (size_t j = 0; j < metapopinfo.pops().size(); j++) this->data[i][j] = NULL; } } template PopSum::~PopSum() { for (int i = 0; i < loci_cnt(); i++) { for (int j = 0; j < pop_cnt(); j++) delete this->data[i][j]; delete [] this->data[i]; delete this->loc_tally[i]; } delete [] this->data; delete [] this->loc_tally; } template int PopSum::initialize(PopMap *pmap) { return 0; } template int PopSum::add_population(map &catalog, PopMap *pmap, size_t pop_index, bool verbose, ofstream &log_fh) { LocusT *loc; Datum **d; LocSum **s; uint locus_id, len; int res; set snp_cols; int incompatible_loci = 0; if (verbose) log_fh << "\n#\n# Recording sites that have incompatible loci -- loci with too many alleles present.\n" << "#\n" << "# Level\tAction\tLocus ID\tChr\tBP\tColumn\tPopID\n#\n"; const MetaPopInfo::Pop& pop = metapopinfo.pops().at(pop_index); for (int i = 0; i < loci_cnt(); i++) { locus_id = rev_locus_index(i); d = pmap->locus(locus_id); s = this->locus(locus_id); loc = catalog[locus_id]; // // Create an array of SumStat objects // len = strlen(loc->con); s[pop_index] = new LocSum(len); // // Check if this locus has already been filtered and is NULL in all individuals. // bool filtered = true; for (uint k = pop.first_sample; k <= pop.last_sample; k++) { if (d[k] != NULL) filtered = false; } if (filtered == true) { for (uint k = 0; k < len; k++) { s[pop_index]->nucs[k].filtered_site = true; } continue; } // // The catalog records which nucleotides are heterozygous. For these nucleotides we will // calculate observed genotype frequencies, allele frequencies, and expected genotype frequencies. // for (uint k = 0; k < loc->snps.size(); k++) { res = this->tally_heterozygous_pos(loc, d, s[pop_index], loc->snps[k]->col, k, pop.first_sample, pop.last_sample); // // If site is incompatible (too many alleles present), log it. // if (res < 0) { s[pop_index]->nucs[loc->snps[k]->col].incompatible_site = true; incompatible_loci++; if (verbose) log_fh << "within_population\t" << "incompatible_locus\t" << loc->id << "\t" << loc->loc.chr << "\t" << loc->sort_bp(loc->snps[k]->col) +1 << "\t" << loc->snps[k]->col << "\t" << pop.name << "\n"; } snp_cols.insert(loc->snps[k]->col); } // // For all other fixed sites, we just need to record them. // for (uint k = 0; k < len; k++) { if (snp_cols.count(k)) continue; this->tally_fixed_pos(loc, d, s[pop_index], k, pop.first_sample, pop.last_sample); } snp_cols.clear(); } cerr << "Population '" << pop.name << "' contained " << incompatible_loci << " incompatible loci -- more than two alleles present.\n"; log_fh << "Population " << pop.name << " contained " << incompatible_loci << " incompatible loci -- more than two alleles present.\n"; return 0; } template int PopSum::tally(map &catalog) { LocusT *loc; LocSum **s; LocTally *ltally; int locus_id, variable_pop; uint16_t p_cnt, q_cnt, len, col; for (int n = 0; n < loci_cnt(); n++) { locus_id = this->rev_locus_index(n); loc = catalog[locus_id]; s = this->locus(locus_id); len = strlen(loc->con); ltally = new LocTally(len); this->loc_tally[n] = ltally; // for (uint i = 0; i < loc->snps.size(); i++) { // uint col = loc->snps[i]->col; for (col = 0; col < len; col++) { ltally->nucs[col].col = col; ltally->nucs[col].bp = loc->sort_bp(col); ltally->nucs[col].loc_id = locus_id; this->tally_ref_alleles(s, col, ltally->nucs[col].allele_cnt, ltally->nucs[col].p_allele, ltally->nucs[col].q_allele, p_cnt, q_cnt); // // Is this site variable? // if (ltally->nucs[col].allele_cnt > 1) ltally->nucs[col].fixed = false; for (int j = 0; j < pop_cnt(); j++) { // // Sum the number of individuals examined at this locus across populations. // ltally->nucs[col].num_indv += s[j]->nucs[col].num_indv; ltally->nucs[col].pop_cnt += s[j]->nucs[col].num_indv > 0 ? 1 : 0; } for (int j = 0; j < pop_cnt(); j++) { // // Sum the most frequent allele across populations. // if (s[j]->nucs[col].p_nuc == ltally->nucs[col].p_allele) ltally->nucs[col].p_freq += s[j]->nucs[col].p * (s[j]->nucs[col].num_indv / (double) ltally->nucs[col].num_indv); else ltally->nucs[col].p_freq += (1 - s[j]->nucs[col].p) * (s[j]->nucs[col].num_indv / (double) ltally->nucs[col].num_indv); // // Sum observed heterozygosity across populations. // ltally->nucs[col].obs_het += s[j]->nucs[col].obs_het * (s[j]->nucs[col].num_indv / (double) ltally->nucs[col].num_indv); } // // We want to report the most frequent allele as the P allele. Reorder the alleles // if necessary. // XXX Possibly unstable for p_freq ~ 0.5. @Nick (July 2016) // if (ltally->nucs[col].p_freq < 0.5) { char a = ltally->nucs[col].p_allele; ltally->nucs[col].p_allele = ltally->nucs[col].q_allele; ltally->nucs[col].q_allele = a; ltally->nucs[col].p_freq = 1 - ltally->nucs[col].p_freq; uint b = p_cnt; p_cnt = q_cnt; q_cnt = b; } // // Check if this is a private allele. Either the site is variable and // the allele exists in one population, or the site is fixed and one // population is homozygous for the private allele. // variable_pop = -1; if (p_cnt == 1 && q_cnt > 1) { for (int j = 0; j < pop_cnt(); j++) if (s[j]->nucs[col].p_nuc == ltally->nucs[col].p_allele || s[j]->nucs[col].q_nuc == ltally->nucs[col].p_allele) variable_pop = j; } else if (p_cnt > 1 && q_cnt == 1) { for (int j = 0; j < pop_cnt(); j++) if (s[j]->nucs[col].p_nuc == ltally->nucs[col].q_allele || s[j]->nucs[col].q_nuc == ltally->nucs[col].q_allele) variable_pop = j; } ltally->nucs[col].priv_allele = variable_pop; } } return 0; } template int PopSum::tally_ref_alleles(LocSum **s, int snp_index, short unsigned int &allele_cnt, char &p_allele, char &q_allele, short unsigned int &p_cnt, short unsigned int &q_cnt) { int nucs[4] = {0}; char nuc[2]; p_allele = 0; q_allele = 0; allele_cnt = 0; for (int j = 0; j < pop_cnt(); j++) { nuc[0] = 0; nuc[1] = 0; nuc[0] = s[j]->nucs[snp_index].p_nuc; nuc[1] = s[j]->nucs[snp_index].q_nuc; for (uint k = 0; k < 2; k++) switch(nuc[k]) { case 'A': case 'a': nucs[0]++; break; case 'C': case 'c': nucs[1]++; break; case 'G': case 'g': nucs[2]++; break; case 'T': case 't': nucs[3]++; break; } } // // Determine how many alleles are present at this position in this population. // We cannot deal with more than two alternative alleles, if there are more than two // in a single population, print a warning and exclude this nucleotide position. // int i; for (i = 0; i < 4; i++) if (nucs[i] > 0) allele_cnt++; if (allele_cnt > 2) { p_allele = 0; q_allele = 0; return 0; } // // Record which nucleotide is the P allele and which is the Q allele. // i = 0; while (p_allele == 0 && i < 4) { if (nucs[i] > 0) { switch(i) { case 0: p_allele = 'A'; break; case 1: p_allele = 'C'; break; case 2: p_allele = 'G'; break; case 3: p_allele = 'T'; break; } } i++; } while (q_allele == 0 && i < 4) { if (nucs[i] > 0) { switch(i) { case 1: q_allele = 'C'; break; case 2: q_allele = 'G'; break; case 3: q_allele = 'T'; break; } } i++; } // // Tabulate the number of populations the p_allele and the q_allele occur in. // p_cnt = 0; q_cnt = 0; for (int j = 0; j < pop_cnt(); j++) { nuc[0] = 0; nuc[1] = 0; nuc[0] = s[j]->nucs[snp_index].p_nuc; nuc[1] = s[j]->nucs[snp_index].q_nuc; for (uint k = 0; k < 2; k++) if (nuc[k] != 0 && nuc[k] == p_allele) p_cnt++; else if (nuc[k] != 0 && nuc[k] == q_allele) q_cnt++; } return 1; } template PopPair *PopSum::Fst(int locus_id, int pop_1, int pop_2, int pos) { LocSum *s_1 = pop(locus_id, pop_1); /////// SLOW! LocSum *s_2 = pop(locus_id, pop_2); PopPair *pair = new PopPair(); // // If this locus only appears in one population do not calculate Fst. // if (s_1->nucs[pos].num_indv == 0 || s_2->nucs[pos].num_indv == 0) return pair; // // Calculate Fst at a locus, sub-population relative to that found in the entire population // Fst = 1 - (Sum_j( (n_j choose 2) * pi_j)) / (pi_all * Sum_j( (n_j choose 2) )) // double n_1, n_2, pi_1, pi_2; n_1 = s_1->nucs[pos].num_indv * 2; n_2 = s_2->nucs[pos].num_indv * 2; pi_1 = s_1->nucs[pos].pi; pi_2 = s_2->nucs[pos].pi; if (pi_1 == 0 && pi_2 == 0 && s_1->nucs[pos].p_nuc == s_2->nucs[pos].p_nuc) return pair; // // Calculate Pi over the entire pooled population. // // First, make sure this site is compatible between the two populations (no more than two alleles). // char nucs[4]; int ncnt[4] = {0}; nucs[0] = s_1->nucs[pos].p_nuc; nucs[1] = s_1->nucs[pos].q_nuc; nucs[2] = s_2->nucs[pos].p_nuc; nucs[3] = s_2->nucs[pos].q_nuc; for (int i = 0; i < 4; i++) switch(nucs[i]) { case 'A': ncnt[0]++; break; case 'C': ncnt[1]++; break; case 'G': ncnt[2]++; break; case 'T': ncnt[3]++; break; } int allele_cnt = 0; for (int i = 0; i < 4; i++) if (ncnt[i] > 0) allele_cnt++; if (allele_cnt > 2) return NULL; double tot_alleles = n_1 + n_2; double p_1 = round(n_1 * s_1->nucs[pos].p); double q_1 = n_1 - p_1; double p_2 = s_1->nucs[pos].p_nuc == s_2->nucs[pos].p_nuc ? s_2->nucs[pos].p : (1 - s_2->nucs[pos].p); p_2 = round(n_2 * p_2); double q_2 = n_2 - p_2; double pi_all = this->pi(tot_alleles, p_1 + p_2, q_1 + q_2); double bcoeff_1 = this->binomial_coeff(n_1, 2); double bcoeff_2 = this->binomial_coeff(n_2, 2); double num = (bcoeff_1 * pi_1) + (bcoeff_2 * pi_2); double den = pi_all * (bcoeff_1 + bcoeff_2); double Fst = 1 - (num / den); pair->alleles = tot_alleles; pair->fst = Fst; pair->pi = pi_all; this->fishers_exact_test(pair, p_1, q_1, p_2, q_2); // cerr << "Locus: " << locus << ", pos: " << pos << "\n" // << " p_1.nuc: " << s_1->nucs[pos].p_nuc << "; q_1.nuc: " << s_1->nucs[pos].q_nuc // << "; p_2.nuc: " << s_2->nucs[pos].p_nuc << "; q_2.nuc: " << s_2->nucs[pos].q_nuc << "\n" // << " Total alleles: " << tot_alleles << "; " << " s_1.p: " << s_1->nucs[pos].p // << "; s_2.p: " << s_2->nucs[pos].p << "\n" // << " p_1: " << p_1 << "; q_1: " << q_1 << " p_2: " << p_2 << "; q_2: " << q_2 << "\n" // << " Pi1: " << pi_1 << "; Pi2: " << pi_2 << "; PiAll: " << pi_all << "\n" // << " N1: " << n_1 << "; N1 choose 2: " << bcoeff_1 << "\n" // << " N2: " << n_2 << "; N2 choose 2: " << bcoeff_2 << "\n" // << " Fst: " << Fst << "\n"; // // Calculate Fst (corrected for different samples sizes) using an AMOVA method, // correcting for unequal sample sizes. // Derived from Weir, _Genetic Data Analysis II_, chapter 5, "F Statistics,", pp166-167. // double p_1_freq = s_1->nucs[pos].p; double q_1_freq = 1 - p_1_freq; double p_2_freq = s_1->nucs[pos].p_nuc == s_2->nucs[pos].p_nuc ? s_2->nucs[pos].p : (1 - s_2->nucs[pos].p); double q_2_freq = 1 - p_2_freq; double p_avg_cor = ( (s_1->nucs[pos].num_indv * p_1_freq) + (s_2->nucs[pos].num_indv * p_2_freq) ) / ( s_1->nucs[pos].num_indv + s_2->nucs[pos].num_indv ); double n_avg_cor = 2 * ((s_1->nucs[pos].num_indv / 2) + (s_2->nucs[pos].num_indv / 2)); pair->amova_fst = ( (s_1->nucs[pos].num_indv * pow((p_1_freq - p_avg_cor), 2) + s_2->nucs[pos].num_indv * pow((p_2_freq - p_avg_cor), 2)) / n_avg_cor ) / (p_avg_cor * (1 - p_avg_cor)); if (log_fst_comp) { pair->comp = new double[18]; pair->comp[0] = n_1; pair->comp[1] = n_2; pair->comp[2] = tot_alleles; pair->comp[3] = p_1; pair->comp[4] = q_1; pair->comp[5] = p_2; pair->comp[6] = q_2; pair->comp[7] = pi_1; pair->comp[8] = pi_2; pair->comp[9] = pi_all; pair->comp[10] = bcoeff_1; pair->comp[11] = bcoeff_2; pair->comp[12] = p_1_freq; pair->comp[13] = q_1_freq; pair->comp[14] = p_2_freq; pair->comp[15] = q_2_freq; pair->comp[16] = p_avg_cor; pair->comp[17] = n_avg_cor; } // // // // Calculate Fst using a pure parametric method (assumes allele counts are real, not // // samples). Jakobsson, Edge, and Rosenberg. "The Relationship Between Fst and the // // Frequency of the Most Frequent Allele." Genetics 193:515-528. Equation 4. // // // double sigma_1 = p_1_freq + q_1_freq; // double sigma_2 = p_2_freq + q_2_freq; // double delta_1 = fabs(p_1_freq - p_2_freq); // double delta_2 = fabs(q_1_freq - q_2_freq); // pair->jakob_fst = (pow(delta_1, 2) + pow(delta_2, 2)) / ( 4 - (pow(sigma_1, 2) + pow(sigma_2, 2)) ); return pair; } template int PopSum::tally_fixed_pos(LocusT *locus, Datum **d, LocSum *s, int pos, uint start, uint end) { double num_indv = 0.0; char p_nuc = 0; for (uint i = start; i <= end; i++) { if (d[i] == NULL || pos >= d[i]->len) continue; // // Before counting this individual, make sure the model definitively called this // position as hEterozygous or hOmozygous. // if (d[i]->model[pos] == 'E') { cerr << "Model: " << d[i]->model << "\n"; cerr << "Warning: heterozygous model call at fixed nucleotide position: " << "locus " << locus->id << " individual " << d[i]->id << "; position: " << pos << "\n"; } num_indv++; p_nuc = locus->con[pos]; } // // Record the results in the PopSum object. // s->nucs[pos].loc_id = locus->id; s->nucs[pos].bp = locus->sort_bp(pos); s->nucs[pos].fixed = true; s->nucs[pos].num_indv = num_indv; s->nucs[pos].alleles = 2 * num_indv; if (num_indv > 0) { s->nucs[pos].p = 1.0; s->nucs[pos].p_nuc = p_nuc; s->nucs[pos].obs_hom = 1.0; s->nucs[pos].obs_het = 0.0; s->nucs[pos].exp_hom = 1.0; s->nucs[pos].exp_het = 0.0; s->nucs[pos].stat[0] = 0.0; // pi s->nucs[pos].stat[1] = -7.0; // fis } return 0; } template int PopSum::tally_heterozygous_pos(LocusT *locus, Datum **d, LocSum *s, int pos, int snp_index, uint start, uint end) { // // Tally up the genotype frequencies. // int nucs[4] = {0}; uint i; char nuc; //cerr << " Calculating summary stats at het locus " << locus->id << " position " << pos << "; snp_index: " << snp_index << "\n"; // // Iterate over each individual in this sub-population. // for (i = start; i <= end; i++) { if (d[i] == NULL || pos >= d[i]->len || d[i]->model[pos] == 'U') continue; // // Pull each allele for this SNP from the observed haplotype. // for (uint j = 0; j < d[i]->obshap.size(); j++) { nuc = d[i]->obshap[j][snp_index]; switch(nuc) { case 'A': case 'a': nucs[0]++; break; case 'C': case 'c': nucs[1]++; break; case 'G': case 'g': nucs[2]++; break; case 'T': case 't': nucs[3]++; break; } } } // // Determine how many alleles are present at this position in this population. // We cannot deal with more than two alternative alleles, if there are more than two // in a single population, print a warning and exclude this nucleotide position. // int allele_cnt = 0; for (i = 0; i < 4; i++) if (nucs[i] > 0) allele_cnt++; if (allele_cnt > 2) return -1; // // Record which nucleotide is the P allele and which is the Q allele. // char p_allele = 0; char q_allele = 0; i = 0; while (p_allele == 0 && i < 4) { if (nucs[i] > 0) { switch(i) { case 0: p_allele = 'A'; break; case 1: p_allele = 'C'; break; case 2: p_allele = 'G'; break; case 3: p_allele = 'T'; break; } } i++; } while (q_allele == 0 && i < 4) { if (nucs[i] > 0) { switch(i) { case 1: q_allele = 'C'; break; case 2: q_allele = 'G'; break; case 3: q_allele = 'T'; break; } } i++; } //cerr << " P Allele: " << p_allele << "; Q Allele: " << q_allele << "\n"; // // Calculate observed genotype frequencies. // double num_indv = 0.0; double obs_het = 0.0; double obs_p = 0.0; double obs_q = 0.0; for (i = start; i <= end; i++) { if (d[i] == NULL || pos >= d[i]->len) continue; // // Before counting this individual, make sure the model definitively called this // position as hEterozygous or hOmozygous. // if (d[i]->model[pos] == 'E' || d[i]->model[pos] == 'O') num_indv++; else continue; if (d[i]->obshap.size() > 1 && this->tally_observed_haplotypes(d[i]->obshap, snp_index) == 2) obs_het++; else if (d[i]->obshap[0][snp_index] == p_allele) obs_p++; else if (d[i]->obshap[0][snp_index] == q_allele) obs_q++; } //cerr << " Num Individuals: " << num_indv << "; Obs Hets: " << obs_het << "; Obs P: " << obs_p << "; Obs Q: " << obs_q << "\n"; if (num_indv == 0) return 0; // // Calculate total number of alleles // double tot_alleles = num_indv * 2; double allele_p = obs_het + (2 * obs_p); double allele_q = obs_het + (2 * obs_q); // // Calculate Pi, equivalent to expected heterozygosity (exp_het) // s->nucs[pos].stat[0] = this->pi(tot_alleles, allele_p, allele_q); if (s->nucs[pos].stat[0] == 0.0) s->nucs[pos].fixed = true; // // Convert to allele frequencies // allele_p = allele_p / tot_alleles; allele_q = allele_q / tot_alleles; //cerr << " P allele frequency: " << allele_p << "; Q allele frequency: " << allele_q << "\n"; // // // // If the minor allele frequency is below the cutoff, set it to zero. // // // if (minor_allele_freq > 0) { // if (allele_p < allele_q) { // if (allele_p < minor_allele_freq) { // s->nucs[pos].pi = 0.0; // s->nucs[pos].fixed = true; // s->nucs[pos].filtered_site = true; // return 0; // } // } else { // if (allele_q < minor_allele_freq) { // s->nucs[pos].pi = 0.0; // s->nucs[pos].fixed = true; // s->nucs[pos].filtered_site = true; // return 0; // } // } // } // // Calculate expected genotype frequencies. // double exp_het = 2 * allele_p * allele_q; // 2pq // double exp_p = allele_p * allele_p; // p^2 // double exp_q = allele_q * allele_q; // q^2 //cerr << " Expected Het: " << exp_het << "; Expected P: " << exp_p << "; Expected Q: " << exp_q << "\n"; obs_het = obs_het / num_indv; obs_p = obs_p / num_indv; obs_q = obs_q / num_indv; //cerr << " Obs Hets Freq: " << obs_het << "; Obs P Freq: " << obs_p << "; Obs Q Freq: " << obs_q << "\n"; // // Record the results in the PopSum object. // s->nucs[pos].loc_id = locus->id; s->nucs[pos].bp = locus->sort_bp(pos); s->nucs[pos].num_indv = num_indv; s->nucs[pos].alleles = tot_alleles; s->nucs[pos].p = allele_p > allele_q ? allele_p : allele_q; s->nucs[pos].p_nuc = allele_p > allele_q ? p_allele : q_allele; s->nucs[pos].q_nuc = allele_p > allele_q ? q_allele : p_allele; s->nucs[pos].obs_hom = 1 - obs_het; s->nucs[pos].obs_het = obs_het; s->nucs[pos].exp_hom = 1 - exp_het; s->nucs[pos].exp_het = exp_het; // // Calculate F_is, the inbreeding coefficient of an individual (I) relative to the subpopulation (S): // Fis = (exp_het - obs_het) / exp_het // double fis = s->nucs[pos].pi == 0 ? -7 : (s->nucs[pos].pi - obs_het) / s->nucs[pos].pi; s->nucs[pos].stat[1] = fis; return 0; } template int PopSum::tally_observed_haplotypes(vector &obshap, int snp_index) { int nucs[4] = {0}; char nuc; // // Pull each allele for this SNP from the observed haplotype. // for (uint j = 0; j < obshap.size(); j++) { nuc = obshap[j][snp_index]; switch(nuc) { case 'A': case 'a': nucs[0]++; break; case 'C': case 'c': nucs[1]++; break; case 'G': case 'g': nucs[2]++; break; case 'T': case 't': nucs[3]++; break; } } int allele_cnt = 0; for (int i = 0; i < 4; i++) if (nucs[i] > 0) allele_cnt++; return allele_cnt; } template int PopSum::fishers_exact_test(PopPair *pair, double p_1, double q_1, double p_2, double q_2) { // | Allele1 | Allele2 | // Fisher's Exact Test: -----+---------+---------+ // Pop1 | p_1 | q_1 | // Pop2 | p_2 | q_2 | // Probability p: // p = ((p_1 + q_1)!(p_2 + q_2)!(p_1 + p_2)!(q_1 + q_2)!) / (n! p_1! q_1! p_2! q_2!) // // According to: // Jerrold H. Zar, "A fast and efficient algorithm for the Fisher exact test." // Behavior Research Methods, Instruments, & Computers 1987, 19(4): 43-44 // // Probability p can be calculated as three binomial coefficients: // Let p_1 + q_1 = r_1; p_2 + q_2 = r_2; p_1 + p_2 = c_1; q_1 + q_2 = c_2 // // p = (r_1 choose p_1)(r_2 choose p_2) / (n choose c_1) // // Fisher's Exact test algorithm implemented according to Sokal and Rohlf, _Biometry_, section 17.4. // double r_1 = p_1 + q_1; double r_2 = p_2 + q_2; double c_1 = p_1 + p_2; double d_1 = p_1 * q_2; double d_2 = p_2 * q_1; double n = r_1 + r_2; double p = 0.0; // char p1_str[32], p2_str[32], q1_str[32], q2_str[32]; // sprintf(p1_str, "% 3.0f", p_1); // sprintf(q1_str, "% 3.0f", q_1); // sprintf(p2_str, "% 3.0f", p_2); // sprintf(q2_str, "% 3.0f", q_2); // // cerr // << " | Allele1 | Allele2 | " << "\n" // << "-----+---------+---------+" << "\n" // << "Pop1 | " << p1_str << " | " << q1_str << " |" << "\n" // << "Pop2 | " << p2_str << " | " << q2_str << " |" << "\n\n"; // // Compute the first tail. // double p1 = p_1; double q1 = q_1; double p2 = p_2; double q2 = q_2; double tail_1 = 0.0; double den = this->binomial_coeff(n, c_1); // // If (p_1*q_2 - p_2*q_1) < 0 decrease cells p_1 and q_2 by one and add one to p_2 and q_1. // Compute p and repeat until one or more cells equal 0. // if (d_1 - d_2 < 0) { do { p = (this->binomial_coeff(r_1, p1) * this->binomial_coeff(r_2, p2)) / den; tail_1 += p; p1--; q2--; p2++; q1++; } while (p1 >= 0 && q2 >= 0); } else { // // Else, if (p_1*q_2 - p_2*q_1) > 0 decrease cells p_2 and q_1 by one and add one to p_1 and q_2. // Compute p and repeat until one or more cells equal 0. // do { p = (this->binomial_coeff(r_1, p1) * this->binomial_coeff(r_2, p2)) / den; tail_1 += p; p2--; q1--; p1++; q2++; } while (p2 >= 0 && q1 >= 0); } // // Compute the second tail. // double tail_2 = 0.0; p = 0; // // If (p_1*q_2 - p_2*q_1) < 0, set to zero the smaller of the two frequencies, adjusting the other values // to keep the marginals the same. // if (d_1 - d_2 < 0) { if (p2 < q1) { q2 += p2; p1 += p2; q1 -= p2; p2 = 0; } else { p1 += q1; q2 += q1; p2 -= q1; q1 = 0; } } else { if (p1 < q2) { q1 += p1; p2 += p1; q2 -= p1; p1 = 0; } else { p2 += q2; q1 += q2; p1 -= q2; q2 = 0; } } // // If (p_1*q_2 - p_2*q_1) < 0 decrease cells p_1 and q_2 by one and add one to p_2 and q_1. // Compute p and repeat until tail_2 > tail_1. // if (d_1 - d_2 < 0) { do { p = (this->binomial_coeff(r_1, p1) * this->binomial_coeff(r_2, p2)) / den; tail_2 += p; p1--; q2--; p2++; q1++; } while (tail_2 < tail_1 && p1 >= 0 && q2 >= 0); tail_2 -= p; } else { // // Else, if (p_1*q_2 - p_2*q_1) > 0 decrease cells p_2 and q_1 by one and add one to p_1 and q_2. // Compute p and repeat until one or more cells equal 0. // do { p = (this->binomial_coeff(r_1, p1) * this->binomial_coeff(r_2, p2)) / den; tail_2 += p; p2--; q1--; p1++; q2++; } while (tail_2 < tail_1 && p2 >= 0 && q1 >= 0); tail_2 -= p; } pair->fet_p = tail_1 + tail_2; if (pair->fet_p > 1.0) pair->fet_p = 1.0; // // Calculate the odds ratio. To account for possible cases were one allele frequency is // zero, we will increment all allele frequencies by one. // if (p_1 == 0 || q_1 == 0 || p_2 == 0 || q_2 == 0) { p_1++; q_1++; p_2++; q_2++; } pair->fet_or = (p_1 * q_2) / (q_1 * p_2); double ln_fet_or = pair->fet_or > 0 ? log(pair->fet_or) : 0.0; // // Calculate the standard error of the natural log of the odds ratio // double se = pair->fet_or > 0 ? sqrt((1 / p_1) + (1 / q_1) + (1 / p_2) + (1 / q_2)) : 0.0; // // Calculate the confidence intervals of the natural log of the odds ratio. // double ln_ci_low = pair->fet_or > 0 ? ln_fet_or - (1.96 * se) : 0; double ln_ci_high = pair->fet_or > 0 ? ln_fet_or + (1.96 * se) : 0; // // Convert the confidence intervals out of natural log space // pair->ci_low = pair->fet_or > 0 ? exp(ln_ci_low) : 0; pair->ci_high = pair->fet_or > 0 ? exp(ln_ci_high) : 0; pair->lod = fabs(log10(pair->fet_or)); return 0; } template double PopSum::pi(double tot_alleles, double p, double q) { // // Calculate Pi, equivalent to expected heterozygosity: // pi = 1 - Sum_i( (n_i choose 2) ) / (n choose 2) // double pi = this->binomial_coeff(p, 2) + this->binomial_coeff(q, 2); pi = pi / binomial_coeff(tot_alleles, 2); pi = 1 - pi; return pi; } template double PopSum::binomial_coeff(double n, double k) { if (n < k) return 0.0; // // Compute the binomial coefficient using the method of: // Y. Manolopoulos, "Binomial coefficient computation: recursion or iteration?", // ACM SIGCSE Bulletin, 34(4):65-67, 2002. // double r = 1.0; double s = (k < n - k) ? n - k + 1 : k + 1; for (double i = n; i >= s; i--) r = r * i / (n - i + 1); return r; } template LocSum **PopSum::locus(int locus_id) { return this->data[popmap.locus_index(locus_id)]; } template LocSum *PopSum::pop(int locus_id, size_t pop_index) { return this->data[popmap.locus_index(locus_id)][pop_index]; } template LocTally *PopSum::locus_tally(int locus_id) { return this->loc_tally[popmap.locus_index(locus_id)]; } #endif // __POPSUM_H__ stacks-1.44/src/populations.cc000644 000765 000024 00000621501 13001006117 017147 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2012-2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // populations -- generate population genetic statistics and output // haplotypes in a population context. // #include #include #include #include "MetaPopInfo.h" #include "export_formats.h" #include "populations.h" using namespace std; typedef MetaPopInfo::Sample Sample; typedef MetaPopInfo::Pop Pop; typedef MetaPopInfo::Group Group; extern int encoded_gtypes[4][4]; // Global variables to hold command-line options. InputMode input_mode = InputMode::stacks; int num_threads = 1; int batch_id = -1; string in_path; string in_vcf_path; string out_path; string pmap_path; string bl_file; string wl_file; string bs_wl_file; string enz; double sigma = 150000.0; double sample_limit = 0.0; int population_limit = 1; bool calc_fstats = false; bool bootstrap = false; bool bootstrap_fst = false; bool bootstrap_pifis = false; bool bootstrap_phist = false; bool bootstrap_div = false; bs_type bootstrap_type = bs_exact; int bootstrap_reps = 100; bool bootstrap_wl = false; bool write_single_snp = false; bool write_random_snp = false; bool merge_sites = false; bool expand_id = false; bool sql_out = false; bool vcf_out = false; bool vcf_haplo_out = false; bool fasta_out = false; bool fasta_strict_out = false; bool genepop_out = false; bool genomic_out = false; bool structure_out = false; bool phase_out = false; bool fastphase_out = false; bool beagle_out = false; bool beagle_phased_out = false; bool plink_out = false; bool hzar_out = false; bool treemix_out = false; bool phylip_out = false; bool phylip_var = false; bool phylip_var_all = false; bool ordered_export = false; bool kernel_smoothed = false; bool loci_ordered = false; bool log_fst_comp = false; bool verbose = false; bool filter_lnl = false; double lnl_limit = 0.0; int min_stack_depth = 0; double merge_prune_lim = 1.0; double minor_allele_freq = 0.0; double max_obs_het = 1.0; double p_value_cutoff = 0.05; corr_type fst_correction = no_correction; set debug_flags; string out_prefix; MetaPopInfo mpopi; set blacklist, bootstraplist; map > whitelist; // // Hold information about restriction enzymes // map renz; map renz_cnt; map renz_len; map renz_olap; int main (int argc, char* argv[]) { #ifndef HAVE_LIBZ cerr << "Stacks was compiled without zlib, and will refuse to parse compressed files.\n"; #endif // // Initialize the globals that need it. // initialize_renz(renz, renz_cnt, renz_len); initialize_renz_olap(renz_olap); srandom(time(NULL)); // // Parse the command line. // parse_command_line(argc, argv); cerr << "Fst kernel smoothing: " << (kernel_smoothed == true ? "on" : "off") << "\n" << "Bootstrap resampling: "; if (bootstrap) cerr << "on, " << (bootstrap_type == bs_exact ? "exact; " : "approximate; ") << bootstrap_reps << " reptitions\n"; else cerr << "off\n"; cerr << "Percent samples limit per population: " << sample_limit << "\n" << "Locus Population limit: " << population_limit << "\n" << "Minimum stack depth: " << min_stack_depth << "\n" << "Log liklihood filtering: " << (filter_lnl == true ? "on" : "off") << "; threshold: " << lnl_limit << "\n" << "Minor allele frequency cutoff: " << minor_allele_freq << "\n" << "Maximum observed heterozygosity cutoff: " << max_obs_het << "\n" << "Applying Fst correction: "; switch(fst_correction) { case p_value: cerr << "P-value correction.\n"; break; case bonferroni_win: cerr << "Bonferroni correction within sliding window.\n"; break; case bonferroni_gen: cerr << "Bonferroni correction across genome wide sites.\n"; break; case no_correction: cerr << "none.\n"; break; } // // Open and initialize the log file. // struct stat path_stat; if (stat(out_path.c_str(), &path_stat) == 0) { // Path exists, check that it is a directory DIR* d = opendir(out_path.c_str()); if (d == NULL) { cerr << "Error: Failed to open '" << out_path << "' as a directory.\n"; return -1; } closedir(d); } else if (mkdir(out_path.c_str(), ACCESSPERMS) != 0) { // Failed to create the directory. cerr << "Error: Failed to create directory '" << out_path << "'.\n"; return -1; } string log_path = out_path + out_prefix + ".populations.log"; ofstream log_fh(log_path.c_str(), ofstream::out); if (log_fh.fail()) { cerr << "Error opening log file '" << log_path << "'\n"; return -1; } init_log(log_fh, argc, argv); log_fh << flush; // // Set the number of OpenMP parallel threads to execute. // #ifdef _OPENMP omp_set_num_threads(num_threads); #endif // // Initialize the catalog and the MetaPopInfo // map catalog; // We need some objects in the main scope for each mode. vector > catalog_matches; VcfHeader *vcf_header = NULL; vector *vcf_records = NULL; // Read the population map file, if any. if (not pmap_path.empty()) { cerr << "Parsing population map...\n"; mpopi.init_popmap(pmap_path); if (mpopi.samples().empty()) { cerr << "Error: Failed to open or parse population map file '" << pmap_path << "'.\n"; return -1; } cerr << "The population map contained " << mpopi.samples().size() << " samples, " << mpopi.pops().size() << " population(s), " << mpopi.groups().size() << " group(s).\n"; } if (input_mode == InputMode::stacks) { // // Stacks mode // if (pmap_path.empty()) { cerr << "No population map specified, building file list...\n"; mpopi.init_directory(in_path); if (mpopi.samples().empty()) { cerr << "Error: Failed to find sample files in directory '" << in_path << "'.\n"; return -1; } } // Check that at least one sample file exists in the directory. bool dir_good = false; for (vector::const_iterator s=mpopi.samples().begin(); s!=mpopi.samples().end(); ++s) { ifstream f; string path = in_path + s->name + ".matches.tsv"; f.open(path); if (f.is_open()) { dir_good = true; break; } #if HAVE_LIBZ path += ".gz"; gzFile g = gzopen(path.c_str(), "rb"); if (g != NULL) { dir_good = true; gzclose(g); break; } #endif } if (!dir_good) { cerr << "Error: Unable to locate any file in input directory '" << in_path << "'.\n"; return -1; } // // Load the catalog // cerr << "Reading the catalog...\n"; string catalog_prefix = in_path + "batch_" + to_string(batch_id) + ".catalog"; bool compressed = false; int res = load_loci(catalog_prefix, catalog, false, false, compressed); if (res == 0) { cerr << "Unable to load the catalog '" << catalog_prefix << "'\n"; return -1; } // // Load the matches // cerr << "Reading matches to the catalog...\n"; vector samples_to_remove; set seen_samples; for (size_t i = 0; i < mpopi.samples().size(); ++i) { catalog_matches.push_back(vector()); vector& m = catalog_matches.back(); load_catalog_matches(in_path + mpopi.samples().at(i).name, m); if (m.size() == 0) { cerr << "Warning: Absent or malformed matches file '" << mpopi.samples()[i].name << ".matches.tsv(.gz)" <<"', excluding this sample from population analysis.\n"; samples_to_remove.push_back(i); catalog_matches.pop_back(); // This introduces an index shift between catalog_matches and [i]/[mpopi], // which will be resolved by a call to MetaPopInfo::delete_samples(). continue; } size_t sample_id = m[0]->sample_id; if (seen_samples.count(sample_id) > 0) { cerr << "Error: sample ID " << sample_id << " occurs twice in this data set, likely the pipeline was run incorrectly.\n"; exit(0); } seen_samples.insert(sample_id); mpopi.set_sample_id(i, sample_id); } mpopi.delete_samples(samples_to_remove); if (mpopi.samples().size() == 0) { cerr << "Error: Couln't find any matches files.\n"; return -1; } // [mpopi] is definitive. } else if (input_mode == InputMode::vcf) { // // VCF mode // // Open the VCF file cerr << "Opening the VCF file...\n"; VcfAbstractParser* parser = Vcf::adaptive_open(in_vcf_path); if (parser == NULL) { cerr << "Error: Unable to open VCF file '" << in_vcf_path << "'.\n"; return -1; } parser->read_header(); if (parser->header().samples().empty()) { cerr << "Error: No samples in VCF file '" << in_vcf_path << "'.\n"; return -1; } // Reconsider the MetaPopInfo in light of the VCF header. if (pmap_path.empty()) { cerr << "No population map specified, creating one from the VCF header...\n"; mpopi.init_names(parser->header().samples()); } else { // Intersect the samples present in the population map and the VCF. vector samples_to_discard; for (size_t i=0; iheader().sample_indexes().count(mpopi.samples()[i].name)) samples_to_discard.push_back(i); if (not samples_to_discard.empty()) { if (samples_to_discard.size() == mpopi.samples().size()) { cerr << "Error: No common samples between the population map and VCF header.\n"; return -1; } cerr << "Warning: of the samples listed in the population map, " << samples_to_discard.size() << " could not be found in the VCF :"; for (const size_t& s : samples_to_discard) cerr << " " << mpopi.samples()[s].name; cerr << "\n"; mpopi.delete_samples(samples_to_discard); } // Create arbitrary sample IDs. for (size_t i = 0; i < mpopi.samples().size(); ++i) mpopi.set_sample_id(i, i+1); //id=i+1 // [mpopi] is definitive. } // Read the SNP records cerr << "Reading the VCF records...\n"; vcf_records = new vector(); vector skipped_notsnp; vector skipped_filter; vcf_records->push_back(VcfRecord()); VcfRecord* rec = & vcf_records->back(); while (parser->next_record(*rec)) { // Check for a SNP. if (not rec->is_snp()) { skipped_notsnp.push_back(parser->line_number()); continue; } // Check for a filtered-out SNP if (not rec->filter.empty() && rec->filter[0] != "PASS") { skipped_filter.push_back(parser->line_number()); continue; } // Save the SNP. vcf_records->push_back(VcfRecord()); rec = & vcf_records->back(); } vcf_records->pop_back(); cerr << "Found " << vcf_records->size() << " SNP records in file '" << in_vcf_path << "'. (Skipped " << skipped_filter.size() << " already filtered-out SNPs and " << skipped_notsnp.size() << " non-SNP records ; more with --verbose.)\n"; if (verbose && not skipped_notsnp.empty()) { log_fh << "The following VCF record lines were determined not to be SNPs and skipped :"; for (vector::const_iterator l=skipped_notsnp.begin(); l!=skipped_notsnp.end(); ++l) log_fh << " " << *l; log_fh << "\n"; } if (vcf_records->size() == 0) { cerr << "Error: No records.\n"; return -1; } catalog = create_catalog(*vcf_records); vcf_header = new VcfHeader(parser->header()); delete parser; } // // Read the blacklist, the whitelist, and the bootstrap-whitelist. // if (bl_file.length() > 0) { load_marker_list(bl_file, blacklist); cerr << "Loaded " << blacklist.size() << " blacklisted markers.\n"; } if (wl_file.length() > 0) { load_marker_column_list(wl_file, whitelist); cerr << "Loaded " << whitelist.size() << " whitelisted markers.\n"; check_whitelist_integrity(catalog, whitelist); } if (bs_wl_file.length() > 0) { load_marker_list(bs_wl_file, bootstraplist); cerr << "Loaded " << bootstraplist.size() << " markers to include when bootstrapping.\n"; } // // Reduce the catalog accordingly, and retrieve the genomic order of loci. // reduce_catalog(catalog, whitelist, blacklist); loci_ordered = order_unordered_loci(catalog); // Report information on the MetaPopInfo. cerr << "Working on " << mpopi.samples().size() << " samples.\n"; cerr << "Working on " << mpopi.pops().size() << " population(s):\n"; for (vector::const_iterator p = mpopi.pops().begin(); p != mpopi.pops().end(); p++) { cerr << " " << p->name << ": "; for (size_t s = p->first_sample; s < p->last_sample; ++s) { cerr << mpopi.samples()[s].name << ", "; } cerr << mpopi.samples()[p->last_sample].name << "\n"; } cerr << "Working on " << mpopi.groups().size() << " group(s) of populations:\n"; for (vector::const_iterator g = mpopi.groups().begin(); g != mpopi.groups().end(); g++) { cerr << " " << g->name << ": "; for (vector::const_iterator p = g->pops.begin(); p != g->pops.end() -1; ++p) { //rem. end()-1 and back() are safe, there's always at least one pop cerr << mpopi.pops()[*p].name << ", "; } cerr << mpopi.pops()[g->pops.back()].name << "\n"; } if (size_t(population_limit) > mpopi.pops().size()) { cerr << "Notice: Population limit (" << population_limit << ")" << " larger than number of popualtions present, adjusting parameter to " << mpopi.pops().size() << "\n"; population_limit = mpopi.pops().size(); } // // Initialize the PopMap // cerr << "Populating observed haplotypes for " << mpopi.samples().size() << " samples, " << catalog.size() << " loci.\n"; PopMap *pmap = new PopMap(mpopi, catalog.size()); if (input_mode == InputMode::stacks) { // Using SStacks matches files... pmap->populate(catalog, catalog_matches); for(vector >::iterator sample = catalog_matches.begin(); sample != catalog_matches.end(); ++sample) for(vector::iterator match = sample->begin(); match != sample->end(); ++match) delete *match; catalog_matches.clear(); } else if (input_mode == InputMode::vcf) { // ...or using VCF records. pmap->populate(catalog, *vcf_records, *vcf_header); delete vcf_records; delete vcf_header; vcf_records = NULL; vcf_header = NULL; } // // Tabulate haplotypes present and in what combinations. // tabulate_haplotypes(catalog, pmap); // // Output a list of heterozygous loci and the associate haplotype frequencies. // if (sql_out) write_sql(catalog, pmap); log_fh << "# Distribution of population loci.\n"; log_haplotype_cnts(catalog, log_fh); apply_locus_constraints(catalog, pmap, log_fh); log_fh << "# Distribution of population loci after applying locus constraints.\n"; log_haplotype_cnts(catalog, log_fh); if (input_mode == InputMode::stacks) { // // Load the output from the SNP calling model (hOm/hEt/Unk) for // each individual at each locus. // cerr << "Loading model outputs for " << mpopi.samples().size() << " samples, " << catalog.size() << " loci.\n"; map::iterator it; map::iterator mit; Datum *d; CSLocus *loc; for (uint i = 0; i < mpopi.samples().size(); i++) { map modres; load_model_results(in_path + mpopi.samples()[i].name, modres); if (modres.size() == 0) { cerr << " Warning: unable to find any model results in file '" << mpopi.samples()[i].name << ".models.tsv(.gz)', excluding this sample from population analysis.\n"; continue; } for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; d = pmap->datum(loc->id, mpopi.samples()[i].id); if (d != NULL) { if (modres.count(d->id) == 0) { cerr << "Fatal error: Unable to find model data for catalog locus " << loc->id << ", sample ID " << mpopi.samples()[i].id << ", sample locus " << d->id << "; likely IDs were mismatched when running pipeline.\n"; exit(0); } d->add_model(modres[d->id]->model); } } for (mit = modres.begin(); mit != modres.end(); mit++) delete mit->second; modres.clear(); } } // // Create the PopSum object and compute the summary statistics. // PopSum *psum = new PopSum(*pmap, mpopi); for (size_t i=0; iadd_population(catalog, pmap, i, verbose, log_fh); } cerr << "Tallying loci across populations..."; psum->tally(catalog); cerr << "done.\n"; // // We have removed loci that were below the -r and -p thresholds. Now we need to // identify individual SNPs that are below the -r threshold or the minor allele // frequency threshold (-a). In these cases we will remove the SNP, but keep the locus. // blacklist.clear(); int pruned_snps = prune_polymorphic_sites(catalog, pmap, psum, whitelist, blacklist, log_fh); cerr << "Pruned " << pruned_snps << " variant sites due to filter constraints (more with --verbose).\n"; // // Create an artificial whitelist if the user requested only the first or a random SNP per locus. // if (write_single_snp) implement_single_snp_whitelist(catalog, psum, whitelist); else if (write_random_snp) implement_random_snp_whitelist(catalog, psum, whitelist); // // Remove the accumulated SNPs // cerr << "Removing " << blacklist.size() << " additional loci for which all variant sites were filtered..."; set empty_list; reduce_catalog(catalog, empty_list, blacklist); reduce_catalog_snps(catalog, whitelist, pmap); int retained = pmap->prune(blacklist); cerr << " retained " << retained << " loci.\n"; // // Merge loci that overlap on a common restriction enzyme cut site. // map > merge_map; if (merge_sites && loci_ordered) merge_shared_cutsite_loci(catalog, pmap, psum, merge_map, log_fh); // // Regenerate summary statistics after pruning SNPs and merging loci. // if (debug_flags.count("VCFCOMP")) vcfcomp_simplify_pmap(catalog, pmap); delete psum; psum = new PopSum(*pmap, mpopi); for (size_t i=0; iadd_population(catalog, pmap, i, verbose, log_fh); } cerr << "Re-tallying loci across populations..."; psum->tally(catalog); cerr << "done.\n"; if (kernel_smoothed) { if (loci_ordered) { for (size_t i=0; i& catalog, PopMap* pmap) { cerr << "DEBUG Deleting information from the pmap & catalog so that they resemble what can be retrieved from a VCF.\n"; // n.b. In this configuration we only have one SNP per locus so we don't have // to worry about what U's imply regarding haplotypes. size_t n_deleted = 0; size_t n_loci = pmap->loci_cnt(); size_t n_samples = pmap->sample_cnt(); for (size_t l=0; lrev_locus_index(l)); if (loc->snps.empty()) continue; if (loc->snps.size() > 1) { cerr << "Error: This requires --write_single_snp.\n"; throw exception(); } size_t col = loc->snps.at(0)->col; Datum** datums = pmap->locus(loc->id); for (size_t s=0; slen) <= col || datums[s]->model[col] == 'U' || datums[s]->obshap.empty() || strcmp(datums[s]->obshap[0], "N") == 0 ) { delete datums[s]; datums[s] = NULL; --loc->cnt; --loc->hcnt; ++n_deleted; } } } cerr << "? Deleted " << n_deleted << " 'Datums'.\n"; set myblacklist; // All NULL. for (auto& l : catalog) { Datum** data = pmap->locus(l.second->id); size_t non_null = 0; for (size_t i=0; i < size_t(pmap->sample_cnt()); ++i) if (data[i]!=NULL) ++non_null; if (non_null == 0) myblacklist.insert(l.second->id); } // Not two alleles. for (auto& l : catalog) { if (l.second->snps.empty() || l.second->snps[0]->rank_2 == 0 || l.second->snps[0]->rank_3 != 0 || l.second->alleles.size() != 2) { myblacklist.insert(l.second->id); } else { // Check the actual number of alleles Datum** data = pmap->locus(l.second->id); set seen_alleles; for (size_t i=0; isample_cnt()); ++i) { Datum* d = data[i]; if (d != NULL) for(char* hapl : d->obshap) seen_alleles.insert(hapl[0]); } if ((int((seen_alleles.count('A') || seen_alleles.count('a'))) + (seen_alleles.count('C') || seen_alleles.count('c')) + (seen_alleles.count('T') || seen_alleles.count('t')) + (seen_alleles.count('G') || seen_alleles.count('g'))) != 2) myblacklist.insert(l.second->id); } } // Same SNP in different loci. for (auto& chr : pmap->ordered_loci) { map > seen_bp0; // (bp, [loc_id's]) for (CSLocus* loc : chr.second) if (not loc->snps.empty()) seen_bp0[loc->sort_bp(loc->snps[0]->col)].push_back(loc->id); for (auto& bp : seen_bp0) if (bp.second.size() > 1) for (size_t loc_id : bp.second) myblacklist.insert(loc_id); } set empty; reduce_catalog(catalog, empty, myblacklist); pmap->prune(myblacklist); cerr << "? Now working on " << catalog.size() << " loci (deleted " << myblacklist.size() << " loci).\n"; } int apply_locus_constraints(map &catalog, PopMap *pmap, ofstream &log_fh) { uint pop_sthg; CSLocus *loc; Datum **d; if (sample_limit == 0 && population_limit == 0 && min_stack_depth == 0) return 0; if (verbose) log_fh << "\n#\n# List of loci removed by first filtering stage of sample and population constraints\n#\n" << "# Action\tLocus ID\tChr\tBP\tColumn\tReason\n"; map::iterator it; uint pop_cnt = mpopi.pops().size(); int *pop_order = new int [pop_cnt]; // Which population each sample belongs to. int *samples = new int [pmap->sample_cnt()]; // For the current locus, how many samples in each population. int *pop_cnts = new int [pop_cnt]; // The total number of samples in each population. int *pop_tot = new int [pop_cnt]; pop_sthg = 0; for (size_t i_pop=0; i_pop blacklist; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; d = pmap->locus(loc->id); for (int i = 0; i < pmap->sample_cnt(); i++) { // // Check that each sample is over the minimum stack depth for this locus. // if (d[i] != NULL && min_stack_depth > 0 && d[i]->tot_depth < min_stack_depth) { below_stack_dep++; delete d[i]; d[i] = NULL; loc->hcnt--; } // // Check that each sample is over the log likelihood threshold. // if (d[i] != NULL && filter_lnl && d[i]->lnl < lnl_limit) { below_lnl_thresh++; delete d[i]; d[i] = NULL; loc->hcnt--; } } // // Tally up the count of samples in this population. // for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] != NULL) pop_cnts[samples[i]]++; } // // Check that the counts for each population are over sample_limit. If not, zero out // the members of that population. // for (uint i = 0; i < pop_cnt; i++) { const Pop& pop = mpopi.pops()[pop_order[i]]; pct = (double) pop_cnts[i] / (double) pop_tot[i]; if (pop_cnts[i] > 0 && pct < sample_limit) { //cerr << "Removing population " << pop_order[i] << " at locus: " << loc->id << "; below sample limit: " << pct << "\n"; for (uint j = pop.first_sample; j <= pop.last_sample; j++) { if (d[j] != NULL) { delete d[j]; d[j] = NULL; loc->hcnt--; } } pop_cnts[i] = 0; } } // // Check that this locus is present in enough populations. // for (uint i = 0; i < pop_cnt; i++) if (pop_cnts[i] > 0) pops++; if (pops < population_limit) { //cerr << "Removing locus: " << loc->id << "; below population limit: " << pops << "\n"; pop_limit = true; } if (pop_limit) { blacklist.insert(loc->id); if (verbose) log_fh << "removed_locus\t" << loc->id << "\t" << loc->loc.chr << "\t" << loc->sort_bp() +1 << "\t" << 0 << "\tfailed_population_limit\n"; } for (uint i = 0; i < pop_cnt; i++) pop_cnts[i] = 0; pop_limit = false; pops = 0; } // // Remove loci // if (min_stack_depth > 0) cerr << "Removed " << below_stack_dep << " samples from loci that are below the minimum stack depth of " << min_stack_depth << "x\n"; if (filter_lnl) cerr << "Removed " << below_lnl_thresh << " samples from loci that are below the log likelihood threshold of " << lnl_limit << "\n"; cerr << "Removing " << blacklist.size() << " loci that did not pass sample/population constraints..."; set whitelist; reduce_catalog(catalog, whitelist, blacklist); int retained = pmap->prune(blacklist); cerr << " retained " << retained << " loci.\n"; delete [] pop_cnts; delete [] pop_tot; delete [] pop_order; delete [] samples; if (retained == 0) exit(0); return 0; } int prune_polymorphic_sites(map &catalog, PopMap *pmap, PopSum *psum, map > &whitelist, set &blacklist, ofstream &log_fh) { map > new_wl; vector pop_prune_list; CSLocus *loc; LocTally *t; LocSum **s; Datum **d; bool sample_prune, maf_prune, het_prune, inc_prune; int size, pruned = 0; if (verbose) log_fh << "\n#\n# List of pruned nucleotide sites\n#\n" << "# Action\tLocus ID\tChr\tBP\tColumn\tReason\n"; // // If the whitelist is populated, use it as a guide for what loci to consider. // // Construct a new whitelist along the way, that is a subset of the existing list. // if (whitelist.size() > 0) { map >::iterator it; for (it = whitelist.begin(); it != whitelist.end(); it++) { // // A locus on the whitelist may have already been filtered out. // if (catalog.count(it->first) == 0) continue; loc = catalog[it->first]; t = psum->locus_tally(loc->id); s = psum->locus(loc->id); // // Check that each SNP in this locus is above the sample_limit and that // each SNP is above the minor allele frequency. If so, add it back to // the whiteliest. // size = it->second.size(); for (uint i = 0; i < loc->snps.size(); i++) { // // If it is not already in the whitelist, ignore it. // if (size > 0 && it->second.count(loc->snps[i]->col) == 0) continue; // // If the site is fixed, ignore it. // if (t->nucs[loc->snps[i]->col].fixed == true) continue; sample_prune = false; maf_prune = false; het_prune = false; inc_prune = false; pop_prune_list.clear(); for (size_t p=0; pnucs[loc->snps[i]->col].incompatible_site) inc_prune = true; else if (s[p]->nucs[loc->snps[i]->col].num_indv == 0 || (double) s[p]->nucs[loc->snps[i]->col].num_indv / (double) psum->pop_size(p) < sample_limit) pop_prune_list.push_back(p); } // // Check how many populations have to be pruned out due to sample limit. If less than // population limit, prune them; if more than population limit, mark locus for deletion. // if ((mpopi.pops().size() - pop_prune_list.size()) < (uint) population_limit) { sample_prune = true; } else { for (size_t p : pop_prune_list) { if (s[p]->nucs[loc->snps[i]->col].num_indv == 0) continue; d = pmap->locus(loc->id); const Pop& pop = mpopi.pops()[p]; for (uint k = pop.first_sample; k <= pop.last_sample; k++) { if (d[k] == NULL || loc->snps[i]->col >= (uint) d[k]->len) continue; if (d[k]->model != NULL) { d[k]->model[loc->snps[i]->col] = 'U'; } } } } if (t->nucs[loc->snps[i]->col].allele_cnt > 1) { // // Test for minor allele frequency. // if ((1 - t->nucs[loc->snps[i]->col].p_freq) < minor_allele_freq) maf_prune = true; // // Test for observed heterozygosity. // if (t->nucs[loc->snps[i]->col].obs_het > max_obs_het) het_prune = true; } if (maf_prune == false && het_prune == false && sample_prune == false && inc_prune == false) { new_wl[loc->id].insert(loc->snps[i]->col); } else { pruned++; if (verbose) { log_fh << "pruned_polymorphic_site\t" << loc->id << "\t" << loc->loc.chr << "\t" << loc->sort_bp(loc->snps[i]->col) +1 << "\t" << loc->snps[i]->col << "\t"; if (inc_prune) log_fh << "incompatible_site\n"; else if (sample_prune) log_fh << "sample_limit\n"; else if (maf_prune) log_fh << "maf_limit\n"; else if (het_prune) log_fh << "obshet_limit\n"; else log_fh << "unknown_reason\n"; } } } // // If no SNPs were retained for this locus, then mark it to be removed entirely. // if (new_wl.count(loc->id) == 0) { if (verbose) log_fh << "removed_locus\t" << loc->id << "\t" << loc->loc.chr << "\t" << loc->sort_bp() +1 << "\t" << 0 << "\tno_snps_remaining\n"; blacklist.insert(loc->id); } } } else { // // Otherwise, just iterate over the catalog. // map::iterator it; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; // // If this locus is fixed, don't try to filter it out. // if (loc->snps.size() == 0) { new_wl.insert(make_pair(loc->id, std::set())); continue; } t = psum->locus_tally(loc->id); s = psum->locus(loc->id); for (uint i = 0; i < loc->snps.size(); i++) { // // If the site is fixed, ignore it. // if (t->nucs[loc->snps[i]->col].fixed == true) continue; sample_prune = false; maf_prune = false; het_prune = false; inc_prune = false; pop_prune_list.clear(); for (size_t p = 0; p < mpopi.pops().size(); p++) { if (s[p]->nucs[loc->snps[i]->col].incompatible_site) inc_prune = true; else if (s[p]->nucs[loc->snps[i]->col].num_indv == 0 || (double) s[p]->nucs[loc->snps[i]->col].num_indv / (double) psum->pop_size(p) < sample_limit) pop_prune_list.push_back(p); } // // Check how many populations have to be pruned out due to sample limit. If less than // population limit, prune them; if more than population limit, mark locus for deletion. // if ((psum->pop_cnt() - pop_prune_list.size()) < (uint) population_limit) { sample_prune = true; } else { for (size_t p : pop_prune_list) { if (s[p]->nucs[loc->snps[i]->col].num_indv == 0) continue; d = pmap->locus(loc->id); const Pop& pop = mpopi.pops()[p]; for (uint k = pop.first_sample; k <= pop.last_sample; k++) { if (d[k] == NULL || loc->snps[i]->col >= (uint) d[k]->len) continue; if (d[k]->model != NULL) { d[k]->model[loc->snps[i]->col] = 'U'; } } } } if (t->nucs[loc->snps[i]->col].allele_cnt > 1) { // // Test for minor allele frequency. // if ((1 - t->nucs[loc->snps[i]->col].p_freq) < minor_allele_freq) maf_prune = true; // // Test for observed heterozygosity. // if (t->nucs[loc->snps[i]->col].obs_het > max_obs_het) het_prune = true; } if (maf_prune == false && het_prune == false && sample_prune == false && inc_prune == false) { new_wl[loc->id].insert(loc->snps[i]->col); } else { pruned++; if (verbose) { log_fh << "pruned_polymorphic_site\t" << loc->id << "\t" << loc->loc.chr << "\t" << loc->sort_bp(loc->snps[i]->col) +1 << "\t" << loc->snps[i]->col << "\t"; if (inc_prune) log_fh << "incompatible_site\n"; else if (sample_prune) log_fh << "sample_limit\n"; else if (maf_prune) log_fh << "maf_limit\n"; else if (het_prune) log_fh << "obshet_limit\n"; else log_fh << "unknown_reason\n"; } } } // // If no SNPs were retained for this locus, then mark it to be removed entirely. // if (new_wl.count(loc->id) == 0) { if (verbose) log_fh << "removed_locus\t" << loc->id << "\t" << loc->loc.chr << "\t" << loc->sort_bp() +1 << "\t" << 0 << "\tno_snps_remaining\n"; blacklist.insert(loc->id); } } } whitelist = new_wl; return pruned; } bool order_unordered_loci(map &catalog) { map::iterator it; CSLocus *loc; set chrs; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (strlen(loc->loc.chr) > 0) chrs.insert(loc->loc.chr); } // // This data is already reference aligned. // if (chrs.size() > 0) return true; cerr << "Catalog is not reference aligned, arbitrarily ordering catalog loci.\n"; uint bp = 1; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; loc->loc.chr = new char[3]; strcpy(loc->loc.chr, "un"); loc->loc.bp = bp; bp += strlen(loc->con); } return false; } int log_haplotype_cnts(map &catalog, ofstream &log_fh) { map::iterator it; map valid, absent, confounded; CSLocus *loc; int missing; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; if (valid.count(loc->hcnt) == 0) valid[loc->hcnt] = 1; else valid[loc->hcnt]++; if (confounded.count(loc->confounded_cnt) == 0) confounded[loc->confounded_cnt] = 1; else confounded[loc->confounded_cnt]++; missing = loc->cnt - loc->hcnt; if (absent.count(missing) == 0) absent[missing] = 1; else absent[missing]++; } map::iterator cnt_it; log_fh << "# Distribution of valid loci matched to catalog locus.\n" << "# Valid samples at locus\tCount\n"; for (cnt_it = valid.begin(); cnt_it != valid.end(); cnt_it++) log_fh << cnt_it->first << "\t" << cnt_it->second << "\n"; log_fh << "# Distribution of confounded loci at catalog locus.\n" << "# Confounded samples at locus\tCount\n"; for (cnt_it = confounded.begin(); cnt_it != confounded.end(); cnt_it++) log_fh << cnt_it->first << "\t" << cnt_it->second << "\n"; log_fh << "# Distribution of missing loci at catalog loci.\n" << "# Absent samples at locus\tCount\n"; for (cnt_it = absent.begin(); cnt_it != absent.end(); cnt_it++) log_fh << cnt_it->first << "\t" << cnt_it->second << "\n"; return 0; } int tabulate_haplotypes(map &catalog, PopMap *pmap) { map::iterator it; vector::iterator hit; Datum **d; CSLocus *loc; double mean, cnt; for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; d = pmap->locus(loc->id); mean = 0.0; cnt = 0.0; for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] == NULL) continue; if (d[i]->obshap.size() > 1) loc->marker = "heterozygous"; mean += d[i]->lnl; cnt++; } if (loc->marker.length() > 0) { create_genotype_map(loc, pmap); call_population_genotypes(loc, pmap); } loc->lnl = mean / cnt; } return 0; } int merge_shared_cutsite_loci(map &catalog, PopMap *pmap, PopSum *psum, map > &merge_map, ofstream &log_fh) { map >::iterator it; CSLocus *cur, *next; Datum **d_1, **d_2; double prune_pct; uint unmergable, tot_loci, tot_samp; uint success = 0; uint failure = 0; uint overlap = 0; uint simple_merge_cnt = 0; uint complex_merge_cnt = 0; uint missing_samps_cnt = 0; uint phase_fail_cnt = 0; uint nomapping_cnt = 0; uint multimapping_cnt = 0; uint multifails_cnt = 0; tot_loci = pmap->loci_cnt(); set loci_to_destroy; map missing_samps_dist; cerr << "To merge adjacent loci at least " << merge_prune_lim * 100 << "% of samples must have both adjacent loci;" << " the remaining " << 100 - (merge_prune_lim * 100) << "% of individuals will be pruned.\n" << "Attempting to merge adjacent loci that share a cutsite..."; if (verbose) log_fh << "\n#\n# List of locus pairs that share a cutsite that failed to merge because they could not be phased.\n#\n"; // // Iterate over each chromosome. // for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { // // Iterate over each ordered locus on this chromosome. // next = it->second[0]; for (uint pos = 1; pos < it->second.size(); pos++) { cur = next; next = it->second[pos]; // // Do these two loci overlap? // +Must occur on opposite strands // +Must overlap according to the length of the cutsite. // if (((cur->loc.strand == strand_minus && next->loc.strand == strand_plus) && ((int) (cur->loc.bp - next->loc.bp + 1) == renz_olap[enz])) || ((cur->loc.strand == strand_plus && next->loc.strand == strand_minus) && ((int) (next->loc.bp - cur->loc.bp + 1) == renz_olap[enz]))) { overlap++; d_1 = pmap->locus(cur->id); d_2 = pmap->locus(next->id); unmergable = 0; tot_samp = 0; // // Check if all members of the population contain these two loci (or are missing both). // for (int i = 0; i < pmap->sample_cnt(); i++) { if (d_1[i] != NULL || d_2[i] != NULL) tot_samp++; if ((d_1[i] != NULL && d_2[i] == NULL) || (d_1[i] == NULL && d_2[i] != NULL)) unmergable++; } prune_pct = (double) (tot_samp - unmergable) / (double) tot_samp; // // If some of the individuals only have one locus and not the other, prune them out. // if (prune_pct < 1.0 && prune_pct >= merge_prune_lim) { for (int i = 0; i < pmap->sample_cnt(); i++) if (d_1[i] != NULL && d_2[i] == NULL) { delete d_1[i]; d_1[i] = NULL; } else if (d_1[i] == NULL && d_2[i] != NULL) { delete d_2[i]; d_2[i] = NULL; } } // // If possible, merge the two loci together. // if (prune_pct < merge_prune_lim) { int pct = (int) (prune_pct * 100); missing_samps_dist[pct]++; if (verbose) log_fh << "Missing samples, Sink Locus: " << cur->id << "; Source Locus: " << next->id << "; " << pct << "% present (" << 100 - pct << "% missing)\n"; missing_samps_cnt++; failure++; continue; } phaset res = merge_and_phase_loci(pmap, cur, next, loci_to_destroy, log_fh); switch(res) { case multiple_fails: if (verbose) log_fh << "Failed to phase, Sink Locus: " << cur->id << "; Source Locus: " << next->id << "; " << "multiple failures\n"; multifails_cnt++; phase_fail_cnt++; failure++; break; case multimapping_fail: if (verbose) log_fh << "Failed to phase, Sink Locus: " << cur->id << "; Source Locus: " << next->id << "; " << "multimapping in one or more individuals\n"; multimapping_cnt++; phase_fail_cnt++; failure++; break; case nomapping_fail: if (verbose) log_fh << "Failed to phase, Sink Locus: " << cur->id << "; Source Locus: " << next->id << "; " << "no mapping in one or more individuals\n"; nomapping_cnt++; phase_fail_cnt++; failure++; break; case complex_phase: if (verbose) log_fh << "Phased Sink Locus: " << cur->id << "; Source Locus: " << next->id << "; " << "a complex phasing operation.\n"; complex_merge_cnt++; success++; merge_map[cur->id] = make_pair(merge_sink, next->id); merge_map[next->id] = make_pair(merge_src, cur->id); break; case simple_merge: if (verbose) log_fh << "Phased Sink Locus: " << cur->id << "; Source Locus: " << next->id << "; " << "a simple merge operation.\n"; simple_merge_cnt++; success++; merge_map[cur->id] = make_pair(merge_sink, next->id); merge_map[next->id] = make_pair(merge_src, cur->id); break; default: cerr << "Warning: Merge failure.\n"; break; } } } } // // Remove those loci that have been merged from both the popualtion map and catalog. // set emptyset; pmap->prune(loci_to_destroy); reduce_catalog(catalog, emptyset, loci_to_destroy); cerr << "done.\n" << "Of " << tot_loci << " loci, " << overlap << " pairs share a cutsite; " << success << " pairs were merged; " << failure << " pairs failed to merge; " << pmap->loci_cnt() << " remaining loci.\n" << " Of those merged, " << simple_merge_cnt << " required only a simple merge without phasing; " << "while " << complex_merge_cnt << " required phasing.\n" << " Of those that failed to merge, " << missing_samps_cnt << " were missing one of the two haplotypes in one or more samples; " << "while " << phase_fail_cnt << " failed to be phased.\n" << " Of those that failed to phase, " << nomapping_cnt << " failed due to a lack of haplotype mappings; " << multimapping_cnt << " failed due to multiple haplotype mappings; " << multifails_cnt << " failed due to both.\n"; log_fh << "\n#\n# Merging adjacent loci with a shared restriction enzyme cutsite\n#\n" << "Of " << tot_loci << " loci, " << overlap << " pairs share a cutsite; " << success << " pairs were merged; " << failure << " pairs failed to merge; " << pmap->loci_cnt() << " remaining loci.\n" << " Of those merged, " << simple_merge_cnt << " required only a simple merge without phasing; " << "while " << complex_merge_cnt << " required phasing.\n" << " Of those that failed to merge, " << missing_samps_cnt << " were missing one of the two haplotypes in one or more samples; " << "while " << phase_fail_cnt << " failed to be phased.\n" << " Of those that failed to phase, " << nomapping_cnt << " failed due to a lack of haplotype mappings; " << multimapping_cnt << " failed due to multiple haplotype mappings; " << multifails_cnt << " failed due to both.\n"; log_fh << "#\n# Distribution of loci with samples missing one of two loci to be merged\n" << "# Percent samples with both loci present\tNumber of cases\n"; map::iterator mit; for (mit = missing_samps_dist.begin(); mit != missing_samps_dist.end(); mit++) log_fh << mit->first << "\t" << mit->second << "\n"; log_fh << "\n"; return 0; } phaset merge_and_phase_loci(PopMap *pmap, CSLocus *cur, CSLocus *next, set &loci_to_destroy, ofstream &log_fh) { Datum **d_1 = pmap->locus(cur->id); Datum **d_2 = pmap->locus(next->id); set phased_results; set phased_haplotypes; string merged_hap; char *h_1, *h_2; int merge_type; if (verbose) log_fh << "Attempting to phase source locus " << cur->id << " with sink locus " << next->id << "\n"; int sample_cnt = 0; int phased_sample_cnt = 0; // // Take a census of the already phased haplotypes. We have phased haplotypes // if for individual i: // 1. d_1 has a single haplotype and d_2 has a single haplotype // 2. d_1 has a single haplotpye and d_2 has multiple haplotypes // 3. d_1 has multiple haplotpyes and d_2 has a single haplotype // // If one or both of the loci have no SNPs, then the haplotype is // recorded as "consensus." Check that condition before we start merging. // if (cur->snps.size() > 0 && next->snps.size() > 0) merge_type = 0; else if (cur->snps.size() == 0) merge_type = 1; else if (next->snps.size() == 0) merge_type = 2; else merge_type = 3; for (int i = 0; i < pmap->sample_cnt(); i++) { if (d_1[i] == NULL || d_2[i] == NULL) continue; else if (d_1[i]->obshap.size() > 1 && d_2[i]->obshap.size() > 1) continue; else { for (uint j = 0; j < d_1[i]->obshap.size(); j++) { for (uint k = 0; k < d_2[i]->obshap.size(); k++) { switch (merge_type) { case 0: merged_hap = string(d_1[i]->obshap[j]) + string(d_2[i]->obshap[k]); break; case 1: merged_hap = string(d_2[i]->obshap[k]); break; case 2: merged_hap = string(d_1[i]->obshap[j]); break; case 3: default: merged_hap = "consensus"; break; } phased_haplotypes.insert(merged_hap); // cerr << "Phasing: '" << d_1[i]->obshap[j] << "' + '" << d_2[i]->obshap[k] << "' => '" << merged_hap << "'\n"; } } phased_sample_cnt++; sample_cnt++; } } // // Indicate that these two loci had a simple merge, with no phasing necessary. // phased_results.insert(simple_merge); // // Now we need to check if we can phase the remaining haplotypes. // for (int i = 0; i < pmap->sample_cnt(); i++) { if (d_1[i] == NULL || d_2[i] == NULL) continue; else if (d_1[i]->obshap.size() > 1 && d_2[i]->obshap.size() > 1) { // cerr << "Attempting to phase individual " << i << ": " << d_1[i]->id << " / " << d_2[i]->id << "\n"; sample_cnt++; // // We should be able to find a sinlge phasing mapping for each haplotype from d_1 to d_2 // that includes all the haplotypes in these two loci. // vector > seen_phased; uint tot_obshap = d_1[i]->obshap.size() + d_2[i]->obshap.size(); uint phased_cnt = 0; for (uint j = 0; j < d_1[i]->obshap.size(); j++) { for (uint k = 0; k < d_2[i]->obshap.size(); k++) { // cerr << " " << d_1[i]->obshap[j] << " + " << d_2[i]->obshap[k]; // // Record each pair of haplotypes that has been seen phased previously. // if (phased_haplotypes.count(string(d_1[i]->obshap[j]) + string(d_2[i]->obshap[k]))) { seen_phased.push_back(make_pair(d_1[i]->obshap[j], d_2[i]->obshap[k])); // cerr << " => " << d_1[i]->obshap[j] << d_2[i]->obshap[k]; } // cerr << "\n"; } } // // Now, we will iterate over all sets of phased haplotypes and look // for combinations that use all four individual haplotypes. // for (uint j = 0; j < seen_phased.size(); j++) { for (uint k = j; k < seen_phased.size(); k++) { set incorporated_haplotypes; // // Count the number of distinct char pointers. If this combination // of haplotypes includes all unphased haplotypes, count it. // incorporated_haplotypes.insert(seen_phased[j].first); incorporated_haplotypes.insert(seen_phased[j].second); incorporated_haplotypes.insert(seen_phased[k].first); incorporated_haplotypes.insert(seen_phased[k].second); if (incorporated_haplotypes.size() == tot_obshap) phased_cnt++; } } // // If one pair of haplotypes is mapped, but the other is not, assume the second pair or // haplotypes must be phased by process of elimination. // if (phased_cnt == 0 && seen_phased.size() == 1) { h_1 = seen_phased[0].first == d_1[i]->obshap[1] ? d_1[i]->obshap[0] : d_1[i]->obshap[1]; h_2 = seen_phased[0].second == d_2[i]->obshap[1] ? d_2[i]->obshap[0] : d_2[i]->obshap[1]; phased_haplotypes.insert(string(h_1) + string(h_2)); phased_cnt++; // cerr << " Phasing: '" << hap_1 << "' + '" << hap_2 << "' => '" << string(hap_1) + string(hap_2) << "'\n"; } if (phased_cnt == 0) { phased_results.insert(nomapping_fail); if (verbose) log_fh << " Locus NOT phased in individual " << i << "; loci " << d_1[i]->id << " / " << d_2[i]->id << " no mapping found.\n"; } else if (phased_cnt == 1) { phased_sample_cnt++; phased_results.insert(complex_phase); } else { phased_results.insert(multimapping_fail); if (verbose) log_fh << " Locus NOT phased in individual " << i << "; loci " << d_1[i]->id << " / " << d_2[i]->id << " multiple mappings found.\n"; } } } if (phased_sample_cnt != sample_cnt) { if (phased_results.count(nomapping_fail) > 0 && phased_results.count(multimapping_fail) > 0) return multiple_fails; else if (phased_results.count(nomapping_fail) > 0) return nomapping_fail; else if (phased_results.count(multimapping_fail) > 0) return multimapping_fail; else { cerr << "WE SHOULD NOT GET HERE\n"; return merge_failure; } } // // Okay, merge these two loci together. // if (!merge_datums(pmap->sample_cnt(), cur->len, d_1, d_2, phased_haplotypes, merge_type)) return merge_failure; // // Merge the catalog entries together. // if (!merge_csloci(cur, next, phased_haplotypes)) return merge_failure; // // Mark the merged locus for destruction. // loci_to_destroy.insert(next->id); if (phased_results.count(complex_phase) > 0) return complex_phase; return simple_merge; } int merge_csloci(CSLocus *sink, CSLocus *src, set &phased_haplotypes) { // // We assume that we are merging two loci: one on the negative strand, one on the // positive. We will keep the sink cslocus and delete the src cslocus. // -> The sink cslocus is assumed to be on the negative strand. // // // 1. Reverse complement the SNP coordinates in the sink locus so that they are // enumerated on the positive strand. Complement the alleles as well. // for (uint j = 0; j < sink->snps.size(); j++) { sink->snps[j]->col = sink->len - sink->snps[j]->col - 1; sink->snps[j]->rank_1 = reverse(sink->snps[j]->rank_1); sink->snps[j]->rank_2 = reverse(sink->snps[j]->rank_2); sink->snps[j]->rank_3 = reverse(sink->snps[j]->rank_3); sink->snps[j]->rank_4 = reverse(sink->snps[j]->rank_4); } // // 2. Adjust the SNP coordinates in the src locus to account for the now, longer length. // for (uint j = 0; j < src->snps.size(); j++) src->snps[j]->col = sink->len + src->snps[j]->col - renz_olap[enz]; // // 3. Combine SNPs between the two catalog loci: add the SNPs from the sink (formerly on the // negative strand) in reverse order, followed by the SNPs from the src. // vector tmpsnp; for (int j = (int) sink->snps.size() - 1; j >= 0; j--) tmpsnp.push_back(sink->snps[j]); for (uint j = 0; j < src->snps.size(); j++) tmpsnp.push_back(src->snps[j]); sink->snps.clear(); for (uint j = 0; j < tmpsnp.size(); j++) sink->snps.push_back(tmpsnp[j]); // // 4. Adjust the genomic location of the sink locus. // uint bp = sink->sort_bp(); sink->loc.bp = bp; sink->loc.strand = strand_plus; // // 5. Adjust the length of the sequence. // sink->len += src->len - renz_olap[enz]; // // 6. Merge the consensus sequence together. // char *new_con = rev_comp(sink->con); delete [] sink->con; sink->con = new_con; new_con = new char[sink->len + 1]; strcpy(new_con, sink->con); delete [] sink->con; sink->con = new_con; new_con += src->len - renz_olap[enz]; strcpy(new_con, src->con); // // 7. Record the now phased haplotypes. // sink->alleles.clear(); set::iterator it; for (it = phased_haplotypes.begin(); it != phased_haplotypes.end(); it++) sink->alleles[*it] = 0; // cerr << "CSLocus " << sink->id << ":\n" // << "Length: " << sink->len << "; Chr: " << sink->loc.chr << "; BP: " << sink->sort_bp() << "; strand: " << (sink->loc.strand == strand_plus ? "+" : "-") << "\n" // << " SNPs:\n"; // for (uint j = 0; j < sink->snps.size(); j++) // cerr << " Col: " << sink->snps[j]->col // << " Rank 1: " << sink->snps[j]->rank_1 // << " Rank 2: " << sink->snps[j]->rank_2 << "\n"; // cerr << " Alleles:\n"; // map::iterator ait; // for (ait = sink->alleles.begin(); ait != sink->alleles.end(); ait++) // cerr << " " << ait->first << "\n"; return 1; } int merge_datums(int sample_cnt, int sink_locus_len, Datum **sink, Datum **src, set &phased_haplotypes, int merge_type) { char tmphap[id_len], *new_hap; uint haplen, model_len, offset; vector tmpsnp; vector tmpobshap; vector tmpobsdep; // // We assume that we are merging two loci: one on the negative strand, one on the // positive. We will keep the sink datum and delete the src datum. // -The sink datum is assumed to be on the negative strand. // for (int i = 0; i < sample_cnt; i++) { if (sink[i] == NULL && src[i] == NULL) continue; else if (sink[i] == NULL || src[i] == NULL) cerr << "Unexpected condition in merging datums: one datum is NULL while the other is not.\n"; // // 1. Reverse complement the SNP coordinates in the sink locus so that they are // enumerated on the positive strand. Complement the alleles as well. // for (uint j = 0; j < sink[i]->snps.size(); j++) { sink[i]->snps[j]->col = sink[i]->len - sink[i]->snps[j]->col - 1; sink[i]->snps[j]->rank_1 = reverse(sink[i]->snps[j]->rank_1); sink[i]->snps[j]->rank_2 = reverse(sink[i]->snps[j]->rank_2); sink[i]->snps[j]->rank_3 = reverse(sink[i]->snps[j]->rank_3); sink[i]->snps[j]->rank_4 = reverse(sink[i]->snps[j]->rank_4); } // // 2. Adjust the SNP coordinates in the src locus to account for the now, longer length. // for (uint j = 0; j < src[i]->snps.size(); j++) src[i]->snps[j]->col = sink[i]->len + src[i]->snps[j]->col - renz_olap[enz]; // // 3. Reverse complement the observed haplotypes in the sink locus. // haplen = strlen(sink[i]->obshap[0]); for (uint j = 0; j < sink[i]->obshap.size(); j++) { for (uint k = 0; k < haplen; k++) tmphap[k] = reverse(sink[i]->obshap[j][haplen - k - 1]); tmphap[haplen] = '\0'; strcpy(sink[i]->obshap[j], tmphap); } // // 4. Combine SNPs between the two datums: add the SNPs from the sink (formerly on the // negative strand) in reverse order, followed by the SNPs from the src. // tmpsnp.clear(); for (int j = (int) sink[i]->snps.size() - 1; j >= 0; j--) tmpsnp.push_back(sink[i]->snps[j]); for (uint j = 0; j < src[i]->snps.size(); j++) tmpsnp.push_back(src[i]->snps[j]); sink[i]->snps.clear(); for (uint j = 0; j < tmpsnp.size(); j++) sink[i]->snps.push_back(tmpsnp[j]); } // // 5. Combine observed haplotypes between the two datums while phasing them. // 5.1 First combine the haplotypes from samples that are already in phase. // string merged_hap; vector to_be_phased; phased_haplotypes.clear(); for (int i = 0; i < sample_cnt; i++) { if (sink[i] == NULL && src[i] == NULL) continue; if (sink[i]->obshap.size() > 1 && src[i]->obshap.size() > 1) { to_be_phased.push_back(i); continue; } else { tmpobshap.clear(); tmpobsdep.clear(); for (uint j = 0; j < sink[i]->obshap.size(); j++) { for (uint k = 0; k < src[i]->obshap.size(); k++) { switch (merge_type) { case 0: merged_hap = string(sink[i]->obshap[j]) + string(src[i]->obshap[k]); break; case 1: merged_hap = string(src[i]->obshap[j]); break; case 2: merged_hap = string(sink[i]->obshap[j]); break; case 3: default: merged_hap = "consensus"; break; } phased_haplotypes.insert(merged_hap); tmpobshap.push_back(merged_hap); tmpobsdep.push_back((sink[i]->depth[j] + src[i]->depth[k]) / 2); } } sink[i]->depth.clear(); for (uint j = 0; j < sink[i]->obshap.size(); j++) delete [] sink[i]->obshap[j]; sink[i]->obshap.clear(); for (uint j = 0; j < tmpobshap.size(); j++) { new_hap = new char[tmpobshap[j].length() + 1]; strcpy(new_hap, tmpobshap[j].c_str()); sink[i]->obshap.push_back(new_hap); sink[i]->depth.push_back(tmpobsdep[j]); } } } // // 5.2 Phase and combine the haplotypes from the remaining samples. // int index; for (uint i = 0; i < to_be_phased.size(); i++) { index = to_be_phased[i]; tmpobshap.clear(); tmpobsdep.clear(); vector > seen_phased; uint tot_obshap = sink[index]->obshap.size() + src[index]->obshap.size(); for (uint j = 0; j < sink[index]->obshap.size(); j++) { for (uint k = 0; k < src[index]->obshap.size(); k++) { if (phased_haplotypes.count(string(sink[index]->obshap[j]) + string(src[index]->obshap[k]))) seen_phased.push_back(make_pair(sink[index]->obshap[j], src[index]->obshap[k])); } } for (uint j = 0; j < seen_phased.size(); j++) { for (uint k = j; k < seen_phased.size(); k++) { set incorporated_haplotypes; incorporated_haplotypes.insert(seen_phased[j].first); incorporated_haplotypes.insert(seen_phased[j].second); incorporated_haplotypes.insert(seen_phased[k].first); incorporated_haplotypes.insert(seen_phased[k].second); if (incorporated_haplotypes.size() == tot_obshap) { tmpobshap.push_back(string(seen_phased[j].first) + string(seen_phased[j].second)); tmpobshap.push_back(string(seen_phased[k].first) + string(seen_phased[k].second)); //tmpobsdep.push_back((sink[index]->depth[j] + src[index]->depth[k]) / 2); } } } sink[index]->depth.clear(); for (uint j = 0; j < sink[index]->obshap.size(); j++) delete [] sink[index]->obshap[j]; sink[index]->obshap.clear(); for (uint j = 0; j < tmpobshap.size(); j++) { new_hap = new char[tmpobshap[j].length() + 1]; strcpy(new_hap, tmpobshap[j].c_str()); sink[index]->obshap.push_back(new_hap); // sink[index]->depth.push_back(tmpobsdep[j]); } } // // 6. Merge model calls; Set the length; combine the two depth and lnl measures together. // string model_calls; char *p; for (int i = 0; i < sample_cnt; i++) { if (sink[i] == NULL && src[i] == NULL) continue; // // Merge the two strings of model calls together. // We need to check if the locus for this individual is shorter than the catalog // locus. If so, we need to expand out the model call array to be the proper length. // reverse_string(sink[i]->model); offset = 0; model_calls.clear(); if (sink_locus_len > sink[i]->len) { offset = sink_locus_len - sink[i]->len; model_calls.assign(offset, 'N'); } model_len = offset + sink[i]->len + src[i]->len - renz_olap[enz]; model_calls.append(sink[i]->model); delete [] sink[i]->model; sink[i]->model = new char[model_len + 1]; strcpy(sink[i]->model, model_calls.c_str()); p = sink[i]->model; p += offset + sink[i]->len - renz_olap[enz]; strcpy(p, src[i]->model); sink[i]->len = model_len; sink[i]->tot_depth = (sink[i]->tot_depth + src[i]->tot_depth) / 2; sink[i]->lnl = (sink[i]->lnl + src[i]->lnl) / 2.0; // // Record which datum was merged into this one. // sink[i]->merge_partner = src[i]->id; } return 1; } int create_genotype_map(CSLocus *locus, PopMap *pmap) { // // Create a genotype map. For any set of haplotypes, this routine will // assign each haplotype to a genotype, e.g. given the haplotypes // 'AC' and 'GT' in the population, this routine will assign 'AC' == 'a' // and 'GT' == 'b'. If an individual is homozygous for 'AC', they will be // assigned an 'aa' genotype. // //cerr << "Creating genotype map for catalog ID " << locus->id << ", marker: " << locus->marker << ".\n"; char gtypes[26] ={'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z'}; Datum **d; map haplotypes; map::iterator k; vector > sorted_haplotypes; d = pmap->locus(locus->id); for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] != NULL) for (uint n = 0; n < d[i]->obshap.size(); n++) haplotypes[d[i]->obshap[n]]++; } // // Check that there are not more haplotypes than we have encodings. // if (haplotypes.size() > 26) return 0; // // Sort the haplotypes map by value // for (k = haplotypes.begin(); k != haplotypes.end(); k++) sorted_haplotypes.push_back(*k); sort(sorted_haplotypes.begin(), sorted_haplotypes.end(), hap_compare); for (uint n = 0, index = 0; n < sorted_haplotypes.size() && index <= 26; n++, index++) { locus->gmap[sorted_haplotypes[n].first] = gtypes[index]; //cerr << "GMAP: " << sorted_haplotypes[n].first << " == " << gtypes[index] << "\n"; } return 0; } int call_population_genotypes(CSLocus *locus, PopMap *pmap) { // // Fetch the array of observed haplotypes from the population // Datum **d = pmap->locus(locus->id); for (int i = 0; i < pmap->sample_cnt(); i++) { if (d[i] == NULL) continue; vector gtypes; string gtype; //cerr << "Sample Id: " << pmap->rev_sample_index(i) << "\n"; for (uint j = 0; j < d[i]->obshap.size(); j++) { // // Impossible allele encountered. // if (locus->gmap.count(d[i]->obshap[j]) == 0) { gtypes.clear(); gtypes.push_back("-"); goto impossible; } gtypes.push_back(locus->gmap[d[i]->obshap[j]]); //cerr << " Observed Haplotype: " << d[i]->obshap[j] << ", Genotype: " << locus->gmap[d[i]->obshap[j]] << "\n"; } impossible: sort(gtypes.begin(), gtypes.end()); for (uint j = 0; j < gtypes.size(); j++) { gtype += gtypes[j]; //cerr << " Adding genotype to string: " << gtypes[j] << "; " << gtype << "\n"; } string m = gtype.length() == 1 ? gtype + gtype : gtype; d[i]->gtype = new char[m.length() + 1]; strcpy(d[i]->gtype, m.c_str()); if (m != "-") locus->gcnt++; //cerr << "Assigning datum, marker: " << locus->marker << ", string: " << m << ", haplotype: " << d[i]->obshap[0] << ", gtype: " << gtype << "\n"; } return 0; } int write_genomic(map &catalog, PopMap *pmap) { string file = out_path + out_prefix + ".genomic.tsv"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening genomic output file '" << file << "'\n"; exit(1); } // // Count the number of markers that have enough samples to output. // map::iterator cit; CSLocus *loc; int num_loci = 0; for (cit = catalog.begin(); cit != catalog.end(); cit++) { loc = cit->second; num_loci += loc->len - renz_len[enz]; } cerr << "Writing " << num_loci << " nucleotide positions to genomic file, '" << file << "'\n"; // // Write the header // fh << num_loci << "\t" << pmap->sample_cnt() << "\n"; // // Output each locus. // map >::iterator it; int a, b; uint rcnt = enz.length() ? renz_cnt[enz] : 0; uint rlen = enz.length() ? renz_len[enz] : 0; char *p; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { for (uint i = 0; i < it->second.size(); i++) { loc = it->second[i]; Datum **d = pmap->locus(loc->id); set snp_locs; string obshap; for (uint i = 0; i < loc->snps.size(); i++) snp_locs.insert(loc->snps[i]->col); uint start = 0; uint end = loc->len; // // Check for the existence of the restriction enzyme cut site, mask off // its output. // for (uint n = 0; n < rcnt; n++) if (strncmp(loc->con, renz[enz][n], rlen) == 0) start += renz_len[enz]; if (start == 0) { p = loc->con + (loc->len - rlen); for (uint n = rcnt; n < rcnt + rcnt; n++) if (strncmp(p, renz[enz][n], rlen) == 0) end -= renz_len[enz]; } uint k = 0; for (uint n = start; n < end; n++) { fh << loc->id << "\t" << loc->loc.chr << "\t" << loc->sort_bp(n) +1; if (snp_locs.count(n) == 0) { for (int j = 0; j < pmap->sample_cnt(); j++) { a = encode_gtype(loc->con[n]); fh << "\t" << encoded_gtypes[a][a]; } } else { for (int j = 0; j < pmap->sample_cnt(); j++) { fh << "\t"; if (d[j] == NULL) fh << "0"; else switch (d[j]->obshap.size()) { case 1: a = encode_gtype(d[j]->obshap[0][k]); fh << encoded_gtypes[a][a]; break; case 2: a = encode_gtype(d[j]->obshap[0][k]); b = encode_gtype(d[j]->obshap[1][k]); fh << encoded_gtypes[a][b]; break; default: fh << "0"; break; } } k++; } fh << "\n"; } } } fh.close(); return 0; } int calculate_haplotype_stats(map &catalog, PopMap *pmap, PopSum *psum) { map >::iterator it; CSLocus *loc; Datum **d; LocStat *l; // // Instantiate the kernel smoothing and bootstrap objects if requested. // KSmooth *ks; OHaplotypes *ord; Bootstrap *bs; if (kernel_smoothed && loci_ordered) { ks = new KSmooth(2); ord = new OHaplotypes(); } // // Open output file and print header. // string file = out_path + out_prefix + ".hapstats.tsv"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening haplotype stats file '" << file << "'\n"; exit(1); } fh.precision(fieldw); fh.setf(std::ios::fixed); // // Write the population members. // for (auto& pop : mpopi.pops()) { fh << "# " << pop.name << "\t"; for (size_t i = pop.first_sample; i <= pop.last_sample; i++) { fh << mpopi.samples()[i].name; if (i < pop.last_sample) fh << ","; } fh << "\n"; } fh << "# Batch ID " << "\t" << "Locus ID" << "\t" << "Chr" << "\t" << "BP" << "\t" << "Pop ID" << "\t" << "N" << "\t" << "Haplotype Cnt" << "\t" << "Gene Diversity" << "\t" << "Smoothed Gene Diversity" << "\t" << "Smoothed Gene Diversity P-value" << "\t" << "Haplotype Diversity" << "\t" << "Smoothed Haplotype Diversity" << "\t" << "Smoothed Haplotype Diversity P-value" << "\t" << "Haplotypes" << "\n"; // // Iterate over the members of each population. // for (auto& pop : mpopi.pops()) { cerr << "Generating haplotype-level summary statistics for population '" << pop.name << "'\n"; map > genome_locstats; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { if (bootstrap_div) bs = new Bootstrap(2); vector &locstats = genome_locstats[it->first]; map locstats_key; ord->order(locstats, locstats_key, it->second); for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; d = pmap->locus(loc->id); if (loc->snps.size() == 0) continue; // cerr << "Looking at locus " << loc->id << "\n"; l = haplotype_diversity(pop.first_sample, pop.last_sample, d); if (l != NULL) { l->loc_id = loc->id; l->bp = loc->sort_bp(); locstats[locstats_key[l->bp]] = l; } } if (kernel_smoothed && loci_ordered) { cerr << " Generating kernel-smoothed statistics on chromosome " << it->first << "\n"; ks->smooth(locstats); } if (bootstrap_div) bs->add_data(locstats); } for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { vector &locstats = genome_locstats[it->first]; if (bootstrap_div) bs->execute(locstats); // // Write results. // for (uint k = 0; k < locstats.size(); k++) { l = locstats[k]; if (l == NULL) continue; fh << batch_id << "\t" << l->loc_id << "\t" << it->first << "\t" << l->bp + 1 << "\t" << pop.name << "\t" << (int) l->alleles << "\t" << l->hap_cnt << "\t" << l->stat[0] << "\t" << l->smoothed[0] << "\t" << l->bs[0] << "\t" << l->stat[1] << "\t" << l->smoothed[1] << "\t" << l->bs[1] << "\t" << l->hap_str << "\n"; } for (uint k = 0; k < locstats.size(); k++) delete locstats[k]; } if (bootstrap_div) delete bs; } if (kernel_smoothed && loci_ordered) { delete ks; delete ord; } fh.close(); return 0; } int nuc_substitution_dist(map &hap_index, double **hdists) { vector haplotypes; map::iterator it; uint i, j; for (it = hap_index.begin(); it != hap_index.end(); it++) haplotypes.push_back(it->first); const char *p, *q; double dist; for (i = 0; i < haplotypes.size(); i++) { for (j = i; j < haplotypes.size(); j++) { dist = 0.0; p = haplotypes[i].c_str(); q = haplotypes[j].c_str(); while (*p != '\0' && *q != '\0') { if (*p != *q) dist++; p++; q++; } hdists[i][j] = dist; hdists[j][i] = dist; } } // // // // Print the distance matrix. // // // cerr << " "; // for (hit = loc_hap_index.begin(); hit != loc_hap_index.end(); hit++) // cerr << "\t" << hit->first; // cerr << "\n"; // for (hit = loc_hap_index.begin(); hit != loc_hap_index.end(); hit++) { // cerr << " " << hit->first; // for (hit_2 = loc_hap_index.begin(); hit_2 != loc_hap_index.end(); hit_2++) // cerr << "\t" << hdists[hit->second][hit_2->second]; // cerr << "\n"; // } // cerr << "\n"; return 0; } int nuc_substitution_identity(map &hap_index, double **hdists) { vector haplotypes; map::iterator it; uint i, j; for (it = hap_index.begin(); it != hap_index.end(); it++) haplotypes.push_back(it->first); double dist; for (i = 0; i < haplotypes.size(); i++) { for (j = i; j < haplotypes.size(); j++) { if (haplotypes[i] == haplotypes[j]) dist = 0.0; else dist = 1.0; hdists[i][j] = dist; hdists[j][i] = dist; } } return 0; } int nuc_substitution_identity_max(map &hap_index, double **hdists) { vector haplotypes; map::iterator it; uint i, j; for (it = hap_index.begin(); it != hap_index.end(); it++) haplotypes.push_back(it->first); for (i = 0; i < haplotypes.size(); i++) { for (j = i; j < haplotypes.size(); j++) { hdists[i][j] = 1.0; hdists[j][i] = 1.0; } } return 0; } int calculate_haplotype_divergence(map &catalog, PopMap *pmap, PopSum *psum) { map >::iterator it; if (bootstrap_phist) cerr << "Calculating halotype F statistics across all populations/groups and bootstrap resampling...\n"; else cerr << "Calculating haplotype F statistics across all populations/groups...\n"; // // Create a list of all the populations we have. // vector pop_ids; for (size_t i=0; i *ks; OHaplotypes *ord; Bootstrap *bs; if (kernel_smoothed && loci_ordered) { ks = new KSmooth(5); ord = new OHaplotypes(); } if (bootstrap_phist) bs = new Bootstrap(5); map > genome_hapstats; uint cnt = 0; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { string chr = it->first; cerr << " Generating haplotype F statistics for " << chr << "..."; map hapstats_key; vector &hapstats = genome_hapstats[chr]; ord->order(hapstats, hapstats_key, it->second); #pragma omp parallel { CSLocus *loc; LocSum **s; Datum **d; HapStat *h; #pragma omp for schedule(dynamic, 1) reduction(+:cnt) for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; s = psum->locus(loc->id); d = pmap->locus(loc->id); if (loc->snps.size() == 0) continue; // // If this locus only appears in one population or there is only a single haplotype, // do not calculate haplotype F stats. // if (fixed_locus(d, pop_ids)) continue; cnt++; // cerr << "Processing locus " << loc->id << "\n"; h = haplotype_amova(d, s, pop_ids); if (h != NULL) { h->stat[4] = haplotype_d_est(d, s, pop_ids); h->loc_id = loc->id; h->bp = loc->sort_bp(); hapstats[hapstats_key[h->bp]] = h; } } } if (bootstrap_phist) bs->add_data(hapstats); cerr << "done.\n"; // // Calculate kernel-smoothed Fst values. // if (kernel_smoothed && loci_ordered) { cerr << " Generating kernel-smoothed haplotype F statistics for " << it->first << "..."; ks->smooth(hapstats); cerr << "done.\n"; } } if (bootstrap_phist) { for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) bs->execute(genome_hapstats[it->first]); } cerr << "done.\n"; if (kernel_smoothed && loci_ordered) { delete ks; delete ord; } if (bootstrap_phist) delete bs; cerr << "Writing haplotype F statistics... "; string file = out_path + out_prefix + ".phistats.tsv"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening haplotype Phi_st file '" << file << "'\n"; exit(1); } fh.precision(fieldw); fh.setf(std::ios::fixed); // // Write the population members. // for (auto& pop : mpopi.pops()) { fh << "# Population " << pop.name << "\t"; for (size_t k = pop.first_sample; k <= pop.last_sample; k++) { fh << mpopi.samples()[k].name; if (k < pop.last_sample) fh << ","; } fh << "\n"; } // // Write the group members. // for (auto& group : mpopi.groups()) { fh << "# Group " << group.name << "\t"; for (size_t i_pop : group.pops) { fh << mpopi.pops()[i_pop].name; if (i_pop != group.pops.back()) fh << ","; } fh << "\n"; } fh << "# Batch ID " << "\t" << "Locus ID" << "\t" << "Chr" << "\t" << "BP" << "\t" << "PopCnt" << "\t"; if (log_fst_comp) fh << "SSD(WP)" << "\t" << "SSD(AP/WG)" << "\t" << "SSD(AG)" << "\t" << "SSD(TOTAL)" << "\t" << "MSD(WP)" << "\t" << "MSD(AP/WG)" << "\t" << "MSD(AG)" << "\t" << "MSD(TOTAL)" << "\t" << "n" << "\t" << "n'" << "\t" << "n''" << "\t" << "Sigma2_a" << "\t" << "Sigma2_b" << "\t" << "Sigma2_c" << "\t" << "Sigma_Total" << "\t"; fh << "phi_st" << "\t" << "Smoothed Phi_st" << "\t" << "Smoothed Phi_st P-value" << "\t" << "Phi_ct" << "\t" << "Smoothed Phi_ct" << "\t" << "Smoothed Phi_ct P-value" << "\t" << "Phi_sc" << "\t" << "Smoothed Phi_sc" << "\t" << "Smoothed Phi_sc P-value" << "\t" << "Fst'" << "\t" << "Smoothed Fst'" << "\t" << "Smoothed Fst' P-value" << "\t" << "D_est" << "\t" << "Smoothed D_est" << "\t" << "Smoothed D_est P-value" << "\n"; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { string chr = it->first; vector &hapstats = genome_hapstats[chr]; for (uint k = 0; k < hapstats.size(); k++) { if (hapstats[k] == NULL) continue; fh << batch_id << "\t" << hapstats[k]->loc_id << "\t" << chr << "\t" << hapstats[k]->bp +1 << "\t" << hapstats[k]->popcnt << "\t"; if (log_fst_comp) fh << hapstats[k]->comp[0] << "\t" << hapstats[k]->comp[1] << "\t" << hapstats[k]->comp[2] << "\t" << hapstats[k]->comp[3] << "\t" << hapstats[k]->comp[4] << "\t" << hapstats[k]->comp[5] << "\t" << hapstats[k]->comp[6] << "\t" << hapstats[k]->comp[7] << "\t" << hapstats[k]->comp[8] << "\t" << hapstats[k]->comp[9] << "\t" << hapstats[k]->comp[10] << "\t" << hapstats[k]->comp[11] << "\t" << hapstats[k]->comp[12] << "\t" << hapstats[k]->comp[13] << "\t" << hapstats[k]->comp[14] << "\t"; fh << hapstats[k]->stat[0] << "\t" << hapstats[k]->smoothed[0] << "\t" << hapstats[k]->bs[0] << "\t" << hapstats[k]->stat[1] << "\t" << hapstats[k]->smoothed[1] << "\t" << hapstats[k]->bs[1] << "\t" << hapstats[k]->stat[2] << "\t" << hapstats[k]->smoothed[2] << "\t" << hapstats[k]->bs[2] << "\t" << hapstats[k]->stat[3] << "\t" << hapstats[k]->smoothed[3] << "\t" << hapstats[k]->bs[3] << "\t" << hapstats[k]->stat[4] << "\t" << hapstats[k]->smoothed[4] << "\t" << hapstats[k]->bs[4] << "\n"; delete hapstats[k]; } } fh.close(); cerr << "wrote " << cnt << " loci to haplotype Phi_st file, '" << file << "'\n"; return 0; } int calculate_haplotype_divergence_pairwise(map &catalog, PopMap *pmap, PopSum *psum) { map >::iterator it; if (bootstrap_phist) cerr << "Calculating pairwise halotype F statistics and bootstrap resampling...\n"; else cerr << "Calculating pairwise haplotype F statistics...\n"; // // Assign all individuals to one group for the pairwise calculations. // vector pop_ids; for (size_t i=0; i *ks; OHaplotypes *ord; Bootstrap *bs; if (kernel_smoothed && loci_ordered) { ks = new KSmooth(5); ord = new OHaplotypes(); } for (uint i = 0; i < mpopi.pops().size(); ++i) { const Pop& pop_i = mpopi.pops()[i]; for (uint j = i + 1; j < mpopi.pops().size(); ++j) { const Pop& pop_j = mpopi.pops()[j]; vector subpop_ids; subpop_ids.push_back(i); subpop_ids.push_back(j); if (bootstrap_phist) bs = new Bootstrap(5); map > genome_hapstats; cerr << " Processing populations '" << pop_i.name << "' and '" << pop_j.name << "'\n"; uint cnt = 0; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { string chr = it->first; cerr << " Generating pairwise haplotype F statistics for " << chr << "..."; map hapstats_key; vector &hapstats = genome_hapstats[chr]; ord->order(hapstats, hapstats_key, it->second); #pragma omp parallel { CSLocus *loc; LocSum **s; Datum **d; HapStat *h; #pragma omp for schedule(dynamic, 1) reduction(+:cnt) for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; s = psum->locus(loc->id); d = pmap->locus(loc->id); if (loc->snps.size() == 0) continue; // // If this locus only appears in one population or there is only a single haplotype, // do not calculate haplotype F stats. // if (fixed_locus(d, subpop_ids)) continue; cnt++; // cerr << "Processing locus " << loc->id << "\n"; h = haplotype_amova(d, s, subpop_ids); if (h != NULL) { h->stat[4] = haplotype_d_est(d, s, subpop_ids); h->loc_id = loc->id; h->bp = loc->sort_bp(); hapstats[hapstats_key[h->bp]] = h; } } } if (bootstrap_phist) bs->add_data(hapstats); cerr << "done.\n"; // // Calculate kernel-smoothed Fst values. // if (kernel_smoothed && loci_ordered) { cerr << " Generating kernel-smoothed Phi_st for " << it->first << "..."; ks->smooth(hapstats); cerr << "done.\n"; } } if (bootstrap_phist) { for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) bs->execute(genome_hapstats[it->first]); } cerr << "done.\n"; if (bootstrap_phist) delete bs; cerr << "Writing haplotype F statistics... "; string file = out_path + out_prefix + ".phistats_" + pop_i.name + "-" + pop_j.name + ".tsv"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening haplotype Phi_st file '" << file << "'\n"; exit(1); } fh.precision(fieldw); fh.setf(std::ios::fixed); // // Write the population members. // for (int k : subpop_ids) { const Pop& pop_k = mpopi.pops()[k]; // This is [pop_i], then [pop_j]. fh << "# Population " << pop_k.name << "\t"; for (size_t n = pop_k.first_sample; n <= pop_k.last_sample; n++) { fh << mpopi.samples()[n].name; if (n < pop_k.last_sample) fh << ","; } fh << "\n"; } fh << "# Batch ID " << "\t" << "Locus ID" << "\t" << "Pop 1 ID" << "\t" << "Pop 2 ID" << "\t" << "Chr" << "\t" << "BP" << "\t"; if (log_fst_comp) fh << "SSD(WP)" << "\t" << "SSD(AP/WG)" << "\t" << "SSD(AG)" << "\t" << "SSD(TOTAL)" << "\t" << "MSD(WP)" << "\t" << "MSD(AP/WG)" << "\t" << "MSD(AG)" << "\t" << "MSD(TOTAL)" << "\t" << "n" << "\t" << "n'" << "\t" << "n''" << "\t" << "Sigma2_a" << "\t" << "Sigma2_b" << "\t" << "Sigma2_c" << "\t" << "Sigma_Total" << "\t"; fh << "phi_st" << "\t" << "Smoothed Phi_st" << "\t" << "Smoothed Phi_st P-value" << "\t" << "Fst'" << "\t" << "Smoothed Fst'" << "\t" << "Smoothed Fst' P-value" << "\t" << "D_est" << "\t" << "Smoothed D_est" << "\t" << "Smoothed D_est P-value" << "\n"; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { string chr = it->first; vector &hapstats = genome_hapstats[chr]; for (uint k = 0; k < hapstats.size(); k++) { if (hapstats[k] == NULL) continue; fh << batch_id << "\t" << hapstats[k]->loc_id << "\t" << pop_i.name << "\t" << pop_j.name << "\t" << chr << "\t" << hapstats[k]->bp +1 << "\t"; if (log_fst_comp) fh << hapstats[k]->comp[0] << "\t" << hapstats[k]->comp[1] << "\t" << hapstats[k]->comp[2] << "\t" << hapstats[k]->comp[3] << "\t" << hapstats[k]->comp[4] << "\t" << hapstats[k]->comp[5] << "\t" << hapstats[k]->comp[6] << "\t" << hapstats[k]->comp[7] << "\t" << hapstats[k]->comp[8] << "\t" << hapstats[k]->comp[9] << "\t" << hapstats[k]->comp[10] << "\t" << hapstats[k]->comp[11] << "\t" << hapstats[k]->comp[12] << "\t" << hapstats[k]->comp[13] << "\t" << hapstats[k]->comp[14] << "\t"; fh << hapstats[k]->stat[0] << "\t" << hapstats[k]->smoothed[0] << "\t" << hapstats[k]->bs[0] << "\t" << hapstats[k]->stat[3] << "\t" << hapstats[k]->smoothed[3] << "\t" << hapstats[k]->bs[3] << "\t" << hapstats[k]->stat[4] << "\t" << hapstats[k]->smoothed[4] << "\t" << hapstats[k]->bs[4] << "\n"; delete hapstats[k]; } } fh.close(); cerr << "wrote " << cnt << " loci to pairwise haplotype file, '" << file << "'\n"; } } if (kernel_smoothed && loci_ordered) { delete ks; delete ord; } return 0; } bool fixed_locus(Datum **d, vector &pop_ids) { set loc_haplotypes; map > pop_haplotypes; for (int pop_id : pop_ids) { const Pop& pop = mpopi.pops()[pop_id]; for (size_t i = pop.first_sample; i <= pop.last_sample; i++) { if (d[i] == NULL) continue; if (d[i]->obshap.size() > 2) { continue; } else if (d[i]->obshap.size() == 1) { if (!uncalled_haplotype(d[i]->obshap[0])) { loc_haplotypes.insert(d[i]->obshap[0]); pop_haplotypes[pop_id].push_back(d[i]->obshap[0]); pop_haplotypes[pop_id].push_back(d[i]->obshap[0]); } } else { for (uint j = 0; j < d[i]->obshap.size(); j++) { if (!uncalled_haplotype(d[i]->obshap[0])) { loc_haplotypes.insert(d[i]->obshap[j]); pop_haplotypes[pop_id].push_back(d[i]->obshap[j]); } } } } } uint valid_pops = 0; for (int pop_id : pop_ids) { if (pop_haplotypes[pop_id].size() > 0) valid_pops++; } // // Check that more than one population has data for this locus. // if (valid_pops <= 1) return true; // // Check that there is more than one haplotype at this locus. // if (loc_haplotypes.size() == 1) return true; return false; } LocStat * haplotype_diversity(int start, int end, Datum **d) { map::iterator hit; vector haplotypes; map hap_freq; map hap_index; double n = 0.0; double gene_diversity = 0.0; double hapl_diversity = 0.0; LocStat *lstat; // // Tabulate the haplotypes in this population. // n = count_haplotypes_at_locus(start, end, d, hap_freq); // cerr << " " << n << " total haplotypes observed.\n"; // // If this haplotype is fixed, don't calculate any statistics. // if (n == 0) return NULL; lstat = new LocStat; // // Store a summary of the haplotype counts to output below. // stringstream sstr; for (hit = hap_freq.begin(); hit != hap_freq.end(); hit++) sstr << hit->first << ":" << hit->second << ";"; lstat->hap_str = sstr.str().substr(0, sstr.str().length() - 1); // // Determine an ordering for the haplotypes. Convert haplotype counts into frequencies. // uint k = 0; for (hit = hap_freq.begin(); hit != hap_freq.end(); hit++) { hap_index[hit->first] = k; haplotypes.push_back(hit->first); k++; // cerr << " Haplotype '" << hit->first << "' occured " << hit->second << " times; "; hit->second = hit->second / n; // cerr << " frequency of " << hit->second << "%\n"; } // // Initialize a two-dimensional array to hold distances between haplotyes. // double **hdists = new double *[hap_index.size()]; for (k = 0; k < hap_index.size(); k++) { hdists[k] = new double[hap_index.size()]; memset(hdists[k], 0, hap_index.size()); } // // Calculate the distances between haplotypes. // nuc_substitution_dist(hap_index, hdists); // // Calculate haplotype diversity, Pi. // for (uint i = 0; i < haplotypes.size(); i++) { for (uint j = 0; j < haplotypes.size(); j++) { hapl_diversity += hap_freq[haplotypes[i]] * hap_freq[haplotypes[j]] * hdists[hap_index[haplotypes[i]]][hap_index[haplotypes[j]]]; } } hapl_diversity = (n / (n-1)) * hapl_diversity; // // Calculate gene diversity. // for (uint i = 0; i < haplotypes.size(); i++) { gene_diversity += hap_freq[haplotypes[i]] * hap_freq[haplotypes[i]]; } gene_diversity = (n / (n - 1)) * (1 - gene_diversity); lstat->alleles = n; lstat->stat[0] = gene_diversity; lstat->stat[1] = hapl_diversity; lstat->hap_cnt = haplotypes.size(); // cerr << " Population " << pop_id << " has haplotype diversity (pi) of " << s[pop_index]->pi << "\n"; for (k = 0; k < hap_index.size(); k++) delete hdists[k]; delete [] hdists; return lstat; } HapStat * haplotype_amova(Datum **d, LocSum **s, vector &pop_ids) { map loc_hap_index; vector loc_haplotypes; map > pop_haplotypes; map > grp_members; vector grps; map::iterator hit, hit_2; HapStat *h; // // Tabulate the occurences of haplotypes at this locus. // for (int pop_id : pop_ids) { const Pop& pop = mpopi.pops()[pop_id]; for (size_t i = pop.first_sample; i <= pop.last_sample; i++) { if (d[i] == NULL) continue; if (d[i]->obshap.size() > 2) { continue; } else if (d[i]->obshap.size() == 1) { if(!uncalled_haplotype(d[i]->obshap[0])) { loc_hap_index[d[i]->obshap[0]]++; loc_haplotypes.push_back(d[i]->obshap[0]); loc_haplotypes.push_back(d[i]->obshap[0]); pop_haplotypes[pop_id].push_back(d[i]->obshap[0]); pop_haplotypes[pop_id].push_back(d[i]->obshap[0]); } } else { for (uint j = 0; j < d[i]->obshap.size(); j++) { if(!uncalled_haplotype(d[i]->obshap[0])) { loc_hap_index[d[i]->obshap[j]]++; loc_haplotypes.push_back(d[i]->obshap[j]); pop_haplotypes[pop_id].push_back(d[i]->obshap[j]); } } } } } // // What is the total number of populations that had valid haplotypes. // double valid_pop_cnt = 0.0; for (int pop_id : pop_ids) { if (pop_haplotypes[pop_id].size() > 0) valid_pop_cnt++; } // // If we filtered a population out at this locus make sure that we still have at least one // representative present in each group. // set uniq_grps; for (size_t pop_id=0; pop_id 0) { uniq_grps.insert(pop.group); grp_members[pop.group].push_back(pop_id); } } set::iterator uit; for (uit = uniq_grps.begin(); uit != uniq_grps.end(); uit++) grps.push_back(*uit); if (grps.size() == 0) return NULL; // cerr << "Groups: "; // for (uint i = 0; i < grps.size(); i++) // cerr << grps[i] << ", "; // cerr << "\n"; // for (git = grp_members.begin(); git != grp_members.end(); git++) { // cerr << "Group " << git->first << ": "; // for (uint i = 0; i < git->second.size(); i++) // cerr << git->second[i] << ", "; // cerr << "\n"; // } // // Determine an ordering for the haplotypes. // uint m = 0; for (hit = loc_hap_index.begin(); hit != loc_hap_index.end(); hit++) { loc_hap_index[hit->first] = m; m++; } // // Initialize a two-dimensional array to hold distances between haplotyes. // double **hdists = new double *[loc_hap_index.size()]; double **hdists_max = new double *[loc_hap_index.size()]; for (uint k = 0; k < loc_hap_index.size(); k++) { hdists[k] = new double[loc_hap_index.size()]; memset(hdists[k], 0, loc_hap_index.size()); hdists_max[k] = new double[loc_hap_index.size()]; memset(hdists_max[k], 0, loc_hap_index.size()); } // // Calculate the distances between haplotypes. // nuc_substitution_dist(loc_hap_index, hdists); // // Calculate the sum of squared distances in each subset: total, within populations, across populations // and withing groups, and across groups. // double ssd_total = amova_ssd_total(loc_haplotypes, loc_hap_index, hdists); double ssd_wp = amova_ssd_wp(grps, grp_members, loc_hap_index, pop_haplotypes, hdists); double ssd_ap_wg = amova_ssd_ap_wg(grps, grp_members, loc_hap_index, pop_haplotypes, hdists, hdists); double ssd_ag = grps.size() > 1 ? amova_ssd_ag(grps, grp_members, loc_hap_index, pop_haplotypes, hdists, ssd_total) : 0.0; // // Calculate n // double n = 0.0; double n_1 = 0.0; double n_2 = 0.0; double s_g = 0.0; double tot_cnt = 0.0; double grp_cnt = 0.0; double num_grps = grps.size(); double a = 0.0; double b = 0.0; for (uint g = 0; g < num_grps; g++) { for (uint r = 0; r < grp_members[grps[g]].size(); r++) { int pop_id_1 = grp_members[grps[g]][r]; tot_cnt += (double) pop_haplotypes[pop_id_1].size(); } } for (uint g = 0; g < num_grps; g++) { grp_cnt = 0.0; for (uint r = 0; r < grp_members[grps[g]].size(); r++) { int pop_id_1 = grp_members[grps[g]][r]; grp_cnt += (double) pop_haplotypes[pop_id_1].size(); } a = 0.0; for (uint r = 0; r < grp_members[grps[g]].size(); r++) { int pop_id_1 = grp_members[grps[g]][r]; a += (double) (pop_haplotypes[pop_id_1].size() * pop_haplotypes[pop_id_1].size()) / grp_cnt; } s_g += a; } n = (tot_cnt - s_g) / (double) (valid_pop_cnt - num_grps); // cerr << " n: "<< n << "\n"; if (num_grps > 1) { // // Calculate n' // a = 0.0; for (uint g = 0; g < num_grps; g++) { for (uint r = 0; r < grp_members[grps[g]].size(); r++) { int pop_id_1 = grp_members[grps[g]][r]; a += ((double) (pop_haplotypes[pop_id_1].size() * pop_haplotypes[pop_id_1].size()) / tot_cnt); } } n_1 = (s_g - a) / (double) (num_grps - 1.0); // cerr << " n': "<< n_1 << "\n"; // // Calculate n'' // for (uint g = 0; g < num_grps; g++) { a = 0.0; for (uint r = 0; r < grp_members[grps[g]].size(); r++) { int pop_id_1 = grp_members[grps[g]][r]; a += pop_haplotypes[pop_id_1].size(); } b += ((a * a) / tot_cnt); } n_2 = (tot_cnt - b) / (double) (num_grps - 1); // cerr << " n'': "<< n_2 << "\n"; } // // Calculate the mean square deviations, equal to SSD divided by degrees of freedom. // double msd_ag = num_grps > 1 ? ssd_ag / (double) (num_grps - 1) : 0.0; double msd_ap_wg = ssd_ap_wg / ((double) (valid_pop_cnt - num_grps)); double msd_wp = ssd_wp / ((double) (loc_haplotypes.size() - valid_pop_cnt)); double msd_total = ssd_total / ((double) (loc_haplotypes.size() - 1)); double sigma_c = msd_wp; double sigma_b = n > 0 ? (msd_ap_wg - sigma_c) / n : 0.0; double sigma_a = 0.0; if (grps.size() > 1) sigma_a = (msd_ag - sigma_c - (n_1 * sigma_b)) / n_2; // Arlequin seems to sum the variance components instead of independently calculating sigma_total: MSD(total) = SSD(total)/degrees.of.freedom double sigma_total = sigma_a + sigma_b + sigma_c; // msd_total; double phi_st = 0.0; double phi_ct = 0.0; double phi_sc = 0.0; if (grps.size() > 1) { phi_st = sigma_total > 0.0 ? (sigma_a + sigma_b) / sigma_total : 0.0; phi_ct = sigma_total > 0.0 ? sigma_a / sigma_total : 0.0; phi_sc = (sigma_a + sigma_b) > 0.0 ? sigma_b / (sigma_b + sigma_c) : 0.0; } else { phi_st = sigma_total > 0.0 ? sigma_b / sigma_total : 0.0; } // cerr << " MSD(AG): " << msd_ag << "; MSD(AP/WG): " << msd_ap_wg << "; MSD(WP): " << msd_wp << "; MSD(TOTAL): " << msd_total << "\n" // << " Sigma_a: " << sigma_a << "; Sigma_b: " << sigma_b << "; Sigma_c: " << sigma_c << "; Sigma_Total: " << sigma_total << "\n" // << " Phi_st: " << phi_st << "; Phi_ct: " << phi_ct << "; Phi_sc: " << phi_sc << "\n"; // // Calculate Fst' = Fst / Fst_max // // First calculate Fst. // // To calculate Fst instead of Phi_st, we need to reset our distance matrix to return 1 if haplotypes are different, 0 otherwise. // nuc_substitution_identity(loc_hap_index, hdists); ssd_wp = amova_ssd_wp(grps, grp_members, loc_hap_index, pop_haplotypes, hdists); ssd_ap_wg = amova_ssd_ap_wg(grps, grp_members, loc_hap_index, pop_haplotypes, hdists, hdists); // // Calculate the mean square deviations, equal to SSD divided by degrees of freedom. // msd_ap_wg = ssd_ap_wg / ((double) (valid_pop_cnt - num_grps)); msd_wp = ssd_wp / ((double) (loc_haplotypes.size() - valid_pop_cnt)); sigma_c = msd_wp; sigma_b = n > 0 ? (msd_ap_wg - sigma_c) / n : 0.0; sigma_total = sigma_b + sigma_c; double fst = sigma_total > 0.0 ? sigma_b / sigma_total : 0.0; // // Now calculate Fst_max. // // Reset our distance matrix to give maximum possible distance between haplotypes // and recalculate sum of squared deviations across groups. // nuc_substitution_identity_max(loc_hap_index, hdists_max); ssd_ap_wg = amova_ssd_ap_wg(grps, grp_members, loc_hap_index, pop_haplotypes, hdists, hdists_max); // // Recalculate the mean square deviations, given maximum divergence between populations. // msd_ap_wg = ssd_ap_wg / ((double) (valid_pop_cnt - num_grps)); sigma_b = n > 0 ? (msd_ap_wg - sigma_c) / n : 0.0; double fst_max = sigma_total > 0.0 ? sigma_b / sigma_total : 0.0; double fst_1 = fst_max > 0.0 ? fst / fst_max : 0.0; // // Cache the results so we can print them in order below, once the parallel code has executed. // h = new HapStat; h->alleles = tot_cnt; h->popcnt = valid_pop_cnt; if (log_fst_comp) { h->comp = new double[15]; h->comp[0] = ssd_wp; h->comp[1] = ssd_ap_wg; h->comp[2] = ssd_ag; h->comp[3] = ssd_total; h->comp[4] = msd_wp; h->comp[5] = msd_ap_wg; h->comp[6] = msd_ag; h->comp[7] = msd_total; h->comp[8] = n; h->comp[9] = n_1; h->comp[10] = n_2; h->comp[11] = sigma_a; h->comp[12] = sigma_b; h->comp[13] = sigma_c; h->comp[14] = sigma_total; } h->stat[0] = phi_st; h->stat[1] = phi_ct; h->stat[2] = phi_sc; h->stat[3] = fst_1; for (uint k = 0; k < loc_hap_index.size(); k++) { delete [] hdists[k]; delete [] hdists_max[k]; } delete [] hdists; delete [] hdists_max; return h; } double amova_ssd_total(vector &loc_haplotypes, map &loc_hap_index, double **hdists) { // // Calculate sum of squared deviations for the total sample, SSD(Total) // double ssd_total = 0.0; for (uint j = 0; j < loc_haplotypes.size(); j++) { for (uint k = 0; k < loc_haplotypes.size(); k++) { ssd_total += hdists[loc_hap_index[loc_haplotypes[j]]][loc_hap_index[loc_haplotypes[k]]]; // cerr << j << "\t" // << k << "\t" // << loc_haplotypes[j] << "\t" // << loc_haplotypes[k] << "\t" // << hdists[loc_hap_index[loc_haplotypes[j]]][loc_hap_index[loc_haplotypes[k]]] << "\n"; } } ssd_total = (1.0 / (double) (2*loc_haplotypes.size())) * ssd_total; // cerr << " ssd_total: "<< ssd_total << "\n"; return ssd_total; } double amova_ssd_wp(vector &grps, map > &grp_members, map &loc_hap_index, map > &pop_haplotypes, double **hdists) { // // Calculate the sum of squared deviations within populations, SSD(WP) // double ssd_wp = 0.0; double ssd = 0.0; int pop_id; for (uint g = 0; g < grps.size(); g++) { for (uint i = 0; i < grp_members[grps[g]].size(); i++) { pop_id = grp_members[grps[g]][i]; ssd = 0.0; for (uint j = 0; j < pop_haplotypes[pop_id].size(); j++) { for (uint k = 0; k < pop_haplotypes[pop_id].size(); k++) { ssd += hdists[loc_hap_index[pop_haplotypes[pop_id][j]]][loc_hap_index[pop_haplotypes[pop_id][k]]]; // cerr << pop_id << "\t" // << j << "\t" // << k << "\t" // << loc_haplotypes[j] << "\t" // << loc_haplotypes[k] << "\t" // << hdists[loc_hap_index[loc_haplotypes[j]]][loc_hap_index[loc_haplotypes[k]]] << "\n"; } } if (pop_haplotypes[pop_id].size() > 0) ssd_wp += (1.0 / (double) (2*pop_haplotypes[pop_id].size())) * ssd; } } // cerr << " ssd_wp: "<< ssd_wp << "\n"; return ssd_wp; } double amova_ssd_ap_wg(vector &grps, map > &grp_members, map &loc_hap_index, map > &pop_haplotypes, double **hdists_1, double **hdists_2) { // // Calculate the sum of squared deviations across populations and within groups, SSD(AP/WG) // double ssd_ap_wg = 0.0; double ssd = 0.0; double ssd_1 = 0.0; double ssd_2 = 0.0; double den = 0.0; int pop_id, pop_id_1, pop_id_2; for (uint g = 0; g < grps.size(); g++) { ssd_1 = 0.0; for (uint r = 0; r < grp_members[grps[g]].size(); r++) { pop_id_1 = grp_members[grps[g]][r]; for (uint j = 0; j < pop_haplotypes[pop_id_1].size(); j++) { for (uint s = 0; s < grp_members[grps[g]].size(); s++) { pop_id_2 = grp_members[grps[g]][s]; for (uint k = 0; k < pop_haplotypes[pop_id_2].size(); k++) { if (pop_id_1 == pop_id_2) ssd_1 += hdists_1[loc_hap_index[pop_haplotypes[pop_id_1][j]]][loc_hap_index[pop_haplotypes[pop_id_2][k]]]; else ssd_1 += hdists_2[loc_hap_index[pop_haplotypes[pop_id_1][j]]][loc_hap_index[pop_haplotypes[pop_id_2][k]]]; } } } } den = 0.0; for (uint r = 0; r < grp_members[grps[g]].size(); r++) { pop_id_1 = grp_members[grps[g]][r]; den += 2 * pop_haplotypes[pop_id_1].size(); } ssd_1 = ssd_1 / den; ssd_2 = 0.0; for (uint r = 0; r < grp_members[grps[g]].size(); r++) { pop_id = grp_members[grps[g]][r]; ssd = 0.0; for (uint j = 0; j < pop_haplotypes[pop_id].size(); j++) { for (uint k = 0; k < pop_haplotypes[pop_id].size(); k++) { ssd += hdists_1[loc_hap_index[pop_haplotypes[pop_id][j]]][loc_hap_index[pop_haplotypes[pop_id][k]]]; } } if (pop_haplotypes[pop_id].size() > 0) ssd_2 += (1.0 / (double) (2*pop_haplotypes[pop_id].size())) * ssd; } ssd_ap_wg += ssd_1 - ssd_2; } // cerr << " ssd_ap_wg: "<< ssd_ap_wg << "\n"; return ssd_ap_wg; } double amova_ssd_ag(vector &grps, map > &grp_members, map &loc_hap_index, map > &pop_haplotypes, double **hdists, double ssd_total) { // // Calculate the sum of squared deviations across groups, SSD(AG) // int pop_id_1, pop_id_2; double ssd_ag = 0.0; double ssd = 0.0; double ssd_1 = 0.0; double den = 0.0; for (uint g = 0; g < grps.size(); g++) { ssd_1 = 0.0; for (uint r = 0; r < grp_members[grps[g]].size(); r++) { pop_id_1 = grp_members[grps[g]][r]; for (uint j = 0; j < pop_haplotypes[pop_id_1].size(); j++) { for (uint s = 0; s < grp_members[grps[g]].size(); s++) { pop_id_2 = grp_members[grps[g]][s]; for (uint k = 0; k < pop_haplotypes[pop_id_2].size(); k++) { ssd_1 += hdists[loc_hap_index[pop_haplotypes[pop_id_1][j]]][loc_hap_index[pop_haplotypes[pop_id_2][k]]]; } } } } den = 0.0; for (uint r = 0; r < grp_members[grps[g]].size(); r++) { pop_id_1 = grp_members[grps[g]][r]; den += 2 * pop_haplotypes[pop_id_1].size(); } ssd += ssd_1 / den; } ssd_ag = ssd_total - ssd; // cerr << " ssd_ag: "<< ssd_ag << "\n"; return ssd_ag; } double haplotype_d_est(Datum **d, LocSum **s, vector &pop_ids) { // // Calculate D_est, fixation index, as described by // Bird, et al., 2011, Detecting and measuring genetic differentiation // +-Equation 11 // and // Jost, 2008, GST and its relatives do not measure differentiation, Molecular Ecology // +- Equation 13, D_est_chao // map loc_haplotypes; map > pop_haplotypes; map pop_totals; map::iterator it; uint pop_cnt = pop_ids.size(); // // Tabulate the occurences of haplotypes at this locus. // for (int pop_id : pop_ids) { const Pop& pop = mpopi.pops()[pop_id]; for (size_t i = pop.first_sample; i <= pop.last_sample; i++) { if (d[i] == NULL) { continue; } else if (d[i]->obshap.size() > 2) { continue; } else if (d[i]->obshap.size() == 1) { loc_haplotypes[d[i]->obshap[0]] += 2; pop_haplotypes[pop_id][d[i]->obshap[0]] += 2; } else { for (uint j = 0; j < d[i]->obshap.size(); j++) { loc_haplotypes[d[i]->obshap[j]]++; pop_haplotypes[pop_id][d[i]->obshap[j]]++; } } } for (it = pop_haplotypes[pop_id].begin(); it != pop_haplotypes[pop_id].end(); it++) pop_totals[pop_id] += it->second; } double x = 0.0; for (it = loc_haplotypes.begin(); it != loc_haplotypes.end(); it++) { double freq_sum_sq = 0.0; double freq_sq_sum = 0.0; for (int pop_id : pop_ids) { freq_sum_sq += (pop_haplotypes[pop_id][it->first] / pop_totals[pop_id]); freq_sq_sum += pow((pop_haplotypes[pop_id][it->first] / pop_totals[pop_id]), 2); } freq_sum_sq = pow(freq_sum_sq, 2); x += (freq_sum_sq - freq_sq_sum) / (pop_cnt - 1); } double y = 0.0; for (it = loc_haplotypes.begin(); it != loc_haplotypes.end(); it++) { for (int pop_id : pop_ids) { y += (pop_haplotypes[pop_id][it->first] * (pop_haplotypes[pop_id][it->first] - 1)) / (pop_totals[pop_id] * (pop_totals[pop_id] - 1)); } } double d_est = 1.0 - (x / y); return d_est; } int calculate_summary_stats(map &catalog, PopMap *pmap, PopSum *psum) { map >::iterator it; CSLocus *loc; LocSum **s; LocTally *t; int len; int pop_cnt = psum->pop_cnt(); // // Calculate the means for each summary statistic. // int *private_cnt; double *num_indv_mean, *p_mean, *obs_het_mean, *obs_hom_mean, *exp_het_mean, *exp_hom_mean, *pi_mean, *fis_mean; double *num_indv_var, *p_var, *obs_het_var, *obs_hom_var, *exp_het_var, *exp_hom_var, *pi_var, *fis_var; double *num_indv_mean_all, *p_mean_all, *obs_het_mean_all, *obs_hom_mean_all, *exp_het_mean_all, *exp_hom_mean_all, *pi_mean_all, *fis_mean_all; double *num_indv_var_all, *p_var_all, *obs_het_var_all, *obs_hom_var_all, *exp_het_var_all, *exp_hom_var_all, *pi_var_all, *fis_var_all; double *n, *n_all, *var_sites; private_cnt = new int[pop_cnt]; n = new double[pop_cnt]; var_sites = new double[pop_cnt]; num_indv_mean = new double[pop_cnt]; num_indv_var = new double[pop_cnt]; p_mean = new double[pop_cnt]; p_var = new double[pop_cnt]; obs_het_mean = new double[pop_cnt]; obs_het_var = new double[pop_cnt]; obs_hom_mean = new double[pop_cnt]; obs_hom_var = new double[pop_cnt]; exp_het_mean = new double[pop_cnt]; exp_het_var = new double[pop_cnt]; exp_hom_mean = new double[pop_cnt]; exp_hom_var = new double[pop_cnt]; pi_mean = new double[pop_cnt]; pi_var = new double[pop_cnt]; fis_mean = new double[pop_cnt]; fis_var = new double[pop_cnt]; n_all = new double[pop_cnt]; num_indv_mean_all = new double[pop_cnt]; num_indv_var_all = new double[pop_cnt]; p_mean_all = new double[pop_cnt]; p_var_all = new double[pop_cnt]; obs_het_mean_all = new double[pop_cnt]; obs_het_var_all = new double[pop_cnt]; obs_hom_mean_all = new double[pop_cnt]; obs_hom_var_all = new double[pop_cnt]; exp_het_mean_all = new double[pop_cnt]; exp_het_var_all = new double[pop_cnt]; exp_hom_mean_all = new double[pop_cnt]; exp_hom_var_all = new double[pop_cnt]; pi_mean_all = new double[pop_cnt]; pi_var_all = new double[pop_cnt]; fis_mean_all = new double[pop_cnt]; fis_var_all = new double[pop_cnt]; for (int j = 0; j < pop_cnt; j++) { private_cnt[j] = 0; n[j] = 0.0; var_sites[j] = 0.0; num_indv_mean[j] = 0.0; num_indv_var[j] = 0.0; p_mean[j] = 0.0; p_var[j] = 0.0; obs_het_mean[j] = 0.0; obs_het_var[j] = 0.0; obs_hom_mean[j] = 0.0; obs_hom_var[j] = 0.0; exp_het_mean[j] = 0.0; exp_het_var[j] = 0.0; exp_hom_mean[j] = 0.0; exp_hom_var[j] = 0.0; pi_mean[j] = 0.0; pi_var[j] = 0.0; fis_mean[j] = 0.0; fis_var[j] = 0.0; n_all[j] = 0.0; num_indv_mean_all[j] = 0.0; num_indv_var_all[j] = 0.0; p_mean_all[j] = 0.0; p_var_all[j] = 0.0; obs_het_mean_all[j] = 0.0; obs_het_var_all[j] = 0.0; obs_hom_mean_all[j] = 0.0; obs_hom_var_all[j] = 0.0; exp_het_mean_all[j] = 0.0; exp_het_var_all[j] = 0.0; exp_hom_mean_all[j] = 0.0; exp_hom_var_all[j] = 0.0; pi_mean_all[j] = 0.0; pi_var_all[j] = 0.0; fis_mean_all[j] = 0.0; fis_var_all[j] = 0.0; } for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; s = psum->locus(loc->id); t = psum->locus_tally(loc->id); len = strlen(loc->con); for (int i = 0; i < len; i++) { // // Compile private alleles // if (t->nucs[i].priv_allele >= 0) private_cnt[t->nucs[i].priv_allele]++; if (t->nucs[i].allele_cnt == 2) { for (int j = 0; j < pop_cnt; j++) { if (s[j]->nucs[i].num_indv == 0) continue; n[j]++; if (s[j]->nucs[i].pi > 0) var_sites[j]++; num_indv_mean[j] += s[j]->nucs[i].num_indv; p_mean[j] += s[j]->nucs[i].p; obs_het_mean[j] += s[j]->nucs[i].obs_het; obs_hom_mean[j] += s[j]->nucs[i].obs_hom; exp_het_mean[j] += s[j]->nucs[i].exp_het; exp_hom_mean[j] += s[j]->nucs[i].exp_hom; pi_mean[j] += s[j]->nucs[i].stat[0]; fis_mean[j] += s[j]->nucs[i].stat[1] != -7.0 ? s[j]->nucs[i].stat[1] : 0.0; n_all[j]++; num_indv_mean_all[j] += s[j]->nucs[i].num_indv; p_mean_all[j] += s[j]->nucs[i].p; obs_het_mean_all[j] += s[j]->nucs[i].obs_het; obs_hom_mean_all[j] += s[j]->nucs[i].obs_hom; exp_het_mean_all[j] += s[j]->nucs[i].exp_het; exp_hom_mean_all[j] += s[j]->nucs[i].exp_hom; pi_mean_all[j] += s[j]->nucs[i].stat[0]; fis_mean_all[j] += s[j]->nucs[i].stat[1] != -7.0 ? s[j]->nucs[i].stat[1] : 0.0; } } else if (t->nucs[i].allele_cnt == 1) { for (int j = 0; j < pop_cnt; j++) { if (s[j]->nucs[i].num_indv == 0) continue; n_all[j]++; num_indv_mean_all[j] += s[j]->nucs[i].num_indv; p_mean_all[j] += s[j]->nucs[i].p; obs_het_mean_all[j] += s[j]->nucs[i].obs_het; obs_hom_mean_all[j] += s[j]->nucs[i].obs_hom; exp_het_mean_all[j] += s[j]->nucs[i].exp_het; exp_hom_mean_all[j] += s[j]->nucs[i].exp_hom; pi_mean_all[j] += s[j]->nucs[i].stat[0]; fis_mean_all[j] += s[j]->nucs[i].stat[1] != -7.0 ? s[j]->nucs[i].stat[1] : 0.0; } } } } } for (int j = 0; j < pop_cnt; j++) { num_indv_mean[j] = num_indv_mean[j] / n[j]; p_mean[j] = p_mean[j] / n[j]; obs_het_mean[j] = obs_het_mean[j] / n[j]; obs_hom_mean[j] = obs_hom_mean[j] / n[j]; exp_het_mean[j] = exp_het_mean[j] / n[j]; exp_hom_mean[j] = exp_hom_mean[j] / n[j]; pi_mean[j] = pi_mean[j] / n[j]; fis_mean[j] = fis_mean[j] / n[j]; num_indv_mean_all[j] = num_indv_mean_all[j] / n_all[j]; p_mean_all[j] = p_mean_all[j] / n_all[j]; obs_het_mean_all[j] = obs_het_mean_all[j] / n_all[j]; obs_hom_mean_all[j] = obs_hom_mean_all[j] / n_all[j]; exp_het_mean_all[j] = exp_het_mean_all[j] / n_all[j]; exp_hom_mean_all[j] = exp_hom_mean_all[j] / n_all[j]; pi_mean_all[j] = pi_mean_all[j] / n_all[j]; fis_mean_all[j] = fis_mean_all[j] / n_all[j]; } string file = out_path + out_prefix + ".sumstats.tsv"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening sumstats file '" << file << "'\n"; exit(1); } fh.precision(fieldw); fh.setf(std::ios::fixed); double p_freq; // // Write the population members. // for (auto& pop : mpopi.pops()) { fh << "# " << pop.name << "\t"; for (size_t i = pop.first_sample; i <= pop.last_sample; i++) { fh << mpopi.samples()[i].name; if (i < pop.last_sample) fh << ","; } fh << "\n"; } cerr << "Writing " << catalog.size() << " loci to summary statistics file, '" << file << "'\n"; fh << "# Batch ID " << "\t" << "Locus ID" << "\t" << "Chr" << "\t" << "BP" << "\t" << "Col" << "\t" << "Pop ID" << "\t" << "P Nuc" << "\t" << "Q Nuc" << "\t" << "N" << "\t" << "P" << "\t" << "Obs Het" << "\t" << "Obs Hom" << "\t" << "Exp Het" << "\t" << "Exp Hom" << "\t" << "Pi" << "\t" << "Smoothed Pi" << "\t" << "Smoothed Pi P-value" << "\t" << "Fis" << "\t" << "Smoothed Fis" << "\t" << "Smoothed Fis P-value" << "\t" << "Private" << "\n"; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { for (uint pos = 0; pos < it->second.size(); pos++) { loc = it->second[pos]; s = psum->locus(loc->id); t = psum->locus_tally(loc->id); len = strlen(loc->con); for (int i = 0; i < len; i++) { // // If this site is fixed in all populations, DON'T output it. If it is variable, // or fixed within populations but variable among, DO output it. // if (t->nucs[i].allele_cnt == 2) { for (int j = 0; j < pop_cnt; j++) { if (s[j]->nucs[i].num_indv == 0) continue; fh << batch_id << "\t" << loc->id << "\t" << loc->loc.chr << "\t" << loc->sort_bp(i) + 1 << "\t" << i << "\t" << mpopi.pops()[j].name << "\t"; // // Output the p and q alleles in the same order in each population. // if (t->nucs[i].p_allele == s[j]->nucs[i].p_nuc) { if (s[j]->nucs[i].q_nuc == 0) fh << s[j]->nucs[i].p_nuc << "\t" << "-"; else fh << s[j]->nucs[i].p_nuc << "\t" << s[j]->nucs[i].q_nuc; p_freq = s[j]->nucs[i].p; } else { if (s[j]->nucs[i].q_nuc == 0) fh << "-\t" << s[j]->nucs[i].p_nuc; else fh << s[j]->nucs[i].q_nuc << "\t" << s[j]->nucs[i].p_nuc; p_freq = 1 - s[j]->nucs[i].p; } fh << "\t" << (int) s[j]->nucs[i].num_indv << "\t" << std::setprecision(8) << p_freq << "\t" << std::setprecision(fieldw) << s[j]->nucs[i].obs_het << "\t" << s[j]->nucs[i].obs_hom << "\t" << s[j]->nucs[i].exp_het << "\t" << s[j]->nucs[i].exp_hom << "\t" << s[j]->nucs[i].stat[0] << "\t" // Pi << s[j]->nucs[i].smoothed[0] << "\t" // Smoothed Pi << s[j]->nucs[i].bs[0] << "\t" // Pi bootstrapped p-value << (s[j]->nucs[i].stat[1] == -7.0 ? 0.0 : s[j]->nucs[i].stat[1]) << "\t" // Fis << s[j]->nucs[i].smoothed[1] << "\t" // Smoothed Fis << s[j]->nucs[i].bs[1] << "\t"; // Fis bootstrapped p-value. (t->nucs[i].priv_allele == j) ? fh << "1\n" : fh << "0\n"; // // Tabulate the residuals to calculate the variance. // num_indv_var[j] += pow((s[j]->nucs[i].num_indv - num_indv_mean[j]), 2); p_var[j] += pow((s[j]->nucs[i].p - p_mean[j]), 2); obs_het_var[j] += pow((s[j]->nucs[i].obs_het - obs_het_mean[j]), 2); obs_hom_var[j] += pow((s[j]->nucs[i].obs_hom - obs_hom_mean[j]), 2); exp_het_var[j] += pow((s[j]->nucs[i].exp_het - exp_het_mean[j]), 2); exp_hom_var[j] += pow((s[j]->nucs[i].exp_hom - exp_hom_mean[j]), 2); pi_var[j] += pow((s[j]->nucs[i].stat[0] - pi_mean[j]), 2); fis_var[j] += pow((s[j]->nucs[i].stat[1] - fis_mean[j]), 2); num_indv_var_all[j] += pow((s[j]->nucs[i].num_indv - num_indv_mean_all[j]), 2); p_var_all[j] += pow((s[j]->nucs[i].p - p_mean_all[j]), 2); obs_het_var_all[j] += pow((s[j]->nucs[i].obs_het - obs_het_mean_all[j]), 2); obs_hom_var_all[j] += pow((s[j]->nucs[i].obs_hom - obs_hom_mean_all[j]), 2); exp_het_var_all[j] += pow((s[j]->nucs[i].exp_het - exp_het_mean_all[j]), 2); exp_hom_var_all[j] += pow((s[j]->nucs[i].exp_hom - exp_hom_mean_all[j]), 2); pi_var_all[j] += pow((s[j]->nucs[i].stat[0] - pi_mean_all[j]), 2); fis_var_all[j] += pow((s[j]->nucs[i].stat[1] - fis_mean_all[j]), 2); } } else if (t->nucs[i].allele_cnt == 1) { for (int j = 0; j < pop_cnt; j++) { if (s[j]->nucs[i].num_indv == 0) continue; num_indv_var_all[j] += pow((s[j]->nucs[i].num_indv - num_indv_mean_all[j]), 2); p_var_all[j] += pow((s[j]->nucs[i].p - p_mean_all[j]), 2); obs_het_var_all[j] += pow((s[j]->nucs[i].obs_het - obs_het_mean_all[j]), 2); obs_hom_var_all[j] += pow((s[j]->nucs[i].obs_hom - obs_hom_mean_all[j]), 2); exp_het_var_all[j] += pow((s[j]->nucs[i].exp_het - exp_het_mean_all[j]), 2); exp_hom_var_all[j] += pow((s[j]->nucs[i].exp_hom - exp_hom_mean_all[j]), 2); pi_var_all[j] += pow((s[j]->nucs[i].stat[0] - pi_mean_all[j]), 2); fis_var_all[j] += pow((s[j]->nucs[i].stat[1] - fis_mean_all[j]), 2); } } } } } // // Calculate the variance. // for (int j = 0; j < pop_cnt; j++) { num_indv_var[j] = num_indv_var[j] / (n[j] - 1); p_var[j] = p_var[j] / (n[j] - 1); obs_het_var[j] = obs_het_var[j] / (n[j] - 1); obs_hom_var[j] = obs_hom_var[j] / (n[j] - 1); exp_het_var[j] = exp_het_var[j] / (n[j] - 1); exp_hom_var[j] = exp_hom_var[j] / (n[j] - 1); pi_var[j] = pi_var[j] / (n[j] - 1); fis_var[j] = fis_var[j] / (n[j] - 1); num_indv_var_all[j] = num_indv_var_all[j] / (n_all[j] - 1); p_var_all[j] = p_var_all[j] / (n_all[j] - 1); obs_het_var_all[j] = obs_het_var_all[j] / (n_all[j] - 1); obs_hom_var_all[j] = obs_hom_var_all[j] / (n_all[j] - 1); exp_het_var_all[j] = exp_het_var_all[j] / (n_all[j] - 1); exp_hom_var_all[j] = exp_hom_var_all[j] / (n_all[j] - 1); pi_var_all[j] = pi_var_all[j] / (n_all[j] - 1); fis_var_all[j] = fis_var_all[j] / (n_all[j] - 1); } fh.close(); file = out_path + out_prefix + ".sumstats_summary.tsv"; fh.open(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening sumstats summary file '" << file << "'\n"; exit(1); } // // Write out summary statistics of the summary statistics. // fh << "# Variant positions\n" << "# Pop ID\t" << "Private\t" << "Num Indv\t" << "Var\t" << "StdErr\t" << "P\t" << "Var\t" << "StdErr\t" << "Obs Het\t" << "Var\t" << "StdErr\t" << "Obs Hom\t" << "Var\t" << "StdErr\t" << "Exp Het\t" << "Var\t" << "StdErr\t" << "Exp Hom\t" << "Var\t" << "StdErr\t" << "Pi\t" << "Var\t" << "StdErr\t" << "Fis\t" << "Var\t" << "StdErr\n"; double *sq_n = new double[pop_cnt]; double *sq_n_all = new double[pop_cnt]; for (int j = 0; j < pop_cnt; j++) { sq_n[j] = sqrt(n[j]); sq_n_all[j] = sqrt(n_all[j]); } for (int j = 0; j < pop_cnt; j++) fh << mpopi.pops()[j].name << "\t" << private_cnt[j] << "\t" << num_indv_mean[j] << "\t" << num_indv_var[j] << "\t" << sqrt(num_indv_var[j]) / sq_n[j] << "\t" << p_mean[j] << "\t" << p_var[j] << "\t" << sqrt(p_var[j]) / sq_n[j] << "\t" << obs_het_mean[j] << "\t" << obs_het_var[j] << "\t" << sqrt(obs_het_var[j]) / sq_n[j] << "\t" << obs_hom_mean[j] << "\t" << obs_hom_var[j] << "\t" << sqrt(obs_hom_var[j]) / sq_n[j] << "\t" << exp_het_mean[j] << "\t" << exp_het_var[j] << "\t" << sqrt(exp_het_var[j]) / sq_n[j] << "\t" << exp_hom_mean[j] << "\t" << exp_hom_var[j] << "\t" << sqrt(exp_hom_var[j]) / sq_n[j] << "\t" << pi_mean[j] << "\t" << pi_var[j] << "\t" << sqrt(pi_var[j]) / sq_n[j] << "\t" << fis_mean[j] << "\t" << fis_var[j] << "\t" << sqrt(num_indv_var[j]) / sq_n[j] << "\n"; fh << "# All positions (variant and fixed)\n" << "# Pop ID\t" << "Private\t" << "Sites\t" << "Variant Sites\t" << "Polymorphic Sites\t" << "% Polymorphic Loci\t" << "Num Indv\t" << "Var\t" << "StdErr\t" << "P\t" << "Var\t" << "StdErr\t" << "Obs Het\t" << "Var\t" << "StdErr\t" << "Obs Hom\t" << "Var\t" << "StdErr\t" << "Exp Het\t" << "Var\t" << "StdErr\t" << "Exp Hom\t" << "Var\t" << "StdErr\t" << "Pi\t" << "Var\t" << "StdErr\t" << "Fis\t" << "Var\t" << "StdErr\n"; for (int j = 0; j < pop_cnt; j++) { fh << mpopi.pops()[j].name << "\t" << private_cnt[j] << "\t" << n_all[j] << "\t" << n[j] << "\t" << var_sites[j] << "\t" << var_sites[j] / n_all[j] * 100 << "\t" << num_indv_mean_all[j] << "\t" << num_indv_var_all[j] << "\t" << sqrt(num_indv_var_all[j]) / sq_n_all[j] << "\t" << p_mean_all[j] << "\t" << p_var_all[j] << "\t" << sqrt(p_var_all[j]) / sq_n_all[j] << "\t" << obs_het_mean_all[j] << "\t" << obs_het_var_all[j] << "\t" << sqrt(obs_het_var_all[j]) / sq_n_all[j] << "\t" << obs_hom_mean_all[j] << "\t" << obs_hom_var_all[j] << "\t" << sqrt(obs_hom_var_all[j]) / sq_n_all[j] << "\t" << exp_het_mean_all[j] << "\t" << exp_het_var_all[j] << "\t" << sqrt(exp_het_var_all[j]) / sq_n_all[j] << "\t" << exp_hom_mean_all[j] << "\t" << exp_hom_var_all[j] << "\t" << sqrt(exp_hom_var_all[j]) / sq_n_all[j] << "\t" << pi_mean_all[j] << "\t" << pi_var_all[j] << "\t" << sqrt(pi_var_all[j]) / sq_n_all[j] << "\t" << fis_mean_all[j] << "\t" << fis_var_all[j] << "\t" << sqrt(num_indv_var_all[j]) / sq_n_all[j] << "\n"; } delete [] private_cnt; delete [] n; delete [] var_sites; delete [] sq_n; delete [] num_indv_mean; delete [] num_indv_var; delete [] p_mean; delete [] p_var; delete [] obs_het_mean; delete [] obs_het_var; delete [] obs_hom_mean; delete [] obs_hom_var; delete [] exp_het_mean; delete [] exp_het_var; delete [] exp_hom_mean; delete [] exp_hom_var; delete [] pi_mean; delete [] pi_var; delete [] fis_mean; delete [] fis_var; delete [] n_all; delete [] sq_n_all; delete [] num_indv_mean_all; delete [] num_indv_var_all; delete [] p_mean_all; delete [] p_var_all; delete [] obs_het_mean_all; delete [] obs_het_var_all; delete [] obs_hom_mean_all; delete [] obs_hom_var_all; delete [] exp_het_mean_all; delete [] exp_het_var_all; delete [] exp_hom_mean_all; delete [] exp_hom_var_all; delete [] pi_mean_all; delete [] pi_var_all; delete [] fis_mean_all; delete [] fis_var_all; fh.close(); return 0; } int write_fst_stats(map &catalog, PopMap *pmap, PopSum *psum, ofstream &log_fh) { // // We want to iterate over each pair of populations and calculate Fst at each // nucleotide of each locus. // if (mpopi.pops().size() == 1) return 0; vector means; // // Instantiate the kernel smoothing object if requested. // OPopPair *ord = new OPopPair(psum, log_fh); KSmooth *ks; Bootstrap *bs; if (kernel_smoothed && loci_ordered) { cerr << "Instantiating the kernel smoothing window, using sigma = " << sigma << " with a sliding window size of " << 6 * sigma << "\n"; ks = new KSmooth(2); } for (uint pop_1 = 0; pop_1 < mpopi.pops().size(); pop_1++) { const Pop& pop_1p = mpopi.pops()[pop_1]; for (uint pop_2 = pop_1 + 1; pop_2 < mpopi.pops().size(); pop_2++) { const Pop& pop_2p = mpopi.pops()[pop_2]; double sum = 0.0; double cnt = 0.0; string file = out_path + out_prefix + ".fst_" + pop_1p.name + "-" + pop_2p.name + ".tsv"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening Fst output file '" << file << "'\n"; exit(1); } fh.precision(fieldw); fh.setf(std::ios::fixed); cerr << "Calculating Fst for populations '" << pop_1p.name << "' and '" << pop_2p.name << "' and writing it to file, '" << file << "'\n"; fh << "# Batch ID" << "\t" << "Locus ID" << "\t" << "Pop 1 ID" << "\t" << "Pop 2 ID" << "\t" << "Chr" << "\t" << "BP" << "\t" << "Column" << "\t" << "Overall Pi" << "\t" << "Fst" << "\t" << "Fisher's P" << "\t" << "Odds Ratio" << "\t" << "CI Low" << "\t" << "CI High" << "\t" << "LOD" << "\t" << "Corrected Fst" << "\t" << "Smoothed Fst" << "\t" << "AMOVA Fst" << "\t" << "Corrected AMOVA Fst" << "\t" << "Smoothed AMOVA Fst" << "\t" << "Smoothed AMOVA Fst P-value" << "\t" << "Window SNP Count"; // // If requested, log Fst component calculations to a file. // if (log_fst_comp) { fh << "\t" << "n_1" << "\t" << "n_2" << "\t" << "tot_alleles" << "\t" << "p_1" << "\t" << "q_1" << "\t" << "p_2" << "\t" << "q_2" << "\t" << "pi_1" << "\t" << "pi_2" << "\t" << "pi_all" << "\t" << "bcoeff_1" << "\t" << "bcoeff_2" << "\t" << "binomial_fst" << "\t" << "p_1_freq" << "\t" << "q_1_freq" << "\t" << "p_2_freq" << "\t" << "q_2_freq" << "\t" << "p_avg_cor" << "\t" << "n_avg_cor" << "\t" << "amova_fst" << "\n"; } else { fh << "\n"; } if (bootstrap_fst) bs = new Bootstrap(2); map >::iterator it; map > genome_pairs; // int snp_dist[max_snp_dist] = {0}; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { string chr = it->first; map pairs_key; vector &pairs = genome_pairs[chr]; // // Order loci between the two populations and calculate Fst // ord->order(pairs, pairs_key, it->second, pop_1, pop_2); // // Apply user-selected correction to the Fst values. // double correction; switch(fst_correction) { case p_value: for (uint i = 0; i < pairs.size(); i++) { if (pairs[i] != NULL) { pairs[i]->stat[0] = pairs[i]->fet_p < p_value_cutoff ? pairs[i]->fst : 0; pairs[i]->stat[1] = pairs[i]->fet_p < p_value_cutoff ? pairs[i]->amova_fst : 0; } } break; case bonferroni_win: correct_fst_bonferroni_win(pairs); break; case bonferroni_gen: correction = p_value_cutoff / catalog.size(); for (uint i = 0; i < pairs.size(); i++) { if (pairs[i] != NULL) { pairs[i]->stat[0] = pairs[i]->fet_p < correction ? pairs[i]->fst : 0; pairs[i]->stat[1] = pairs[i]->fet_p < correction ? pairs[i]->amova_fst : 0; } } break; case no_correction: for (uint i = 0; i < pairs.size(); i++) { if (pairs[i] != NULL) { pairs[i]->stat[0] = pairs[i]->fst; pairs[i]->stat[1] = pairs[i]->amova_fst; } } break; } // // If bootstrapping is enabled, record all Fst values. // if (bootstrap_fst) bs->add_data(pairs); // // Calculate kernel-smoothed Fst values. // if (kernel_smoothed && loci_ordered) { cerr << " Generating kernel-smoothed Fst for " << it->first << ".\n"; ks->smooth(pairs); } } // // If bootstrap resampling method is approximate, generate our single, empirical distribution. // map > approx_fst_dist; // if (bootstrap_fst && bootstrap_type == bs_approx) // bootstrap_fst_approximate_dist(fst_samples, allele_depth_samples, weights, snp_dist, approx_fst_dist); for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { string chr = it->first; vector &pairs = genome_pairs[chr]; // // Bootstrap resample this chromosome. // if (bootstrap_fst && bootstrap_type == bs_exact) { cerr << " Bootstrap resampling kernel-smoothed Fst for " << it->first << ".\n"; bs->execute(pairs); } for (uint i = 0; i < pairs.size(); i++) { if (pairs[i] == NULL) continue; // // Calculate Fst P-value from approximate distribution. // // if (bootstrap_fst && bootstrap_type == bs_approx) // pairs[i]->bs[0] = bootstrap_approximate_pval(pairs[i]->snp_cnt, pairs[i]->stat[0], approx_fst_dist); cnt++; sum += pairs[i]->stat[1]; // Corrected AMOVA Fst fh << batch_id << "\t" << pairs[i]->loc_id << "\t" << pop_1p.name << "\t" << pop_2p.name << "\t" << chr << "\t" << pairs[i]->bp +1 << "\t" << pairs[i]->col << "\t" << pairs[i]->pi << "\t" << pairs[i]->fst << "\t" << std::setprecision(9) << pairs[i]->fet_p << "\t" << pairs[i]->fet_or << "\t" << pairs[i]->ci_low << "\t" << pairs[i]->ci_high << "\t" << pairs[i]->lod << "\t" << pairs[i]->stat[0] << "\t" << pairs[i]->smoothed[0] << "\t" << pairs[i]->amova_fst << "\t" << pairs[i]->stat[1] << "\t" << pairs[i]->smoothed[1] << "\t" << pairs[i]->bs[1] << "\t" << pairs[i]->snp_cnt; if (log_fst_comp) { fh << "\t" << pairs[i]->comp[0] << "\t" << pairs[i]->comp[1] << "\t" << pairs[i]->comp[2] << "\t" << pairs[i]->comp[3] << "\t" << pairs[i]->comp[4] << "\t" << pairs[i]->comp[5] << "\t" << pairs[i]->comp[6] << "\t" << pairs[i]->comp[7] << "\t" << pairs[i]->comp[8] << "\t" << pairs[i]->comp[9] << "\t" << pairs[i]->comp[10] << "\t" << pairs[i]->comp[11] << "\t" << pairs[i]->fst << "\t" << pairs[i]->comp[12] << "\t" << pairs[i]->comp[13] << "\t" << pairs[i]->comp[14] << "\t" << pairs[i]->comp[15] << "\t" << pairs[i]->comp[16] << "\t" << pairs[i]->comp[17] << "\t" << pairs[i]->amova_fst << "\n"; } else { fh << "\n"; } delete pairs[i]; } } cerr << "Pop 1: " << pop_1p.name << "; Pop 2: " << pop_2p.name << "; mean Fst: " << (sum / cnt) << "\n"; means.push_back(sum / cnt); cerr << "Pooled populations '" << pop_1p.name << "' and '" << pop_2p.name << "' contained: " << ord->incompatible_loci << " incompatible loci; " << ord->multiple_loci << " nucleotides covered by more than one RAD locus.\n"; fh.close(); if (bootstrap_fst) delete bs; } } // // Write out the mean Fst measure of each pair of populations. // string file = out_path + out_prefix + ".fst_summary.tsv"; ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening generic output file '" << file << "'\n"; exit(1); } // // Write out X-axis header. // for (auto& pop : mpopi.pops()) fh << "\t" << pop.name; fh << "\n"; uint n = 0; for (uint i = 0; i < mpopi.pops().size() - 1; i++) { fh << mpopi.pops()[i].name; for (uint k = 0; k <= i; k++) fh << "\t"; for (uint j = i + 1; j < mpopi.pops().size(); j++) { fh << "\t" << means[n]; n++; } fh << "\n"; } fh.close(); delete ord; if (kernel_smoothed && loci_ordered) { delete ks; } return 0; } int correct_fst_bonferroni_win(vector &pairs) { int limit = 3 * sigma; int limit_l, limit_u; double correction; uint cnt, pos_l, pos_u; pos_l = 0; pos_u = 0; for (uint pos_c = 0; pos_c < pairs.size(); pos_c++) { if (pairs[pos_c] == NULL) continue; limit_l = pairs[pos_c]->bp - limit > 0 ? pairs[pos_c]->bp - limit : 0; limit_u = pairs[pos_c]->bp + limit; while (pos_l < pairs.size()) { if (pairs[pos_l] == NULL) { pos_l++; } else { if (pairs[pos_l]->bp < limit_l) pos_l++; else break; } } while (pos_u < pairs.size()) { if (pairs[pos_u] == NULL) { pos_u++; } else { if (pairs[pos_u]->bp < limit_u) pos_u++; else break; } } cnt = 0; for (uint i = pos_l; i < pos_u; i++) { if (pairs[i] == NULL) continue; cnt++; } correction = p_value_cutoff / cnt; pairs[pos_c]->stat[0] = pairs[pos_c]->fet_p < correction ? pairs[pos_c]->fst : 0; } return 0; } int kernel_smoothed_popstats(map &catalog, PopMap *pmap, PopSum *psum, int pop_id, ofstream &log_fh) { // int snp_dist[max_snp_dist] = {0}; // int sites_per_snp = 0; // int tot_windows = 0; map >::iterator it; map > genome_sites; // // Instantiate the kernel smoothing object if requested. // KSmooth *ks = new KSmooth(2); OSumStat *ord = new OSumStat(psum, log_fh); Bootstrap *bs; if (bootstrap_pifis) bs = new Bootstrap(2); for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { vector &sites = genome_sites[it->first]; ord->order(sites, it->second, pop_id); if (bootstrap_pifis) bs->add_data(sites); } cerr << " Population '" << mpopi.pops()[pop_id].name << "' contained " << ord->multiple_loci << " nucleotides covered by more than one RAD locus.\n"; for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { if (bootstrap_pifis) cerr << " Smoothing and bootstrapping chromosome " << it->first << "\n"; else cerr << " Smoothing chromosome " << it->first << "\n"; vector &sites = genome_sites[it->first]; ks->smooth(sites); if (bootstrap_pifis && bootstrap_type == bs_exact) bs->execute_mixed(sites); } delete ks; delete ord; if (bootstrap_pifis) delete bs; // // // // If bootstrap resampling method is approximate, generate our single, empirical distribution. // // // map > approx_fis_dist; // map > approx_pi_dist; // if (bootstrap && bootstrap_type == bs_approx) { // sites_per_snp = sites_per_snp / tot_windows; // // cerr << "Sites per snp: " << sites_per_snp << "\n"; // bootstrap_popstats_approximate_dist(fis_samples, pi_samples, allele_depth_samples, // weights, snp_dist, sites_per_snp, // approx_fis_dist, approx_pi_dist); // for (it = pmap->ordered_loci.begin(); it != pmap->ordered_loci.end(); it++) { // for (uint pos = 0; pos < it->second.size(); pos++) { // loc = it->second[pos]; // len = strlen(loc->con); // lsum = psum->pop(loc->id, pop_id); // for (int k = 0; k < len; k++) // if (lsum->nucs[k].num_indv > 0 && bootstrap && lsum->nucs[k].pi > 0) { // // // // Calculate Fis/Pi p-values from approximate distribution. // // // lsum->nucs[k].wFis_pval = bootstrap_approximate_pval(lsum->nucs[k].snp_cnt, lsum->nucs[k].wFis, approx_fis_dist); // lsum->nucs[k].wPi_pval = bootstrap_approximate_pval(lsum->nucs[k].snp_cnt, lsum->nucs[k].wPi, approx_pi_dist); // } // } // } // } return 0; } int bootstrap_popstats_approximate_dist(vector &fis_samples, vector &pi_samples, vector &allele_samples, double *weights, int *snp_dist, int sites_per_snp, map > &approx_fis_dist, map > &approx_pi_dist) { // // Allocate an array of bootstrap resampling objects. // int win_size = 6 * sigma + 1; int win_cntr = win_size / 2; // // Initialize the Fst distribution map. // for (int i = 0; i < max_snp_dist; i++) { if (snp_dist[i] == 0.0) continue; // cerr << "SNP Dist: " << i << " snps occurred " << snp_dist[i] << "\n"; approx_fis_dist[i] = vector (); approx_fis_dist[i].reserve(bootstrap_reps); approx_pi_dist[i] = vector (); approx_pi_dist[i].reserve(bootstrap_reps); } vector poss; poss.reserve(max_snp_dist); double weighted_fis, weighted_pi, sum_fis, sum_pi, final_weight_fis, final_weight_pi; // int index_1, index_2; int pos, index_3, dist, start, end; int half = sites_per_snp / 2; for (int i = 0; i < max_snp_dist; i++) { if (snp_dist[i] == 0.0) continue; cerr << " Generating NULL distribution for " << i << " SNPs...\n"; // #pragma omp parallel private(poss, pos, index_1, index_2, index_3, dist, sum_fis, sum_pi, weighted_fis, weighted_pi, final_weight_fis, final_weight_pi) #pragma omp parallel private(poss, pos, index_3, dist, sum_fis, sum_pi, weighted_fis, weighted_pi, final_weight_fis, final_weight_pi) { BSample *bs = new BSample[win_size]; // // Populate the BSample objects. // for (int n = 0; n < win_size; n++) bs[n].bp = n + 1; vector fiss, pis; // // Bootstrap this bitch. // #pragma omp for schedule(dynamic, 1) for (int j = 0; j < bootstrap_reps; j++) { // cerr << " Bootsrap rep " << j << "\n"; // // First SNP is always placed at the center of the window. // pos = win_cntr; // index_1 = (int) (fis_samples.size() * (random() / (RAND_MAX + 1.0))); // index_2 = (int) (pi_samples.size() * (random() / (RAND_MAX + 1.0))); index_3 = (int) (allele_samples.size() * (random() / (RAND_MAX + 1.0))); // // Fill in the area around the SNP with fixed sites. // start = pos - half > 0 ? pos - half : 0; end = pos + half < win_size ? pos + half : win_size; for (int n = start; n < end; n++) { // bs[n].f = 0; // bs[n].pi = 0; bs[n].alleles = bs[pos].alleles; poss.push_back(n); } // bs[pos].f = fis_samples[index_1]; // bs[pos].pi = pi_samples[index_2]; bs[pos].alleles = allele_samples[index_3]; // cerr << " Placing SNP at position: " << pos << "; with data from " << index_1 << " filling area from " << start << " to " << end << "\n"; // // Randomly select the positions and values for each SNP to populate the window // for (int k = 0; k < i - 1; k++) { pos = (int) (win_size * (random() / (RAND_MAX + 1.0))); // index_1 = (int) (fis_samples.size() * (random() / (RAND_MAX + 1.0))); // index_2 = (int) (pi_samples.size() * (random() / (RAND_MAX + 1.0))); index_3 = (int) (allele_samples.size() * (random() / (RAND_MAX + 1.0))); poss.push_back(pos); // // Fill in the area around the SNP with fixed sites. // start = pos - half > 0 ? pos - half : 0; end = pos + half < win_size ? pos + half : win_size; for (int n = start; n < end; n++) { // bs[n].f = 0; // bs[n].pi = 0; bs[n].alleles = bs[pos].alleles; poss.push_back(n); } // bs[pos].f = fis_samples[index_1]; // bs[pos].pi = pi_samples[index_2]; bs[pos].alleles = allele_samples[index_3]; // cerr << " Placing SNP at position: " << pos << "; with data from " << index_1 << " filling area from " << start << " to " << end << "\n"; } weighted_fis = 0.0; sum_fis = 0.0; weighted_pi = 0.0; sum_pi = 0.0; for (int n = 0; n < win_size; n++) { // if (bs[n].pi < 0.0) // continue; // // Calculate weighted Fst at this position. // dist = bs[n].bp > bs[win_cntr].bp ? bs[n].bp - bs[win_cntr].bp : bs[win_cntr].bp - bs[n].bp; final_weight_fis = (bs[n].alleles - 1) * weights[dist]; // weighted_fis += bs[n].f * final_weight_fis; sum_fis += final_weight_fis; final_weight_pi = (bs[n].alleles - 1) * weights[dist]; // weighted_pi += bs[n].pi * final_weight_pi; sum_pi += final_weight_pi; } fiss.push_back(weighted_fis / sum_fis); pis.push_back(weighted_pi / sum_pi); // cerr << " New weighted fis value: " << weighted_fis / sum_fis << "; size: " << fiss.size() << "\n"; for (uint n = 0; n < poss.size(); n++) { // bs[poss[n]].f = 0.0; // bs[poss[n]].pi = -1.0; } poss.clear(); } // #pragma omp critical // { // vector &f = approx_fis_dist[i]; // for (uint n = 0; n < fiss.size(); n++) // f.push_back(fiss[n]); // vector &p = approx_pi_dist[i]; // for (uint n = 0; n < pis.size(); n++) // p.push_back(pis[n]); // } delete [] bs; } sort(approx_fis_dist[i].begin(), approx_fis_dist[i].end()); sort(approx_pi_dist[i].begin(), approx_pi_dist[i].end()); } return 0; } int bootstrap_fst_approximate_dist(vector &fst_samples, vector &allele_samples, double *weights, int *snp_dist, map > &approx_fst_dist) { // // Allocate an array of bootstrap resampling objects. // int win_size = 6 * sigma + 1; int win_cntr = win_size / 2; // // Initialize the Fst distribution map. // for (int i = 0; i < max_snp_dist; i++) { if (snp_dist[i] == 0.0) continue; // cerr << "SNP Dist: " << i << " snps occurred " << snp_dist[i] << "\n"; approx_fst_dist[i] = vector (); approx_fst_dist[i].reserve(bootstrap_reps); } vector poss; poss.reserve(max_snp_dist); double weighted_fst, sum, final_weight; //int index_1; int pos, index_2, dist; for (int i = 0; i < max_snp_dist; i++) { if (snp_dist[i] == 0.0) continue; cerr << " Generating NULL distribution for " << i << " SNPs...\n"; // #pragma omp parallel private(poss, pos, index_1, index_2, dist, sum, weighted_fst, final_weight) #pragma omp parallel private(poss, pos, index_2, dist, sum, weighted_fst, final_weight) { BSample *bs = new BSample[win_size]; // // Populate the BSample objects. // for (int n = 0; n < win_size; n++) bs[n].bp = n + 1; vector fsts; // // Bootstrap this bitch. // #pragma omp for schedule(dynamic, 1) for (int j = 0; j < bootstrap_reps; j++) { // cerr << "Bootsrap rep " << j << "\n"; // // First SNP is always placed at the center of the window. // pos = win_cntr; // index_1 = (int) (fst_samples.size() * (random() / (RAND_MAX + 1.0))); index_2 = (int) (allele_samples.size() * (random() / (RAND_MAX + 1.0))); // bs[pos].f = fst_samples[index_1]; bs[pos].alleles = allele_samples[index_2]; // // Randomly select the positions and values for each SNP to populate the window // for (int k = 0; k < i - 1; k++) { pos = (int) (win_size * (random() / (RAND_MAX + 1.0))); // index_1 = (int) (fst_samples.size() * (random() / (RAND_MAX + 1.0))); index_2 = (int) (allele_samples.size() * (random() / (RAND_MAX + 1.0))); // bs[pos].f = fst_samples[index_1]; // bs[pos].alleles = allele_samples[index_2]; // cerr << " " << j << ": Placing SNP at position: " << pos << " with data from index " << index_1 << "\n"; poss.push_back(pos); } weighted_fst = 0.0; sum = 0.0; for (int n = 0; n < win_size; n++) { // if (bs[n].f == 0.0) // continue; // // Calculate weighted Fst at this position. // dist = bs[n].bp > bs[win_cntr].bp ? bs[n].bp - bs[win_cntr].bp : bs[win_cntr].bp - bs[n].bp; final_weight = (bs[n].alleles - 1) * weights[dist]; // weighted_fst += bs[n].f * final_weight; sum += final_weight; } fsts.push_back(weighted_fst / sum); // cerr << " New weighted Fst value: " << weighted_fst / sum << "; size: " << fsts.size() << "\n"; // for (uint n = 0; n < poss.size(); n++) // bs[poss[n]].f = 0.0; poss.clear(); } // #pragma omp critical // { // vector &f = approx_fst_dist[i]; // for (uint n = 0; n < fsts.size(); n++) // f.push_back(fsts[n]); // } delete [] bs; } sort(approx_fst_dist[i].begin(), approx_fst_dist[i].end()); } return 0; } double bootstrap_approximate_pval(int snp_cnt, double stat, map > &approx_dist) { if (approx_dist.count(snp_cnt) == 0) return 1.0; vector::iterator up; vector &dist = approx_dist[snp_cnt]; double pos; up = upper_bound(dist.begin(), dist.end(), stat); if (up == dist.begin()) pos = 1; else if (up == dist.end()) pos = dist.size(); else pos = up - dist.begin() + 1; double res = 1.0 - (pos / (double) dist.size()); // cerr << "Generated Approx Smoothed Fst Distribution:\n"; // for (uint n = 0; n < dist.size(); n++) // cerr << " n: " << n << "; Fst: " << dist[n] << "\n"; // cerr << "Comparing Fst value: " << stat // << " at position " << (up - dist.begin()) << " out of " // << dist.size() << " positions (converted position: " << pos << "); pvalue: " << res << ".\n"; return res; } int write_generic(map &catalog, PopMap *pmap, bool write_gtypes) { string file = out_path + out_prefix + (write_gtypes ? ".genotypes.tsv" : ".haplotypes.tsv"); ofstream fh(file.c_str(), ofstream::out); if (fh.fail()) { cerr << "Error opening generic output file '" << file << "'\n"; exit(1); } // // Count the number of markers that have enough samples to output. // map::iterator it; CSLocus *loc; int num_loci = catalog.size(); cerr << "Writing " << num_loci << " loci to " << (write_gtypes ? "genotype" : "observed haplotype") << " file, '" << file << "'\n"; // // Write the header // fh << "Catalog ID\t"; if (expand_id) fh << "\t"; if (write_gtypes) fh << "Marker\t"; fh << "Cnt\t"; for (int i = 0; i < pmap->sample_cnt(); i++) { fh << mpopi.samples()[i].name; if (i < pmap->sample_cnt() - 1) fh << "\t"; } fh << "\n"; // // Output each locus. // for (it = catalog.begin(); it != catalog.end(); it++) { loc = it->second; stringstream id; loc->annotation.length() > 0 ? id << loc->id << "|" << loc->annotation : id << loc->id; fh << id.str(); if (expand_id) { if (loc->annotation.length() > 0) id << "\t" << loc->id << "\t" << loc->annotation; else if (strlen(loc->loc.chr) > 0) id << "\t" << loc->id << "\t" << loc->loc.chr << "_" << loc->loc.bp +1; else id << "\t" << loc->id << "\t"; } if (write_gtypes) fh << "\t" << loc->marker; write_gtypes ? fh << "\t" << loc->gcnt : fh << "\t" << loc->hcnt; Datum **d = pmap->locus(loc->id); string obshap; for (int i = 0; i < pmap->sample_cnt(); i++) { fh << "\t"; if (d[i] == NULL) fh << "-"; else { if (write_gtypes) { fh << d[i]->gtype; } else { obshap = ""; for (uint j = 0; j < d[i]->obshap.size(); j++) obshap += string(d[i]->obshap[j]) + "/"; obshap = obshap.substr(0, obshap.length()-1); fh << obshap; } } } fh << "\n"; } fh.close(); return 0; } int load_marker_list(string path, set &list) { char line[id_len]; ifstream fh(path.c_str(), ifstream::in); if (fh.fail()) { cerr << "Error opening white/black list file '" << path << "'\n"; exit(1); } size_t line_num = 0; while (fh.getline(line, id_len)) { ++line_num; // // Skip blank & commented lines ; correct windows-style line ends. // size_t len = strlen(line); if (len == 0) { continue; } else if (line[len-1] == '\r') { line[len-1] = '\0'; --len; if (len == 0) continue; } char* p = line; while (isspace(*p) && *p != '\0') ++p; if (*p == '#') continue; // // Parse the blacklist // char* e; int marker = (int) strtol(line, &e, 10); if (*e == '\0') { list.insert(marker); } else { cerr << "Error: Unable to parse blacklist '" << path << "' at line " << line_num << ".\n"; throw exception(); } } fh.close(); if (list.size() == 0) { cerr << "Unable to load any markers from '" << path << "'\n"; exit(1); } return 0; } int load_marker_column_list(string path, map > &list) { char line[id_len]; ifstream fh(path.c_str(), ifstream::in); if (fh.fail()) { cerr << "Error opening white/black list file '" << path << "'\n"; exit(1); } vector parts; uint col; char *e; uint line_num = 1; while (fh.getline(line, id_len)) { // // Skip blank & commented lines ; correct windows-style line ends. // size_t len = strlen(line); if (len == 0) { continue; } else if (line[len-1] == '\r') { line[len-1] = '\0'; --len; if (len == 0) continue; } char* p = line; while (isspace(*p) && *p != '\0') ++p; if (*p == '#') continue; // // Parse the whitelist, we expect: // [] // parse_tsv(line, parts); if (parts.size() > 2) { cerr << "Too many columns in whitelist " << path << "' at line " << line_num << "\n"; exit(1); } else if (parts.size() == 2) { int marker = (int) strtol(parts[0].c_str(), &e, 10); if (*e != '\0') { cerr << "Unable to parse whitelist, '" << path << "' at line " << line_num << "\n"; exit(1); } col = (int) strtol(parts[1].c_str(), &e, 10); if (*e != '\0') { cerr << "Unable to parse whitelist, '" << path << "' at line " << line_num << "\n"; exit(1); } list[marker].insert(col); } else { int marker = (int) strtol(parts[0].c_str(), &e, 10); if (*e != '\0') { cerr << "Unable to parse whitelist, '" << path << "' at line " << line_num << "\n"; exit(1); } list.insert(make_pair(marker, std::set())); } line_num++; } fh.close(); if (list.size() == 0) { cerr << "Unable to load any markers from '" << path << "'\n"; help(); } return 0; } bool hap_compare(pair a, pair b) { return (a.second > b.second); } int parse_command_line(int argc, char* argv[]) { while (1) { static struct option long_options[] = { {"help", no_argument, NULL, 'h'}, {"version", no_argument, NULL, 'v'}, {"verbose", no_argument, NULL, 'd'}, {"sql", no_argument, NULL, 's'}, {"vcf", no_argument, NULL, 1004}, {"vcf_haplotypes", no_argument, NULL, 'n'}, {"fasta", no_argument, NULL, 'F'}, {"fasta_strict", no_argument, NULL, 'J'}, {"structure", no_argument, NULL, 'S'}, {"fastphase", no_argument, NULL, 'A'}, {"phase", no_argument, NULL, 'C'}, {"beagle", no_argument, NULL, 'E'}, {"beagle_phased", no_argument, NULL, 'H'}, {"plink", no_argument, NULL, 'K'}, {"genomic", no_argument, NULL, 'g'}, {"genepop", no_argument, NULL, 'G'}, {"phylip", no_argument, NULL, 'Y'}, {"phylip_var", no_argument, NULL, 'L'}, {"phylip_var_all", no_argument, NULL, 'T'}, {"hzar", no_argument, NULL, 'Z'}, {"treemix", no_argument, NULL, 'U'}, {"merge_sites", no_argument, NULL, 'D'}, {"window_size", required_argument, NULL, 'w'}, {"threads", required_argument, NULL, 't'}, {"batch_id", required_argument, NULL, 'b'}, {"in_path", required_argument, NULL, 'P'}, {"out_path", required_argument, NULL, 'O'}, {"in_vcf", required_argument, NULL, 'V'}, {"progeny", required_argument, NULL, 'r'}, {"min_depth", required_argument, NULL, 'm'}, {"renz", required_argument, NULL, 'e'}, {"popmap", required_argument, NULL, 'M'}, {"whitelist", required_argument, NULL, 'W'}, {"blacklist", required_argument, NULL, 'B'}, {"write_single_snp", no_argument, NULL, 'I'}, {"write_random_snp", no_argument, NULL, 'j'}, {"ordered_export", no_argument, NULL, 1002}, {"kernel_smoothed", no_argument, NULL, 'k'}, {"fstats", no_argument, NULL, '6'}, {"log_fst_comp", no_argument, NULL, 'l'}, {"bootstrap_type", required_argument, NULL, 1001}, {"bootstrap_reps", required_argument, NULL, 1003}, {"bootstrap_wl", required_argument, NULL, 'Q'}, {"bootstrap", no_argument, NULL, '1'}, {"bootstrap_fst", no_argument, NULL, '2'}, {"bootstrap_phist", no_argument, NULL, '3'}, {"bootstrap_div", no_argument, NULL, '4'}, {"bootstrap_pifis", no_argument, NULL, '5'}, {"min_populations", required_argument, NULL, 'p'}, {"min_maf", required_argument, NULL, 'a'}, {"max_obs_het", required_argument, NULL, 'q'}, {"lnl_lim", required_argument, NULL, 'c'}, {"merge_prune_lim", required_argument, NULL, 'i'}, {"fst_correction", required_argument, NULL, 'f'}, {"p_value_cutoff", required_argument, NULL, 'u'}, {"debug_flags", required_argument, NULL, 1000}, {0, 0, 0, 0} }; // getopt_long stores the option index here. int c = getopt_long(argc, argv, "ACDEFGHJKLNSTUV:YZ123456dghjklnsva:b:c:e:f:i:m:o:p:q:r:t:u:w:B:I:M:O:P:R:Q:W:", long_options, NULL); // Detect the end of the options. if (c == -1) break; switch (c) { case 'h': help(); break; case 'd': verbose = true; break; case 't': num_threads = atoi(optarg); break; case 'P': in_path = optarg; if (!in_path.empty() && in_path.back() != '/') in_path += "/"; break; case 'O': out_path = optarg; if (!out_path.empty() && out_path.back() != '/') out_path += "/"; break; case 'V': in_vcf_path = optarg; break; case 'M': pmap_path = optarg; break; case 'D': merge_sites = true; break; case 'i': merge_prune_lim = is_double(optarg); if (merge_prune_lim > 1.0) merge_prune_lim = merge_prune_lim / 100; if (merge_prune_lim < 0 || merge_prune_lim > 1.0) { cerr << "Unable to parse the merge sites pruning limit.\n"; help(); } break; case 'q': max_obs_het = is_double(optarg); if (max_obs_het > 1) max_obs_het = max_obs_het / 100; if (max_obs_het < 0 || max_obs_het > 1.0) { cerr << "Unable to parse the maximum observed heterozygosity.\n"; help(); } break; case 'b': batch_id = is_integer(optarg); if (batch_id < 0) { cerr << "Batch ID (-b) must be an integer, e.g. 1, 2, 3\n"; help(); } break; case 'r': sample_limit = atof(optarg); if (sample_limit > 1) sample_limit = sample_limit / 100; if (sample_limit > 1.0) { cerr << "Unable to parse the sample limit frequency\n"; help(); } break; case 'p': population_limit = atoi(optarg); break; case 'k': kernel_smoothed = true; calc_fstats = true; break; case '6': calc_fstats = true; break; case 'l': log_fst_comp = true; break; case '1': bootstrap = true; bootstrap_fst = true; bootstrap_phist = true; bootstrap_pifis = true; bootstrap_div = true; break; case '2': bootstrap_fst = true; break; case '3': bootstrap_phist = true; break; case '4': bootstrap_div = true; break; case '5': bootstrap_pifis = true; break; case 1001: if (strcasecmp(optarg, "exact") == 0) bootstrap_type = bs_exact; else if (strcasecmp(optarg, "approx") == 0) bootstrap_type = bs_approx; else { cerr << "Unknown bootstrap type specified '" << optarg << "'\n"; help(); } break; case 1003: bootstrap_reps = atoi(optarg); break; case 'Q': bs_wl_file = optarg; bootstrap_wl = true; break; case 'c': lnl_limit = is_double(optarg); filter_lnl = true; break; case 'I': write_single_snp = true; break; case 'j': write_random_snp = true; break; case 1002: ordered_export = true; break; case 's': sql_out = true; break; case 1004: vcf_out = true; break; case 'n': vcf_haplo_out = true; break; case 'F': fasta_out = true; break; case 'J': fasta_strict_out = true; break; case 'G': genepop_out = true; break; case 'S': structure_out = true; break; case 'A': fastphase_out = true; break; case 'C': phase_out = true; break; case 'E': beagle_out = true; break; case 'H': beagle_phased_out = true; break; case 'K': plink_out = true; break; case 'Z': hzar_out = true; break; case 'Y': phylip_out = true; break; case 'L': phylip_var = true; break; case 'T': phylip_var_all = true; break; case 'U': treemix_out = true; break; case 'g': genomic_out = true; break; case 'W': wl_file = optarg; break; case 'B': bl_file = optarg; break; case 'm': min_stack_depth = atoi(optarg); break; case 'a': minor_allele_freq = atof(optarg); if (minor_allele_freq > 1) minor_allele_freq = minor_allele_freq / 100; if (minor_allele_freq < 0 || minor_allele_freq > 0.5) { cerr << "Unable to parse the minor allele frequency.\n"; help(); } break; case 'f': if (strcasecmp(optarg, "p_value") == 0) fst_correction = p_value; else if (strcasecmp(optarg, "bonferroni_win") == 0) fst_correction = bonferroni_win; else if (strcasecmp(optarg, "bonferroni_gen") == 0) fst_correction = bonferroni_gen; else { cerr << "Unknown Fst correction specified '" << optarg << "'\n"; help(); } break; case 'u': p_value_cutoff = atof(optarg); break; case 'e': enz = optarg; if (renz.count(enz) == 0) { cerr << "Unrecognized restriction enzyme specified: '" << enz.c_str() << "'.\n"; help(); } break; case 'w': sigma = atof(optarg); break; case 'v': version(); exit(0); break; case '?': // getopt_long already printed an error message. help(); break; case 1000: { static const set known_debug_flags = {"VCFCOMP"}; stringstream ss (optarg); string s; while (std::getline(ss, s, ',')) { if (known_debug_flags.count(s)) { debug_flags.insert(s); } else { cerr << "DEBUG> Error: Unknown error flag '" << s << "'.\n"; return -1; } } cerr << "DEBUG> Debug flag(s) : '" << optarg << "'.\n"; if (debug_flags.count("VCFCOMP") && not write_random_snp) { write_single_snp = true; cerr << "DEBUG> Added --write_single_snp.\n"; } break; } default: cerr << "Unknown command line option: '" << (char) c << "'\n"; help(); abort(); } } // // Check argument constrains. // if (not in_path.empty() && not in_vcf_path.empty()) { cerr << "Error: Please specify either '-P' or '--in_vcf', not both.\n"; help(); } else if (not in_path.empty()) { input_mode = InputMode::stacks; } else if (not in_vcf_path.empty()) { input_mode = InputMode::vcf; } else { cerr << "Error: One of '--in_path' or '--in_vcf' is required.\n"; help(); } if (input_mode == InputMode::stacks) { if (pmap_path.empty()) cerr << "A population map was not specified, all samples will be read from '" << in_path << "' as a single popultaion.\n"; if (batch_id < 0) { cerr << "You must specify a batch ID.\n"; help(); } if (out_path.empty()) out_path = in_path; out_prefix = string("batch_") + to_string(batch_id); } else if (input_mode == InputMode::vcf) { if (out_path.empty()) { cerr << "Error: Malformed arguments: input mode 'vcf' requires an output directory (--out_path).\n"; help(); } // Determine out_prefix string fname = in_vcf_path; if (in_vcf_path.find_last_of('/') != string::npos && in_vcf_path.back() != '/') fname = in_vcf_path.substr(in_vcf_path.find_last_of('/')+1); size_t trim = 0; if (fname.length() > 4 && fname.substr(fname.length()-4) == ".vcf") trim = 4; else if (fname.length() > 7 && fname.substr(fname.length()-7) == ".vcf.gz") trim = 7; out_prefix = fname.substr(0, fname.length()-trim); } // Other if (write_single_snp && write_random_snp) { cerr << "Please specify either '--write_single_snp' or '--write_random_snp', not both.\n"; help(); } if (merge_sites == true && enz.length() == 0) { cerr << "You must specify the restriction enzyme associated with this data set to merge overlaping cutsites.\n"; help(); } return 0; } void version() { cerr << "populations " << VERSION << "\n\n"; } void help() { cerr << "populations " << VERSION << "\n" << "Usage:\n" << "populations -P dir -b batch_id [-O dir] [-M popmap] (filters) [--fstats] [-k [--window_size=150000] [--bootstrap [-N 100]]] (output formats)\n" << "populations -V vcf -O dir [-M popmap] (filters) [--fstats] [-k [--sigma=150000] [--bootstrap [-N 100]]] (output formats)\n" << "\n" << " -P,--in_path: path to the directory containing the Stacks files.\n" << " -b,--batch_id: Batch ID to examine when exporting from the catalog (required by -P).\n" << " -V,--in_vcf: path to an input VCF file.\n" << " -O,--out_path: path to a directory where to write the output files. (Required by -V; otherwise defaults to value of -P.)\n" << " -M,--popmap: path to a population map. (Format is 'SAMPLE1\tPOP1\\n...'.)\n" << " -t,--threads: number of threads to run in parallel sections of code.\n" << " -s,--sql_out: output a file to import results into an SQL database.\n" << "\n" << "Data Filtering:\n" << " -p [int]: minimum number of populations a locus must be present in to process a locus.\n" << " -r [float]: minimum percentage of individuals in a population required to process a locus for that population.\n" << " --min_maf [float]: specify a minimum minor allele frequency required to process a nucleotide site at a locus (0 < min_maf < 0.5).\n" << " --max_obs_het [float]: specify a maximum observed heterozygosity required to process a nucleotide site at a locus.\n" << " -m [int]: specify a minimum stack depth required for individuals at a locus.\n" << " --lnl_lim [float]: filter loci with log likelihood values below this threshold.\n" << " --write_single_snp: restrict data analysis to only the first SNP per locus.\n" << " --write_random_snp: restrict data analysis to one random SNP per locus.\n" << " -B: path to a file containing Blacklisted markers to be excluded from the export.\n" << " -W: path to a file containing Whitelisted markers to include in the export.\n" << "\n" << "Merging and Phasing:\n" << " -e,--renz: restriction enzyme name.\n" << " --merge_sites: merge loci that were produced from the same restriction enzyme cutsite (requires reference-aligned data).\n" << " --merge_prune_lim: when merging adjacent loci, if at least X% samples posses both loci prune the remaining samples out of the analysis.\n" << "\n" << "Fstats:\n" << " --fstats: enable SNP and haplotype-based F statistics.\n" << " --fst_correction: specify a correction to be applied to Fst values: 'p_value', 'bonferroni_win', or 'bonferroni_gen'. Default: off.\n" << " --p_value_cutoff [float]: maximum p-value to keep an Fst measurement. Default: 0.05. (Also used as base for Bonferroni correction.)\n" << "\n" << "Kernel-smoothing algorithm:\n" << " -k,--kernel_smoothed: enable kernel-smoothed Pi, Fis, Fst, Fst', and Phi_st calculations.\n" << " --sigma [int]: standard deviation of the kernel smoothing weight distribution. Default 150kb.\n" << " --bootstrap: turn on boostrap resampling for all smoothed statistics.\n" << " -N,--bootstrap_reps [int]: number of bootstrap resamplings to calculate (default 100).\n" << " --bootstrap_pifis: turn on boostrap resampling for smoothed SNP-based Pi and Fis calculations.\n" << " --bootstrap_fst: turn on boostrap resampling for smoothed Fst calculations based on pairwise population comparison of SNPs.\n" << " --bootstrap_div: turn on boostrap resampling for smoothed haplotype diveristy and gene diversity calculations based on haplotypes.\n" << " --bootstrap_phist: turn on boostrap resampling for smoothed Phi_st calculations based on haplotypes.\n" << " --bootstrap_wl [path]: only bootstrap loci contained in this whitelist.\n" << "\n" << "File output options:\n" << " --ordered_export: if data is reference aligned, exports will be ordered; only a single representative of each overlapping site.\n" << " --genomic: output each nucleotide position (fixed or polymorphic) in all population members to a file (requires --renz).\n" << " --fasta: output full sequence for each unique haplotype, from each sample locus in FASTA format, regardless of plausibility.\n" << " --fasta_strict: output full sequence for each haplotype, from each sample locus in FASTA format, only for biologically plausible loci.\n" << " --vcf: output SNPs in Variant Call Format (VCF).\n" << " --vcf_haplotypes: output haplotypes in Variant Call Format (VCF).\n" << " --genepop: output results in GenePop format.\n" << " --structure: output results in Structure format.\n" << " --phase: output genotypes in PHASE format.\n" << " --fastphase: output genotypes in fastPHASE format.\n" << " --beagle: output genotypes in Beagle format.\n" << " --beagle_phased: output haplotypes in Beagle format.\n" << " --plink: output genotypes in PLINK format.\n" << " --hzar: output genotypes in Hybrid Zone Analysis using R (HZAR) format.\n" << " --phylip: output nucleotides that are fixed-within, and variant among populations in Phylip format for phylogenetic tree construction.\n" << " --phylip_var: include variable sites in the phylip output encoded using IUPAC notation.\n" << " --phylip_var_all: include all sequence as well as variable sites in the phylip output encoded using IUPAC notation.\n" << " --treemix: output SNPs in a format useable for the TreeMix program (Pickrell and Pritchard).\n" << "\n" << "Additional options:\n" << " -h,--help: display this help messsage.\n" << " -v,--version: print program version.\n" << " --verbose: turn on additional logging.\n" << (" --log_fst_comp: log components of Fst/Phi_st calculations to a file.\n"); // << " --bootstrap_type [exact|approx]: enable bootstrap resampling for population statistics (reference genome required).\n" exit(0); } stacks-1.44/src/populations.h000644 000765 000024 00000016043 12750704270 017027 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2012-2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __POPULATIONS_H__ #define __POPULATIONS_H__ #ifdef _OPENMP #include // OpenMP library #endif #include // Process command-line options #include // Open/Read contents of a directory #include #include #include #include #include using std::pair; using std::make_pair; #include using std::string; #include #include using std::ifstream; using std::ofstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include using std::setw; using std::setprecision; using std::fixed; #include using std::stringstream; #include using std::vector; #include using std::map; #include using std::set; #include "constants.h" #include "stacks.h" #include "locus.h" #include "renz.h" #include "PopMap.h" #include "PopSum.h" #include "utils.h" #include "log_utils.h" #include "catalog_utils.h" #include "sql_utilities.h" #include "genotype_dictionaries.h" #include "ordered.h" #include "smoothing.h" #include "bootstrap.h" enum corr_type {p_value, bonferroni_win, bonferroni_gen, no_correction}; enum bs_type {bs_exact, bs_approx, bs_none}; enum merget {merge_sink, merge_src}; enum phaset {merge_failure, simple_merge, complex_phase, nomapping_fail, multimapping_fail, multiple_fails}; enum class InputMode {stacks, vcf}; const int max_snp_dist = 500; void help( void ); void version( void ); int parse_command_line(int, char**); int build_file_list(); int load_marker_list(string, set &); int load_marker_column_list(string, map > &); int apply_locus_constraints(map &, PopMap *, ofstream &); int prune_polymorphic_sites(map &, PopMap *, PopSum *, map > &, set &, ofstream &); int log_haplotype_cnts(map &, ofstream &); bool order_unordered_loci(map &); int merge_shared_cutsite_loci(map &, PopMap *, PopSum *, map > &, ofstream &); phaset merge_and_phase_loci(PopMap *, CSLocus *, CSLocus *, set &, ofstream &); int merge_datums(int, int, Datum **, Datum **, set &, int); int merge_csloci(CSLocus *, CSLocus *, set &); int tabulate_haplotypes(map &, PopMap *); int create_genotype_map(CSLocus *, PopMap *); int call_population_genotypes(CSLocus *, PopMap *); int translate_genotypes(map &, map > &, map &, PopMap *, map &, set &); // This function doesn't exist (March 24, 2016) int correct_fst_bonferroni_win(vector &); int bootstrap_fst_approximate_dist(vector &, vector &, double *, int *, map > &); // not used (March 23, 2016) int kernel_smoothed_popstats(map &, PopMap *, PopSum *, int, ofstream &); int bootstrap_popstats_approximate_dist(vector &, vector &, vector &, double *, int *, int, map > &, map > &); // not used (March 23, 2016) double bootstrap_approximate_pval(int, double, map > &); int calculate_summary_stats(map &, PopMap *, PopSum *); int calculate_haplotype_stats(map &, PopMap *, PopSum *); int kernel_smoothed_hapstats(vector &, PopSum *, int, double *); int calculate_haplotype_divergence(map &, PopMap *, PopSum *); int calculate_haplotype_divergence_pairwise(map &, PopMap *, PopSum *); bool fixed_locus(Datum **, vector &); int nuc_substitution_dist(map &, double **); int nuc_substitution_identity(map &, double **); int nuc_substitution_identity_max(map &, double **); HapStat *haplotype_amova(Datum **, LocSum **, vector &); double amova_ssd_total(vector &, map &, double **); double amova_ssd_wp(vector &, map > &, map &, map > &, double **); double amova_ssd_ap_wg(vector &, map > &, map &, map > &, double **, double **); double amova_ssd_ag(vector &, map > &, map &, map > &, double **, double); double haplotype_d_est(Datum **, LocSum **, vector &); LocStat *haplotype_diversity(int, int, Datum **); double count_haplotypes_at_locus(int, int, Datum**, map&); //int tally_ref_alleles(LocSum **, int, int, char &, char &); //unused; also commented out in the .cc //int load_snp_calls(string, PopMap *); //no implementation //bool compare_pop_map(pair, pair); //no implementation; the function is in [sql_utilities.h] bool hap_compare(pair, pair); void vcfcomp_simplify_pmap (map& catalog, PopMap* pmap); inline bool uncalled_haplotype(const char *haplotype) { for (const char *p = haplotype; *p != '\0'; p++) if (*p == 'N' || *p == 'n') return true; return false; } inline double count_haplotypes_at_locus(int start, int end, Datum **d, map &hap_cnts) { double n = 0.0; for (int i = start; i <= end; i++) { if (d[i] == NULL) // No data, ignore this sample. continue; const vector& haps = d[i]->obshap; if (haps.size() > 2) { // Too many haplotypes, ignore this sample. continue; } else if (haps.size() == 1) { // Homozygote. if(!uncalled_haplotype(d[i]->obshap[0])) { n += 2; hap_cnts[d[i]->obshap[0]] += 2; } } else { // Heterozygote. for (uint j = 0; j < d[i]->obshap.size(); j++) { if(!uncalled_haplotype(d[i]->obshap[0])) { n++; hap_cnts[d[i]->obshap[j]]++; } } } } return n; } #endif // __POPULATIONS_H__ stacks-1.44/src/process_radtags.cc000644 000765 000024 00000133456 12774470747 020022 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2011-2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // process_radtags -- clean raw reads using a sliding window approach; // split reads by barcode, check RAD cutsite is intact, correct barcodes/cutsites // within one basepair, truncate reads on request. // #include #include "process_radtags.h" // // Global variables to hold command-line options. // FileT in_file_type = FileT::unknown; FileT out_file_type = FileT::unknown; string in_file; string in_file_p1; string in_file_p2; string in_path_1; string in_path_2; string out_path; string barcode_file; string renz_1; string renz_2; char *adapter_1; char *adapter_2; barcodet barcode_type = null_null; bool retain_header = false; bool filter_adapter = false; bool paired = false; bool clean = false; bool quality = false; bool recover = false; bool interleaved = false; bool merge = false; bool discards = false; bool overhang = false; bool filter_illumina = false; bool check_radtag = true; uint truncate_seq = 0; int barcode_dist_1 = 1; int barcode_dist_2 = -1; double win_size = 0.15; uint score_limit = 10; uint len_limit = 0; uint num_threads = 1; // // How to shift FASTQ-encoded quality scores from ASCII down to raw scores // score = encoded letter - 64; Illumina version 1.3 - 1.5 // score = encoded letter - 33; Sanger / Illumina version 1.6+ int qual_offset = 33; // // Handle variable-size barcodes. // uint min_bc_size_1 = 0; uint max_bc_size_1 = 0; uint min_bc_size_2 = 0; uint max_bc_size_2 = 0; // // Kmer data for adapter filtering. // int kmer_size = 5; int distance = 1; int adp_1_len = 0; int adp_2_len = 0; AdapterHash adp_1_kmers, adp_2_kmers; map renz; map renz_cnt; map renz_len; int main (int argc, char* argv[]) { initialize_renz(renz, renz_cnt, renz_len); parse_command_line(argc, argv); // // If input files are gzipped, output gziped files, unless the user chooses an output type. // if (out_file_type == FileT::unknown) { if (in_file_type == FileT::gzfastq || in_file_type == FileT::bam) out_file_type = FileT::gzfastq; else out_file_type = FileT::fastq; } if (paired) cerr << "Processing paired-end data.\n"; else cerr << "Processing single-end data.\n"; cerr << "Using Phred+" << qual_offset << " encoding for quality scores.\n"; if (truncate_seq > 0) cerr << "Reads will be truncated to " << truncate_seq << "bp\n"; if (filter_illumina) cerr << "Discarding reads marked as 'failed' by Illumina's chastity/purity filters.\n"; if (filter_adapter) { cerr << "Filtering reads for adapter sequence:\n"; if (adapter_1 != NULL) { cerr << " " << adapter_1 << "\n"; init_adapter_seq(kmer_size, adapter_1, adp_1_len, adp_1_kmers); } if (adapter_2 != NULL) { cerr << " " << adapter_2 << "\n"; init_adapter_seq(kmer_size, adapter_2, adp_2_len, adp_2_kmers); } cerr << " " << distance << " mismatches allowed to adapter sequence.\n"; } vector > files; vector barcodes; set se_bc, pe_bc; map pair_1_fhs, pair_2_fhs, rem_1_fhs, rem_2_fhs; map pair_1_gzfhs, pair_2_gzfhs, rem_1_gzfhs, rem_2_gzfhs; map > counters; map > barcode_log; build_file_list(files); load_barcodes(barcode_file, barcodes, se_bc, pe_bc, min_bc_size_1, max_bc_size_1, min_bc_size_2, max_bc_size_2); if (recover && barcode_type != null_null) { if (barcode_type == index_null || barcode_type == inline_null) cerr << "Will attempt to recover barcodes with at most " << barcode_dist_1 << " mismatches.\n"; else cerr << "Will attempt to recover barcodes with at most " << barcode_dist_1 << " / " << barcode_dist_2 << " mismatches.\n"; } if (out_file_type == FileT::gzfastq || out_file_type == FileT::gzfasta) open_files(files, barcodes, pair_1_gzfhs, pair_2_gzfhs, rem_1_gzfhs, rem_2_gzfhs, counters); else open_files(files, barcodes, pair_1_fhs, pair_2_fhs, rem_1_fhs, rem_2_fhs, counters); int result = 1; for (uint i = 0; i < files.size(); i++) { cerr << "Processing file " << i+1 << " of " << files.size() << " [" << files[i].first.c_str() << "]\n"; counters[files[i].first]["total"] = 0; counters[files[i].first]["ill_filtered"] = 0; counters[files[i].first]["adapter"] = 0; counters[files[i].first]["low_quality"] = 0; counters[files[i].first]["noradtag"] = 0; counters[files[i].first]["ambiguous"] = 0; counters[files[i].first]["retained"] = 0; counters[files[i].first]["recovered"] = 0; if (paired) { if (out_file_type == FileT::gzfastq || out_file_type == FileT::gzfasta) result = process_paired_reads(files[i].first, files[i].second, se_bc, pe_bc, pair_1_gzfhs, pair_2_gzfhs, rem_1_gzfhs, rem_2_gzfhs, counters[files[i].first], barcode_log); else result = process_paired_reads(files[i].first, files[i].second, se_bc, pe_bc, pair_1_fhs, pair_2_fhs, rem_1_fhs, rem_2_fhs, counters[files[i].first], barcode_log); } else { if (out_file_type == FileT::gzfastq || out_file_type == FileT::gzfasta) result = process_reads(files[i].first, se_bc, pe_bc, pair_1_gzfhs, counters[files[i].first], barcode_log); else result = process_reads(files[i].first, se_bc, pe_bc, pair_1_fhs, counters[files[i].first], barcode_log); } cerr << " " << counters[files[i].first]["total"] << " total reads; "; if (filter_illumina) cerr << "-" << counters[files[i].first]["ill_filtered"] << " failed Illumina reads; "; cerr << "-" << counters[files[i].first]["ambiguous"] << " ambiguous barcodes; " << "-" << counters[files[i].first]["noradtag"] << " ambiguous RAD-Tags; " << "+" << counters[files[i].first]["recovered"] << " recovered; " << "-" << counters[files[i].first]["low_quality"] << " low quality reads; " << counters[files[i].first]["retained"] << " retained reads.\n"; if (filter_adapter) cerr << " " << counters[files[i].first]["adapter"] << " reads with adapter sequence.\n"; if (!result) { cerr << "Error processing reads.\n"; break; } } cerr << "Closing files, flushing buffers...\n"; if (out_file_type == FileT::gzfastq || out_file_type == FileT::gzfasta) { close_file_handles(pair_1_gzfhs); if (paired) { close_file_handles(rem_1_gzfhs); close_file_handles(rem_2_gzfhs); close_file_handles(pair_2_gzfhs); } } else { close_file_handles(pair_1_fhs); if (paired) { close_file_handles(rem_1_fhs); close_file_handles(rem_2_fhs); close_file_handles(pair_2_fhs); } } print_results(argc, argv, barcodes, counters, barcode_log); return 0; } template int process_paired_reads(string prefix_1, string prefix_2, set &se_bc, set &pe_bc, map &pair_1_fhs, map &pair_2_fhs, map &rem_1_fhs, map &rem_2_fhs, map &counter, map > &barcode_log) { Input *fh_1, *fh_2; Read *r_1, *r_2; ofstream *discard_fh_1, *discard_fh_2; int return_val = 1; string path_1 = in_path_1 + prefix_1; string path_2 = in_path_2 + prefix_2; if (interleaved) cerr << " Reading data from:\n " << path_1 << "\n"; else cerr << " Reading data from:\n " << path_1 << " and\n " << path_2 << "\n"; if (in_file_type == FileT::fastq) { fh_1 = new Fastq(path_1.c_str()); fh_2 = interleaved ? fh_1 : new Fastq(path_2.c_str()); } else if (in_file_type == FileT::gzfastq) { fh_1 = new GzFastq(path_1.c_str()); fh_2 = interleaved ? fh_1 : new GzFastq(path_2.c_str()); } else if (in_file_type == FileT::bam) { fh_1 = new BamUnAln(path_1.c_str()); fh_2 = fh_1; } else if (in_file_type == FileT::bustard) { fh_1 = new Bustard(path_1.c_str()); fh_2 = interleaved ? fh_1 : new Bustard(path_2.c_str()); } // // Open a file for recording discarded reads // if (discards) { path_1 = out_path + prefix_1 + ".discards"; discard_fh_1 = new ofstream(path_1.c_str(), ifstream::out); if (discard_fh_1->fail()) { cerr << "Error opening discard output file '" << path_1 << "'\n"; exit(1); } path_2 = out_path + prefix_2 + ".discards"; discard_fh_2 = new ofstream(path_2.c_str(), ifstream::out); if (discard_fh_1->fail()) { cerr << "Error opening discard output file '" << path_2 << "'\n"; exit(1); } } // // Read in the first record, initializing the Seq object s. Then // initialize the Read object r, then loop, using the same objects. // Seq *s_1 = fh_1->next_seq(); Seq *s_2 = fh_2->next_seq(); if (s_1 == NULL || s_2 == NULL) { cerr << "Attempting to read first pair of input records, unable to allocate " << "Seq object (Was the correct input type specified?).\n"; exit(1); } r_1 = new Read(strlen(s_1->seq), 1, min_bc_size_1, win_size); r_2 = new Read(strlen(s_2->seq), 2, min_bc_size_2, win_size); // // Set len_limit so that if we encounter reads already shorter than truncate_seq limit // they will be discarded. // if (truncate_seq > 0) len_limit = truncate_seq; BarcodePair bc; // // If no barcodes were specified, set the barcode object to be the input file names. // if (max_bc_size_1 == 0) bc.set(prefix_1, prefix_2); long i = 1; cerr << " Processing RAD-Tags..."; do { if (i % 1000000 == 0) cerr << i/1000000 << "M..."; parse_input_record(s_1, r_1); parse_input_record(s_2, r_2); counter["total"] += 2; if (barcode_type != null_null && barcode_type != inline_null && barcode_type != index_null) bc.set(r_1->se_bc, r_2->pe_bc); else if (barcode_type != null_null) bc.set(r_1->se_bc); process_barcode(r_1, r_2, bc, pair_1_fhs, se_bc, pe_bc, barcode_log, counter); // // Adjust the size of the read to accommodate truncating the sequence and variable // barcode lengths. With standard Illumina data we want to output constant length // reads even as the barcode size may change. Other technologies, like IonTorrent // need to be truncated uniformly. // if (truncate_seq > 0) { if (truncate_seq + r_1->inline_bc_len <= r_1->len) r_1->set_len(truncate_seq + r_1->inline_bc_len); if (truncate_seq + r_2->inline_bc_len <= r_2->len) r_2->set_len(truncate_seq + r_2->inline_bc_len); } else { if (barcode_type == inline_null || barcode_type == inline_inline || barcode_type == inline_index) r_1->set_len(r_1->len - (max_bc_size_1 - r_1->inline_bc_len)); if (barcode_type == index_inline || barcode_type == inline_inline) r_2->set_len(r_2->len - (max_bc_size_2 - r_2->inline_bc_len)); } if (r_1->retain) process_singlet(r_1, renz_1, false, barcode_log[bc], counter); if (r_2->retain) process_singlet(r_2, renz_2, true, barcode_log[bc], counter); int result_1 = 1; int result_2 = 1; if (r_1->retain && r_2->retain) { if (retain_header) { result_1 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(pair_1_fhs[bc], s_1, r_1) : write_fasta(pair_1_fhs[bc], s_1, r_1); result_2 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(pair_2_fhs[bc], s_2, r_2) : write_fasta(pair_2_fhs[bc], s_2, r_2); } else { result_1 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(pair_1_fhs[bc], r_1, overhang) : write_fasta(pair_1_fhs[bc], r_1, overhang); result_2 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(pair_2_fhs[bc], r_2, overhang) : write_fasta(pair_2_fhs[bc], r_2, overhang); } } else if (r_1->retain && !r_2->retain) { // // Write to the remainder file. // if (retain_header) result_1 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(rem_1_fhs[bc], s_1, r_1) : write_fasta(rem_1_fhs[bc], s_1, r_1); else result_1 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(rem_1_fhs[bc], r_1, overhang) : write_fasta(rem_1_fhs[bc], r_1, overhang); } else if (!r_1->retain && r_2->retain) { // // Write to the remainder file. // if (retain_header) result_2 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(rem_2_fhs[bc], s_2, r_2) : write_fasta(rem_2_fhs[bc], s_2, r_2); else result_2 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(rem_2_fhs[bc], r_2, overhang) : write_fasta(rem_2_fhs[bc], r_2, overhang); } if (!result_1 || !result_2) { cerr << "Error writing to output file for '" << bc.str() << "'\n"; return_val = -1; break; } if (discards && !r_1->retain) result_1 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(discard_fh_1, s_1) : write_fasta(discard_fh_1, s_1); if (discards && !r_2->retain) result_2 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(discard_fh_2, s_2) : write_fasta(discard_fh_2, s_2); if (!result_1 || !result_2) { cerr << "Error writing to discard file for '" << bc.str() << "'\n"; return_val = -1; break; } delete s_1; delete s_2; i++; } while ((s_1 = fh_1->next_seq()) != NULL && (s_2 = fh_2->next_seq()) != NULL); cerr << "\n"; if (discards) { delete discard_fh_1; delete discard_fh_2; } delete fh_1; if (interleaved == false) delete fh_2; delete r_1; delete r_2; return return_val; } template int process_reads(string prefix, set &se_bc, set &pe_bc, map &pair_1_fhs, map &counter, map > &barcode_log) { Input *fh; Read *r; ofstream *discard_fh; int return_val = 1; string path = in_path_1 + prefix; if (in_file_type == FileT::fastq) fh = new Fastq(path.c_str()); else if (in_file_type == FileT::gzfastq) fh = new GzFastq(path.c_str()); else if (in_file_type == FileT::bam) fh = new BamUnAln(path.c_str()); else if (in_file_type == FileT::bustard) fh = new Bustard(path.c_str()); // // Open a file for recording discarded reads // if (discards) { path = out_path + prefix + ".discards"; discard_fh = new ofstream(path.c_str(), ifstream::out); if (discard_fh->fail()) { cerr << "Error opening discard output file '" << path << "'\n"; exit(1); } } // // Read in the first record, initializing the Seq object s. Then // initialize the Read object r, then loop, using the same objects. // Seq *s = fh->next_seq(); if (s == NULL) { cerr << "Attempting to read first input record, unable to allocate " << "Seq object (Was the correct input type specified?).\n"; exit(1); } // // Set len_limit so that if we encounter reads already shorter than truncate_seq limit // they will be discarded. // if (truncate_seq > 0) len_limit = truncate_seq; r = new Read(strlen(s->seq), 1, min_bc_size_1, win_size); BarcodePair bc; // // If no barcodes were specified, set the barcode object to be the input file name so // that reads are written to an output file of the same name as the input file. // if (max_bc_size_1 == 0) bc.set(prefix); //cerr << "Length: " << r->len << "; Window length: " << r->win_len << "; Stop position: " << r->stop_pos << "\n"; long i = 1; cerr << " Processing RAD-Tags..."; do { if (i % 1000000 == 0) cerr << i/1000000 << "M..."; counter["total"]++; parse_input_record(s, r); if (barcode_type == inline_null || barcode_type == index_null) bc.set(r->se_bc); else if (barcode_type == index_inline || barcode_type == inline_index) bc.set(r->se_bc, r->pe_bc); process_barcode(r, NULL, bc, pair_1_fhs, se_bc, pe_bc, barcode_log, counter); // // Adjust the size of the read to accommodate truncating the sequence and variable // barcode lengths. With standard Illumina data we want to output constant length // reads even as the barcode size may change. Other technologies, like IonTorrent // need to be truncated uniformly. // if (truncate_seq > 0) { if (truncate_seq + r->inline_bc_len <= r->len) r->set_len(truncate_seq + r->inline_bc_len); } else { if (barcode_type == inline_null || barcode_type == inline_inline || barcode_type == inline_index) r->set_len(r->len - (max_bc_size_1 - r->inline_bc_len)); } if (r->retain) process_singlet(r, renz_1, false, barcode_log[bc], counter); int result = 1; if (r->retain) { if (retain_header) result = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(pair_1_fhs[bc], s, r) : write_fasta(pair_1_fhs[bc], s, r); else result = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(pair_1_fhs[bc], r, overhang) : write_fasta(pair_1_fhs[bc], r, overhang); } if (!result) { cerr << "Error writing to output file for '" << bc.str() << "'\n"; return_val = -1; break; } if (discards && !r->retain) result = out_file_type == FileT::fastq ? write_fastq(discard_fh, s) : write_fasta(discard_fh, s); if (!result) { cerr << "Error writing to discard file for '" << bc.str() << "'\n"; return_val = -1; break; } delete s; i++; } while ((s = fh->next_seq()) != NULL); cerr << "\n"; if (discards) delete discard_fh; // // Close the file and delete the Input object. // delete fh; delete r; return return_val; } inline int process_singlet(Read *href, string res_enz, bool paired_end, map &bc_log, map &counter) { char *p; if (filter_illumina && href->filter) { counter["ill_filtered"]++; href->retain = 0; return 0; } // // If this read is already shorter than our length limit, discard it. // if (len_limit > 0 && (href->len - href->inline_bc_len) < len_limit) { counter["low_quality"]++; if (barcode_type != null_null) bc_log["low_qual"]++; href->retain = 0; return 0; } // // Is the RADTAG intact? // if (check_radtag && res_enz.length() > 0) { bool rad_cor = false; for (int i = 0; i < renz_cnt[res_enz]; i++) { p = href->seq + href->inline_bc_len; if (strncmp(p, renz[res_enz][i], renz_len[res_enz]) == 0) rad_cor = true; } if (rad_cor == false) { // // Try to correct the RAD-Tag. // if (!correct_radtag(href, res_enz, counter)) { if (barcode_type != null_null) bc_log["noradtag"]++; counter["noradtag"]++; href->retain = 0; return 0; } } } // // Drop this sequence if it has any uncalled nucleotides. // if (clean) { for (char *p = href->seq + href->inline_bc_len; *p != '\0'; p++) if (*p == '.' || *p == 'N') { counter["low_quality"]++; href->retain = 0; if (barcode_type != null_null) bc_log["low_qual"]++; return 0; } } // // Drop this sequence if it has low quality scores. // if (quality && check_quality_scores(href, qual_offset, score_limit, len_limit, href->inline_bc_len) <= 0) { counter["low_quality"]++; if (barcode_type != null_null) bc_log["low_qual"]++; href->retain = 0; return 0; } // // Drop this sequence if it contains adapter sequence. // if (filter_adapter) { int res = 1; if (paired_end == true && adp_2_len > 0) res = filter_adapter_seq(href, adapter_2, adp_2_len, adp_2_kmers, kmer_size, distance, len_limit); if (paired_end == false && adp_1_len > 0) res = filter_adapter_seq(href, adapter_1, adp_1_len, adp_1_kmers, kmer_size, distance, len_limit); if (res <= 0) { // cerr << "Sequence " << href->seq << " contains adapter.\n"; counter["adapter"]++; href->retain = 0; return 0; } } if (barcode_type != null_null) bc_log["retained"]++; counter["retained"]++; return 0; } int correct_radtag(Read *href, string res_enz, map &counter) { if (recover == false) return 0; // // If the RAD-Tag sequence is off by no more than a single nucleotide, correct it. // int d = 0; for (int i = 0; i < renz_cnt[res_enz]; i++) { d = dist(renz[res_enz][i], href->seq + href->inline_bc_len); if (d <= 1) { // // Correct the read. // strncpy(href->seq + href->inline_bc_len, renz[res_enz][i], renz_len[res_enz]); counter["recovered"]++; return 1; } } return 0; } int dist(const char *res_enz, char *seq) { const char *p; char *q; int dist = 0; for (p = res_enz, q = seq; *p != '\0'; p++, q++) if (*p != *q) dist++; return dist; } int print_results(int argc, char **argv, vector &barcodes, map > &counters, map > &barcode_log) { map >::iterator it; string log_path = out_path + "process_radtags.log"; ofstream log(log_path.c_str()); if (log.fail()) { cerr << "Unable to open log file '" << log_path << "'\n"; return 0; } cerr << "Outputing details to log: '" << log_path << "'\n\n"; init_log(log, argc, argv); log << "File\t" << "Retained Reads\t"; if (filter_illumina) log << "Illumina Filtered\t"; if (filter_adapter) log << "Adapter Seq" << "\t"; log << "Low Quality\t" << "Ambiguous Barcodes\t" << "Ambiguous RAD-Tag\t" << "Total\n"; for (it = counters.begin(); it != counters.end(); it++) { log << it->first << "\t" << it->second["retained"] << "\t"; if (filter_illumina) log << it->second["ill_filtered"] << "\t"; if (filter_adapter) log << it->second["adapter"] << "\t"; log << it->second["low_quality"] << "\t" << it->second["ambiguous"] << "\t" << it->second["noradtag"] << "\t" << it->second["total"] << "\n"; } map c; c["total"] = 0; c["low_quality"] = 0; c["adapter"] = 0; c["ill_filtered"] = 0; c["ambiguous"] = 0; c["noradtag"] = 0; // // Total up the individual counters // for (it = counters.begin(); it != counters.end(); it++) { c["total"] += it->second["total"]; c["ill_filtered"] += it->second["ill_filtered"]; c["low_quality"] += it->second["low_quality"]; c["adapter"] += it->second["adapter"]; c["ambiguous"] += it->second["ambiguous"]; c["noradtag"] += it->second["noradtag"]; c["retained"] += it->second["retained"]; } std::ostream cerr_bis (cerr.rdbuf()); cerr_bis << std::fixed << std::setprecision(1); auto print_nreads = [&cerr_bis,&c] (long n, const string& legend) { size_t nspaces = std::to_string(c["total"]).length() - std::to_string(n).length(); cerr_bis << string(nspaces, ' ') << n << " " << legend << " (" << (double) n / c["total"] * 100 << "%)\n"; }; cerr_bis << c["total"] << " total sequences\n"; if (filter_illumina) print_nreads(c["ill_filtered"], "failed Illumina filtered reads"); if (filter_adapter) print_nreads(c["adapter"], "reads contained adapter sequence"); print_nreads(c["ambiguous"], "ambiguous barcode drops"); print_nreads(c["low_quality"], "low quality read drops"); print_nreads(c["noradtag"], "ambiguous RAD-Tag drops"); print_nreads(c["retained"], "retained reads"); log << "\n" << "Total Sequences\t" << c["total"] << "\n"; if (filter_illumina) log << "Failed Illumina filtered reads\t" << c["ill_filtered"] << "\n"; if (filter_adapter) log << "Reads containing adapter sequence\t" << c["adapter"] << "\n"; log << "Ambiguous Barcodes\t" << c["ambiguous"] << "\n" << "Low Quality\t" << c["low_quality"] << "\n" << "Ambiguous RAD-Tag\t" << c["noradtag"] << "\n" << "Retained Reads\t" << c["retained"] << "\n"; if (max_bc_size_1 == 0) return 0; // // Where barcode filenames specified? // bool bc_names = false; for (uint i = 0; i < barcodes.size(); i++) if (barcodes[i].name_exists()) { bc_names = true; break; } // // Print out barcode information. // log << "\n" << "Barcode\t"; if (bc_names) log << "Filename\t"; log << "Total\t" << "No RadTag\t" << "Low Quality\t" << "Retained\n"; set barcode_list; for (uint i = 0; i < barcodes.size(); i++) { barcode_list.insert(barcodes[i]); log << barcodes[i] << "\t"; if (bc_names) log << barcodes[i].name << "\t"; if (barcode_log.count(barcodes[i]) == 0) log << "0\t" << "0\t" << "0\t" << "0\n"; else log << barcode_log[barcodes[i]]["total"] << "\t" << barcode_log[barcodes[i]]["noradtag"] << "\t" << barcode_log[barcodes[i]]["low_qual"] << "\t" << barcode_log[barcodes[i]]["retained"] << "\n"; } log << "\n" << "Sequences not recorded\n" << "Barcode\t" << "Total\n"; // // Sort unused barcodes by number of occurances. // map >::iterator bit; vector > bcs; for (bit = barcode_log.begin(); bit != barcode_log.end(); bit++) bcs.push_back(make_pair(bit->first, bit->second["total"])); sort(bcs.begin(), bcs.end(), compare_barcodes); for (uint i = 0; i < bcs.size(); i++) { if (barcode_list.count(bcs[i].first)) continue; if (bcs[i].second == 0) continue; log << bcs[i].first << "\t" << bcs[i].second << "\n"; } log.close(); return 0; } int compare_barcodes(pair a, pair b) { return a.second > b.second; } int parse_command_line(int argc, char* argv[]) { FileT ftype; int c; while (1) { static struct option long_options[] = { {"help", no_argument, NULL, 'h'}, {"version", no_argument, NULL, 'v'}, {"quality", no_argument, NULL, 'q'}, {"clean", no_argument, NULL, 'c'}, {"recover", no_argument, NULL, 'r'}, {"discards", no_argument, NULL, 'D'}, {"paired", no_argument, NULL, 'P'}, {"interleaved", no_argument, NULL, 'I'}, {"merge", no_argument, NULL, 'm'}, {"disable_rad_check", no_argument, NULL, 'R'}, {"filter_illumina", no_argument, NULL, 'F'}, {"retain_header", no_argument, NULL, 'H'}, {"null_index", no_argument, NULL, 'U'}, {"index_null", no_argument, NULL, 'u'}, {"inline_null", no_argument, NULL, 'V'}, {"index_index", no_argument, NULL, 'W'}, {"inline_inline", no_argument, NULL, 'x'}, {"index_inline", no_argument, NULL, 'Y'}, {"inline_index", no_argument, NULL, 'Z'}, {"barcode_dist_1", required_argument, NULL, 'B'}, {"barcode_dist_2", required_argument, NULL, 'C'}, {"infile_type", required_argument, NULL, 'i'}, {"outfile_type", required_argument, NULL, 'y'}, {"file", required_argument, NULL, 'f'}, {"file_p1", required_argument, NULL, '1'}, {"file_p2", required_argument, NULL, '2'}, {"path", required_argument, NULL, 'p'}, {"outpath", required_argument, NULL, 'o'}, {"truncate", required_argument, NULL, 't'}, {"renz_1", required_argument, NULL, 'e'}, {"renz_2", required_argument, NULL, 'z'}, {"barcodes", required_argument, NULL, 'b'}, {"window_size", required_argument, NULL, 'w'}, {"score_limit", required_argument, NULL, 's'}, {"encoding", required_argument, NULL, 'E'}, {"len_limit", required_argument, NULL, 'L'}, {"adapter_1", required_argument, NULL, 'A'}, {"adapter_2", required_argument, NULL, 'G'}, {"adapter_mm", required_argument, NULL, 'T'}, {0, 0, 0, 0} }; // getopt_long stores the option index here. int option_index = 0; c = getopt_long(argc, argv, "HuUVWxYZhvRFIcqrDPmB:C:i:y:f:o:t:e:z:b:1:2:p:s:w:E:L:A:G:T:", long_options, &option_index); // Detect the end of the options. if (c == -1) break; switch (c) { case 'h': help(); break; case 'i': if (strcasecmp(optarg, "bustard") == 0) in_file_type = FileT::bustard; else if (strcasecmp(optarg, "bam") == 0) in_file_type = FileT::bam; else if (strcasecmp(optarg, "gzfastq") == 0) in_file_type = FileT::gzfastq; else in_file_type = FileT::fastq; break; case 'y': if (strcasecmp(optarg, "fastq") == 0) out_file_type = FileT::fastq; else if (strcasecmp(optarg, "gzfastq") == 0) out_file_type = FileT::gzfastq; else if (strcasecmp(optarg, "fasta") == 0) out_file_type = FileT::fasta; else if (strcasecmp(optarg, "gzfasta") == 0) out_file_type = FileT::gzfasta; break; case 'E': if (strcasecmp(optarg, "phred64") == 0) qual_offset = 64; else if (strcasecmp(optarg, "phred33") == 0) qual_offset = 33; else { cerr << "Unknown quality score encoding, '" << optarg << "'\n"; help(); } break; case 'f': in_file = optarg; ftype = FileT::fastq; break; case 'p': in_path_1 = optarg; in_path_2 = in_path_1; ftype = FileT::fastq; break; case '1': paired = true; in_file_p1 = optarg; ftype = FileT::fastq; break; case '2': paired = true; in_file_p2 = optarg; ftype = FileT::fastq; break; case 'P': paired = true; break; case 'I': interleaved = true; break; case 'B': barcode_dist_1 = is_integer(optarg); break; case 'C': barcode_dist_2 = is_integer(optarg); break; case 'o': out_path = optarg; break; case 'q': quality = true; break; case 'c': clean = true; break; case 'r': recover = true; break; case 't': truncate_seq = is_integer(optarg); break; case 'e': renz_1 = optarg; break; case 'z': renz_2 = optarg; break; case 'b': barcode_file = optarg; if (barcode_type == null_null) barcode_type = inline_null; break; case 'm': merge = true; break; case 'D': discards = true; break; case 'R': check_radtag = false; break; case 'F': filter_illumina = true; break; case 'U': barcode_type = null_index; break; case 'u': barcode_type = index_null; break; case 'V': barcode_type = inline_null; break; case 'W': barcode_type = index_index; break; case 'x': barcode_type = inline_inline; break; case 'Y': barcode_type = index_inline; break; case 'Z': barcode_type = inline_index; break; case 'A': adapter_1 = new char[strlen(optarg) + 1]; strcpy(adapter_1, optarg); filter_adapter = true; break; case 'G': adapter_2 = new char[strlen(optarg) + 1]; strcpy(adapter_2, optarg); filter_adapter = true; break; case 'T': distance = is_integer(optarg); break; case 'H': retain_header = true; break; case 'L': len_limit = is_integer(optarg); break; case 'w': win_size = is_double(optarg); break; case 's': score_limit = is_integer(optarg); break; case 'v': version(); break; case '?': // getopt_long already printed an error message. help(); break; default: cerr << "Unknown command line option '" << (char) c << "'\n"; help(); abort(); } } if (in_file.length() == 0 && in_path_1.length() == 0 && in_file_p1.length() == 0) { cerr << "You must specify an input file of a directory path to a set of input files.\n"; help(); } if (in_file.length() > 0 && in_path_1.length() > 0) { cerr << "You must specify either a single input file (-f) or a directory path (-p), not both.\n"; help(); } if (in_file.length() > 0 && (in_file_p1.length() > 0 || in_file_p2.length() > 0)) { cerr << "You must specify either a single input file (-f) or a set of paired files (-1, -2), not both.\n"; help(); } if (in_path_1.length() > 0 && (in_file_p1.length() > 0 || in_file_p2.length() > 0)) { cerr << "You must specify either a file path (-p) or a set of paired files (-1, -2), not both.\n"; help(); } if (in_path_1.length() > 0 && in_path_1.at(in_path_1.length() - 1) != '/') in_path_1 += "/"; if (in_path_2.length() > 0 && in_path_2.at(in_path_2.length() - 1) != '/') in_path_2 += "/"; if (out_path.length() == 0) out_path = "."; if (out_path.at(out_path.length() - 1) != '/') out_path += "/"; if (in_file_type == FileT::unknown) in_file_type = ftype; if (in_file_type == FileT::bam && paired == true && interleaved == false) { cerr << "You may only specify a BAM input file for paired-end data if the read pairs are interleaved.\n"; help(); } if (in_file_type == FileT::bam && (barcode_type != inline_null && barcode_type != inline_inline && barcode_type != null_null)) { cerr << "For BAM input files only inline or unbarcoded data can be processed.\n"; help(); } if (barcode_file.length() == 0 && barcode_type != null_null) { cerr << "You specified a barcode type without providing a file containing barcodes.\n"; help(); } if (barcode_file.length() == 0) cerr << "No barcodes specified, files will not be demultiplexed.\n"; if (barcode_file.length() > 0 && merge) { cerr << "You may specify a set of barcodes, or that all files should be merged, not both.\n"; help(); } if (check_radtag && renz_1.length() == 0) { cerr << "You must specify the restriction enzyme used.\n"; help(); } if (check_radtag && renz.count(renz_1) == 0) { cerr << "Unrecognized restriction enzyme specified: '" << renz_1.c_str() << "'.\n"; help(); } if (check_radtag && renz_2.length() > 0 && renz.count(renz_2) == 0) { cerr << "Unrecognized restriction enzyme specified: '" << renz_2.c_str() << "'.\n"; help(); } if (score_limit > 40) { cerr << "Score limit must be between 0 and 40.\n"; help(); } if (win_size < 0 || win_size >= 1) { cerr << "Window size is a fraction between 0 and 1.\n"; help(); } if (recover && barcode_type != null_null) { if (barcode_type != index_null && barcode_type != inline_null && barcode_dist_2 < 0) barcode_dist_2 = barcode_dist_1; } return 0; } void version() { std::cerr << "process_radtags " << VERSION << "\n\n"; exit(0); } void help() { std::cerr << "process_radtags " << VERSION << "\n" << "process_radtags [-f in_file | -p in_dir [-P] [-I] | -1 pair_1 -2 pair_2] -b barcode_file -o out_dir -e enz [-c] [-q] [-r] [-t len] [-D] [-w size] [-s lim] [-h]\n" << " f: path to the input file if processing single-end sequences.\n" << " i: input file type, either 'bustard' for the Illumina BUSTARD format, 'bam', 'fastq' (default), or 'gzfastq' for gzipped FASTQ.\n" << " y: output type, either 'fastq', 'gzfastq', 'fasta', or 'gzfasta' (default is to match the input file type).\n" << " p: path to a directory of files.\n" << " P: files contained within directory specified by '-p' are paired.\n" << " I: specify that the paired-end reads are interleaved in single files.\n" << " 1: first input file in a set of paired-end sequences.\n" << " 2: second input file in a set of paired-end sequences.\n" << " o: path to output the processed files.\n" << " b: path to a file containing barcodes for this run.\n" << " c: clean data, remove any read with an uncalled base.\n" << " q: discard reads with low quality scores.\n" << " r: rescue barcodes and RAD-Tags.\n" << " t: truncate final read length to this value.\n" << " E: specify how quality scores are encoded, 'phred33' (Illumina 1.8+, Sanger, default) or 'phred64' (Illumina 1.3 - 1.5).\n" << " D: capture discarded reads to a file.\n" << " w: set the size of the sliding window as a fraction of the read length, between 0 and 1 (default 0.15).\n" << " s: set the score limit. If the average score within the sliding window drops below this value, the read is discarded (default 10).\n" << " h: display this help messsage." << "\n\n" << " Barcode options:\n" << " --inline_null: barcode is inline with sequence, occurs only on single-end read (default).\n" << " --index_null: barcode is provded in FASTQ header (Illumina i5 or i7 read).\n" << " --null_index: barcode is provded in FASTQ header (Illumina i7 read if both i5 and i7 read are provided).\n" << " --inline_inline: barcode is inline with sequence, occurs on single and paired-end read.\n" << " --index_index: barcode is provded in FASTQ header (Illumina i5 and i7 reads).\n" << " --inline_index: barcode is inline with sequence on single-end read and occurs in FASTQ header (from either i5 or i7 read).\n" << " --index_inline: barcode occurs in FASTQ header (Illumina i5 or i7 read) and is inline with single-end sequence (for single-end data) on paired-end read (for paired-end data).\n\n" << " Restriction enzyme options:\n" << " -e , --renz_1 : provide the restriction enzyme used (cut site occurs on single-end read)\n" << " --renz_2 : if a double digest was used, provide the second restriction enzyme used (cut site occurs on the paired-end read).\n" << " Currently supported enzymes include:\n" << " "; map::iterator it; uint cnt = renz_cnt.size(); it = renz_cnt.begin(); for (uint i = 1; i <= cnt; i++) { std::cerr << "'" << it->first << "'"; if (i < cnt - 1) std::cerr << ", "; else if (i == cnt - 1) std::cerr << ", or "; if (i % 8 == 0) std::cerr << "\n "; it++; } std::cerr << "\n" << " Adapter options:\n" << " --adapter_1 : provide adaptor sequence that may occur on the single-end read for filtering.\n" << " --adapter_2 : provide adaptor sequence that may occur on the paired-read for filtering.\n" << " --adapter_mm : number of mismatches allowed in the adapter sequence.\n\n" << " Output options:\n" << " --retain_header: retain unmodified FASTQ headers in the output.\n" << " --merge: if no barcodes are specified, merge all input files into a single output file.\n\n" << " Advanced options:\n" << " --filter_illumina: discard reads that have been marked by Illumina's chastity/purity filter as failing.\n" << " --disable_rad_check: disable checking if the RAD site is intact.\n" << " --len_limit : specify a minimum sequence length (useful if your data has already been trimmed).\n" << " --barcode_dist_1: the number of allowed mismatches when rescuing single-end barcodes (default 1).\n" << " --barcode_dist_2: the number of allowed mismatches when rescuing paired-end barcodes (defaults to --barcode_dist_1).\n"; exit(0); } stacks-1.44/src/process_radtags.h000644 000765 000024 00000005741 12764047443 017647 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __PROCESS_RADTAGS_H__ #define __PROCESS_RADTAGS_H__ #include #include // Process command-line options #include // Open/Read contents of a directory #include #include #include #include #include #include using std::stringstream; using std::istream; using std::ofstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include using std::vector; #include using std::map; #include using std::set; #include using std::pair; #include "constants.h" #include "renz.h" #include "clean.h" #include "file_io.h" #include "write.h" #include "utils.h" #include "log_utils.h" #include "BustardI.h" // Reading input files in Tab-separated Bustard format #include "FastqI.h" // Reading input files in FASTQ format #include "gzFastq.h" // Reading gzipped input files in FASTQ format #include "BamUnalignedI.h" // Reading data from unaligned BAM files void help( void ); void version( void ); int parse_command_line(int, char **); template int process_reads(string, set &, set &, map &, map &, map > &); template int process_paired_reads(string, string, set &, set &, map &, map &, map &, map &, map &, map > &); int process_singlet(Read *, string, bool, map &, map &); int correct_radtag(Read *, string, map &); int check_quality_scores(Read *, bool); int dist(const char *, char *); int print_results(int, char **, vector &, map > &, map > &); int compare_barcodes(pair, pair); #endif // __PROCESS_RADTAGS_H__ stacks-1.44/src/process_shortreads.cc000644 000765 000024 00000124471 12764047443 020540 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2011-2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // process_shortreads -- clean raw reads using a sliding window approach; // split reads by barcode if barcodes provided, correct barcodes // within one basepair, truncate reads on request. // #include "process_shortreads.h" // // Global variables to hold command-line options. // FileT in_file_type = FileT::unknown; FileT out_file_type = FileT::unknown; string in_file; string in_file_p1; string in_file_p2; string in_path_1; string in_path_2; string out_path; string barcode_file; char *adapter_1; char *adapter_2; barcodet barcode_type = null_null; bool retain_header = false; bool filter_adapter = false; bool paired = false; bool clean = false; bool quality = false; bool recover = false; bool interleaved = false; bool merge = false; bool discards = false; bool overhang = true; bool matepair = false; bool filter_illumina = false; bool trim_reads = true; uint truncate_seq = 0; int barcode_dist_1 = 1; int barcode_dist_2 = -1; double win_size = 0.15; int score_limit = 10; int len_limit = 31; int num_threads = 1; // // How to shift FASTQ-encoded quality scores from ASCII down to raw scores // score = encoded letter - 64; Illumina version 1.3 - 1.5 // score = encoded letter - 33; Sanger / Illumina version 1.6+ int qual_offset = 33; // // Handle variable-size barcodes. // uint min_bc_size_1 = 0; uint max_bc_size_1 = 0; uint min_bc_size_2 = 0; uint max_bc_size_2 = 0; // // Kmer data for adapter filtering. // int kmer_size = 5; int distance = 1; int adp_1_len = 0; int adp_2_len = 0; AdapterHash adp_1_kmers, adp_2_kmers; int main (int argc, char* argv[]) { parse_command_line(argc, argv); // // If input files are gzipped, output gziped files, unless the user chooses an output type. // if (out_file_type == FileT::unknown) { if (in_file_type == FileT::gzfastq || in_file_type == FileT::bam) out_file_type = FileT::gzfastq; else out_file_type = FileT::fastq; } cerr << "Using Phred+" << qual_offset << " encoding for quality scores.\n" << "Reads trimmed shorter than " << len_limit << " nucleotides will be discarded.\n"; if (truncate_seq > 0) cerr << "Reads will be truncated to " << truncate_seq << "bp\n"; if (filter_illumina) cerr << "Discarding reads marked as 'failed' by Illumina's chastity/purity filters.\n"; if (filter_adapter) { cerr << "Filtering reads for adapter sequence:\n"; if (adapter_1 != NULL) { cerr << " " << adapter_1 << "\n"; init_adapter_seq(kmer_size, adapter_1, adp_1_len, adp_1_kmers); } if (adapter_2 != NULL) { cerr << " " << adapter_2 << "\n"; init_adapter_seq(kmer_size, adapter_2, adp_2_len, adp_2_kmers); } cerr << " " << distance << " mismatches allowed to adapter sequence.\n"; } vector > files; vector barcodes; set se_bc, pe_bc; map pair_1_fhs, pair_2_fhs, rem_1_fhs, rem_2_fhs; map pair_1_gzfhs, pair_2_gzfhs, rem_1_gzfhs, rem_2_gzfhs; map > counters; map > barcode_log; build_file_list(files); load_barcodes(barcode_file, barcodes, se_bc, pe_bc, min_bc_size_1, max_bc_size_1, min_bc_size_2, max_bc_size_2); if (recover && barcode_type != null_null) { if (barcode_type == index_null || barcode_type == inline_null) cerr << "Will attempt to recover barcodes with at most " << barcode_dist_1 << " mismatches.\n"; else cerr << "Will attempt to recover barcodes with at most " << barcode_dist_1 << " / " << barcode_dist_2 << " mismatches.\n"; } if (out_file_type == FileT::gzfastq || out_file_type == FileT::gzfasta) open_files(files, barcodes, pair_1_gzfhs, pair_2_gzfhs, rem_1_gzfhs, rem_2_gzfhs, counters); else open_files(files, barcodes, pair_1_fhs, pair_2_fhs, rem_1_fhs, rem_2_fhs, counters); int result = 1; for (uint i = 0; i < files.size(); i++) { cerr << "Processing file " << i+1 << " of " << files.size() << " [" << files[i].first.c_str() << "]\n"; counters[files[i].first]["total"] = 0; counters[files[i].first]["ill_filtered"] = 0; counters[files[i].first]["low_quality"] = 0; counters[files[i].first]["trimmed"] = 0; counters[files[i].first]["adapter"] = 0; counters[files[i].first]["ambiguous"] = 0; counters[files[i].first]["retained"] = 0; counters[files[i].first]["orphaned"] = 0; counters[files[i].first]["recovered"] = 0; if (paired) { if (out_file_type == FileT::gzfastq || out_file_type == FileT::gzfasta) result = process_paired_reads(files[i].first, files[i].second, se_bc, pe_bc, pair_1_gzfhs, pair_2_gzfhs, rem_1_gzfhs, rem_2_gzfhs, counters[files[i].first], barcode_log); else result = process_paired_reads(files[i].first, files[i].second, se_bc, pe_bc, pair_1_fhs, pair_2_fhs, rem_1_fhs, rem_2_fhs, counters[files[i].first], barcode_log); } else { if (out_file_type == FileT::gzfastq || out_file_type == FileT::gzfasta) result = process_reads(files[i].first, se_bc, pe_bc, pair_1_gzfhs, counters[files[i].first], barcode_log); else result = process_reads(files[i].first, se_bc, pe_bc, pair_1_fhs, counters[files[i].first], barcode_log); } cerr << " " << counters[files[i].first]["total"] << " total reads; "; if (filter_illumina) cerr << "-" << counters[files[i].first]["ill_filtered"] << " failed Illumina reads; "; cerr << "-" << counters[files[i].first]["ambiguous"] << " ambiguous barcodes; " << "+" << counters[files[i].first]["recovered"] << " recovered; " << "-" << counters[files[i].first]["low_quality"] << " low quality reads; " << counters[files[i].first]["retained"] << " retained reads.\n" << " "; if (filter_adapter) cerr << counters[files[i].first]["adapter"] << " reads with adapter sequence; "; cerr << counters[files[i].first]["trimmed"] << " trimmed reads; " << counters[files[i].first]["orphaned"] << " orphaned paired-ends.\n"; if (!result) { cerr << "Error processing reads.\n"; break; } } cerr << "Closing files, flushing buffers...\n"; if (out_file_type == FileT::gzfastq || out_file_type == FileT::gzfasta) { close_file_handles(pair_1_gzfhs); if (paired) { close_file_handles(rem_1_gzfhs); close_file_handles(rem_2_gzfhs); close_file_handles(pair_2_gzfhs); } } else { close_file_handles(pair_1_fhs); if (paired) { close_file_handles(rem_1_fhs); close_file_handles(rem_2_fhs); close_file_handles(pair_2_fhs); } } print_results(argc, argv, barcodes, counters, barcode_log); return 0; } template int process_paired_reads(string prefix_1, string prefix_2, set &se_bc, set &pe_bc, map &pair_1_fhs, map &pair_2_fhs, map &rem_1_fhs, map &rem_2_fhs, map &counter, map > &barcode_log) { Input *fh_1, *fh_2; Read *r_1, *r_2; ofstream *discard_fh_1, *discard_fh_2; int return_val = 1; string path_1 = in_path_1 + prefix_1; string path_2 = in_path_2 + prefix_2; if (interleaved) cerr << " Reading data from:\n " << path_1 << "\n"; else cerr << " Reading data from:\n " << path_1 << " and\n " << path_2 << "\n"; if (in_file_type == FileT::fastq) { fh_1 = new Fastq(path_1.c_str()); fh_2 = interleaved ? fh_1 : new Fastq(path_2.c_str()); } else if (in_file_type == FileT::gzfastq) { fh_1 = new GzFastq(path_1.c_str()); fh_2 = interleaved ? fh_1 : new GzFastq(path_2.c_str()); } else if (in_file_type == FileT::bam) { fh_1 = new BamUnAln(path_1.c_str()); fh_2 = fh_1; } else if (in_file_type == FileT::bustard) { fh_1 = new Bustard(path_1.c_str()); fh_2 = interleaved ? fh_1 : new Bustard(path_2.c_str()); } // // Open a file for recording discarded reads // if (discards) { path_1 = out_path + prefix_1 + ".discards"; discard_fh_1 = new ofstream(path_1.c_str(), ifstream::out); if (discard_fh_1->fail()) { cerr << "Error opening discard output file '" << path_1 << "'\n"; exit(1); } path_2 = out_path + prefix_2 + ".discards"; discard_fh_2 = new ofstream(path_2.c_str(), ifstream::out); if (discard_fh_1->fail()) { cerr << "Error opening discard output file '" << path_2 << "'\n"; exit(1); } } // // Read in the first record, initializing the Seq object s. Then // initialize the Read object r, then loop, using the same objects. // Seq *s_1 = fh_1->next_seq(); Seq *s_2 = fh_2->next_seq(); if (s_1 == NULL || s_2 == NULL) { cerr << "Attempting to read first pair of input records, unable to allocate " << "Seq object (Was the correct input type specified?).\n"; exit(1); } r_1 = new Read(strlen(s_1->seq), 1, min_bc_size_1, win_size); r_2 = new Read(strlen(s_2->seq), 2, min_bc_size_2, win_size); BarcodePair bc; // // If no barcodes were specified, set the barcode object to be the input file names. // if (max_bc_size_1 == 0) bc.set(prefix_1, prefix_2); long i = 1; do { if (i % 10000 == 0) cerr << " Processing short read " << i << " \r"; parse_input_record(s_1, r_1); parse_input_record(s_2, r_2); counter["total"] += 2; if (barcode_type != null_null && barcode_type != inline_null && barcode_type != index_null) bc.set(r_1->se_bc, r_2->pe_bc); else if (barcode_type != null_null) bc.set(r_1->se_bc); process_barcode(r_1, r_2, bc, pair_1_fhs, se_bc, pe_bc, barcode_log, counter); // // Adjust the size of the read to accommodate truncating the sequence and variable // barcode lengths. With standard Illumina data we want to output constant length // reads even as the barcode size may change. Other technologies, like IonTorrent // need to be truncated uniformly. // if (truncate_seq > 0) { if (truncate_seq + r_1->inline_bc_len <= r_1->len) r_1->set_len(truncate_seq + r_1->inline_bc_len); if (truncate_seq + r_2->inline_bc_len <= r_2->len) r_2->set_len(truncate_seq + r_2->inline_bc_len); } else { if (barcode_type == inline_null || barcode_type == inline_inline || barcode_type == inline_index) r_1->set_len(r_1->len - (max_bc_size_1 - r_1->inline_bc_len)); if (barcode_type == inline_index || barcode_type == index_index) r_2->set_len(r_2->len - (max_bc_size_2 - r_2->inline_bc_len)); } if (r_1->retain) process_singlet(r_1, false, barcode_log[bc], counter); if (r_2->retain) process_singlet(r_2, true, barcode_log[bc], counter); if (matepair) { rev_complement(r_1->seq, r_1->inline_bc_len, overhang); reverse_qual(r_1->phred, r_1->inline_bc_len, overhang); } int result_1 = 1; int result_2 = 1; if (r_1->retain && r_2->retain) { if (retain_header) { result_1 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(pair_1_fhs[bc], s_1, r_1) : write_fasta(pair_1_fhs[bc], s_1, r_1); result_2 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(pair_2_fhs[bc], s_2, r_2) : write_fasta(pair_2_fhs[bc], s_2, r_2); } else { result_1 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(pair_1_fhs[bc], r_1, overhang) : write_fasta(pair_1_fhs[bc], r_1, overhang); result_2 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(pair_2_fhs[bc], r_2, overhang) : write_fasta(pair_2_fhs[bc], r_2, overhang); } } else if (r_1->retain && !r_2->retain) { // // Write to a remainder file. // if (retain_header) result_1 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(rem_1_fhs[bc], s_1, r_1) : write_fasta(rem_1_fhs[bc], s_1, r_1); else result_1 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(rem_1_fhs[bc], r_1, overhang) : write_fasta(rem_1_fhs[bc], r_1, overhang); } else if (!r_1->retain && r_2->retain) { // Write to a remainder file. if (retain_header) result_2 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(rem_2_fhs[bc], s_2, r_2) : write_fasta(rem_2_fhs[bc], s_2, r_2); else result_2 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(rem_2_fhs[bc], r_2, overhang) : write_fasta(rem_2_fhs[bc], r_2, overhang); } if (!result_1 || !result_2) { cerr << "Error writing to output file for '" << bc.str() << "'\n"; return_val = -1; break; } if (discards && !r_1->retain) result_1 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(discard_fh_1, s_1) : write_fasta(discard_fh_1, s_1); if (discards && !r_2->retain) result_2 = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(discard_fh_2, s_2) : write_fasta(discard_fh_2, s_2); delete s_1; delete s_2; if (!result_1 || !result_2) { cerr << "Error writing to discard file for '" << bc.str() << "'\n"; return_val = -1; break; } i++; } while ((s_1 = fh_1->next_seq()) != NULL && (s_2 = fh_2->next_seq()) != NULL); if (discards) { delete discard_fh_1; delete discard_fh_2; } delete fh_1; if (interleaved == false) delete fh_2; return return_val; } template int process_reads(string prefix, set &se_bc, set &pe_bc, map &pair_1_fhs, map &counter, map > &barcode_log) { Input *fh; Read *r; ofstream *discard_fh; int return_val = 1; string path = in_path_1 + prefix; if (in_file_type == FileT::fastq) fh = new Fastq(path.c_str()); else if (in_file_type == FileT::gzfastq) fh = new GzFastq(path.c_str()); else if (in_file_type == FileT::bam) fh = new BamUnAln(path.c_str()); else if (in_file_type == FileT::bustard) fh = new Bustard(path.c_str()); // // Open a file for recording discarded reads // if (discards) { path = path + ".discards"; discard_fh = new ofstream(path.c_str(), ifstream::out); if (discard_fh->fail()) { cerr << "Error opening discard output file '" << path << "'\n"; exit(1); } } // // Read in the first record, initializing the Seq object s. Then // initialize the Read object r, then loop, using the same objects. // Seq *s = fh->next_seq(); if (s == NULL) { cerr << "Attempting to read first input record, unable to allocate " << "Seq object (Was the correct input type specified?).\n"; exit(1); } r = new Read(strlen(s->seq), 1, min_bc_size_1, win_size); BarcodePair bc; // // If no barcodes were specified, set the barcode object to be the input file name so // that reads are written to an output file of the same name as the input file. // if (max_bc_size_1 == 0) bc.set(prefix); //cerr << "Length: " << r->len << "; Window length: " << r->win_len << "; Stop position: " << r->stop_pos << "\n"; long i = 1; do { if (i % 10000 == 0) cerr << " Processing short read " << i << " \r"; counter["total"]++; parse_input_record(s, r); if (barcode_type == inline_null || barcode_type == index_null) bc.set(r->se_bc); else if (barcode_type == index_inline || barcode_type == inline_index) bc.set(r->se_bc, r->pe_bc); process_barcode(r, NULL, bc, pair_1_fhs, se_bc, pe_bc, barcode_log, counter); // // Adjust the size of the read to accommodate truncating the sequence and variable // barcode lengths. With standard Illumina data we want to output constant length // reads even as the barcode size may change. Other technologies, like IonTorrent // need to be truncated uniformly. // if (truncate_seq > 0) { if (truncate_seq + r->inline_bc_len <= r->len) r->set_len(truncate_seq + r->inline_bc_len); } else { if (barcode_type == inline_null || barcode_type == inline_inline || barcode_type == inline_index) r->set_len(r->len - (max_bc_size_1 - r->inline_bc_len)); } if (r->retain) process_singlet(r, false, barcode_log[bc], counter); int result = 1; if (r->retain) { if (retain_header) result = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(pair_1_fhs[bc], s, r) : write_fasta(pair_1_fhs[bc], s, r); else result = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(pair_1_fhs[bc], r, overhang) : write_fasta(pair_1_fhs[bc], r, overhang); } if (!result) { cerr << "Error writing to output file for '" << bc.str() << "'\n"; return_val = -1; break; } if (discards && !r->retain) result = (out_file_type == FileT::fastq || out_file_type == FileT::gzfastq) ? write_fastq(discard_fh, s) : write_fasta(discard_fh, s); if (!result) { cerr << "Error writing to discard file for '" << bc.str() << "'\n"; return_val = -1; break; } delete s; i++; } while ((s = fh->next_seq()) != NULL); if (discards) delete discard_fh; // // Close the file and delete the Input object. // delete fh; return return_val; } inline int process_singlet(Read *href, bool paired_end, map &bc_log, map &counter) { if (filter_illumina && href->filter) { counter["ill_filtered"]++; href->retain = 0; return 0; } // // Drop this sequence if it has any uncalled nucleotides // if (clean) { for (char *p = href->seq + href->inline_bc_len; *p != '\0'; p++) if (*p == '.' || *p == 'N') { counter["low_quality"]++; href->retain = 0; return 0; } } bool adapter_trim = false; bool quality_trim = false; // // Drop or trim this sequence if it has low quality scores // if (quality) { int res = check_quality_scores(href, qual_offset, score_limit, len_limit, href->inline_bc_len); if (trim_reads) { if (res == 0) { counter["low_quality"]++; href->retain = 0; return 0; } else if (res < 0) { quality_trim = true; } } else { if (res <= 0) { counter["low_quality"]++; href->retain = 0; return 0; } } } // // Drop or trim this sequence if it contains adapter sequence. // if (filter_adapter) { int res = 1; if (paired_end == true && adp_2_len > 0) res = filter_adapter_seq(href, adapter_2, adp_2_len, adp_2_kmers, kmer_size, distance, len_limit); if (paired_end == false && adp_1_len > 0) res = filter_adapter_seq(href, adapter_1, adp_1_len, adp_1_kmers, kmer_size, distance, len_limit); if (res == 0) { counter["adapter"]++; href->retain = 0; return 0; } else if (res < 0) { counter["adapter"]++; adapter_trim = true; } } if (adapter_trim || quality_trim) counter["trimmed"]++; if (barcode_type != null_null) bc_log["retained"]++; counter["retained"]++; return 0; } int dist(const char *res_enz, char *seq) { const char *p; char *q; int dist = 0; for (p = res_enz, q = seq; *p != '\0'; p++, q++) if (*p != *q) dist++; return dist; } int print_results(int argc, char **argv, vector &barcodes, map > &counters, map > &barcode_log) { map >::iterator it; string log_path = out_path + "process_shortreads.log"; ofstream log(log_path.c_str()); if (log.fail()) { cerr << "Unable to open log file '" << log_path << "'\n"; return 0; } cerr << "Outputing details to log: '" << log_path << "'\n\n"; init_log(log, argc, argv); log << "File\t" << "Retained Reads\t"; if (filter_illumina) log << "Illumina Filtered\t"; if (filter_adapter) log << "Adapter Seq" << "\t"; log << "Low Quality\t" << "Ambiguous Barcodes\t" << "Trimmed Reads\t" << "Orphaned paired-end reads\t" << "Total\n"; for (it = counters.begin(); it != counters.end(); it++) { log << it->first << "\t" << it->second["retained"] << "\t"; if (filter_illumina) log << it->second["ill_filtered"] << "\t"; if (filter_adapter) log << it->second["adapter"] << "\t"; log << it->second["low_quality"] << "\t" << it->second["ambiguous"] << "\t" << it->second["trimmed"] << "\t" << it->second["orphaned"] << "\t" << it->second["total"] << "\n"; } map c; c["total"] = 0; c["low_quality"] = 0; c["adapter"] = 0; c["ill_filtered"] = 0; c["ambiguous"] = 0; c["trimmed"] = 0; c["orphaned"] = 0; // // Total up the individual counters // for (it = counters.begin(); it != counters.end(); it++) { c["total"] += it->second["total"]; c["ill_filtered"] += it->second["ill_filtered"]; c["adapter"] += it->second["adapter"]; c["low_quality"] += it->second["low_quality"]; c["ambiguous"] += it->second["ambiguous"]; c["trimmed"] += it->second["trimmed"]; c["orphaned"] += it->second["orphaned"]; c["retained"] += it->second["retained"]; } cerr << c["total"] << " total sequences;\n"; if (filter_illumina) cerr << " " << c["ill_filtered"] << " failed Illumina filtered reads;\n"; if (filter_adapter) cerr << " " << c["adapter"] << " reads contained adapter sequence;\n"; cerr << " " << c["ambiguous"] << " ambiguous barcode drops;\n" << " " << c["low_quality"] << " low quality read drops;\n" << " " << c["trimmed"] << " trimmed reads;\n" << " " << c["orphaned"] << " orphaned paired-end reads;\n" << c["retained"] << " retained reads.\n"; log << "\n" << "Total Sequences\t" << c["total"] << "\n"; if (filter_illumina) log << "Failed Illumina filtered reads\t" << c["ill_filtered"] << "\n"; if (filter_adapter) log << "Reads containing adapter sequence\t" << c["adapter"] << "\n"; log << "Ambiguous Barcodes\t" << c["ambiguous"] << "\n" << "Low Quality\t" << c["low_quality"] << "\n" << "Trimmed Reads\t" << c["trimmed"] << "\n" << "Orphaned Paired-ends\t" << c["orphaned"] << "\n" << "Retained Reads\t" << c["retained"] << "\n"; if (max_bc_size_1 == 0) return 0; // // Where barcode filenames specified? // bool bc_names = false; for (uint i = 0; i < barcodes.size(); i++) if (barcodes[i].name_exists()) { bc_names = true; break; } // // Print out barcode information. // log << "\n" << "Barcode\t"; if (bc_names) log << "Filename\t"; log << "Total\t" << "Retained\n"; set barcode_list; for (uint i = 0; i < barcodes.size(); i++) { barcode_list.insert(barcodes[i]); log << barcodes[i] << "\t"; if (bc_names) log << barcodes[i].name << "\t"; if (barcode_log.count(barcodes[i]) == 0) log << "0\t" << "0\t" << "0\n"; else log << barcode_log[barcodes[i]]["total"] << "\t" << barcode_log[barcodes[i]]["retained"] << "\n"; } log << "\n" << "Sequences not recorded\n" << "Barcode\t" << "Total\n"; // // Sort unused barcodes by number of occurances. // map >::iterator bit; vector > bcs; for (bit = barcode_log.begin(); bit != barcode_log.end(); bit++) bcs.push_back(make_pair(bit->first, bit->second["total"])); sort(bcs.begin(), bcs.end(), compare_barcodes); for (uint i = 0; i < bcs.size(); i++) { if (barcode_list.count(bcs[i].first)) continue; if (bcs[i].second == 0) continue; log << bcs[i].first << "\t" << bcs[i].second << "\n"; } log.close(); return 0; } int compare_barcodes(pair a, pair b) { return a.second > b.second; } int parse_command_line(int argc, char* argv[]) { FileT ftype; int c; while (1) { static struct option long_options[] = { {"help", no_argument, NULL, 'h'}, {"version", no_argument, NULL, 'v'}, {"quality", no_argument, NULL, 'q'}, {"clean", no_argument, NULL, 'c'}, {"recover", no_argument, NULL, 'r'}, {"discards", no_argument, NULL, 'D'}, {"paired", no_argument, NULL, 'P'}, {"interleaved", no_argument, NULL, 'I'}, {"merge", no_argument, NULL, 'm'}, {"mate-pair", no_argument, NULL, 'M'}, {"no_overhang", no_argument, NULL, 'O'}, {"filter_illumina", no_argument, NULL, 'F'}, {"retain_header", no_argument, NULL, 'H'}, {"no_read_trimming", no_argument, NULL, 'N'}, {"null_index", no_argument, NULL, 'U'}, {"index_null", no_argument, NULL, 'u'}, {"inline_null", no_argument, NULL, 'V'}, {"index_index", no_argument, NULL, 'W'}, {"inline_inline", no_argument, NULL, 'x'}, {"index_inline", no_argument, NULL, 'Y'}, {"inline_index", no_argument, NULL, 'Z'}, {"barcode_dist_1", required_argument, NULL, 'B'}, {"barcode_dist_2", required_argument, NULL, 'C'}, {"infile_type", required_argument, NULL, 'i'}, {"outfile_type", required_argument, NULL, 'y'}, {"file", required_argument, NULL, 'f'}, {"file_p1", required_argument, NULL, '1'}, {"file_p2", required_argument, NULL, '2'}, {"path", required_argument, NULL, 'p'}, {"outpath", required_argument, NULL, 'o'}, {"truncate", required_argument, NULL, 't'}, {"barcodes", required_argument, NULL, 'b'}, {"window_size", required_argument, NULL, 'w'}, {"score_limit", required_argument, NULL, 's'}, {"encoding", required_argument, NULL, 'E'}, {"len_limit", required_argument, NULL, 'L'}, {"adapter_1", required_argument, NULL, 'A'}, {"adapter_2", required_argument, NULL, 'G'}, {"adapter_mm", required_argument, NULL, 'T'}, {0, 0, 0, 0} }; // getopt_long stores the option index here. int option_index = 0; c = getopt_long(argc, argv, "hHvcqrINFuVWxYZOPmDi:y:f:o:t:B:C:b:1:2:p:s:w:E:L:A:G:T:", long_options, &option_index); // Detect the end of the options. if (c == -1) break; switch (c) { case 'h': help(); break; case 'i': if (strcasecmp(optarg, "bustard") == 0) in_file_type = FileT::bustard; else if (strcasecmp(optarg, "bam") == 0) in_file_type = FileT::bam; else if (strcasecmp(optarg, "gzfastq") == 0) in_file_type = FileT::gzfastq; else in_file_type = FileT::fastq; break; case 'y': if (strcasecmp(optarg, "fastq") == 0) out_file_type = FileT::fastq; else if (strcasecmp(optarg, "gzfastq") == 0) out_file_type = FileT::gzfastq; else if (strcasecmp(optarg, "fasta") == 0) out_file_type = FileT::fasta; else if (strcasecmp(optarg, "gzfasta") == 0) out_file_type = FileT::gzfasta; break; case 'E': if (strcasecmp(optarg, "phred64") == 0) qual_offset = 64; else if (strcasecmp(optarg, "phred33") == 0) qual_offset = 33; break; case 'f': in_file = optarg; ftype = FileT::fastq; break; case 'p': in_path_1 = optarg; in_path_2 = in_path_1; ftype = FileT::fastq; break; case '1': paired = true; in_file_p1 = optarg; ftype = FileT::fastq; break; case '2': paired = true; in_file_p2 = optarg; ftype = FileT::fastq; break; case 'P': paired = true; break; case 'I': interleaved = true; break; case 'B': barcode_dist_1 = is_integer(optarg); break; case 'C': barcode_dist_2 = is_integer(optarg); break; case 'o': out_path = optarg; break; case 'm': merge = true; break; case 'M': matepair = true; break; case 'D': discards = true; break; case 'q': quality = true; break; case 'c': clean = true; break; case 'r': recover = true; break; case 'O': overhang = false; break; case 'F': filter_illumina = true; break; case 'H': retain_header = true; break; case 'N': trim_reads = false; break; case 't': truncate_seq = is_integer(optarg); break; case 'b': barcode_file = optarg; if (barcode_type == null_null) barcode_type = inline_null; break; case 'U': barcode_type = null_index; break; case 'u': barcode_type = index_null; break; case 'V': barcode_type = inline_null; break; case 'W': barcode_type = index_index; break; case 'x': barcode_type = inline_inline; break; case 'Y': barcode_type = index_inline; break; case 'Z': barcode_type = inline_index; break; case 'A': adapter_1 = new char[strlen(optarg) + 1]; strcpy(adapter_1, optarg); filter_adapter = true; break; case 'G': adapter_2 = new char[strlen(optarg) + 1]; strcpy(adapter_2, optarg); filter_adapter = true; break; case 'T': distance = is_integer(optarg); break; case 'L': len_limit = is_integer(optarg); break; case 'w': win_size = is_double(optarg); break; case 's': score_limit = is_integer(optarg); break; case 'v': version(); break; case '?': // getopt_long already printed an error message. help(); break; default: cerr << "Unknown command line option '" << (char) c << "'\n"; help(); abort(); } } if (in_file.length() == 0 && in_path_1.length() == 0 && in_file_p1.length() == 0) { cerr << "You must specify an input file of a directory path to a set of input files.\n"; help(); } if (in_file.length() > 0 && in_path_1.length() > 0) { cerr << "You must specify either a single input file (-f) or a directory path (-p), not both.\n"; help(); } if (in_file.length() > 0 && (in_file_p1.length() > 0 || in_file_p2.length() > 0)) { cerr << "You must specify either a single input file (-f) or a set of paired files (-1, -2), not both.\n"; help(); } if (in_path_1.length() > 0 && (in_file_p1.length() > 0 || in_file_p2.length() > 0)) { cerr << "You must specify either a file path (-p) or a set of paired files (-1, -2), not both.\n"; help(); } if (in_path_1.length() > 0 && in_path_1.at(in_path_1.length() - 1) != '/') in_path_1 += "/"; if (in_path_2.length() > 0 && in_path_2.at(in_path_2.length() - 1) != '/') in_path_2 += "/"; if (out_path.length() == 0) out_path = "."; if (out_path.at(out_path.length() - 1) != '/') out_path += "/"; if (barcode_file.length() == 0) { overhang = false; cerr << "No barcodes specified, files will not be demultiplexed.\n"; } if (barcode_file.length() > 0 && merge) { cerr << "You may specify a set of barcodes, or that all files should be merged, not both.\n"; help(); } if (in_file_type == FileT::unknown) in_file_type = ftype; if (in_file_type == FileT::bam && paired == true && interleaved == false) { cerr << "You may only specify a BAM input file for paired-end data if the read pairs are interleaved.\n"; help(); } if (in_file_type == FileT::bam && (barcode_type != inline_null && barcode_type != inline_inline && barcode_type != null_null)) { cerr << "For BAM input files only inline or unbarcoded data can be processed.\n"; help(); } if (score_limit < 0 || score_limit > 40) { cerr << "Score limit must be between 0 and 40.\n"; help(); } if (win_size < 0 || win_size >= 1) { cerr << "Window size is a fraction between 0 and 1.\n"; help(); } if (recover && barcode_type != null_null) { if (barcode_type != index_null && barcode_type != inline_null && barcode_dist_2 < 0) barcode_dist_2 = barcode_dist_1; } return 0; } void version() { std::cerr << "process_shortreads " << VERSION << "\n\n"; exit(0); } void help() { std::cerr << "process_shortreads " << VERSION << "\n" << "process_shortreads [-f in_file | -p in_dir [-P] [-I] | -1 pair_1 -2 pair_2] -b barcode_file -o out_dir [-i type] [-y type] [-c] [-q] [-r] [-E encoding] [-t len] [-D] [-w size] [-s lim] [-h]\n" << " f: path to the input file if processing single-end seqeunces.\n" << " i: input file type, either 'bustard' for the Illumina BUSTARD format, 'bam', 'fastq' (default), or 'gzfastq' for gzipped FASTQ.\n" << " p: path to a directory of single-end Illumina files.\n" << " 1: first input file in a set of paired-end sequences.\n" << " 2: second input file in a set of paired-end sequences.\n" << " P: specify that input is paired (for use with '-p').\n" << " I: specify that the paired-end reads are interleaved in single files.\n" << " o: path to output the processed files.\n" << " y: output type, either 'fastq' or 'fasta' (default fastq).\n" << " b: a list of barcodes for this run.\n" << " c: clean data, remove any read with an uncalled base.\n" << " q: discard reads with low quality scores.\n" << " r: rescue barcodes.\n" << " t: truncate final read length to this value.\n" << " E: specify how quality scores are encoded, 'phred33' (Illumina 1.8+, Sanger) or 'phred64' (Illumina 1.3 - 1.5, default).\n" << " D: capture discarded reads to a file.\n" << " w: set the size of the sliding window as a fraction of the read length, between 0 and 1 (default 0.15).\n" << " s: set the score limit. If the average score within the sliding window drops below this value, the read is discarded (default 10).\n" << " h: display this help messsage.\n\n" << " Barcode options:\n" << " --inline_null: barcode is inline with sequence, occurs only on single-end read (default).\n" << " --index_null: barcode is provded in FASTQ header (Illumina i5 or i7 read).\n" << " --null_index: barcode is provded in FASTQ header (Illumina i7 read if both i5 and i7 read are provided).\n" << " --inline_inline: barcode is inline with sequence, occurs on single and paired-end read.\n" << " --index_index: barcode is provded in FASTQ header (Illumina i5 and i7 reads).\n" << " --inline_index: barcode is inline with sequence on single-end read and occurs in FASTQ header (from either i5 or i7 read).\n" << " --index_inline: barcode occurs in FASTQ header (Illumina i5 or i7 read) and is inline with single-end sequence (for single-end data) on paired-end read (for paired-end data).\n\n" << " Adapter options:\n" << " --adapter_1 : provide adaptor sequence that may occur on the first read for filtering.\n" << " --adapter_2 : provide adaptor sequence that may occur on the paired-read for filtering.\n" << " --adapter_mm : number of mismatches allowed in the adapter sequence.\n\n" << " Output options:\n" << " --retain_header: retain unmodified FASTQ headers in the output.\n" << " --merge: if no barcodes are specified, merge all input files into a single output file (or single pair of files).\n\n" << " Advanced options:\n" << " --no_read_trimming: do not trim low quality reads, just discard them.\n" << " --len_limit : when trimming sequences, specify the minimum length a sequence must be to keep it (default 31bp).\n" << " --filter_illumina: discard reads that have been marked by Illumina's chastity/purity filter as failing.\n" << " --barcode_dist: provide the distace between barcodes to allow for barcode rescue (default 2)\n" << " --mate-pair: raw reads are circularized mate-pair data, first read will be reverse complemented.\n" << " --no_overhang: data does not contain an overhang nucleotide between barcode and seqeunce.\n"; exit(0); } stacks-1.44/src/process_shortreads.h000644 000765 000024 00000005561 12764047443 020400 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2011-2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __PROCESS_SHORTREADS_H__ #define __PROCESS_SHORTREADS_H__ #include #include // Process command-line options #include // Open/Read contents of a directory #include #include #include #include #include #include using std::stringstream; using std::istream; using std::ofstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include using std::vector; #include using std::map; #include using std::set; #include using std::pair; #include "constants.h" #include "clean.h" #include "file_io.h" #include "utils.h" #include "log_utils.h" #include "write.h" #include "BustardI.h" // Reading input files in Tab-separated Bustard format #include "FastqI.h" // Reading input files in FASTQ format #include "gzFastq.h" // Reading gzipped input files in FASTQ format #include "BamUnalignedI.h" // Reading data from unaligned BAM files void help( void ); void version( void ); int parse_command_line(int, char **); template int process_reads(string, set &, set &, map &, map &, map > &); template int process_paired_reads(string, string, set &, set &, map &, map &, map &, map &, map &, map > &); int process_singlet(Read *, bool, map &, map &); int dist(const char *, char *); int print_results(int, char **, vector &, map > &, map > &); int compare_barcodes(pair, pair); #endif // __PROCESS_SHORTREADS_H__ stacks-1.44/src/pstacks.cc000644 000765 000024 00000102026 13001006117 016236 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // pstacks -- search an existing set of stacks for polymorphisms // #include "pstacks.h" // // Global variables to hold command-line options. // FileT in_file_type; string in_file; FileT out_file_type; string out_path; int sql_id = 0; int min_stack_cov = 3; double req_pct_aln = 0.85; bool keep_sec_alns = false; int num_threads = 1; // // For use with the multinomial model to call fixed nucleotides. // modelt model_type = snp; double alpha = 0.05; double bound_low = 0.0; double bound_high = 1.0; double p_freq = 0.5; double barcode_err_freq = 0.0; double heterozygote_limit = -3.84; double homozygote_limit = 3.84; int main (int argc, char* argv[]) { parse_command_line(argc, argv); cerr << "Min depth of coverage to report a stack: " << min_stack_cov << "\n" << "Model type: "; switch (model_type) { case snp: cerr << "SNP\n"; break; case fixed: cerr << "Fixed\n"; break; case bounded: cerr << "Bounded; lower epsilon bound: " << bound_low << "; upper bound: " << bound_high << "\n"; break; } cerr << "Alpha significance level for model: " << alpha << "\n"; // // Set limits to call het or homozygote according to chi-square distribution with one // degree of freedom: // http://en.wikipedia.org/wiki/Chi-squared_distribution#Table_of_.CF.872_value_vs_p-value // if (alpha == 0.1) { heterozygote_limit = -2.71; homozygote_limit = 2.71; } else if (alpha == 0.05) { heterozygote_limit = -3.84; homozygote_limit = 3.84; } else if (alpha == 0.01) { heterozygote_limit = -6.64; homozygote_limit = 6.64; } else if (alpha == 0.001) { heterozygote_limit = -10.83; homozygote_limit = 10.83; } // // Set the number of OpenMP parallel threads to execute. // #ifdef _OPENMP omp_set_num_threads(num_threads); #endif HashMap* radtags = new HashMap(); map unique; load_radtags(in_file, *radtags); reduce_radtags(*radtags, unique); for (auto& stack : *radtags) for (Seq* read : stack.second) delete read; delete radtags; //dump_stacks(unique); map merged; populate_merged_tags(unique, merged); //dump_merged_stacks(merged); // Call the consensus sequence again, now that remainder tags have been merged. cerr << "Identifying polymorphic sites and calling consensus sequences..."; call_consensus(merged, unique, true); cerr << "done.\n"; count_raw_reads(unique, merged); calc_coverage_distribution(unique, merged); cerr << "Writing loci, SNPs, alleles to '" << out_path << "...'\n"; write_results(merged, unique); return 0; } int call_alleles(MergedStack *mtag, vector &reads) { int row; int height = reads.size(); string allele; char base; vector::iterator snp; DNANSeq *d; for (row = 0; row < height; row++) { allele.clear(); uint snp_cnt = 0; for (snp = mtag->snps.begin(); snp != mtag->snps.end(); snp++) { if ((*snp)->type != snp_type_het) continue; snp_cnt++; d = reads[row]; base = (*d)[(*snp)->col]; // // Check to make sure the nucleotide at the location of this SNP is // of one of the two possible states the multinomial model called. // if (base == (*snp)->rank_1 || base == (*snp)->rank_2) allele += base; else break; } if (snp_cnt > 0 && allele.length() == snp_cnt) mtag->alleles[allele]++; } return 0; } int call_consensus(map &merged, map &unique, bool invoke_model) { // // OpenMP can't parallelize random access iterators, so we convert // our map to a vector of integer keys. // map::iterator it; vector keys; for (it = merged.begin(); it != merged.end(); it++) keys.push_back(it->first); int i; #pragma omp parallel private(i) { #pragma omp for schedule(dynamic) for (i = 0; i < (int) keys.size(); i++) { MergedStack *mtag; PStack *utag; mtag = merged[keys[i]]; // // Create a two-dimensional array, each row containing one read. For // each unique tag that has been merged together, add the sequence for // that tag into our array as many times as it originally occurred. // vector::iterator j; vector reads; for (j = mtag->utags.begin(); j != mtag->utags.end(); j++) { utag = unique[*j]; for (uint k = 0; k < utag->count; k++) { reads.push_back(utag->seq); } } // // Iterate over each column of the array and call the consensus base. // int row, col; int length = reads[0]->size(); int height = reads.size(); string con; map nuc; map::iterator max, n; DNANSeq *d; for (col = 0; col < length; col++) { nuc['A'] = 0; nuc['C'] = 0; nuc['G'] = 0; nuc['T'] = 0; nuc['N'] = 0; for (row = 0; row < height; row++) { d = reads[row]; if (nuc.count((*d)[col])) nuc[(*d)[col]]++; } // // Find the base with a plurality of occurances and call it. // max = nuc.end(); for (n = nuc.begin(); n != nuc.end(); n++) { if (n->first == 'N') continue; if (max == nuc.end() || n->second > max->second) max = n; } con += max->second == 0 ? 'N' : max->first; // // Search this column for the presence of a SNP // if (invoke_model) switch(model_type) { case snp: call_multinomial_snp(mtag, col, nuc, true); break; case bounded: call_bounded_multinomial_snp(mtag, col, nuc, true); break; case fixed: call_multinomial_fixed(mtag, col, nuc); break; } } if (invoke_model) { call_alleles(mtag, reads); if (model_type == fixed) { // // Mask nucleotides that are not fixed. // vector::iterator s; for (s = mtag->snps.begin(); s != mtag->snps.end(); s++) { if ((*s)->type == snp_type_unk) con.replace((*s)->col, 1, "N"); } } } mtag->add_consensus(con.c_str()); // // If SNPs were called at this locus but no alleles could be determined, // blacklist this tag. This can occur if there are two many uncalled bases // in the locus (Ns), such that haplotypes can't be consistently read // due to the presence of the Ns in the reads. // if (mtag->alleles.empty()) for (uint j = 0; j < mtag->snps.size(); j++) if (mtag->snps[j]->type == snp_type_het) { mtag->blacklisted = 1; break; } } } return 0; } double calc_coverage_distribution(map &unique, map &merged) { map::iterator it; vector::iterator k; PStack *tag; double depth = 0.0; double total = 0.0; double sum = 0.0; double mean = 0.0; double max = 0.0; double stdev = 0.0; for (it = merged.begin(); it != merged.end(); it++) { depth = 0.0; for (k = it->second->utags.begin(); k != it->second->utags.end(); k++) { tag = unique[*k]; depth += tag->count; } if (depth < min_stack_cov) continue; if (depth > max) max = depth; sum += depth; total++; } mean = sum / total; // // Calculate the standard deviation // for (it = merged.begin(); it != merged.end(); it++) { depth = 0.0; for (k = it->second->utags.begin(); k != it->second->utags.end(); k++) { tag = unique[*k]; depth += tag->count; } if (depth < min_stack_cov) continue; sum += pow((depth - mean), 2); } stdev = sqrt(sum / (total - 1)); cerr << " Mean coverage depth is " << mean << "; Std Dev: " << stdev << "; Max: " << max << "\n"; return mean; } int count_raw_reads(map &unique, map &merged) { map::iterator it; vector::iterator k; PStack *tag; long int m = 0; for (it = merged.begin(); it != merged.end(); it++) { for (k = it->second->utags.begin(); k != it->second->utags.end(); k++) { tag = unique[*k]; m += tag->count; } m += it->second->remtags.size(); } cerr << " Number of utilized reads " << m << "\n"; return 0; } int write_results(map &m, map &u) { map::iterator i; vector::iterator j; vector::iterator k; vector::iterator s; map::iterator t; MergedStack *tag_1; PStack *tag_2; stringstream sstr; bool gzip = (in_file_type == FileT::bam) ? true : false; // // Parse the input file name to create the output files // size_t pos_1 = in_file.find_last_of("/"); size_t pos_2 = in_file.find_last_of("."); string tag_file = out_path + in_file.substr(pos_1 + 1, (pos_2 - pos_1 - 1)) + ".tags.tsv"; string snp_file = out_path + in_file.substr(pos_1 + 1, (pos_2 - pos_1 - 1)) + ".snps.tsv"; string all_file = out_path + in_file.substr(pos_1 + 1, (pos_2 - pos_1 - 1)) + ".alleles.tsv"; string mod_file = out_path + in_file.substr(pos_1 + 1, (pos_2 - pos_1 - 1)) + ".models.tsv"; if (gzip) { tag_file += ".gz"; snp_file += ".gz"; all_file += ".gz"; mod_file += ".gz"; } // // Open the output files for writing. // gzFile gz_tags, gz_snps, gz_alle, gz_mods; ofstream tags, snps, alle, mods; if (gzip) { gz_tags = gzopen(tag_file.c_str(), "wb"); if (!gz_tags) { cerr << "Error: Unable to open gzipped tag file '" << tag_file << "': " << strerror(errno) << ".\n"; exit(1); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_tags, libz_buffer_size); #endif gz_mods = gzopen(mod_file.c_str(), "wb"); if (!gz_mods) { cerr << "Error: Unable to open gzipped tag file '" << tag_file << "': " << strerror(errno) << ".\n"; exit(1); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_mods, libz_buffer_size); #endif gz_snps = gzopen(snp_file.c_str(), "wb"); if (!gz_snps) { cerr << "Error: Unable to open gzipped snps file '" << snp_file << "': " << strerror(errno) << ".\n"; exit(1); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_snps, libz_buffer_size); #endif gz_alle = gzopen(all_file.c_str(), "wb"); if (!gz_alle) { cerr << "Error: Unable to open gzipped alleles file '" << all_file << "': " << strerror(errno) << ".\n"; exit(1); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_alle, libz_buffer_size); #endif } else { tags.open(tag_file.c_str()); if (tags.fail()) { cerr << "Error: Unable to open tag file for writing.\n"; exit(1); } mods.open(mod_file.c_str()); if (mods.fail()) { cerr << "Error: Unable to open tag file for writing.\n"; exit(1); } snps.open(snp_file.c_str()); if (snps.fail()) { cerr << "Error: Unable to open SNPs file for writing.\n"; exit(1); } alle.open(all_file.c_str()); if (alle.fail()) { cerr << "Error: Unable to open allele file for writing.\n"; exit(1); } } // // Record the version of Stacks used and the date generated as a comment in the catalog. // // Obtain the current date. // stringstream log; time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%F %T", timeinfo); log << "# pstacks version " << VERSION << "; generated on " << date << "\n"; if (gzip) { gzputs(gz_tags, log.str().c_str()); gzputs(gz_mods, log.str().c_str()); gzputs(gz_snps, log.str().c_str()); gzputs(gz_alle, log.str().c_str()); } else { tags << log.str(); mods << log.str(); snps << log.str(); alle << log.str(); } int id; char *buf; // = new char[m.begin()->second->len + 1]; int wrote = 0; int excluded = 0; int blacklisted = 0; for (i = m.begin(); i != m.end(); i++) { tag_1 = i->second; float total = 0; for (k = tag_1->utags.begin(); k != tag_1->utags.end(); k++) total += u[*k]->count; if (total < min_stack_cov) { excluded++; continue; } // // Calculate the log likelihood of this merged stack. // tag_1->gen_matrix(u); tag_1->calc_likelihood(); wrote++; if (tag_1->blacklisted) blacklisted++; // First write the consensus sequence sstr << "0" << "\t" << sql_id << "\t" << tag_1->id << "\t" << tag_1->loc.chr << "\t" << tag_1->loc.bp << "\t" << (tag_1->loc.strand == strand_plus ? "+" : "-") << "\t" << "consensus\t" << "\t\t" << tag_1->con << "\t" << tag_1->deleveraged << "\t" << tag_1->blacklisted << "\t" << tag_1->lumberjackstack << "\t" << tag_1->lnl << "\n"; // // Write a sequence recording the output of the SNP model for each nucleotide. // sstr << "0" << "\t" << sql_id << "\t" << tag_1->id << "\t" << "\t" << "\t" << "\t" << "model\t" << "\t" << "\t"; for (s = tag_1->snps.begin(); s != tag_1->snps.end(); s++) { switch((*s)->type) { case snp_type_het: sstr << "E"; break; case snp_type_hom: sstr << "O"; break; default: sstr << "U"; break; } } sstr << "\t" << "\t" << "\t" << "\t" << "\n"; if (gzip) gzputs(gz_tags, sstr.str().c_str()); else tags << sstr.str(); if (gzip) gzputs(gz_mods, sstr.str().c_str()); else mods << sstr.str(); sstr.str(""); // Now write out the components of each unique tag merged into this one. id = 0; for (k = tag_1->utags.begin(); k != tag_1->utags.end(); k++) { tag_2 = u[*k]; buf = tag_2->seq->seq(); for (j = tag_2->map.begin(); j != tag_2->map.end(); j++) { sstr << "0" << "\t" << sql_id << "\t" << tag_1->id << "\t\t\t\t" << "primary\t" << id << "\t" << *j << "\t" << buf << "\t\t\t\t\n"; if (gzip) gzputs(gz_tags, sstr.str().c_str()); else tags << sstr.str(); sstr.str(""); } id++; delete [] buf; } // // Write out the model calls for each nucleotide in this locus. // for (s = tag_1->snps.begin(); s != tag_1->snps.end(); s++) { sstr << "0" << "\t" << sql_id << "\t" << tag_1->id << "\t" << (*s)->col << "\t"; switch((*s)->type) { case snp_type_het: sstr << "E\t"; break; case snp_type_hom: sstr << "O\t"; break; default: sstr << "U\t"; break; } sstr << std::fixed << std::setprecision(2) << (*s)->lratio << "\t" << (*s)->rank_1 << "\t" << (*s)->rank_2 << "\t\t\n"; } if (gzip) gzputs(gz_snps, sstr.str().c_str()); else snps << sstr.str(); sstr.str(""); // Write the expressed alleles seen for the recorded SNPs and // the percentage of tags a particular allele occupies. // char pct[id_len]; for (t = tag_1->alleles.begin(); t != tag_1->alleles.end(); t++) { sprintf(pct, "%.2f", ((t->second/total) * 100)); sstr << "0" << "\t" << sql_id << "\t" << tag_1->id << "\t" << t->first << "\t" << pct << "\t" << t->second << "\n"; } if (gzip) gzputs(gz_alle, sstr.str().c_str()); else alle << sstr.str(); sstr.str(""); } if (gzip) { gzclose(gz_tags); gzclose(gz_mods); gzclose(gz_snps); gzclose(gz_alle); } else { tags.close(); mods.close(); snps.close(); alle.close(); } cerr << " Wrote " << wrote << " loci, excluded " << excluded << " loci due to insuffient depth of coverage; blacklisted " << blacklisted << " loci.\n"; return 0; } int populate_merged_tags(map &unique, map &merged) { map::iterator i; map::iterator it_new, it_old; map > locations; map >::iterator k; set::iterator s; char id[id_len]; PStack *u; MergedStack *m; int global_id = 1; // // Create a map of each unique Stack that has been aligned to the same genomic location. // for (i = unique.begin(); i != unique.end(); i++) { snprintf(id, id_len - 1, "%s|%d|%s", i->second->loc.chr, i->second->loc.bp, i->second->loc.strand == strand_plus ? "+" : "-"); locations[id].insert(i->second->id); } it_old = merged.begin(); for (k = locations.begin(); k != locations.end(); k++) { m = new MergedStack; m->id = global_id; // // Record the consensus and physical location for this stack. // s = k->second.begin(); m->add_consensus(unique[*s]->seq); m->loc.set(unique[*s]->loc.chr, unique[*s]->loc.bp, unique[*s]->loc.strand); // // Record the individual stacks that were aligned together. // for (; s != k->second.end(); s++) { u = unique[*s]; m->count += u->count; m->utags.push_back(u->id); } // // Insert the new MergedStack giving a hint as to which position // to insert it at. // it_new = merged.insert(it_old, pair(global_id, m)); it_old = it_new; global_id++; } cerr << " Merged " << unique.size() << " unique Stacks into " << merged.size() << " loci.\n"; return 0; } // // This function assumes that there may be identical reads, mapped to multiple // places in the genome. In this case, reads are broken down by read ID // and split into different Stack objects. // int reduce_radtags(HashMap &radtags, map &unique) { HashMap::iterator it; vector::iterator sit; PStack *u; int global_id = 1; for (it = radtags.begin(); it != radtags.end(); it++) { // // Make sure there aren't any reads of identical sequence that have been mapped to // different genomic locations. // map locations; map::iterator lit; for (sit = (*it).second.begin(); sit != (*it).second.end(); sit++) locations[(*sit)->loc_str]++; for (lit = locations.begin(); lit != locations.end(); lit++) { // // Populate a Stack object for this unique radtag. // u = new PStack; u->id = global_id; u->count = lit->second; u->add_seq(&it->first); // // Record the physical location of this stack. // for (sit = (*it).second.begin(); sit != (*it).second.end(); sit++) { if (strcmp((*sit)->loc_str, lit->first.c_str()) == 0) { u->add_id((*sit)->id); u->loc.set((*sit)->loc.chr, (*sit)->loc.bp, (*sit)->loc.strand); } } unique[u->id] = u; global_id++; } } cerr << " " << radtags.size() << " unique stacks were aligned to " << unique.size() << " genomic locations.\n"; return 0; } // // We expect tags to have already been aligned to a reference genome. Therefore, the tags // are identified by their chromosome and basepair location. // int load_radtags(string in_file, HashMap &radtags) { Input *fh = NULL; Seq c; if (in_file_type == FileT::bowtie) fh = new Bowtie(in_file.c_str()); else if (in_file_type == FileT::sam) fh = new Sam(in_file.c_str()); else if (in_file_type == FileT::bam) fh = new Bam(in_file.c_str()); else if (in_file_type == FileT::tsv) fh = new Tsv(in_file.c_str()); cerr << "Parsing " << in_file.c_str() << "\n"; int secondary = 0; int supplementary = 0; int below_req_aln = 0; int i = 0; cerr << "Loading aligned sequences..."; while ((fh->next_seq(c)) != 0) { if (i % 1000000 == 0 && i>0) cerr << i/1000000 << "M..."; i++; switch (c.aln_type) { case sec_aln: secondary++; if (!keep_sec_alns) continue; break; case sup_aln: supplementary++; continue; break; case pri_aln: default: break; } if (c.pct_aln < req_pct_aln) { below_req_aln++; continue; } HashMap::iterator element = radtags.insert({DNANSeq(strlen(c.seq), c.seq), vector()}).first; element->second.push_back(new Seq(c)); Seq& the_seq = *element->second.back(); if (the_seq.seq != NULL) { delete[] the_seq.seq; the_seq.seq = NULL; } if (the_seq.qual != NULL) { delete[] the_seq.qual; the_seq.qual = NULL; } } cerr << "done\n"; if (i == 0) { cerr << "Error: Unable to load data from '" << in_file.c_str() << "'.\n"; exit(-1); } cerr << "Loaded " << i << " sequence reads; " << "identified " << radtags.size() << " unique stacks from those reads.\n" << " Discarded " << below_req_aln << " reads where the aligned percentage of the read was too low.\n"; if (keep_sec_alns) cerr << " Kept " << secondary << " secondarily aligned reads (reads may be present in the data set more than once).\n"; else cerr << " Discarded " << secondary << " secondarily aligned reads (primary alignments were retained).\n"; cerr << " Discarded " << supplementary << " supplementary aligned (chimeric) reads.\n"; // // Close the file and delete the Input object. // delete fh; return 0; } int dump_stacks(map &u) { map::iterator it; vector::iterator fit; vector >::iterator pit; vector::iterator mit; for (it = u.begin(); it != u.end(); it++) { cerr << "Stack ID: " << (*it).second->id << "\n" << " Seq: " << (*it).second->seq->seq() << "\n" << " IDs: "; for (fit = (*it).second->map.begin(); fit != (*it).second->map.end(); fit++) cerr << *fit << " "; cerr << "\n\n"; } return 0; } int dump_merged_stacks(map &m) { map::iterator it; vector >::iterator pit; vector::iterator fit; for (it = m.begin(); it != m.end(); it++) { cerr << "MergedStack ID: " << it->second->id << "\n" << " Consensus: "; if (it->second->con != NULL) cerr << it->second->con << "\n"; else cerr << "\n"; cerr << " IDs: "; for (fit = it->second->utags.begin(); fit != it->second->utags.end(); fit++) cerr << (*fit) << " "; cerr << "\n" << " Distances: "; for (pit = it->second->dist.begin(); pit != it->second->dist.end(); pit++) cerr << (*pit).first << ": " << (*pit).second << ", "; cerr << "\n\n"; } return 0; } int parse_command_line(int argc, char* argv[]) { int c; while (1) { static struct option long_options[] = { {"help", no_argument, NULL, 'h'}, {"version", no_argument, NULL, 'v'}, {"infile_type", required_argument, NULL, 't'}, {"outfile_type", required_argument, NULL, 'y'}, {"file", required_argument, NULL, 'f'}, {"outpath", required_argument, NULL, 'o'}, {"id", required_argument, NULL, 'i'}, {"min_cov", required_argument, NULL, 'm'}, {"pct_aln", required_argument, NULL, 'a'}, {"keep_sec_aln", required_argument, NULL, 'k'}, {"num_threads", required_argument, NULL, 'p'}, {"bc_err_freq", required_argument, NULL, 'e'}, {"model_type", required_argument, NULL, 'T'}, {"bound_low", required_argument, NULL, 'L'}, {"bound_high", required_argument, NULL, 'U'}, {"alpha", required_argument, NULL, 'A'}, {0, 0, 0, 0} }; // getopt_long stores the option index here. int option_index = 0; c = getopt_long(argc, argv, "hkvOT:a:A:L:U:f:o:i:e:p:m:s:f:t:y:", long_options, &option_index); // Detect the end of the options. if (c == -1) break; switch (c) { case 'h': help(); break; case 't': if (strcmp(optarg, "bowtie") == 0) in_file_type = FileT::bowtie; else if (strcmp(optarg, "sam") == 0) in_file_type = FileT::sam; else if (strcmp(optarg, "bam") == 0) in_file_type = FileT::bam; else if (strcmp(optarg, "tsv") == 0) in_file_type = FileT::tsv; else in_file_type = FileT::unknown; break; case 'y': if (strcmp(optarg, "sam") == 0) out_file_type = FileT::sam; else out_file_type = FileT::sql; break; case 'f': in_file = optarg; break; case 'o': out_path = optarg; break; case 'i': sql_id = is_integer(optarg); if (sql_id < 0) { cerr << "SQL ID (-i) must be an integer, e.g. 1, 2, 3\n"; help(); } break; case 'm': min_stack_cov = atoi(optarg); break; case 'a': req_pct_aln = is_double(optarg); if (req_pct_aln > 1) req_pct_aln = req_pct_aln / 100; if (req_pct_aln < 0 || req_pct_aln > 1.0) { cerr << "Unable to parse the required alignment percentage.\n"; help(); } break; case 'k': keep_sec_alns = true; break; case 'e': barcode_err_freq = atof(optarg); break; case 'T': if (strcmp(optarg, "snp") == 0) { model_type = snp; } else if (strcmp(optarg, "fixed") == 0) { model_type = fixed; } else if (strcmp(optarg, "bounded") == 0) { model_type = bounded; } else { cerr << "Unknown model type specified '" << optarg << "'\n"; help(); } case 'L': bound_low = atof(optarg); break; case 'U': bound_high = atof(optarg); break; case 'A': alpha = atof(optarg); break; case 'p': num_threads = atoi(optarg); break; case 'v': version(); break; case '?': // getopt_long already printed an error message. help(); break; default: cerr << "Unknown command line option '" << (char) c << "'\n"; help(); abort(); } } if (alpha != 0.1 && alpha != 0.05 && alpha != 0.01 && alpha != 0.001) { cerr << "SNP model alpha significance level must be either 0.1, 0.05, 0.01, or 0.001.\n"; help(); } if (bound_low != 0 && (bound_low < 0 || bound_low >= 1.0)) { cerr << "SNP model lower bound must be between 0.0 and 1.0.\n"; help(); } if (bound_high != 1 && (bound_high <= 0 || bound_high > 1.0)) { cerr << "SNP model upper bound must be between 0.0 and 1.0.\n"; help(); } if (bound_low > 0 || bound_high < 1.0) { model_type = bounded; } if (in_file.length() == 0 || in_file_type == FileT::unknown) { cerr << "You must specify an input file of a supported type.\n"; help(); } if (out_path.length() == 0) out_path = "."; if (out_path.at(out_path.length() - 1) != '/') out_path += "/"; if (model_type == fixed && barcode_err_freq == 0) { cerr << "You must specify the barcode error frequency.\n"; help(); } return 0; } void version() { std::cerr << "pstacks " << VERSION << "\n\n"; exit(0); } void help() { std::cerr << "pstacks " << VERSION << "\n" << "pstacks -t file_type -f file_path [-o path] [-i id] [-m min_cov] [-p num_threads] [-h]" << "\n" << " t: input file Type. Supported types: bowtie, sam, or bam.\n" << " f: input file path.\n" << " o: output path to write results.\n" << " i: SQL ID to insert into the output to identify this sample.\n" << " m: minimum depth of coverage to report a stack (default 3).\n" << " p: enable parallel execution with num_threads threads.\n" << " h: display this help messsage.\n" << " --pct_aln : require read alignments to use at least this percentage of the read (default 85%).\n" << " --keep_sec_alns: keep secondary alignments (default: false, only keep primary alignments).\n" << " Model options:\n" << " --model_type : either 'snp' (default), 'bounded', or 'fixed'\n" << " For the SNP or Bounded SNP model:\n" << " --alpha : chi square significance level required to call a heterozygote or homozygote, either 0.1, 0.05 (default), 0.01, or 0.001.\n" << " For the Bounded SNP model:\n" << " --bound_low : lower bound for epsilon, the error rate, between 0 and 1.0 (default 0).\n" << " --bound_high : upper bound for epsilon, the error rate, between 0 and 1.0 (default 1).\n" << " For the Fixed model:\n" << " --bc_err_freq : specify the barcode error frequency, between 0 and 1.0.\n"; exit(0); } stacks-1.44/src/pstacks.h000644 000765 000024 00000005534 12764047443 016134 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __PSTACKS_H__ #define __PSTACKS_H__ #ifdef _OPENMP #include // OpenMP library #endif #include #include // Support for gzipped output files. #include // Process command-line options #include #include #include #include using std::ofstream; using std::stringstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include // std::setprecision #include using std::unordered_map; #include using std::vector; #include using std::map; #include using std::set; #include using std::pair; #include "config.h" #include "constants.h" #include "stacks.h" // Major data structures for holding stacks #include "mstack.h" #include "kmers.h" #include "utils.h" #include "models.h" // Contains maximum likelihood statistical models. #include "Tsv.h" // Reading input files in Tab-separated values format #include "BowtieI.h" // Reading input files in Bowtie format #include "SamI.h" // Reading input files in SAM format #include "BamI.h" // Reading input files in BAM format #include "DNANSeq.h" const int barcode_size = 5; #ifdef HAVE_SPARSEHASH typedef sparse_hash_map > HashMap; #else typedef unordered_map > HashMap; #endif void help( void ); void version( void ); int parse_command_line(int, char**); int load_radtags(string, HashMap &); int reduce_radtags(HashMap &, map &); int populate_merged_tags(map &, map &); int call_consensus(map &, map &, bool); int call_alleles(MergedStack *, vector &); int count_raw_reads(map &, map &); double calc_coverage_distribution(map &, map &); int write_results(map &, map &); // // Debugging // int dump_stacks(map &); int dump_merged_stacks(map &); #endif // __PSTACKS_H__ stacks-1.44/src/renz.cc000644 000765 000024 00000022030 12750704270 015557 0ustar00catchenstaff000000 000000 #include "renz.h" using namespace std; // // First line of static array contains each enzyme's cut sites. Second // line is the reverse complement of each cut site. // const char *aciI[] = {"CGC", "CGG", // C/CGC, AciI "GCG", "CCG"}; const char *ageI[] = {"CCGGT", // A/CCGGT, AgeI "ACCGG"}; const char *aluI[] = {"CT", // AG/CT, AluI "AG"}; const char *apeKI[] = {"CAGC", "CTGC", // G/CWGC, ApeKI; W=A or T "GTCG", "GACG"}; const char *apoI[] = {"AATTC", "AATTT", // R/AATTY, ApoI (also known as XapI) "GAATT", "AAATT"}; const char *aseI[] = {"TAAT", // AT/TAAT, AseI "ATTA"}; const char *bamHI[] = {"GATCC", // G/GATCC, BamHI "GGATC"}; const char *bfaI[] = {"TAG", // C/TAG, BfaI "CTA"}; const char *bgIII[] = {"GATCT", // A/GATCT, BgIII "AGATC"}; const char *bsaHI[] = {"CGCC", "CGTC", // GR/CGYC, BsaHI "GGCG", "GACG"}; const char *bspDI[] = {"CGAT", // AT/CGAT, BspDI "ATCG"}; const char *bstYI[] = {"GATCC", "GATCT", // R/GATCY, BstYI (also known as PsuI) "GGATC", "AGATC"}; const char *claI[] = {"CGAT", // AT/CGAT, ClaI "ATCG"}; const char *csp6I[] = {"TAC", // G/TAC, Csp6I "GTA"}; const char *ddeI[] = {"TAAG", "TCAG", "TGAG", "TTAG", // C/TNAG, DdeI "CTTA", "CTGA", "CTCA", "CTAA"}; const char *dpnII[] = {"GATC", // GATC, DpnII "GATC"}; const char *eaeI[] = {"GGCCA", "GGCCG", // Y/GGCCR, EaeI "TGGCC", "CGGCC"}; const char *ecoRI[] = {"AATTC", // G/AATTC, EcoRI "GAATT"}; const char *ecoRV[] = {"ATC", // GAT/ATC, EcoRV "GAT"}; const char *ecoT22I[] = {"TGCAT", // A/TGCAT, EcoT22I "ATGCA"}; const char *hindIII[] = {"AGCTT", // A/AGCTT, HindIII "TCGAA"}; const char *hpaII[] = {"CGG", // C/CGG, HpaII "CCG"}; const char *kpnI[] = {"GTACC", // C/CATGG, KpnI "GGTAC"}; const char *mluCI[] = {"AATT", // AATT, MluCI "AATT"}; const char *mseI[] = {"TAA", // T/TAA, MseI "TTA"}; const char *mspI[] = {"CGG", // C/CGG, MspI "CCG"}; const char *ncoI[] = {"CATGG", // C/CATGG, NcoI "CCATG"}; const char *ndeI[] = {"TATG", // CA/TATG, NdeI "CATA"}; const char *nheI[] = {"CTAGC", // G/CTAGC, NheI "GCTAG"}; const char *nlaIII[] = {"CATG", // CATG, NlaIII "CATG"}; const char *notI[] = {"GGCCGC", // GC/GGCCGC, NotI "GCGGCC"}; const char *nsiI[] = {"TGCAT", // ATGCA/T, NsiI "ATGCA"}; const char *pstI[] = {"TGCAG", // CTGCA/G, PstI "CTGCA"}; const char *rsaI[] = {"AC", // GT/AC, RsaI "GT"}; const char *sacI[] = {"AGCTC", // GAGCT/C, SacI "GAGCT"}; const char *sau3AI[] = {"GATC", // GATC, Sau3AI "GATC"}; const char *sbfI[] = {"TGCAGG", // CCTGCA/GG, SbfI "CCTGCA"}; const char *sexAI[] = {"CCAGGT", "CCTGGT", // A/CCWGGT, SexAI; W=A or T "ACCTGG", "ACCAGG"}; const char *sgrAI[] = {"CCGGCG", "CCGGTG", // CR/CCGGYG, SgrAI; R=A or G; Y=C or T "CGCCGG", "CACCGG"}; const char *speI[] = {"CTAGT", // A/CTAGT, SpeI "ACTAG"}; const char *sphI[] = {"CATGC", // GCATG/C, SphI "GCATG"}; const char *taqI[] = {"CGA", // T/CGA, TaqI "TCG"}; const char *xbaI[] = {"CTAGA", // T/CTAGA, XbaI "TCTAG"}; const char *xhoI[] = {"TCGAG", // C/TCGAG, XhoI "CTCGA"}; void initialize_renz(map &renz, map &renz_cnt, map &renz_len) { renz["sbfI"] = sbfI; // CCTGCA/GG, SbfI renz["pstI"] = pstI; // CTGCA/G, PstI renz["notI"] = notI; // GC/GGCCGC, NotI renz["ecoRI"] = ecoRI; // G/AATTC, EcoRI renz["sgrAI"] = sgrAI; // CR/CCGGYG, SgrAI; R=A or G; Y=C or T renz["apeKI"] = apeKI; // G/CWGC, ApeKI; W=A or T renz["hindIII"] = hindIII; // A/AGCTT, HindIII renz["dpnII"] = dpnII; // GATC, DpnII renz["sphI"] = sphI; // GCATG/C, SphI renz["nlaIII"] = nlaIII; // CATG, NlaIII renz["mluCI"] = mluCI; // AATT, MluCI renz["ecoT22I"] = ecoT22I; // A/TGCAT, EcoT22I renz["ndeI"] = ndeI; // CA/TATG, NdeI renz["nsiI"] = nsiI; // ATGCA/T, NsiI renz["mseI"] = mseI; // T/TAA, MseI renz["mspI"] = mspI; // C/CGG, MspI renz["sexAI"] = sexAI; // A/CCWGGT, SexAI; W=A or T renz["sau3AI"] = sau3AI; // GATC, Sau3AI renz["bamHI"] = bamHI; // G/GATCC, BamHI renz["xbaI"] = xbaI; // T/CTAGA, XbaI renz["eaeI"] = eaeI; // Y/GGCCR, EaeI renz["taqI"] = taqI; // T/CGA, TaqI renz["claI"] = claI; // AT/CGAT, ClaI renz["nheI"] = nheI; // G/CTAGC, NheI renz["speI"] = speI; // A/CTAGT, SpeI renz["apoI"] = apoI; // R/AATTY, ApoI, XapI renz["bstYI"] = bstYI; // R/GATCY, BstYI, PsuI renz["xhoI"] = xhoI; // C/TCGAG, XhoI renz["sacI"] = sacI; // GAGCT/C, SacI renz["bgIII"] = bgIII; // A/GATCT, BgIII renz["ecoRV"] = ecoRV; // GAT/ATC, EcoRV renz["kpnI"] = kpnI; // C/CATGG, KpnI renz["ddeI"] = ddeI; // C/TNAG, DdeI renz["aluI"] = aluI; // AG/CT, AluI renz["ageI"] = ageI; // A/CCGGT, AgeI renz["rsaI"] = rsaI; // GT/AC, RsaI renz["aciI"] = aciI; // C/CGC, AciI renz["bfaI"] = bfaI; // C/TAG, BfaI renz["aseI"] = aseI; // AT/TAAT, AseI renz["bspDI"] = bspDI; // AT/CGAT, BspDI renz["csp6I"] = csp6I; // G/TAC, Csp6I renz["bsaHI"] = bsaHI; // GR/CGYC, BsaHI renz["hpaII"] = hpaII; // C/CGG, HpaII renz["ncoI"] = ncoI; // C/CATGG, NcoI renz_cnt["sbfI"] = 1; renz_cnt["pstI"] = 1; renz_cnt["notI"] = 1; renz_cnt["ecoRI"] = 1; renz_cnt["sgrAI"] = 2; renz_cnt["apeKI"] = 2; renz_cnt["hindIII"] = 1; renz_cnt["dpnII"] = 1; renz_cnt["sphI"] = 1; renz_cnt["nlaIII"] = 1; renz_cnt["mluCI"] = 1; renz_cnt["ecoT22I"] = 1; renz_cnt["ndeI"] = 1; renz_cnt["nsiI"] = 1; renz_cnt["mseI"] = 1; renz_cnt["mspI"] = 1; renz_cnt["sexAI"] = 2; renz_cnt["sau3AI"] = 1; renz_cnt["bamHI"] = 1; renz_cnt["xbaI"] = 1; renz_cnt["eaeI"] = 2; renz_cnt["taqI"] = 1; renz_cnt["claI"] = 1; renz_cnt["nheI"] = 1; renz_cnt["speI"] = 1; renz_cnt["apoI"] = 2; renz_cnt["bstYI"] = 2; renz_cnt["xhoI"] = 1; renz_cnt["sacI"] = 1; renz_cnt["bgIII"] = 1; renz_cnt["ecoRV"] = 1; renz_cnt["kpnI"] = 1; renz_cnt["ddeI"] = 4; renz_cnt["aluI"] = 1; renz_cnt["ageI"] = 1; renz_cnt["rsaI"] = 1; renz_cnt["aciI"] = 2; renz_cnt["bfaI"] = 1; renz_cnt["aseI"] = 1; renz_cnt["bspDI"] = 1; renz_cnt["csp6I"] = 1; renz_cnt["bsaHI"] = 2; renz_cnt["hpaII"] = 1; renz_cnt["ncoI"] = 1; renz_len["sbfI"] = 6; renz_len["pstI"] = 5; renz_len["notI"] = 6; renz_len["ecoRI"] = 5; renz_len["sgrAI"] = 6; renz_len["apeKI"] = 4; renz_len["hindIII"] = 5; renz_len["dpnII"] = 4; renz_len["sphI"] = 5; renz_len["nlaIII"] = 4; renz_len["mluCI"] = 4; renz_len["ecoT22I"] = 5; renz_len["ndeI"] = 4; renz_len["nsiI"] = 5; renz_len["mseI"] = 3; renz_len["mspI"] = 3; renz_len["sexAI"] = 6; renz_len["sau3AI"] = 4; renz_len["bamHI"] = 5; renz_len["xbaI"] = 5; renz_len["eaeI"] = 5; renz_len["taqI"] = 3; renz_len["claI"] = 4; renz_len["nheI"] = 5; renz_len["speI"] = 5; renz_len["apoI"] = 5; renz_len["bstYI"] = 5; renz_len["xhoI"] = 5; renz_len["sacI"] = 5; renz_len["bgIII"] = 5; renz_len["ecoRV"] = 3; renz_len["kpnI"] = 5; renz_len["ddeI"] = 4; renz_len["aluI"] = 2; renz_len["ageI"] = 5; renz_len["rsaI"] = 2; renz_len["aciI"] = 3; renz_len["bfaI"] = 3; renz_len["aseI"] = 4; renz_len["bspDI"] = 4; renz_len["csp6I"] = 3; renz_len["bsaHI"] = 4; renz_len["hpaII"] = 3; renz_len["ncoI"] = 5; } void initialize_renz_olap(map &renz_olap) { renz_olap["sbfI"] = 4; } stacks-1.44/src/renz.h000644 000765 000024 00000002075 12750704270 015430 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2011-2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __RENZ_H__ #define __RENZ_H__ #include #include using std::map; using std::string; void initialize_renz_olap(map &renz_olap); void initialize_renz(map &renz, map &renz_cnt, map &renz_len); #endif // __RENZ_H__ stacks-1.44/src/rxstacks.cc000644 000765 000024 00000165113 12774470747 016474 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2013-2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // rxstacks -- make model call corrections and haplotype corrections // across a population of samples. // #include "MetaPopInfo.h" #include "rxstacks.h" typedef MetaPopInfo::Sample Sample; // Global variables to hold command-line options. int num_threads = 1; int batch_id = 0; string in_path; string out_path; FileT in_file_type = FileT::sql; double confounded_limit = 0.75; bool filter_confounded = false; bool prune_haplotypes = false; int max_haplotype_cnt = 0; bool lnl_dist = false; bool filter_lnl = false; double lnl_limit = 0.0; bool verbose = false; // // For use with the multinomial model to call fixed nucleotides. // modelt model_type = snp; double alpha = 0.1; double bound_low = 0.0; double bound_high = 1.0; double p_freq = 0.5; double barcode_err_freq = 0.0; double heterozygote_limit = -2.71; double homozygote_limit = 2.71; const int barcode_size = 5; int main (int argc, char* argv[]) { parse_command_line(argc, argv); cerr << "Log liklihood filtering: " << (filter_lnl == true ? "on" : "off") << "; threshold: " << lnl_limit << "\n" << "Prune haplotypes: " << (prune_haplotypes == true ? "yes" : "no") << "\n" << "Filter confounded loci: " << (filter_confounded == true ? "yes" : "no") << "\n"; // // Set limits to call het or homozygote according to chi-square distribution with one // degree of freedom: // http://en.wikipedia.org/wiki/Chi-squared_distribution#Table_of_.CF.872_value_vs_p-value // if (alpha == 0.1) { heterozygote_limit = -2.71; homozygote_limit = 2.71; } else if (alpha == 0.05) { heterozygote_limit = -3.84; homozygote_limit = 3.84; } else if (alpha == 0.01) { heterozygote_limit = -6.64; homozygote_limit = 6.64; } else if (alpha == 0.001) { heterozygote_limit = -10.83; homozygote_limit = 10.83; } // // Set the number of OpenMP parallel threads to execute. // if (verbose) num_threads = 1; #ifdef _OPENMP omp_set_num_threads(num_threads); #endif MetaPopInfo mpopi; mpopi.init_directory(in_path); if (mpopi.samples().empty()) { cerr << "Error: Failed to find sample files in directory '" << in_path << "'.\n"; return -1; } // // Open and initialize the log files. // ofstream log_fh, log_snp_fh, log_hap_fh; init_log(argc, argv, log_fh, log_snp_fh, log_hap_fh); // // Load the catalog // stringstream catalog_file; map catalog; bool compressed = false; int res; catalog_file << in_path << "batch_" << batch_id << ".catalog"; if ((res = load_loci(catalog_file.str(), catalog, false, false, compressed)) == 0) { cerr << "Unable to load the catalog '" << catalog_file.str() << "'\n"; return 0; } in_file_type = compressed == true ? FileT::gzsql : FileT::sql; // // Let's fill in the SNP model calls to include both hets and homozygotes to // make it easier to iterate over them later. // fill_catalog_snps(catalog); // // Load matches to the catalog // vector > catalog_matches; vector samples_to_remove; set seen_samples; for (size_t i=0; i()); vector& m = catalog_matches.back(); const Sample& sample = mpopi.samples()[i]; load_catalog_matches(in_path + sample.name, m); if (m.size() == 0) { cerr << "Warning: Absent or malformed matches file '" << sample.name << ".matches.tsv(.gz)" <<"', excluding this sample from population analysis.\n"; samples_to_remove.push_back(i); catalog_matches.pop_back(); // n.b. Index shift will be resolved by the call to MetaPopInfo::delete_samples(). continue; } size_t sample_id = m[0]->sample_id; if (seen_samples.count(sample_id) > 0) { cerr << "Error: sample ID " << sample_id << " occurs twice in this data set, likely the pipeline was run incorrectly.\n"; return -1; } seen_samples.insert(sample_id); mpopi.set_sample_id(i, sample_id); } mpopi.delete_samples(samples_to_remove); if (mpopi.samples().size() == 0) { cerr << "Error: Couln't find any matches files.\n"; return -1; } // [mpopi] is definitive. cerr << "Working on " << mpopi.samples().size() << " samples.\n"; // // Create the population map // cerr << "Populating observed haplotypes for " << mpopi.samples().size() << " samples, " << catalog.size() << " loci.\n"; PopMap *pmap = new PopMap(mpopi, catalog.size()); pmap->populate(catalog, catalog_matches); // // Sum haplotype counts across the population for each catalog locus. // sum_haplotype_counts(catalog, pmap); // // Calculate mean log likelihood across the population for each catalog locus. // calc_lnl_means(catalog, pmap); int catalog_id, tag_id; // // Process samples matched to the catalog, one by one. // for (uint i = 0; i < mpopi.samples().size(); i++) { const Sample& sample = mpopi.samples()[i]; cerr << "Loading stacks from sample " << sample.name << " [" << i+1 << " of " << mpopi.samples().size() << "]...\n"; ////// ////// ////// if (sample_id != 176) continue; ////// ////// map stacks; int res; if ((res = load_loci(in_path + sample.name, stacks, true, true, compressed)) == 0) { cerr << "Unable to load sample file '" << sample.name << "'\n"; continue; } cerr << "Making corrections to sample " << sample.name << "..."; set > uniq_matches; set >::iterator it; vector > matches; // // There are multiple matches per stack, but we only need to process // each stack once to make corrections. // for (uint j = 0; j < catalog_matches[i].size(); j++) { catalog_id = catalog_matches[i][j]->cat_id; tag_id = catalog_matches[i][j]->tag_id; uniq_matches.insert(make_pair(catalog_id, tag_id)); } // // Put the catalog/tag ID pairs into a vector for parallel processing. // for (it = uniq_matches.begin(); it != uniq_matches.end(); it++) matches.push_back(*it); unsigned long int nuc_cnt = 0; unsigned long int unk_hom_cnt = 0; unsigned long int unk_het_cnt = 0; unsigned long int het_unk_cnt = 0; unsigned long int hom_unk_cnt = 0; unsigned long int het_hom_cnt = 0; unsigned long int hom_het_cnt = 0; unsigned long int conf_loci_cnt = 0; unsigned long int pruned_hap_cnt = 0; unsigned long int pruned_mst_hap_cnt = 0; unsigned long int blacklist_cnt = 0; unsigned long int lnl_cnt = 0; #pragma omp parallel private(catalog_id, tag_id) { Datum *d; Locus *loc; CSLocus *cloc; uint seq_len; string seq, model; char *adj_seq; vector > cigar; vector reads; #pragma omp for schedule(dynamic, 1) reduction(+:nuc_cnt) reduction(+:unk_hom_cnt) reduction(+:unk_het_cnt) \ reduction(+:hom_unk_cnt) reduction(+:het_unk_cnt) reduction(+:hom_het_cnt) reduction(+:het_hom_cnt) \ reduction(+:conf_loci_cnt) reduction(+:pruned_hap_cnt) reduction(+:pruned_mst_hap_cnt) reduction(+:blacklist_cnt) reduction(+:lnl_cnt) for (uint j = 0; j < matches.size(); j++) { catalog_id = matches[j].first; tag_id = matches[j].second; //if (tag_id == 10970) { // cerr << "Hit the tag.\n"; //} //// if (catalog_id != 3080) continue; if (catalog.count(catalog_id) == 0) continue; cloc = catalog[catalog_id]; loc = stacks[tag_id]; if (filter_confounded && ((double) cloc->confounded_cnt / (double) cloc->cnt > confounded_limit)) { // cerr << "Catalog locus " << cloc->id << " is confounded; confounded cnt: " // << cloc->confounded_cnt << "; total: " << cloc->cnt // << "; freq: " << (double) cloc->confounded_cnt / (double)cloc->cnt << "\n"; loc->blacklisted = true; conf_loci_cnt++; continue; } d = pmap->datum(catalog_id, sample.id); if (d == NULL) continue; if (filter_lnl && cloc->lnl < lnl_limit) { loc->blacklisted = true; lnl_cnt++; continue; } // // If this locus was matched to the catalog using a gapped alignment, adjust the sequence // and SNP positions. // if (d->cigar != NULL) { seq_len = parse_cigar(d->cigar, cigar); seq = apply_cigar_to_seq(loc->con, cigar); model = apply_cigar_to_model_seq(loc->model, cigar); loc->add_consensus(seq.c_str()); loc->add_model(model.c_str()); adjust_and_add_snps_for_gaps(cigar, loc); for (uint k = 0; k < loc->reads.size(); k++) { reads.push_back(loc->reads[k]); adj_seq = new char[seq_len + 1]; apply_cigar_to_seq(adj_seq, seq_len, loc->reads[k], cigar); loc->reads[k] = adj_seq; } } prune_nucleotides(cloc, loc, d, log_snp_fh, nuc_cnt, unk_hom_cnt, unk_het_cnt, hom_unk_cnt, het_unk_cnt, hom_het_cnt, het_hom_cnt); // // Prune haplotypes from this locus. // if (prune_haplotypes) { prune_mst_haplotypes(cloc, d, loc, pruned_mst_hap_cnt, log_hap_fh); prune_locus_haplotypes(cloc, d, loc, pruned_hap_cnt, log_hap_fh); if (loc->blacklisted) blacklist_cnt++; } // // If this locus was matched to the catalog using a gapped alignment, de-adjust the sequence // and SNP positions for writing back to disk. // if (d->cigar != NULL) { seq = remove_cigar_from_seq(loc->con, cigar); model = remove_cigar_from_seq(loc->model, cigar); loc->add_consensus(seq.c_str()); loc->add_model(model.c_str()); remove_snps_from_gaps(cigar, loc); for (uint k = 0; k < loc->reads.size(); k++) { adj_seq = loc->reads[k]; loc->reads[k] = reads[k]; delete [] adj_seq; } reads.clear(); } } } cerr << "done.\n"; unsigned long int total = unk_hom_cnt + unk_het_cnt + hom_unk_cnt + het_unk_cnt + hom_het_cnt + het_hom_cnt; cerr << "Total nucleotides processed: " << nuc_cnt << "\n" << " Total nucleotides converted: " << total << "\n" << " Converted from unknown to homozygous: " << unk_hom_cnt << " nucleotides.\n" << " Converted from unknown to heterozygous: " << unk_het_cnt << " nucleotides.\n" << " Converted from homozygous to unknown: " << hom_unk_cnt << " nucleotides.\n" << " Converted from heterozygous to unknown: " << het_unk_cnt << " nucleotides.\n" << " Converted from homozygous to heterozygous: " << hom_het_cnt << " nucleotides.\n" << " Converted from heterozygous to homozygous: " << het_hom_cnt << " nucleotides.\n" << "Pruned: " << pruned_mst_hap_cnt << " haplotypes using a tree method.\n" << "Pruned: " << pruned_hap_cnt << " haplotypes using a rare haplotype method.\n" << "Blacklisted: " << blacklist_cnt << " loci due to inability to call haplotypes.\n" << "Blacklisted: " << lnl_cnt << " loci due to log likelihoods below threshold.\n" << "Blacklisted: " << conf_loci_cnt << " confounded loci.\n"; log_fh << sample.name << "\t" << nuc_cnt << "\t" << total << "\t" << unk_hom_cnt << "\t" << unk_het_cnt << "\t" << hom_unk_cnt << "\t" << het_unk_cnt << "\t" << hom_het_cnt << "\t" << het_hom_cnt << "\t" << blacklist_cnt << "\t" << conf_loci_cnt << "\t" << lnl_cnt << "\t" << pruned_hap_cnt << "\t" << pruned_mst_hap_cnt << "\n"; cerr << "Writing modified stacks, SNPs, alleles to '" << out_path << "'..."; // // Rewrite stacks, model outputs, and haplotypes. // write_results(sample.name, stacks); // // Free up memory // cerr << "Freeing memory..."; map::iterator stack_it; for (stack_it = stacks.begin(); stack_it != stacks.end(); stack_it++) delete stack_it->second; cerr << "done.\n"; } log_fh.close(); if (verbose) { log_snp_fh.close(); log_hap_fh.close(); } return 0; } int dist(string hap_1, string hap_2) { int dist = 0; const char *p = hap_1.c_str(); const char *q = hap_2.c_str(); const char *p_end = p + hap_1.length(); const char *q_end = q + hap_2.length(); // // Count the number of characters that are different // between the two sequences. // while (p < p_end && q < q_end) { dist += (*p == *q) ? 0 : 1; p++; q++; } return dist; } int calc_lnl_means(map &catalog, PopMap *pmap) { map::iterator it; CSLocus *cloc; Datum **d; uint cnt, mid, tot; double median, mean; vector lnls; ofstream log_fh; if (lnl_dist) { stringstream log; log << "batch_" << batch_id << ".rxstacks_lnls.tsv"; string log_path = out_path + log.str(); log_fh.open(log_path.c_str(), ofstream::out); if (log_fh.fail()) { cerr << "Error opening log file '" << log_path << "'\n"; exit(1); } log_fh << "# Catalog Locus\tMean\tMedian\n"; } tot = 0; for (it = catalog.begin(); it != catalog.end(); it++) { cloc = it->second; d = pmap->locus(cloc->id); cnt = pmap->sample_cnt(); mean = 0.0; lnls.clear(); for (uint i = 0; i < cnt; i++) { if (d[i] == NULL) continue; lnls.push_back(d[i]->lnl); mean += d[i]->lnl; } if (lnls.size() == 0) continue; sort(lnls.begin(), lnls.end()); mid = lnls.size() / 2; median = lnls.size() % 2 == 0 ? lnls[mid] + lnls[mid+1] / 2.0 : lnls[mid+1]; mean = mean / (double) lnls.size(); cloc->lnl = mean; // // If the mean log likelihood for this catalog locus is below the threshold, count it as // its constituent components will be filtered as encountered later. // if (filter_lnl && cloc->lnl < lnl_limit) tot++; if (lnl_dist) log_fh << cloc->id << "\t" << mean << "\t" << median << "\n"; } if (lnl_dist) log_fh.close(); // // Print number of catalog loci that are confounded and will be removed. // cerr << tot << " catalog loci will be removed from the analysis due to log likelihoods below the threshold.\n"; return 0; } int sum_haplotype_counts(map &catalog, PopMap *pmap) { map::iterator it; CSLocus *cloc; Datum **d; uint cnt; for (it = catalog.begin(); it != catalog.end(); it++) { cloc = it->second; d = pmap->locus(cloc->id); cnt = pmap->sample_cnt(); for (uint i = 0; i < cnt; i++) { if (d[i] == NULL) continue; if (d[i]->obshap.size() == 1) { if (cloc->hap_cnts.count(d[i]->obshap[0]) == 0) cloc->hap_cnts[d[i]->obshap[0]] = 2; else cloc->hap_cnts[d[i]->obshap[0]] += 2; } else { for (uint j = 0; j < d[i]->obshap.size(); j++) if (cloc->hap_cnts.count(d[i]->obshap[j]) == 0) cloc->hap_cnts[d[i]->obshap[j]] = 1; else cloc->hap_cnts[d[i]->obshap[j]] += 1; } } } return 0; } int prune_mst_haplotypes(CSLocus *cloc, Datum *d, Locus *loc, unsigned long &pruned_hap_cnt, ofstream &log_fh) { // // Create a minimum spanning tree in order to determine the minimum distance // between each node in the list. // MinSpanTree *mst = new MinSpanTree; map::iterator it; vector keys; vector haps; Node *n; for (it = cloc->hap_cnts.begin(); it != cloc->hap_cnts.end(); it++) { n = mst->add_node(it->first); haps.push_back(it->first); keys.push_back(n->id); } // // We are going to connect nodes in the graph when a SNP occurs in one // of the positions of the haplotype. // Node *n_1, *n_2; uint snp_pos = 0; for (uint i = 0; i < cloc->snps.size(); i++) { if (cloc->snps[i]->type != snp_type_het) continue; for (uint j = 0; j < haps.size(); j++) { for (uint k = j + 1; k < haps.size(); k++) { // // If these two haplotypes differ by this SNP (and only this SNP), connect them in the graph. // if (haps[j].at(snp_pos) != haps[k].at(snp_pos) && dist(haps[j], haps[k]) == 1) { n_1 = mst->node(haps[j]); n_2 = mst->node(haps[k]); n_1->add_edge(n_2, 1); n_2->add_edge(n_1, 1); } } } snp_pos++; } // // Build the minimum spanning tree. // mst->build_tree(); // // Sort the haplotypes by read depth in this sample // vector > haplotypes; for (uint i = 0; i < d->obshap.size(); i++) haplotypes.push_back(make_pair(string(d->obshap[i]), (double) d->depth[i])); uint size = haplotypes.size(); // // Sort according to haplotype frequency. // sort(haplotypes.begin(), haplotypes.end(), compare_pair_haplotype_rev); if (size <= 2) return 0; // // Pull out the two most frequently occuring haplotypes. // string hap_1, hap_2; double hap_1_depth, hap_2_depth; hap_1 = haplotypes[size - 1].first; hap_1_depth = haplotypes[size - 1].second; haplotypes.pop_back(); if (haplotypes[size - 2].second > haplotypes[size - 3].second) { hap_2 = haplotypes[size - 2].first; hap_2_depth = haplotypes[size - 2].second; haplotypes.pop_back(); } else { hap_2 = ""; hap_2_depth = 0.0; } // // For each remaining haplotpye, check if it can be merged into a node (haplotype) no // more than one nucleotide apart. If there is more than one, merge it into the more // frequently occuring haplotype. // string hap, src_hap, dest_hap, label; double max, weighted; for (uint i = 0; i < haplotypes.size(); i++) { // // Find the current haplotype in the MST. // n_1 = mst->node(haplotypes[i].first); max = 0.0; hap = ""; weighted = 0.0; // // Check any potential edges in the graph for merging. // for (uint j = 0; j < n_1->edges.size(); j++) { label = n_1->edges[j]->child->label; if (label == hap_1) { weighted = (double) cloc->hap_cnts[label] * log(hap_1_depth); // cerr << "Cloc hap: " << label << "; popcnt: " << cloc->hap_cnts[label] << "; hap depth: " << hap_1_depth << "; weighted: " << weighted << "\n"; } else if (label == hap_2) { weighted = (double) cloc->hap_cnts[label] * log(hap_2_depth); // cerr << "Cloc hap: " << label << "; popcnt: " << cloc->hap_cnts[label] << "; hap depth: " << hap_2_depth << "; weighted: " << weighted << "\n"; } else continue; if (weighted == max) { // // There is more than one identical possibility, we can do no more. // hap = ""; break; } else if (weighted > max) { max = weighted; hap = label; } } if (hap.length() == 0) continue; src_hap = convert_catalog_haplotype_to_sample(haplotypes[i].first, cloc, loc); dest_hap = convert_catalog_haplotype_to_sample(hap, cloc, loc); if (verbose) { #pragma omp critical log_fh << cloc->id << "\t" << loc->sample_id << "\t" << loc->id << "\t" << src_hap << "\t" << haplotypes[i].first << "\t" << dest_hap << "\t" << hap << "\t" << "mst" << "\n"; } pruned_hap_cnt++; // // Remove the haplotype. // it = loc->alleles.find(src_hap); if (it != loc->alleles.end()) { loc->alleles.erase(it); } // // Add to the count of the merged-to haplotype. // if (loc->alleles.count(dest_hap) > 0) { loc->alleles[dest_hap]++; } else { cerr << "Error finding allele\n"; } } // // Update the matched haplotypes in the Datum object, so the haplotype pruner can // operate on newly generated, spurious haplotypes. // generate_matched_haplotypes(cloc, loc, d); return 0; } int prune_locus_haplotypes(CSLocus *cloc, Datum *d, Locus *loc, unsigned long &pruned_hap_cnt, ofstream &log_fh) { if (d->obshap.size() < 2) return 0; // // Identify the two most frequent haplotypes in this sample. // vector > haplotypes; double weighted_hap; for (uint i = 0; i < d->obshap.size(); i++) { // // Lookup the number of occurrences of this haplotype in the // population as well as the depth of the haplotype in this indiviudal. // We will weight the occurrences of the haplotype in the population by the natural log // of the read depth of the haplotype in this individual, storing the result. // weighted_hap = (double) cloc->hap_cnts[d->obshap[i]] * log((double) d->depth[i]); haplotypes.push_back(make_pair(string(d->obshap[i]), weighted_hap)); } // // Sort according to haplotype frequency. // sort(haplotypes.begin(), haplotypes.end(), compare_pair_haplotype); if (haplotypes.size() == 0) { cerr << "Error processing catalog locus " << cloc->id << "\n"; return -1; } // // Prune out excess haplotypes. // for (uint i = 2; i < haplotypes.size(); i++) { // // Make sure that those haplotypes we want to discard occur at a frequency lower // than the second most frequent haplotype, instead of being tied for second. // if (haplotypes[i].second >= haplotypes[1].second || (max_haplotype_cnt > 0 && haplotypes[i].second > max_haplotype_cnt)) continue; remove_haplotype(cloc, loc, haplotypes[i].first, pruned_hap_cnt, log_fh, "rare_step_1"); haplotypes.erase(haplotypes.begin() + i); } // // If there are more than two haplotypes remaining and the second, third, etc // haplotype exist only in this individual, prune them out. // if (haplotypes.size() > 2) { int stop_pos = haplotypes.size() - 1; int start_pos = stop_pos; double score = haplotypes[stop_pos].second; while (start_pos > 1) { if (cloc->hap_cnts[haplotypes[start_pos].first] == 1 && haplotypes[start_pos].second == score) start_pos--; else break; } if (start_pos < stop_pos) { for (int i = start_pos; i <= stop_pos; i++) remove_haplotype(cloc, loc, haplotypes[i].first, pruned_hap_cnt, log_fh, "rare_step_1"); } } // // Update the matched haplotypes in the Datum object, so the haplotype pruner can // operate on newly generated, spurious haplotypes. // generate_matched_haplotypes(cloc, loc, d); return 0; } string convert_catalog_haplotype_to_sample(string cat_haplotype, CSLocus *cloc, Locus *loc) { int cat_snp = 0; int cat_idx = -1; int loc_snp = 0; int loc_idx = -1; int k = -1; int j = -1; string hap; do { j++; loc_idx++; // // Advance to a het in the sample locus. // while (j < (int) loc->snps.size() && loc->snps[j]->type != snp_type_het) j++; if (j >= (int) loc->snps.size()) break; loc_snp = loc->snps[j]->col; do { k++; cat_idx++; // // Advance to the het in the catalog locus that corresponds to the sample locus. // while (k < (int) cloc->snps.size() && cloc->snps[k]->type != snp_type_het) k++; if (k >= (int) cloc->snps.size()) break; cat_snp = cloc->snps[k]->col; } while (cat_snp < loc_snp); // // Extract out the nucleotide from the catalog haplotype that matches the sample // haplotype. For example, catalog haplotype may be 'ACGTG' while sample haplotype // is 'CT'. // if (j < (int) loc->snps.size() && k < (int) cloc->snps.size() && cat_snp == loc_snp) { hap += cat_haplotype.at(cat_idx); } else { cerr << "Error processing catalog locus " << cloc->id << "\n"; return ""; } } while (j < (int) loc->snps.size()); return hap; } int remove_haplotype(CSLocus *cloc, Locus *loc, string haplotype, unsigned long &pruned_hap_cnt, ofstream &log_fh, string alg_type) { map::iterator it; string hap = ""; hap = convert_catalog_haplotype_to_sample(haplotype, cloc, loc); if (verbose) { #pragma omp critical log_fh << cloc->id << "\t" << loc->sample_id << "\t" << loc->id << "\t" << hap << "\t" << haplotype << "\t" << "\t" << "\t" << alg_type << "\n"; } // // Remove the haplotype. // it = loc->alleles.find(hap); if (it != loc->alleles.end()) { loc->alleles.erase(it); pruned_hap_cnt++; } // // Decrement the count for this haplotype in the catalog locus. // if (cloc->hap_cnts.count(haplotype) > 0) cloc->hap_cnts[haplotype]--; return 0; } int prune_nucleotides(CSLocus *cloc, Locus *loc, Datum *d, ofstream &log_fh, unsigned long int &nuc_cnt, unsigned long int &unk_hom_cnt, unsigned long int &unk_het_cnt, unsigned long int &hom_unk_cnt, unsigned long int &het_unk_cnt, unsigned long int &hom_het_cnt, unsigned long int &het_hom_cnt) { map nucs; set cnucs; set::iterator it; set rows; nuc_cnt += loc->len; for (uint i = 0; i < loc->snps.size() && i < cloc->snps.size(); i++) { // // Safety checks. // if (loc->snps[i]->col != cloc->snps[i]->col) cerr << "Warning: comparing mismatched SNPs in catalog locus " << cloc->id << " and sample " << loc->sample_id << ", locus " << loc->id << "; col: " << loc->snps[i]->col << "\n"; if (loc->snps[i]->type == snp_type_het && cloc->snps[i]->type == snp_type_hom) cerr << "Warning: sample locus is variable while catalog locus is fixed; catalog locus " << cloc->id << " and sample " << loc->sample_id << ", locus " << loc->id << "; col: " << loc->snps[i]->col << "\n"; // // Either there is an unknown call in locus, or, there is a snp in the catalog and any state in the locus. // if ((loc->snps[i]->type == snp_type_unk) || (cloc->snps[i]->type == snp_type_het && loc->snps[i]->type == snp_type_hom)) { // cerr << " Looking at SNP call in tag " << loc->id << " at position " << i << "; col: " << loc->snps[i]->col << "\n" // << " Catalog column: " << cloc->snps[i]->col << " (" << i << "); Sample column: " << loc->snps[i]->col << " (" << i << ")\n" // << " Sample has model call type: " << (loc->snps[i]->type == snp_type_unk ? "Unknown" : "Homozygous") << "; nucleotides: '" // << loc->snps[i]->rank_1 << "' and '" << loc->snps[i]->rank_2 << "'; model call: " << loc->model[loc->snps[i]->col] << "\n" // << " Catalog has model call type: " << (cloc->snps[i]->type == snp_type_het ? "Heterozygous" : "Homozygous") << "; nucleotides: '" // << cloc->snps[i]->rank_1 << "' and '" << cloc->snps[i]->rank_2 << "'\n"; if (loc->snps[i]->rank_1 == 'N' || cloc->snps[i]->rank_1 == 'N' || loc->snps[i]->rank_1 == '-' || cloc->snps[i]->rank_1 == '-') continue; cnucs.insert(cloc->snps[i]->rank_1); if (cloc->snps[i]->rank_2 != 0) cnucs.insert(cloc->snps[i]->rank_2); if (cloc->snps[i]->rank_3 != 0) cnucs.insert(cloc->snps[i]->rank_3); if (cloc->snps[i]->rank_4 != 0) cnucs.insert(cloc->snps[i]->rank_4); // cerr << " Catalog has nucleotides: "; // for (it = cnucs.begin(); it != cnucs.end(); it++) // cerr << *it << ", "; // cerr << "\n"; // // Tally the number of occurances of each nucleotide also present in the // catalog in order to fuel the snp calling model. // // Note reads that contain nucleotides not present in the catalog so they // can be excluded when calling haplotypes from the read. // nucs['A'] = 0; nucs['C'] = 0; nucs['G'] = 0; nucs['T'] = 0; nucs['N'] = 0; for (uint k = 0; k < loc->reads.size(); k++) { if (cnucs.count(loc->reads[k][i]) > 0) nucs[loc->reads[k][i]]++; else if (loc->reads[k][i] != 'N') rows.insert(k); } // // Test pruned data for homozygosity or heterozygosity. // invoke_model(loc, i, nucs); cnucs.clear(); } } if (verbose) { #pragma omp critical log_model_calls(loc, log_fh, unk_hom_cnt, unk_het_cnt, hom_unk_cnt, het_unk_cnt, hom_het_cnt, het_hom_cnt); } else { log_model_calls(loc, log_fh, unk_hom_cnt, unk_het_cnt, hom_unk_cnt, het_unk_cnt, hom_het_cnt, het_hom_cnt); } // // Re-call alleles. // loc->alleles.clear(); call_alleles(loc, rows); // // If SNPs were called at this locus but no alleles could be determined, // blacklist this tag. This can occur if a fixed nucleotide isn't captured in // the catalog and all the reads are removed for the purpose of reading haplotypes. // if (loc->alleles.size() <= 1) for (uint j = 0; j < loc->snps.size(); j++) if (loc->snps[j]->type == snp_type_het) { loc->blacklisted = 1; break; } // // Update the matched haplotypes in the Datum object, so the haplotype pruner can // operate on newly generated, spurious haplotypes. // generate_matched_haplotypes(cloc, loc, d); return 0; } int invoke_model(Locus *loc, int col, map &nucs) { // // Search this column for the presence of a SNP // switch(model_type) { case snp: call_multinomial_snp(loc, col, nucs); break; case bounded: call_bounded_multinomial_snp(loc, col, nucs); break; default: break; } return 0; } int call_alleles(Locus *loc, set &rows) { int row; int height = loc->reads.size(); string allele; char base; vector::iterator snp; for (row = 0; row < height; row++) { // // If a read had a nucleotide not present in the catalog, do not call // a haplotype from it. // if (rows.count(row) > 0) continue; allele.clear(); uint snp_cnt = 0; for (snp = loc->snps.begin(); snp != loc->snps.end(); snp++) { if ((*snp)->type != snp_type_het) continue; snp_cnt++; base = loc->reads[row][(*snp)->col]; // // Check to make sure the nucleotide at the location of this SNP is // of one of the two possible states the multinomial model called. // if (base == (*snp)->rank_1 || base == (*snp)->rank_2) allele += base; else break; } if (snp_cnt > 0 && allele.length() == snp_cnt) loc->alleles[allele]++; } return 0; } int generate_matched_haplotypes(CSLocus *cloc, Locus *loc, Datum *d) { // // Free the existing matched haplotypes. // for (uint i = 0; i < d->obshap.size(); i++) delete [] d->obshap[i]; d->obshap.clear(); d->depth.clear(); // // Construct a set of haplotypes from the locus relative to the catalog locus. // (The locus already has a set of haplotypes, however, they don't necessarily // account for all the SNPs in the catalog, so we will augment them with sequence // from the consensus.) // vector > merged_snps; map > columns; map >::iterator c; vector >::iterator k; for (uint i = 0; i < cloc->snps.size(); i++) { if (cloc->snps[i]->type != snp_type_het) continue; columns[cloc->snps[i]->col] = make_pair("catalog", cloc->snps[i]); } for (uint i = 0; i < loc->snps.size(); i++) { if (loc->snps[i]->type != snp_type_het) continue; // // Is this column already represented in the catalog? // if (columns.count(loc->snps[i]->col)) columns[loc->snps[i]->col] = make_pair("both", loc->snps[i]); else columns[loc->snps[i]->col] = make_pair("query", loc->snps[i]); } for (c = columns.begin(); c != columns.end(); c++) merged_snps.push_back((*c).second); // // Sort the SNPs by column // sort(merged_snps.begin(), merged_snps.end(), compare_pair_snp); map::iterator b; string old_allele, new_allele; int pos; for (b = loc->alleles.begin(); b != loc->alleles.end(); b++) { old_allele = b->first; new_allele = ""; pos = 0; for (k = merged_snps.begin(); k != merged_snps.end(); k++) { // // If the SNPs from the catalog haplotype beyond the length of the query, add Ns // if (k->first == "catalog") { new_allele += (k->second->col > loc->len - 1) ? 'N' : loc->con[k->second->col]; } else { new_allele += old_allele[pos]; pos++; } } char *h = new char[new_allele.length() + 1]; strcpy(h, new_allele.c_str()); d->obshap.push_back(h); d->depth.push_back(b->second); // loc->alleles[new_allele] = b->second; // cerr << "Adding haplotype: " << new_allele << " [" << b->first << "]\n"; } return 0; } int log_model_calls(Locus *loc, ofstream &log_fh, unsigned long int &unk_hom_cnt, unsigned long int &unk_het_cnt, unsigned long int &hom_unk_cnt, unsigned long int &het_unk_cnt, unsigned long int &hom_het_cnt, unsigned long int &het_hom_cnt) { // // Log model call changes // for (uint j = 0; j < loc->snps.size(); j++) { switch(loc->model[j]) { case 'U': switch(loc->snps[j]->type) { case snp_type_het: if (verbose) log_fh << loc->sample_id << "\t" << loc->id << "\t" << loc->snps[j]->col << "\t" << 'U' << "\t" << 'E' << "\n"; unk_het_cnt++; break; case snp_type_hom: if (verbose) log_fh << loc->sample_id << "\t" << loc->id << "\t" << loc->snps[j]->col << "\t" << 'U' << "\t" << 'O' << "\n"; unk_hom_cnt++; break; case snp_type_unk: default: break; } break; case 'E': switch(loc->snps[j]->type) { case snp_type_het: break; case snp_type_hom: if (verbose) log_fh << loc->sample_id << "\t" << loc->id << "\t" << loc->snps[j]->col << "\t" << 'E' << "\t" << 'O' << "\n"; het_hom_cnt++; break; case snp_type_unk: default: if (verbose) log_fh << loc->sample_id << "\t" << loc->id << "\t" << loc->snps[j]->col << "\t" << 'E' << "\t" << 'U' << "\n"; het_unk_cnt++; break; } break; case 'O': default: switch(loc->snps[j]->type) { case snp_type_het: if (verbose) log_fh << loc->sample_id << "\t" << loc->id << "\t" << loc->snps[j]->col << "\t" << 'O' << "\t" << 'E' << "\n"; hom_het_cnt++; break; case snp_type_hom: break; case snp_type_unk: default: if (verbose) log_fh << loc->sample_id << "\t" << loc->id << "\t" << loc->snps[j]->col << "\t" << 'O' << "\t" << 'U' << "\n"; hom_unk_cnt++; break; } break; } } return 0; } int write_results(string file, map &m) { map::iterator i; vector::iterator j; vector::iterator k; map::iterator t; Locus *tag_1; stringstream sstr; bool gzip = (in_file_type == FileT::gzsql) ? true : false; // // Parse the input file name to create the output files // string tag_file = out_path + file + ".tags.tsv"; string snp_file = out_path + file + ".snps.tsv"; string all_file = out_path + file + ".alleles.tsv"; string mod_file = out_path + file + ".models.tsv"; if (gzip) { tag_file += ".gz"; snp_file += ".gz"; all_file += ".gz"; mod_file += ".gz"; } // // Open the output files for writing. // gzFile gz_tags, gz_snps, gz_alle, gz_mods; ofstream tags, snps, alle, mods; if (gzip) { gz_tags = gzopen(tag_file.c_str(), "wb"); if (!gz_tags) { cerr << "Error: Unable to open gzipped tag tag file '" << tag_file << "': " << strerror(errno) << ".\n"; exit(1); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_tags, libz_buffer_size); #endif gz_mods = gzopen(mod_file.c_str(), "wb"); if (!gz_mods) { cerr << "Error: Unable to open gzipped model file '" << mod_file << "': " << strerror(errno) << ".\n"; exit(1); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_mods, libz_buffer_size); #endif gz_snps = gzopen(snp_file.c_str(), "wb"); if (!gz_snps) { cerr << "Error: Unable to open gzipped catalog snps file '" << snp_file << "': " << strerror(errno) << ".\n"; exit(1); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_snps, libz_buffer_size); #endif gz_alle = gzopen(all_file.c_str(), "wb"); if (!gz_alle) { cerr << "Error: Unable to open gzipped catalog alleles file '" << all_file << "': " << strerror(errno) << ".\n"; exit(1); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_alle, libz_buffer_size); #endif } else { tags.open(tag_file.c_str()); if (tags.fail()) { cerr << "Error: Unable to open tag file for writing.\n"; exit(1); } mods.open(mod_file.c_str()); if (mods.fail()) { cerr << "Error: Unable to open model file for writing.\n"; exit(1); } snps.open(snp_file.c_str()); if (snps.fail()) { cerr << "Error: Unable to open catalog SNPs file for writing.\n"; exit(1); } alle.open(all_file.c_str()); if (alle.fail()) { cerr << "Error: Unable to open catalog alleles file for writing.\n"; exit(1); } } // // Record the version of Stacks used and the date generated as a comment in the catalog. // // Obtain the current date. // stringstream log; time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%F %T", timeinfo); log << "# rxstacks version " << VERSION << "; generated on " << date << "\n"; if (gzip) { gzputs(gz_tags, log.str().c_str()); gzputs(gz_mods, log.str().c_str()); gzputs(gz_snps, log.str().c_str()); gzputs(gz_alle, log.str().c_str()); } else { tags << log.str(); mods << log.str(); snps << log.str(); alle << log.str(); } int wrote = 0; for (i = m.begin(); i != m.end(); i++) { tag_1 = i->second; wrote++; // First write the consensus sequence sstr << "0" << "\t" << tag_1->sample_id << "\t" << tag_1->id << "\t" << tag_1->loc.chr << "\t" << tag_1->loc.bp << "\t" << (tag_1->loc.strand == strand_plus ? "+" : "-") << "\t" << "consensus" << "\t" << "\t" << "\t" << tag_1->con << "\t" << (tag_1->deleveraged == true ? "1" : "0") << "\t" << (tag_1->blacklisted == true ? "1" : "0") << "\t" << (tag_1->lumberjackstack == true ? "1" : "0") << "\t" << tag_1->lnl << "\n"; // // Write a sequence recording the output of the SNP model for each nucleotide. // sstr << "0" << "\t" << tag_1->sample_id << "\t" << tag_1->id << "\t" << "\t" << "\t" << "\t" << "model" << "\t" << "\t" << "\t"; for (uint j = 0; j < tag_1->snps.size(); j++) { switch(tag_1->snps[j]->type) { case snp_type_het: sstr << "E"; break; case snp_type_hom: sstr << "O"; break; default: sstr << "U"; break; } } sstr << "\t" << "\t" << "\t" << "\t" << "\n"; if (gzip) gzputs(gz_tags, sstr.str().c_str()); else tags << sstr.str(); if (gzip) gzputs(gz_mods, sstr.str().c_str()); else mods << sstr.str(); sstr.str(""); // // Now write out each read from this locus. // for (uint j = 0; j < tag_1->reads.size(); j++) { sstr << "0" << "\t" << tag_1->sample_id << "\t" << tag_1->id << "\t\t\t\t"; if (tag_1->comp_type[j] == primary) { sstr << "primary" << "\t" << tag_1->comp_cnt[j] << "\t"; } else { sstr << "secondary" << "\t\t"; } sstr << tag_1->comp[j] << "\t" << tag_1->reads[j] << "\t\t\t\t\n"; } if (gzip) gzputs(gz_tags, sstr.str().c_str()); else tags << sstr.str(); sstr.str(""); // // Write out the model calls for each nucleotide in this locus. // for (uint j = 0; j < tag_1->snps.size(); j++) { sstr << "0" << "\t" << tag_1->sample_id << "\t" << tag_1->id << "\t" << tag_1->snps[j]->col << "\t"; switch(tag_1->snps[j]->type) { case snp_type_het: sstr << "E\t"; break; case snp_type_hom: sstr << "O\t"; break; default: sstr << "U\t"; break; } sstr << std::fixed << std::setprecision(3) << tag_1->snps[j]->lratio << "\t" << tag_1->snps[j]->rank_1 << "\t" << (tag_1->snps[j]->rank_2 == 0 ? '-' : tag_1->snps[j]->rank_2) << "\t\t\n"; } if (gzip) gzputs(gz_snps, sstr.str().c_str()); else snps << sstr.str(); sstr.str(""); // // Write the expressed alleles seen for the recorded SNPs and // the percentage of tags a particular allele occupies. // char pct[id_len]; for (t = tag_1->alleles.begin(); t != tag_1->alleles.end(); t++) { sprintf(pct, "%.2f", ((t->second/double(tag_1->reads.size())) * 100)); sstr << "0" << "\t" << tag_1->sample_id << "\t" << tag_1->id << "\t" << t->first << "\t" << pct << "\t" << t->second << "\n"; } if (gzip) gzputs(gz_alle, sstr.str().c_str()); else alle << sstr.str(); sstr.str(""); } if (gzip) { gzclose(gz_tags); gzclose(gz_mods); gzclose(gz_snps); gzclose(gz_alle); } else { tags.close(); mods.close(); snps.close(); alle.close(); } cerr << "wrote " << wrote << " loci.\n"; return 0; } int fill_catalog_snps(map &catalog) { map::iterator it; CSLocus *cloc; for (it = catalog.begin(); it != catalog.end(); it++) { cloc = it->second; queue snps; for (uint j = 0; j < cloc->snps.size(); j++) snps.push(cloc->snps[j]); cloc->snps.clear(); for (uint j = 0; j < cloc->len; j++) { if (snps.size() > 0 && snps.front()->col == j) { cloc->snps.push_back(snps.front()); snps.pop(); } else { SNP *snp = new SNP; snp->type = snp_type_hom; snp->col = j; snp->lratio = 0; snp->rank_1 = cloc->con[j]; snp->rank_2 = 0; cloc->snps.push_back(snp); } } } return 0; } int init_log(int argc, char **argv, ofstream &log_fh, ofstream &log_snp_fh, ofstream &log_hap_fh) { stringstream log; stringstream sstr; // // Open the log files. // log << out_path << "batch_" << batch_id << ".rxstacks.log"; log_fh.open(log.str().c_str(), ofstream::out); if (log_fh.fail()) { cerr << "Error opening log file '" << log.str() << "'\n"; exit(1); } if (verbose) { log.str(""); log << out_path << "batch_" << batch_id << ".rxstacks.snps.log"; log_snp_fh.open(log.str().c_str(), ofstream::out); if (log_snp_fh.fail()) { cerr << "Error opening log file '" << log.str() << "'\n"; exit(1); } log.str(""); log << out_path << "batch_" << batch_id << ".rxstacks.haplotypes.log"; log_hap_fh.open(log.str().c_str(), ofstream::out); if (log_hap_fh.fail()) { cerr << "Error opening log file '" << log.str() << "'\n"; exit(1); } } // // Obtain the current date. // time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%F %T", timeinfo); sstr << "#"; for (int i = 0; i < argc; i++) sstr << " " << argv[i]; sstr << "\n" << "# rxstacks executed " << date; log_fh << sstr.str() << "\n" << "# Sample\t" << "Total nucs\t" << "Total nucs converted\t" << "Unk to Hom\t" << "Unk to Het\t" << "Hom to Unk\t" << "Het to Unk\t" << "Hom to Het\t" << "Het to Hom\t" << "Confounded loci\t" << "Lnl Filtered loci\t" << "Pruned Haplotypes\t" << "MST-Pruned Haplotypes\n"; if (verbose) { log_snp_fh << sstr.str() << "\n" << "# Sample Id\t" << "Locus ID\t" << "SNP Col\t" << "Orig Value\t" << "Corr Value\n"; log_hap_fh << sstr.str() << "\n" << "# Catalog Locus\t" << "Sample\t" << "Sample Locus\t" << "Sample Haplotype\t" << "Catalog Haplotype\t" << "Corrected Sample Haplotype\t" << "Corrected Catalog Haplotype\t" << "Algorithm\n"; } return 0; } int parse_command_line(int argc, char* argv[]) { int c; while (1) { static struct option long_options[] = { {"help", no_argument, NULL, 'h'}, {"version", no_argument, NULL, 'v'}, {"conf_filter", no_argument, NULL, 'F'}, {"prune_haplo", no_argument, NULL, 'H'}, {"lnl_filter", no_argument, NULL, 'G'}, {"lnl_dist", no_argument, NULL, 'D'}, {"verbose", no_argument, NULL, 'V'}, {"num_threads", required_argument, NULL, 't'}, {"batch_id", required_argument, NULL, 'b'}, {"in_path", required_argument, NULL, 'P'}, {"outpath", required_argument, NULL, 'o'}, {"model_type", required_argument, NULL, 'T'}, {"bound_low", required_argument, NULL, 'L'}, {"bound_high", required_argument, NULL, 'U'}, {"alpha", required_argument, NULL, 'A'}, {"conf_lim", required_argument, NULL, 'C'}, {"max_haplo", required_argument, NULL, 'M'}, {"lnl_lim", required_argument, NULL, 'I'}, {0, 0, 0, 0} }; // getopt_long stores the option index here. int option_index = 0; c = getopt_long(argc, argv, "hvVFGDHo:t:b:P:T:L:U:A:C:I:", long_options, &option_index); // Detect the end of the options. if (c == -1) break; switch (c) { case 'h': help(); break; case 't': num_threads = atoi(optarg); break; case 'P': in_path = optarg; break; case 'b': batch_id = is_integer(optarg); if (batch_id < 0) { cerr << "Batch ID (-b) must be an integer, e.g. 1, 2, 3\n"; help(); } break; case 'o': out_path = optarg; break; case 'T': if (strcmp(optarg, "snp") == 0) { model_type = snp; } else if (strcmp(optarg, "fixed") == 0) { model_type = fixed; } else if (strcmp(optarg, "bounded") == 0) { model_type = bounded; } else { cerr << "Unknown model type specified '" << optarg << "'\n"; help(); } case 'L': bound_low = atof(optarg); break; case 'U': bound_high = atof(optarg); break; case 'A': alpha = atof(optarg); break; case 'F': filter_confounded = true; break; case 'C': confounded_limit = is_double(optarg); filter_confounded = true; break; case 'H': prune_haplotypes = true; break; case 'M': max_haplotype_cnt = is_integer(optarg); break; case 'G': filter_lnl = true; break; case 'I': lnl_limit = is_double(optarg); filter_lnl = true; break; case 'D': lnl_dist = true; break; case 'V': verbose = true; break; case 'v': version(); break; case '?': // getopt_long already printed an error message. help(); break; default: help(); abort(); } } if (in_path.length() == 0) { cerr << "You must specify a path to the directory containing Stacks output files.\n"; help(); } if (out_path.length() == 0) { cerr << "No output path specified, files in '" << in_path << "' will be overwritten.\n"; out_path = in_path; } if (in_path.at(in_path.length() - 1) != '/') in_path += "/"; if (out_path.at(out_path.length() - 1) != '/') out_path += "/"; if (batch_id == 0) { cerr << "You must specify a batch ID.\n"; help(); } if (alpha != 0.1 && alpha != 0.05 && alpha != 0.01 && alpha != 0.001) { cerr << "SNP model alpha significance level must be either 0.1, 0.05, 0.01, or 0.001.\n"; help(); } if (bound_low != 0 && (bound_low < 0 || bound_low >= 1.0)) { cerr << "SNP model lower bound must be between 0.0 and 1.0.\n"; help(); } if (bound_high != 1 && (bound_high <= 0 || bound_high > 1.0)) { cerr << "SNP model upper bound must be between 0.0 and 1.0.\n"; help(); } if (bound_low > 0 || bound_high < 1.0) { model_type = bounded; } if (filter_confounded == true && (confounded_limit < 0 || confounded_limit > 1.0)) { cerr << "Confounded locus limit is a percentage and must be between 0.0 and 1.0.\n"; help(); } return 0; } void version() { std::cerr << "rxstacks " << VERSION << "\n\n"; exit(0); } void help() { std::cerr << "rxstacks " << VERSION << "\n" << "rxstacks -b batch_id -P path [-o path] [-t threads] [-v] [-h]" << "\n" << " b: Batch ID to examine when exporting from the catalog.\n" << " P: path to the Stacks output files.\n" << " o: output path to write results.\n" << " t: number of threads to run in parallel sections of code.\n" << " v: print program version." << "\n" << " h: display this help messsage." << "\n\n" << " Filtering options:\n" << " --lnl_filter: filter catalog loci based on the mean log likelihood of the catalog locus in the population.\n" << " --lnl_lim : minimum log likelihood required to keep a catalog locus.\n" << " --lnl_dist: print distribution of mean log likelihoods for catalog loci.\n" << " --conf_filter: filter confounded loci.\n" << " --conf_lim : between 0.0 and 1.0 (default 0.75), proportion of loci in population that must be confounded relative to the catalog locus.\n" << " --prune_haplo: prune out non-biological haplotypes unlikely to occur in the population.\n" << " --max_haplo : only consider haplotypes for pruning if they occur in fewer than max_haplo_cnt samples.\n" << " Model options:\n" << " --model_type : either 'snp' (default), 'bounded', or 'fixed'\n" << " For the SNP or Bounded SNP model:\n" << " --alpha : chi square significance level required to call a heterozygote or homozygote, either 0.1 (default), 0.05, 0.01, or 0.001.\n" << " For the Bounded SNP model:\n" << " --bound_low : lower bound for epsilon, the error rate, between 0 and 1.0 (default 0).\n" << " --bound_high : upper bound for epsilon, the error rate, between 0 and 1.0 (default 1).\n" << " Logging Options:\n" << " --verbose: extended logging, including coordinates of all changed nucleotides (forces single-threaded execution).\n"; exit(0); } stacks-1.44/src/rxstacks.h000644 000765 000024 00000006332 12764047443 016323 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2013, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __RXSTACKS_H__ #define __RXSTACKS_H__ #ifdef _OPENMP #include // OpenMP library #endif #include // Process command-line options #include // Open/Read contents of a directory #include #include #include #include #include using std::pair; using std::make_pair; #include using std::string; #include #include using std::ifstream; using std::ofstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include // std::setprecision #include using std::stringstream; #include using std::vector; #include using std::queue; #include using std::map; #include using std::set; #include "constants.h" #include "stacks.h" #include "locus.h" #include "PopMap.h" #include "PopSum.h" #include "utils.h" #include "sql_utilities.h" #include "aln_utils.h" #include "models.h" #include "mst.h" void help( void ); void version( void ); int parse_command_line(int, char**); int build_file_list(vector > &); int init_log(int, char **, ofstream &, ofstream &, ofstream &); int sum_haplotype_counts(map &, PopMap *); int prune_mst_haplotypes(CSLocus *, Datum *, Locus *, unsigned long int &, ofstream &); int prune_locus_haplotypes(CSLocus *, Datum *, Locus *, unsigned long int &, ofstream &); string convert_catalog_haplotype_to_sample(string, CSLocus *, Locus *); int remove_haplotype(CSLocus *, Locus *, string, unsigned long &, ofstream &, string); int dist(string, string); int measure_error(CSLocus *, Locus *, Datum *, ofstream &); int calc_lnl_means(map &, PopMap *); int prune_nucleotides(CSLocus *, Locus *, Datum *, ofstream &, unsigned long int &, unsigned long int &, unsigned long int &, unsigned long int &, unsigned long int &, unsigned long int &, unsigned long int &); int invoke_model(Locus *, int, map &); int call_alleles(Locus *, set &); int generate_matched_haplotypes(CSLocus *, Locus *, Datum *); int fill_catalog_snps(map &); int log_model_calls(Locus *, ofstream &, unsigned long int &, unsigned long int &, unsigned long int &, unsigned long int &, unsigned long int &, unsigned long int &); int write_results(string, map &); #endif // __RXSTACKS_H__ stacks-1.44/src/SamI.h000644 000765 000024 00000021236 13001012336 015264 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __SAMI_H__ #define __SAMI_H__ // // Code to parse Sam format. This format is created for // reads that have been aligned to a reference genome. It takes the tab-separated form: // // ... ... // // One record per line. // #include "input.h" class Sam: public Input { int parse_cigar(const char *, vector > &, bool); int find_start_bp_neg(int, vector > &); int find_start_bp_pos(int, vector > &); int edit_gaps(vector > &, char *); public: Sam(const char *path) : Input(path) {} ~Sam() {} Seq *next_seq(); int next_seq(Seq& s); }; Seq* Sam::next_seq() { Seq* s = new Seq(); if(next_seq(*s) != 1) { delete s; s = NULL; } return s; } int Sam::next_seq(Seq& s) { vector parts; int flag = 0; int sflag = 0; int aflag = 0; uint len; // // Read a record from the file and place it in a Seq object, skipping header // definitions and unaligned sequences. // do { this->fh.getline(this->line, max_len); if (!this->fh.good()) return 0; len = strlen(this->line); if (this->line[len - 1] == '\r') this->line[len - 1] = '\0'; parse_tsv(this->line, parts); // // According to SAM spec FLAGs are the second field, // if FLAG bit 0x4 is set, sequence is not mapped. // flag = atoi(parts[1].c_str()); flag = flag & 4; flag = flag >> 2; } while (parts[0][0] == '@' || flag == 1); // // Check which strand this is aligned to: // SAM reference: FLAG bit 0x10 - sequence is reverse complemented // flag = atoi(parts[1].c_str()); sflag = flag & 16; sflag = sflag >> 4; // // Check if this is the primary or secondary alignment. // alnt aln_type = pri_aln; aflag = flag & 256; aflag = aflag >> 8; if (aflag) aln_type = sec_aln; // // Check if this is a supplemenatry (chimeric) alignment (not yet defined in Bam.h). // aflag = flag & 2048; aflag = aflag >> 11; if (aflag) aln_type = sup_aln; // // If the read was aligned on the reverse strand (and is therefore reverse complemented) // alter the start point of the alignment to reflect the right-side of the read, at the // end of the RAD cut site. // // To accomplish this, we must parse the alignment CIGAR string // vector > cigar; this->parse_cigar(parts[5].c_str(), cigar, sflag); int bp = sflag ? this->find_start_bp_neg(atoi(parts[3].c_str()), cigar) : this->find_start_bp_pos(atoi(parts[3].c_str()), cigar); // // Sam format has a 1-based offset for chrmosome/basepair positions, adjust it to match // the Stacks, 0-based offset. // bp--; // // Calculate the percentage of the sequence that was aligned to the reference. // len = 0; for (uint i = 0; i < cigar.size(); i++) switch (cigar[i].first) { case 'M': case 'I': case '=': case 'X': len += cigar[i].second; break; case 'D': case 'S': case 'H': case 'N': break; default: cerr << "Error parsing CIGAR string '" << cigar[i].second << cigar[i].first << "'.\n"; break; } double pct_aln = double(len) / double(parts[9].length()); s = Seq(parts[0].c_str(), parts[9].c_str(), parts[10].c_str(), // Read ID, Sequence, Quality parts[2].c_str(), bp, sflag ? strand_minus : strand_plus, // Chr, BasePair, Strand aln_type, pct_aln); // Alignment type (primary or secondary), percent of read aligned if (cigar.size() > 0) this->edit_gaps(cigar, s.seq); return 1; } int Sam::parse_cigar(const char *cigar_str, vector > &cigar, bool orientation) { char buf[id_len]; int dist; const char *p, *q; p = cigar_str; if (*p == '*') return 0; while (*p != '\0') { q = p + 1; while (*q != '\0' && isdigit(*q)) q++; strncpy(buf, p, q - p); buf[q-p] = '\0'; dist = atoi(buf); // // If aligned to the negative strand, sequence has been reverse complemented and // CIGAR string should be interpreted in reverse. // if (orientation == strand_plus) cigar.push_back(make_pair(*q, dist)); else cigar.insert(cigar.begin(), make_pair(*q, dist)); p = q + 1; } return 0; } int Sam::find_start_bp_neg(int aln_bp, vector > &cigar) { uint size = cigar.size(); char op; uint dist; for (uint i = 0; i < size; i++) { op = cigar[i].first; dist = cigar[i].second; switch(op) { case 'I': case 'H': break; case 'S': if (i < size - 1) aln_bp += dist; break; case 'M': case '=': case 'X': case 'D': case 'N': aln_bp += dist; break; default: break; } } return aln_bp - 1; } int Sam::find_start_bp_pos(int aln_bp, vector > &cigar) { char op; uint dist; op = cigar[0].first; dist = cigar[0].second; if (op == 'S') aln_bp -= dist; return aln_bp; } int Sam::edit_gaps(vector > &cigar, char *seq) { char *buf; uint size = cigar.size(); char op; uint dist, bp, len, buf_len, buf_size, j, k, stop; len = strlen(seq); bp = 0; buf = new char[len + 1]; buf_size = len + 1; for (uint i = 0; i < size; i++) { op = cigar[i].first; dist = cigar[i].second; switch(op) { case 'S': stop = bp + dist; stop = stop > len ? len : stop; while (bp < stop) { seq[bp] = 'N'; bp++; } break; case 'D': // // A deletion has occured in the read relative to the reference genome. // Pad the read with sufficent Ns to match the deletion, shifting the existing // sequence down. Trim the final length to keep the read length consistent. // k = bp >= len ? len : bp; strncpy(buf, seq + k, buf_size - 1); buf[buf_size - 1] = '\0'; buf_len = strlen(buf); stop = bp + dist; stop = stop > len ? len : stop; while (bp < stop) { seq[bp] = 'N'; bp++; } j = bp; k = 0; while (j < len && k < buf_len) { seq[j] = buf[k]; k++; j++; } break; case 'I': // // An insertion has occurred in the read relative to the reference genome. Delete the // inserted bases and pad the end of the read with Ns. // if (bp >= len) break; k = bp + dist > len ? len : bp + dist; strncpy(buf, seq + k, buf_size - 1); buf[buf_size - 1] = '\0'; buf_len = strlen(buf); j = bp; k = 0; while (j < len && k < buf_len) { seq[j] = buf[k]; k++; j++; } stop = j + dist; stop = stop > len ? len : stop; while (j < stop) { seq[j] = 'N'; j++; } break; case 'M': case '=': case 'X': bp += dist; break; default: break; } } delete [] buf; return 0; } #endif // __SAMI_H__ stacks-1.44/src/smoothing.h000644 000765 000024 00000012133 12764047443 016464 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2014, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __SMOOTHING_H__ #define __SMOOTHING_H__ #include #include "smoothing_utils.h" template class KSmooth { uint size; // Number of elements expected in the StatT class to smooth. double *weights; // Weight matrix to apply while smoothing. public: KSmooth(int size) { this->size = size; this->weights = calc_weights(); } ~KSmooth() { delete [] this->weights; } int smooth(vector &popstats); }; template int KSmooth::smooth(vector &popstats) { // // To generate smooth genome-wide distributions of Fst, we calculate a kernel-smoothing // moving average of Fst values along each ordered chromosome. // // For each genomic region centered on a nucleotide position c, the contribution of the population // genetic statistic at position p to the region average was weighted by the Gaussian function: // exp( (-1 * (p - c)^2) / (2 * sigma^2)) // // In addition, we weight each position according to (n_k - 1), where n_k is the number of alleles // sampled at that location. // // By default, sigma = 150Kb, for computational efficiency, only calculate average out to 3sigma. // #pragma omp parallel { int limit = 3 * sigma; int dist; uint pos_l, pos_u; double sum, final_weight; PopStat *c, *p; pos_l = 0; pos_u = 0; #pragma omp for schedule(dynamic, 1) for (uint pos_c = 0; pos_c < popstats.size(); pos_c++) { c = popstats[pos_c]; if (c == NULL) continue; for (uint i = 0; i < this->size; i++) c->smoothed[i] = 0.0; sum = 0.0; determine_window_limits(popstats, c->bp, pos_l, pos_u); for (uint pos_p = pos_l; pos_p < pos_u; pos_p++) { p = popstats[pos_p]; if (p == NULL) continue; dist = p->bp > c->bp ? p->bp - c->bp : c->bp - p->bp; if (dist > limit || dist < 0) { #pragma omp critical { cerr << "ERROR: current basepair is out of the sliding window.\n" << " Calculating sliding window; start position: " << pos_l << ", " << (popstats[pos_l] == NULL ? -1 : popstats[pos_l]->bp +1) << "bp; end position: " << pos_u << ", " << (popstats[pos_u] == NULL ? -1 : popstats[pos_u]->bp +1) << "bp; center: " << pos_c << ", " << popstats[pos_c]->bp +1 << "bp\n" << " Current position: " << pos_p << ", " << popstats[pos_p]->bp +1 << "; Dist: " << dist << "\n" << " Window positions:\n"; for (uint j = pos_l; j < pos_u; j++) { p = popstats[j]; if (p == NULL) continue; cerr << " Position: " << j << "; " << p->bp +1 << "bp\n"; } //exit(0); } continue; } // sites_cnt++; final_weight = (p->alleles - 1) * this->weights[dist]; for (uint i = 0; i < this->size; i++) c->smoothed[i] += p->stat[i] * final_weight; sum += final_weight; // if (c->loc_id == 9314) { // cerr << " id: " << p->loc_id // << "; dist: " << dist // << "; weight: " << weights[dist] // << "; final_weight: " << final_weight // << "; fst': " << p->stat[3] // << "; sum: " << sum // << "; smoothed: " << c->smoothed[3] << "\n"; // } } // sites_per_snp += (sites_cnt / snp_cnt); // tot_windows++; // // if (snp_cnt < max_snp_dist) { // #pragma omp atomic // snp_dist[snp_cnt]++; // } // c->snp_cnt = snp_cnt; for (uint i = 0; i < this->size; i++) c->smoothed[i] /= sum; } } return 0; } #endif // __SMOOTHING_H__ stacks-1.44/src/smoothing_utils.h000644 000765 000024 00000004326 12764047443 017711 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2014, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __SMOOTHING_UTILS_H__ #define __SMOOTHING_UTILS_H__ #include #include using std::vector; extern double sigma; inline double * calc_weights() { // // Calculate weights for window smoothing operations. // // For each genomic region centered on a nucleotide position c, the contribution of the population // genetic statistic at position p to the region average was weighted by the Gaussian function: // exp( (-1 * (p - c)^2) / (2 * sigma^2)) // int limit = 3 * sigma; double *weights = new double[limit + 1]; for (int i = 0; i <= limit; i++) weights[i] = exp((-1 * pow(i, 2)) / (2 * pow(sigma, 2))); return weights; } template inline int determine_window_limits(vector &sites, uint center_bp, uint &pos_l, uint &pos_u) { int limit = 3 * sigma; int limit_l = center_bp - limit > 0 ? center_bp - limit : 0; int limit_u = center_bp + limit; while (pos_l < sites.size()) { if (sites[pos_l] == NULL) { pos_l++; } else { if (sites[pos_l]->bp < limit_l) pos_l++; else break; } } while (pos_u < sites.size()) { if (sites[pos_u] == NULL) { pos_u++; } else { if (sites[pos_u]->bp < limit_u) pos_u++; else break; } } return 0; } #endif // __SMOOTHING_UTILS_H__ stacks-1.44/src/sql_utilities.cc000644 000765 000024 00000017203 12750704270 017501 0ustar00catchenstaff000000 000000 #include "sql_utilities.h" using namespace std; int load_catalog_matches(string sample, vector &matches) { CatMatch *m; string f; vector parts; long int line_num; ifstream fh; gzFile gz_fh; char *line = (char *) malloc(sizeof(char) * max_len); int size = max_len; int cnt = 0; bool gzip = false; int fh_status = 1; f = sample + ".matches.tsv"; fh.open(f.c_str(), ifstream::in); if (fh.fail()) { // // Test for a gzipped file. // f = sample + ".matches.tsv.gz"; gz_fh = gzopen(f.c_str(), "rb"); if (!gz_fh) return 0; #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_fh, libz_buffer_size); #endif gzip = true; } cerr << " Parsing " << f.c_str() << "\n"; line_num = 1; while (fh_status) { fh_status = (gzip == true) ? read_gzip_line(gz_fh, &line, &size) : read_line(fh, &line, &size); line_num++; if (!fh_status && strlen(line) == 0) continue; if (is_comment(line)) continue; parse_tsv(line, parts); cnt = parts.size(); if (cnt != num_matches_fields && cnt != num_matches_fields - 1) { cerr << "Error parsing " << f.c_str() << " at line: " << line_num << ". (" << parts.size() << " fields).\n"; return 0; } m = new CatMatch; m->batch_id = atoi(parts[1].c_str()); m->cat_id = atoi(parts[2].c_str()); m->sample_id = atoi(parts[3].c_str()); m->tag_id = atoi(parts[4].c_str()); m->haplotype = new char[parts[5].length() + 1]; strcpy(m->haplotype, parts[5].c_str()); m->depth = atoi(parts[6].c_str()); m->lnl = is_double(parts[7].c_str()); if (cnt == num_matches_fields && parts[8].length() > 0) { m->cigar = new char[parts[8].length() + 1]; strcpy(m->cigar, parts[8].c_str()); } matches.push_back(m); } if (gzip) gzclose(gz_fh); else fh.close(); return 0; } int load_model_results(string sample, map &modres) { string f; vector parts; long int line_num; ifstream fh; gzFile gz_fh; char *line = (char *) malloc(sizeof(char) * max_len); int size = max_len; bool gzip = false; bool open_fail = false; int fh_status = 1; // // Parse the models file (if it exists), otherwise parse the tag file to // pull in the model calls for each locus. // gzip = false; fh_status = 1; line_num = 1; f = sample + ".models.tsv"; fh.open(f.c_str(), ifstream::in); if (fh.fail()) open_fail = true; if (open_fail) { // // Test for a gzipped MODELs file. // f = sample + ".models.tsv.gz"; gz_fh = gzopen(f.c_str(), "rb"); if (!gz_fh) { open_fail = true; } else { open_fail = false; #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_fh, libz_buffer_size); #endif gzip = true; } } if (open_fail) { // // Test for a TAGs file. // f = sample + ".tags.tsv"; fh.open(f.c_str(), ifstream::in); if (fh.fail()) open_fail = true; else open_fail = false; } if (open_fail) { // // Test for a gzipped TAGs file. // f = sample + ".tags.tsv.gz"; gz_fh = gzopen(f.c_str(), "rb"); if (!gz_fh) { open_fail = true; } else { open_fail = false; #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_fh, libz_buffer_size); #endif gzip = true; } } if (open_fail) { cerr << " Unable to open '" << sample << "'\n"; return 0; } cerr << " Parsing " << f.c_str() << "\n"; ModRes *mod; uint tag_id, samp_id; while (fh_status) { fh_status = (gzip == true) ? read_gzip_line(gz_fh, &line, &size) : read_line(fh, &line, &size); line_num++; if (!fh_status && strlen(line) == 0) continue; if (is_comment(line)) continue; parse_tsv(line, parts); if (parts.size() != num_tags_fields) { cerr << "Error parsing " << f.c_str() << " at line: " << line_num << ". (" << parts.size() << " fields).\n"; return 0; } // // Read the model sequence, a series of letters specifying if the model called a // homozygous base (O), a heterozygous base (E), or if the base type was unknown (U). // if (parts[6] != "model") continue; samp_id = atoi(parts[1].c_str()); tag_id = atoi(parts[2].c_str()); mod = new ModRes(samp_id, tag_id, parts[9].c_str()); modres[tag_id] = mod; } if (gzip) gzclose(gz_fh); else fh.close(); delete [] line; return 1; } int load_snp_calls(string sample, map &snpres) { string f; int id, samp_id; vector parts; long int line_num; SNP *snp; SNPRes *snpr; ifstream fh; gzFile gz_fh; char *line = (char *) malloc(sizeof(char) * max_len); int size = max_len; bool gzip = false; int fh_status = 1; // // Parse the SNP file // f = sample + ".snps.tsv"; fh.open(f.c_str(), ifstream::in); if (fh.fail()) { // // Test for a gzipped file. // f = sample + ".snps.tsv.gz"; gz_fh = gzopen(f.c_str(), "rb"); if (!gz_fh) { cerr << " Unable to open '" << sample << "'\n"; return 0; } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_fh, libz_buffer_size); #endif gzip = true; } cerr << " Parsing " << f.c_str() << "\n"; line_num = 1; while (fh_status) { fh_status = (gzip == true) ? read_gzip_line(gz_fh, &line, &size) : read_line(fh, &line, &size); if (!fh_status && strlen(line) == 0) continue; if (is_comment(line)) continue; parse_tsv(line, parts); if (parts.size() != num_snps_fields && parts.size() != num_snps_fields - 2) { cerr << "Error parsing " << f.c_str() << " at line: " << line_num << ". (" << parts.size() << " fields).\n"; return 0; } samp_id = atoi(parts[1].c_str()); id = atoi(parts[2].c_str()); snp = new SNP; snp->col = atoi(parts[3].c_str()); if (parts[4] == "O") snp->type = snp_type_hom; else if (parts[4] == "E") snp->type = snp_type_het; else snp->type = snp_type_unk; snp->lratio = atof(parts[5].c_str()); snp->rank_1 = parts[6].at(0); snp->rank_2 = parts[7].at(0) == '-' ? 0 : parts[7].at(0); if (parts.size() == 10) { if (parts[8].length() == 0 || parts[8].at(0) == '-') snp->rank_3 = 0; else snp->rank_3 = parts[8].at(0); if (parts[9].length() == 0 || parts[9].at(0) == '-') snp->rank_4 = 0; else snp->rank_4 = parts[9].at(0); } if (snpres.count(id) == 0) { snpr = new SNPRes(samp_id, id); snpres[id] = snpr; } snpres[id]->snps.push_back(snp); line_num++; } if (gzip) gzclose(gz_fh); else fh.close(); delete [] line; return 1; } stacks-1.44/src/sql_utilities.h000644 000765 000024 00000031411 12774470747 017357 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // sql_utilities.h -- template routines to read and write Stacks SQL file formats. // #ifndef __SQL_UTILITIES_H__ #define __SQL_UTILITIES_H__ #include "input.h" #include "utils.h" // // The expected number of tab-separated fields in our SQL input files. // const uint num_tags_fields = 14; const uint num_snps_fields = 10; const uint num_alleles_fields = 6; const uint num_matches_fields = 9; int load_catalog_matches(string sample, vector &matches); int load_model_results(string sample, map &modres); int load_snp_calls(string sample, map &snpres); template int load_loci(string sample, map &loci, bool store_reads, bool load_all_model_calls, bool &compressed) { LocusT *c; SNP *snp; string f; char *cmp; const char *p, *q; int len; vector parts; set blacklisted; long int line_num; ifstream fh; gzFile gz_fh; char *line = (char *) malloc(sizeof(char) * max_len); int size = max_len; bool gzip = false; bool open_fail = true; int fh_status = 1; // // // // First, try to parse the models file to pull in the consensus sequence and model string // // for each locus. If the models file is not available or we are requested to store the // // reads from each stack, fall back to the tags file. // // // if (!store_reads) { // f = sample + ".models.tsv"; // fh.open(f.c_str(), ifstream::in); // open_fail = fh.fail() ? true : false; // } // if (!store_reads && open_fail) { // // // // Test for a gzipped MODELs file. // // // f = sample + ".models.tsv.gz"; // gz_fh = gzopen(f.c_str(), "rb"); // if (!gz_fh) { // open_fail = true; // } else { // open_fail = false; // #if ZLIB_VERNUM >= 0x1240 // gzbuffer(gz_fh, libz_buffer_size); // #endif // gzip = true; // } // } if (open_fail) { // // Test for a TAGs file. // f = sample + ".tags.tsv"; fh.open(f.c_str(), ifstream::in); if (fh.fail()) open_fail = true; else open_fail = false; } if (open_fail) { // // Test for a gzipped TAGs file. // f = sample + ".tags.tsv.gz"; gz_fh = gzopen(f.c_str(), "rb"); if (!gz_fh) { open_fail = true; } else { open_fail = false; #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_fh, libz_buffer_size); #endif gzip = true; } } if (open_fail) { cerr << " Unable to open '" << sample << "'\n"; return 0; } cerr << " Parsing " << f.c_str() << "\n"; uint id; line_num = 1; while (fh_status) { fh_status = (gzip == true) ? read_gzip_line(gz_fh, &line, &size) : read_line(fh, &line, &size); if (!fh_status && strlen(line) == 0) continue; if (is_comment(line)) continue; parse_tsv(line, parts); if (parts.size() != num_tags_fields) { cerr << "Error parsing " << f.c_str() << " at line: " << line_num << ". (" << parts.size() << " fields).\n"; return 0; } id = atoi(parts[2].c_str()); if (parts[6] != "consensus") { if (blacklisted.count(id)) continue; // // Make sure this locus has already been defined (consensus sequence SHOULD always // be specified first in the file for a particular locus). // if (loci.count(id) > 0) { // // Read the model sequence, a series of letters specifying if the model called a // homozygous base (O), a heterozygous base (E), or if the base type was unknown (U). // if (parts[6] == "model") { loci[id]->model = new char[parts[9].length() + 1]; strcpy(loci[id]->model, parts[9].c_str()); } else { // // Otherwise, we expect a primary or secondary read, record these if specified. // loci[id]->depth++; if (store_reads) { char *read = new char[parts[9].length() + 1]; strcpy(read, parts[9].c_str()); loci[id]->reads.push_back(read); char *read_id = new char[parts[8].length() + 1]; strcpy(read_id, parts[8].c_str()); loci[id]->comp.push_back(read_id); // // Store the internal stack number for this read. // loci[id]->comp_cnt.push_back(atoi(parts[7].c_str())); // // Store the read type. // if (parts[6] == "primary") loci[id]->comp_type.push_back(primary); else loci[id]->comp_type.push_back(secondary); } } continue; } else { cerr << "Error parsing " << f.c_str() << " at line: " << line_num << ". (stack " << id << " does not exist).\n"; return 0; } } // // Do not include blacklisted tags in the catalog. They are tags that are composed // of noise and/or repetitive sequence. // if (parts[11] == "1") { blacklisted.insert(id); continue; } c = new LocusT; c->sample_id = atoi(parts[1].c_str()); c->id = id; c->add_consensus(parts[9].c_str()); // // Read in the flags // c->deleveraged = (parts[10] == "1" ? true : false); c->lumberjackstack = (parts[12] == "1" ? true : false); // // Read in the log likelihood of the locus. // c->lnl = is_double(parts[13].c_str()); // // Parse the physical genome location of this locus. // c->loc.set(parts[3].c_str(), atoi(parts[4].c_str()), (parts[5] == "+" ? strand_plus : strand_minus)); // // Parse the components of this stack (either the Illumina ID, or the catalog constituents) // q = parts[8].c_str(); while (*q != '\0') { for (p = q; *q != ',' && *q != '\0'; q++); len = q - p; cmp = new char[len + 1]; strncpy(cmp, p, len); cmp[len] = '\0'; c->comp.push_back(cmp); if (*q != '\0') q++; } loci[c->id] = c; line_num++; } if (gzip) gzclose(gz_fh); else fh.close(); // // Next, parse the SNP file and load model calls. // gzip = false; fh_status = 1; line_num = 1; f = sample + ".snps.tsv"; fh.open(f.c_str(), ifstream::in); if (fh.fail()) { // // Test for a gzipped file. // f = sample + ".snps.tsv.gz"; gz_fh = gzopen(f.c_str(), "rb"); if (!gz_fh) { cerr << " Unable to open '" << sample << "'\n"; return 0; } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_fh, libz_buffer_size); #endif gzip = true; compressed = true; } cerr << " Parsing " << f.c_str() << "\n"; while (fh_status) { fh_status = (gzip == true) ? read_gzip_line(gz_fh, &line, &size) : read_line(fh, &line, &size); if (!fh_status && strlen(line) == 0) continue; if (is_comment(line)) continue; parse_tsv(line, parts); if (parts.size() != num_snps_fields && parts.size() != num_snps_fields - 2) { cerr << "Error parsing " << f.c_str() << " at line: " << line_num << ". (" << parts.size() << " fields).\n"; return 0; } id = atoi(parts[2].c_str()); if (blacklisted.count(id)) continue; // // Only load heterozygous model calls. // if (load_all_model_calls == false && parts[4] != "E") continue; snp = new SNP; snp->col = atoi(parts[3].c_str()); snp->lratio = atof(parts[5].c_str()); snp->rank_1 = parts[6].at(0); snp->rank_2 = parts[7].at(0) == '-' ? 0 : parts[7].at(0); if (parts[4] == "E") snp->type = snp_type_het; else if (parts[4] == "O") snp->type = snp_type_hom; else snp->type = snp_type_unk; if (parts.size() == 10) { if (parts[8].length() == 0 || parts[8].at(0) == '-') snp->rank_3 = 0; else snp->rank_3 = parts[8].at(0); if (parts[9].length() == 0 || parts[9].at(0) == '-') snp->rank_4 = 0; else snp->rank_4 = parts[9].at(0); } if (loci.count(id) > 0) { loci[id]->snps.push_back(snp); } else { cerr << "Error parsing " << f.c_str() << " at line: " << line_num << ". SNP asks for nonexistent locus with ID: " << id << "\n"; return 0; } line_num++; } if (gzip) gzclose(gz_fh); else fh.close(); // // Finally, parse the Alleles file // gzip = false; fh_status = 1; line_num = 1; f = sample + ".alleles.tsv"; fh.open(f.c_str(), ifstream::in); if (fh.fail()) { // // Test for a gzipped file. // f = sample + ".alleles.tsv.gz"; gz_fh = gzopen(f.c_str(), "rb"); if (!gz_fh) { cerr << " Unable to open '" << sample << "'\n"; return 0; } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_fh, libz_buffer_size); #endif gzip = true; compressed = true; } cerr << " Parsing " << f.c_str() << "\n"; while (fh_status) { fh_status = (gzip == true) ? read_gzip_line(gz_fh, &line, &size) : read_line(fh, &line, &size); if (!fh_status && strlen(line) == 0) continue; if (is_comment(line)) continue; parse_tsv(line, parts); if (parts.size() != num_alleles_fields) { cerr << "Error parsing " << f.c_str() << " at line: " << line_num << ". (" << parts.size() << " fields).\n"; return 0; } id = atoi(parts[2].c_str()); if (blacklisted.count(id)) continue; if (loci.count(id) > 0) { loci[id]->alleles[parts[3]] = atoi(parts[5].c_str()); } else { cerr << "Error parsing " << f.c_str() << " at line: " << line_num << ". SNP asks for nonexistent locus with ID: " << id << "\n"; return 0; } line_num++; } if (gzip) gzclose(gz_fh); else fh.close(); // // Populate the strings member with the sequence for each allele for each Locus. // typename map::iterator i; for (i = loci.begin(); i != loci.end(); i++) i->second->populate_alleles(); free(line); return 1; } template int dump_loci(map &u) { typename map::iterator i; vector::iterator s; for (i = u.begin(); i != u.end(); i++) { cerr << "Locus ID: " << i->second->id << "\n" << " Consensus: " << i->second->con << "\n" << " Genomic Location: " << i->second->loc.chr << "; " << i->second->loc.bp +1 << "bp\n" << " SNPs:\n"; for (s = i->second->snps.begin(); s != i->second->snps.end(); s++) cerr << " Col: " << (*s)->col << " rank 1: " << (*s)->rank_1 << " rank 2: " << (*s)->rank_2 << "\n"; cerr << "\n"; } return 0; } #endif // __SQL_UTILITIES_H__ stacks-1.44/src/sstacks.cc000644 000765 000024 00000135620 12775212561 016272 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // sstacks -- search for occurances of stacks in a catalog of stacks. // #include "sstacks.h" // Global variables to hold command-line options. queue samples; string catalog_path; string out_path; FileT in_file_type = FileT::sql; int num_threads = 1; int batch_id = 0; int samp_id = 0; int catalog = 0; bool verify_haplotypes = true; bool impute_haplotypes = true; bool require_uniq_haplotypes = false; bool gapped_alignments = false; searcht search_type = sequence; double min_match_len = 0.80; double max_gaps = 2.0; int gapped_kmer_len = 19; int main (int argc, char* argv[]) { parse_command_line(argc, argv); uint sample_cnt = samples.size(); // // Set the number of OpenMP parallel threads to execute. // #ifdef _OPENMP omp_set_num_threads(num_threads); #endif map catalog; bool compressed = false; int res; if (search_type == sequence) cerr << "Searching for matches by sequence identity...\n"; else if (search_type == genomic_loc) cerr << "Searching for matches by genomic location...\n"; catalog_path += ".catalog"; res = load_loci(catalog_path, catalog, false, false, compressed); if (res == 0) { cerr << "Unable to parse catalog, '" << catalog_path << "'\n"; return 0; } KmerHashMap kmer_map; map > allele_map; vector kmer_map_keys; if (gapped_alignments) { cerr << "Populating kmer dictionary for gapped alignments..."; populate_kmer_hash(catalog, kmer_map, kmer_map_keys, allele_map, gapped_kmer_len); cerr << "done.\n"; } string sample_path; int i = 1; while (!samples.empty()) { map sample; sample_path = samples.front(); samples.pop(); cerr << "Processing sample '" << sample_path << "' [" << i << " of " << sample_cnt << "]\n"; res = load_loci(sample_path, sample, false, false, compressed); if (res == 0) { cerr << "Unable to parse '" << sample_path << "'\n"; return 0; } in_file_type = compressed == true ? FileT::gzsql : FileT::sql; // // Assign the ID for this sample data. // samp_id = sample.begin()->second->sample_id; //dump_loci(catalog); //dump_loci(sample); if (search_type == sequence) { cerr << "Searching for sequence matches...\n"; find_matches_by_sequence(catalog, sample); if (gapped_alignments) { cerr << "Searching for gapped alignments...\n"; search_for_gaps(catalog, sample, kmer_map, allele_map, min_match_len); } } else if (search_type == genomic_loc) { cerr << "Searching for matches by genomic location...\n"; find_matches_by_genomic_loc(catalog, sample); } write_matches(sample_path, sample); i++; // // Free memory associated with the sample. // for (map::iterator j = sample.begin(); j != sample.end(); j++) delete j->second; sample.clear(); } if (gapped_alignments) free_kmer_hash(kmer_map, kmer_map_keys); // // Free memory associated with the catalog. // for (map::iterator j = catalog.begin(); j != catalog.end(); j++) delete j->second; catalog.clear(); return 0; } int find_matches_by_genomic_loc(map &sample_1, map &sample_2) { // // Calculate the distance (number of mismatches) between each pair // of Radtags. We expect all radtags to be the same length; // map::iterator i; map::iterator j; int k; char id[id_len]; // // Build a hash map out of the first sample (usually the catalog) // // // Create a map of the genomic locations of stacks in sample_1 // cerr << " Creating map of genomic locations..."; map > locations; for (j = sample_1.begin(); j != sample_1.end(); j++) { snprintf(id, id_len - 1, "%s|%d|%c", j->second->loc.chr, j->second->loc.bp, j->second->loc.strand == strand_plus ? '+' : '-'); locations[id].insert(j->second->id); } cerr << "done.\n"; // OpenMP can't parallelize random access iterators, so we convert // our map to a vector of integer keys. vector keys; for (i = sample_2.begin(); i != sample_2.end(); i++) keys.push_back(i->first); // // Initialize some counters // unsigned long matches = 0; unsigned long nomatch = 0; unsigned long nosnps = 0; unsigned long tot_hap = 0; unsigned long ver_hap = 0; #pragma omp parallel private(i, j, k, id) { unsigned long verified; #pragma omp for reduction(+:matches) reduction(+:tot_hap) reduction(+:ver_hap) reduction(+:nomatch) reduction(+:nosnps) for (k = 0; k < (int) keys.size(); k++) { i = sample_2.find(keys[k]); snprintf(id, id_len - 1, "%s|%d|%c", i->second->loc.chr, i->second->loc.bp, i->second->loc.strand == strand_plus ? '+' : '-'); if (locations.count(id) > 0) { Locus *tag; set::iterator loc_it; vector >::iterator q; matches++; for (loc_it = locations[id].begin(); loc_it != locations[id].end(); loc_it++) { tag = sample_1[*loc_it]; // // Generate haplotypes for query tag relative to the catalog tag. // set query_haplotypes; generate_query_haplotypes(tag, i->second, query_haplotypes); tot_hap += query_haplotypes.size() > 0 ? query_haplotypes.size() : 1; if (verify_haplotypes) { verified = verify_genomic_loc_match(tag, i->second, query_haplotypes, nosnps); ver_hap += verified; if (verified == 0) nomatch++; } else { i->second->add_match(tag->id, tag->strings.begin()->first); } } } } } cerr << keys.size() << " stacks matched against the catalog containing " << sample_1.size() << " loci.\n" << " " << matches << " matching loci, " << nomatch << " contained no verified haplotypes.\n" << " " << nosnps << " loci contained SNPs unaccounted for in the catalog and were excluded.\n" << " " << tot_hap << " total haplotypes examined from matching loci, " << ver_hap << " verified.\n"; return 0; } int verify_genomic_loc_match(Locus *s1_tag, QLocus *s2_tag, set &query_haplotypes, unsigned long &nosnps) { vector::iterator i, j; // // We have found a match between the genomic location of s1 and s2. We now want // to verify that the haplotypes are consistent between the tags, i.e. they // have the same number and types of SNPs. // // // 1. First, if there are no SNPs present in either the query or catalog, just // check that the strings match. // uint min_len = s1_tag->len > s2_tag->len ? s2_tag->len : s1_tag->len; if (s1_tag->snps.size() == 0 && s2_tag->snps.size() == 0 && strncmp(s1_tag->con, s2_tag->con, min_len) == 0) { s2_tag->add_match(s1_tag->id, "consensus"); return 1; } // // 2. Second, we will check that the query locus (s2_tag) does not have any SNPs // lacking in the catalog tag (s1_tag). // bool found; for (j = s2_tag->snps.begin(); j != s2_tag->snps.end(); j++) { found = false; // // SNP occurs in a column that is beyond the length of the catalog // if ((*j)->col > min_len - 1) continue; for (i = s1_tag->snps.begin(); i != s1_tag->snps.end(); i++) { if ((*i)->col == (*j)->col) found = true; } // // Query locus posses a SNP not present in the catalog. // if (found == false) { nosnps++; return 0; } } // // Finally, check that one of the constructed alleles matches the allele // passed in on the stack. // string cat_haplotype; vector >::iterator c; set::iterator a; uint matches = 0; for (a = query_haplotypes.begin(); a != query_haplotypes.end(); a++) { if (impute_haplotypes) { int res = impute_haplotype(*a, s1_tag->strings, cat_haplotype); if (res > 0) { // // If the matching haplotype was imputed, record the depths of the query alleles // under the new, imputed alleles. // if (s2_tag->alleles.count(cat_haplotype) == 0) { if (s2_tag->alleles.count(*a) > 0) s2_tag->alleles[cat_haplotype] = s2_tag->alleles[*a]; else s2_tag->alleles[cat_haplotype] = s2_tag->depth; } //cerr << s2_tag->id << "; Adding cat haplotype: " << cat_haplotype << " based on depth of " << *a << ", " << s2_tag->alleles[cat_haplotype] << "\n"; s2_tag->add_match(s1_tag->id, cat_haplotype); matches++; } else if (res < 0) { cerr << " Failure imputing haplotype for catalog locus: " << s1_tag->id << " and query tag: " << s2_tag->id << "\n"; } } else { for (c = s1_tag->strings.begin(); c != s1_tag->strings.end(); c++) if (*a == c->first) { //cerr << " Adding match between " << s1_tag->id << " and " << c->first << "\n"; s2_tag->add_match(s1_tag->id, c->first); matches++; } } } return matches; } // int impute_haplotype(string query_haplotype, // vector > &cat_haplotypes, // string &match) { // uint max_len = query_haplotype.length() > cat_haplotypes[0].first.length() ? // query_haplotype.length() : // cat_haplotypes[0].first.length(); // //cerr << "Query len: " << query_haplotype.length() << "; Max length: " << max_len << "\n"; // vector cur, next; // for (uint i = 0; i < cat_haplotypes.size(); i++) // cur.push_back(cat_haplotypes[i].first); // match = ""; // // // // Examine the haplotypes one SNP at a time. If we are able to uniquely // // determine the catalog haplotype that the query haplotype corresponds // // to, return it. // // // uint j = 0; // while (cur.size() > 1 && j < max_len) { // for (uint i = 0; i < cur.size(); i++) { // //cerr << "Comparing query[" << j << "]: '" << query_haplotype[j] << "' to catalog '" << cur[i][j] << "'\n"; // if (query_haplotype[j] == cur[i][j]) { // //cerr << " Keeping this haplotype.\n"; // next.push_back(cur[i]); // } // } // cur = next; // next.clear(); // j++; // } // // // // If there is only one left, make sure what we have of the haplotype does match // // and its not simply an erroneously called haplotype. // // // if (cur.size() == 1 && // strncmp(cur[0].c_str(), query_haplotype.c_str(), max_len) == 0) { // match = cur[0]; // return 1; // } // // // // If, after examining all the available SNPs in the query haplotype, there is // // still more than a single possible catalog haplotype, then we can't impute it. // // // return 0; // } int impute_haplotype(string query_haplotype, vector > &cat_haplotypes, string &match) { if (cat_haplotypes.size() == 0) { cerr << "Warning: malformed catalog tag: missing haplotype information.\n"; return -1; } //cerr << "Examining " << query_haplotype << "\n"; uint max_len = query_haplotype.length() > cat_haplotypes[0].first.length() ? query_haplotype.length() : cat_haplotypes[0].first.length(); //cerr << "Query len: " << query_haplotype.length() << "; Max length: " << max_len << "\n"; vector cur, next; uint match_cnt, no_n_cnt; for (uint i = 0; i < cat_haplotypes.size(); i++) cur.push_back(cat_haplotypes[i].first); match = ""; // // Examine the haplotypes one SNP at a time. If we are able to uniquely // determine the catalog haplotype that the query haplotype corresponds // to, return it. // uint j = 0; while (cur.size() > 1 && j < max_len) { for (uint i = 0; i < cur.size(); i++) { //cerr << "Comparing query[" << j << "]: '" << query_haplotype << "' to catalog '" << cur[i] << "'\n"; if (require_uniq_haplotypes && (query_haplotype[j] == cur[i][j] || query_haplotype[j] == 'N')) { //cerr << " Keeping this haplotype.\n"; next.push_back(cur[i]); } else if (query_haplotype[j] == cur[i][j]) { //cerr << " Keeping this haplotype.\n"; next.push_back(cur[i]); } //else { //cerr << " Discarding this haplotype.\n"; //} } cur = next; next.clear(); j++; } // // If there is only one left, make sure what we have of the haplotype does match // and its not simply an erroneously called haplotype. // no_n_cnt = 0; match_cnt = 0; if (cur.size() == 1) { if (require_uniq_haplotypes) { for (uint k = 0; k < max_len; k++) if (query_haplotype[k] != 'N') no_n_cnt++; for (uint k = 0; k < max_len; k++) if (cur[0][k] == query_haplotype[k]) match_cnt++; if (match_cnt == no_n_cnt) { //cerr << "Keeping " << query_haplotype << "\n"; match = cur[0]; return 1; } } else { if (strncmp(cur[0].c_str(), query_haplotype.c_str(), max_len) == 0) { match = cur[0]; return 1; } } } // // If, after examining all the available SNPs in the query haplotype, there is // still more than a single possible catalog haplotype, then we can't impute it. // return 0; } int generate_query_haplotypes(Locus *s1_tag, QLocus *s2_tag, set &query_haplotypes) { // // Construct a set of haplotypes from the query locus relative to the catalog locus. // (The query locus already has a set of haplotypes, however, they don't necessarily // account for all the SNPs in the catalog, so we will augment them with sequence // from the consensus.) // if (s1_tag->snps.size() == 0 && s2_tag->snps.size() == 0) return 0; vector > merged_snps; map > columns; map >::iterator c; vector >::iterator k; vector::iterator i; for (i = s1_tag->snps.begin(); i != s1_tag->snps.end(); i++) columns[(*i)->col] = make_pair("catalog", *i); for (i = s2_tag->snps.begin(); i != s2_tag->snps.end(); i++) { // // Is this column already represented in the catalog? // if (columns.count((*i)->col)) columns[(*i)->col] = make_pair("both", *i); else columns[(*i)->col] = make_pair("query", *i); } for (c = columns.begin(); c != columns.end(); c++) merged_snps.push_back((*c).second); // // Sort the SNPs by column // sort(merged_snps.begin(), merged_snps.end(), compare_pair_snp); map converted_alleles; map::iterator b; string old_allele, new_allele; int pos; for (b = s2_tag->alleles.begin(); b != s2_tag->alleles.end(); b++) { old_allele = b->first; new_allele = ""; pos = 0; for (k = merged_snps.begin(); k != merged_snps.end(); k++) { // // If the SNPs from the catalog haplotype beyond the length of the query, add Ns // if (k->first == "catalog") { new_allele += (k->second->col > s2_tag->len - 1) ? 'N' : s2_tag->con[k->second->col]; } else { new_allele += old_allele[pos]; pos++; } } query_haplotypes.insert(new_allele); converted_alleles[new_allele] = b->second; // cerr << "Adding haplotype: " << new_allele << " [" << b->first << "]\n"; } if (s2_tag->alleles.size() == 0) { new_allele = ""; for (k = merged_snps.begin(); k != merged_snps.end(); k++) { new_allele += (k->second->col > s2_tag->len - 1) ? 'N' : s2_tag->con[k->second->col]; } query_haplotypes.insert(new_allele); // cerr << "Adding haplotype 2: " << new_allele << "\n"; } else { s2_tag->alleles.clear(); for (b = converted_alleles.begin(); b != converted_alleles.end(); b++) s2_tag->alleles[b->first] = b->second; } return 0; } int find_matches_by_sequence(map &sample_1, map &sample_2) { map::iterator i; uint min_tag_len; // // We don't assume all radtags will be the same length. // min_tag_len = sample_1.begin()->second->len > sample_2.begin()->second->len ? sample_2.begin()->second->len : sample_1.begin()->second->len; // // Build a hash map out of the catalog, using only the minimum length // substring of the longest reads; // HashMap sample_1_map; vector sample_1_map_keys; populate_hash(sample_1, sample_1_map, sample_1_map_keys, min_tag_len); // // OpenMP can't parallelize random access iterators, so we convert // our map to a vector of integer keys. // vector keys; for (i = sample_2.begin(); i != sample_2.end(); i++) keys.push_back(i->first); // // Initialize some counters // unsigned long matches = 0; unsigned long mmatch = 0; unsigned long nosnps = 0; unsigned long nomatch = 0; unsigned long no_haps = 0; unsigned long tot_hap = 0; unsigned long ver_hap = 0; #pragma omp parallel { #pragma omp for reduction(+:matches) reduction(+:tot_hap) reduction(+:ver_hap) reduction(+:nomatch) reduction(+:mmatch) for (uint k = 0; k < keys.size(); k++) { QLocus *query = sample_2[keys[k]]; // // Iterate through the haplotypes for this tag in sample_2 // HashMap::iterator hit; vector >::iterator q; // Query records allele_type/search string pairs vector >::iterator c; // Hash map records id/allele_type pairs map > haplo_hits; set loci_hit; for (q = query->strings.begin(); q != query->strings.end(); q++) { // cerr << " Looking for haplotype: " << q->first << " with sequence " << q->second.substr(0, min_tag_len) << "\n"; hit = sample_1_map.find(q->second.substr(0, min_tag_len).c_str()); if (hit != sample_1_map.end()) { tot_hap++; // cerr << " Found a match for " << hit->first << "\n"; for (c = hit->second.begin(); c != hit->second.end(); c++) { // // Record the catalog loci hit by the haplotypes of this query locus. // loci_hit.insert(c->first); // // Record the haplotypes hit between the query and catalog loci. // haplo_hits[q->first].push_back(c->second); if (verify_haplotypes == false) query->add_match(c->first, c->second); } } } if (loci_hit.size() == 0) nomatch++; else if (loci_hit.size() > 0) matches++; if (verify_haplotypes && loci_hit.size() > 0) { uint verified = verify_sequence_match(sample_1, query, loci_hit, haplo_hits, min_tag_len, mmatch, nosnps); ver_hap += verified; if (verified == 0) no_haps++; } } } // // Free memory associated with the hash. // for (HashMap::iterator i = sample_1_map.begin(); i != sample_1_map.end(); i++) i->second.clear(); sample_1_map.clear(); for (uint i = 0; i < sample_1_map_keys.size(); i++) delete [] sample_1_map_keys[i]; sample_1_map_keys.clear(); cerr << keys.size() << " stacks compared against the catalog containing " << sample_1.size() << " loci.\n" << " " << matches << " matching loci, " << no_haps << " contained no verified haplotypes.\n" << " " << mmatch << " loci matched more than one catalog locus and were excluded.\n" << " " << nosnps << " loci contained SNPs unaccounted for in the catalog and were excluded.\n" << " " << tot_hap << " total haplotypes examined from matching loci, " << ver_hap << " verified.\n"; return 0; } int verify_sequence_match(map &sample_1, QLocus *query, set &loci_hit, map > &haplo_hits, uint min_tag_len, unsigned long &mmatch, unsigned long &nosnps) { // // 1. Check that this query locus matches just a single catalog locus. // if (loci_hit.size() > 1) { mmatch++; return 0; } Locus *cat = sample_1[*(loci_hit.begin())]; // // 2. Make sure the query has no SNPs unaccounted for in the catalog. // vector::iterator i, j; bool found; for (i = query->snps.begin(); i != query->snps.end(); i++) { found = false; // // SNP occurs in a column that is beyond the length of the catalog // if ((*i)->col > min_tag_len - 1) continue; for (j = cat->snps.begin(); j != cat->snps.end(); j++) { if ((*i)->col == (*j)->col) found = true; } // // Query locus posses a SNP not present in the catalog. // if (found == false) { nosnps++; return 0; } } // // 3. We want a one-to-one correspondance between a query haplotype and a // catalog haplotype. This relationship fails when the catalog and query seqeunces // are different lengths and the full length haplotype can not be determined. // map >::iterator it; map cat_hap, query_hap; for (it = haplo_hits.begin(); it != haplo_hits.end(); it++) { query_hap[it->first] = it->second.size(); for (uint j = 0; j < it->second.size(); j++) cat_hap[it->second[j]]++; } uint verified = 0; for (it = haplo_hits.begin(); it != haplo_hits.end(); it++) for (uint j = 0; j < it->second.size(); j++) { if (cat_hap[it->second[j]] == 1 && query_hap[it->first] == 1) { verified++; query->add_match(cat->id, it->second[j]); // // If the matching haplotype was imputed, record the depths of the query alleles // under the new, imputed alleles. // if (query->alleles.count(it->second[j]) == 0) { if (query->alleles.count(it->first) > 0) query->alleles[it->second[j]] = query->alleles[it->first]; else query->alleles[it->second[j]] = query->depth; } } } return verified; } int search_for_gaps(map &catalog, map &sample, KmerHashMap &kmer_map, map > &allele_map, double min_match_len) { // // Search for loci that can be merged with a gapped alignment. // // // OpenMP can't parallelize random access iterators, so we convert // our map to a vector of integer keys. // map::iterator it; vector keys; for (it = sample.begin(); it != sample.end(); it++) keys.push_back(it->first); // // Calculate the number of k-mers we will generate. If kmer_len == 0, // determine the optimal length for k-mers. // int con_len = strlen(sample[keys[0]]->con); int num_kmers = con_len - gapped_kmer_len + 1; // // Calculate the minimum number of matching k-mers required for a possible sequence match. // int min_hits = (round((double) con_len * min_match_len) - (gapped_kmer_len * max_gaps)) - gapped_kmer_len + 1; cerr << " Searching with a k-mer length of " << gapped_kmer_len << " (" << num_kmers << " k-mers per read); " << min_hits << " k-mer hits required.\n"; uint gapped_aln = 0; uint matches = 0; uint mmatches = 0; uint nomatches = 0; uint ver_hap = 0; uint tot_hap = 0; uint bad_aln = 0; uint no_haps = 0; uint nosnps = 0; #pragma omp parallel { QLocus *query; Locus *tag_2; KmerHashMap::iterator h; AlignRes aln_res; vector kmers; set uniq_kmers; vector hits; vector > ordered_hits; uint hit_cnt, index, prev_id, allele_id, hits_size, stop, top_hit; pair cat_hit; string query_allele, query_seq, cat_allele, cat_seq; map > haplo_hits; set loci_hit; vector > cigar; GappedAln *aln = new GappedAln(); initialize_kmers(gapped_kmer_len, num_kmers, kmers); #pragma omp for schedule(dynamic) reduction(+:matches) reduction(+:nomatches) reduction(+:mmatches) reduction(+:gapped_aln) reduction(+:ver_hap) \ reduction(+:tot_hap) reduction(+:bad_aln) reduction(+:no_haps) for (uint i = 0; i < keys.size(); i++) { query = sample[keys[i]]; // // If we already matched this locus to the catalog without using gapped alignments, skip it now. // if (query->matches.size() > 0) continue; gapped_aln++; map > query_hits; loci_hit.clear(); for (vector >::iterator allele = query->strings.begin(); allele != query->strings.end(); allele++) { // cerr << "Allele: " << query_allele << "\n"; query_allele = allele->first; query_seq = allele->second; tot_hap++; generate_kmers_lazily(allele->second.c_str(), gapped_kmer_len, num_kmers, kmers); // // We want to create a list of unique kmers to search with; otherwise, repetitive kmers will // generate, multiple, spurious hits in sequences with multiple copies of the same kmer. // uniq_kmers.clear(); for (int j = 0; j < num_kmers; j++) uniq_kmers.insert(kmers[j]); hits.clear(); ordered_hits.clear(); // // Lookup the occurances of each k-mer in the kmer_map // for (set::iterator j = uniq_kmers.begin(); j != uniq_kmers.end(); j++) { h = kmer_map.find(j->c_str()); if (h != kmer_map.end()) for (uint k = 0; k < h->second.size(); k++) hits.push_back(h->second[k]); } // // Sort the vector of indexes; provides the number of hits to each allele/locus // and orders them largest to smallest. // sort(hits.begin(), hits.end()); // // Iterate through the list of hits and collapse them down by number of kmer hits per allele. // hits_size = hits.size(); if (hits_size == 0) continue; prev_id = hits[0]; index = 0; do { hit_cnt = 0; allele_id = prev_id; while (index < hits_size && (uint) hits[index] == prev_id) { hit_cnt++; index++; } if (index < hits_size) prev_id = hits[index]; if (hit_cnt >= (uint) min_hits) ordered_hits.push_back(make_pair(allele_id, hit_cnt)); } while (index < hits_size); if (ordered_hits.size() == 0) continue; // // Process the hits from most kmer hits to least kmer hits. // sort(ordered_hits.begin(), ordered_hits.end(), compare_pair_intint); // // Only try to align the sequences with the most kmers in common. // top_hit = ordered_hits[0].second; stop = 1; for (uint j = 1; j < ordered_hits.size(); j++) if ((uint)ordered_hits[j].second < top_hit) { stop = j; break; } for (uint j = 0; j < stop; j++) { cat_hit = allele_map.at(ordered_hits[j].first); hit_cnt = ordered_hits[j].second; tag_2 = catalog[cat_hit.second]; cat_allele = cat_hit.first; cat_seq = ""; for (uint k = 0; k < tag_2->strings.size(); k++) if (tag_2->strings[k].first == cat_hit.first) { cat_seq = tag_2->strings[k].second; break; } aln->init(tag_2->len, query->len); // cerr << "Attempting to align: cat id " << tag_2->id << " with locus id " << query->id << "\n" // << "Cat allele: " << cat_allele << "; seq: " << cat_seq << "\n" // << "Allele: " << query_allele << "; seq: " << allele->second << "\n"; if (aln->align(cat_seq, query_seq)) { aln->parse_cigar(cigar); aln_res = aln->result(); // // At this point in the analysis, all possible alleles we want to detect must already // be present in the catalog. Therefore, we should reject any alignment that implies a // change in the catalog sequence (with the presence of an deletion element) as // spurious. // if (aln_res.cigar.find('D') != string::npos) continue; // // If the alignment has too many gaps, skip it. // If the alignment doesn't span enough of the two sequences, skip it. // if (aln_res.gap_cnt <= (max_gaps + 1) && aln_res.pct_id >= min_match_len && dist(cat_seq.c_str(), query_seq.c_str(), cigar) == 0) { // cerr << "Adding match: " << aln_res.cigar << "\n"; loci_hit.insert(tag_2->id); query_hits[query_allele][cat_allele] = aln_res; } } } } if (verify_gapped_match(catalog, query, loci_hit, query_hits, mmatches, nosnps, no_haps, bad_aln, ver_hap)) matches++; else nomatches++; } // // Free the k-mers we generated for this query and the alignment class. // for (uint j = 0; j < kmers.size(); j++) delete [] kmers[j]; kmers.clear(); delete aln; } cerr << "Out of " << keys.size() << " query loci, " << gapped_aln << " gapped alignments attempted.\n" << " " << matches << " loci matched one catalog locus; " << tot_hap << " total haplotypes examined, " << ver_hap << " verified.\n" << " " << nomatches << " loci matched no catalog locus;\n" << " " << mmatches << " loci matched more than one catalog locus and were excluded.\n" << " " << nosnps << " loci contained SNPs unaccounted for in the catalog and were excluded.\n" << " " << no_haps << " loci had no verified haplotypes.\n" << " " << bad_aln << " loci had inconsistent alignments to a catalog locus and were excluded.\n"; return 0; } bool verify_gapped_match(map &catalog, QLocus *query, set &loci_hit, map > &query_hits, uint &mmatch, uint &nosnps, uint &no_haps, uint &bad_aln, uint &ver_hits) { // // 1. Check that this query locus matches just a single catalog locus. // if (loci_hit.size() == 0) { return false; } else if (loci_hit.size() > 1) { mmatch++; return false; } int cat_id = *(loci_hit.begin()); Locus *cat = catalog[cat_id]; map >::iterator query_it; map::iterator cat_it; AlignRes aln_res; string query_allele, cat_allele, converted_query_allele, qseq; int query_len; uint verified = 0; // // 2. Check if there was a consistent alignment between the alleles to the catalog locus. // set cigars; for (query_it = query_hits.begin(); query_it != query_hits.end(); query_it++) { map &cat_hits = query_it->second; for (cat_it = cat_hits.begin(); cat_it != cat_hits.end(); cat_it++) cigars.insert(cat_it->second.cigar); } if (cigars.size() > 1) { bad_aln++; return false; } // // 3. Make sure the query has no SNPs unaccounted for in the catalog. // vector > cigar; query_len = parse_cigar(invert_cigar(*cigars.begin()).c_str(), cigar); adjust_snps_for_gaps(cigar, query); qseq = apply_cigar_to_seq(query->con, cigar); query->add_consensus(qseq.c_str()); int min_tag_len = (uint)query_len > cat->len ? query_len : cat->len; vector::iterator i, j; bool found; for (i = query->snps.begin(); i != query->snps.end(); i++) { found = false; // // SNP occurs in a column that is beyond the length of the catalog // if ((int)(*i)->col > min_tag_len - 1) continue; for (j = cat->snps.begin(); j != cat->snps.end(); j++) { if ((*i)->col == (*j)->col) found = true; } // // Query locus posses a SNP not present in the catalog. // if (found == false) { nosnps++; return false; } } // // 4. Assign the allele hits after verifying there is a match between catalog and query allele. // for (query_it = query_hits.begin(); query_it != query_hits.end(); query_it++) { query_allele = query_it->first; map &cat_hits = query_it->second; if (cat_hits.size() == 1 && cat_hits.begin()->first == "consensus") { cat_it = cat_hits.begin(); cat_allele = cat_it->first; aln_res = cat_it->second; verified++; query->add_match(cat_id, cat_allele, query_allele, 0, invert_cigar(aln_res.cigar)); continue; } converted_query_allele = generate_query_allele(cat, query, query_allele); cat_it = cat_hits.find(converted_query_allele); if (cat_it != cat_hits.end()) { cat_allele = cat_it->first; aln_res = cat_it->second; verified++; query->add_match(cat_id, cat_allele, query_allele, 0, invert_cigar(aln_res.cigar)); } // } else { // // // // Check for alleles that have Ns in the query, but not in the catalog, but otherwise match. // // // uint match_cnt = 0; // for (cat_it = cat_hits.begin(); cat_it != cat_hits.end(); cat_it++) { // cat_allele = cat_it->first; // if (match_alleles(cat_allele, converted_query_allele)) { // match_cnt++; // aln_res = cat_it->second; // } // } // if (match_cnt == 1) { // verified++; // query->add_match(cat_id, cat_allele, query_allele, 0, invert_cigar(aln_res.cigar)); // } // } } if (verified > 0) { ver_hits += verified; } else { no_haps++; return false; } return true; } bool match_alleles(allele_type catalog_allele, allele_type query_allele) { const char *q = catalog_allele.c_str(); const char *p = query_allele.c_str(); const char *stop = p + query_allele.length(); while (p < stop) { if (*p != 'N' && *p != *q) return false; p++; q++; } return true; } string generate_query_allele(Locus *ctag, Locus *qtag, allele_type allele) { string new_allele = ""; if (qtag->snps.size() == 0) { for (uint i = 0; i < ctag->snps.size(); i++) new_allele += ctag->snps[i]->col > qtag->len - 1 ? 'N' : qtag->con[ctag->snps[i]->col]; } else { uint pos = 0; uint index = 0; for (uint i = 0; i < ctag->snps.size(); i++) { if (index < qtag->snps.size() && qtag->snps[index]->col == ctag->snps[i]->col) { new_allele += allele[pos]; index++; pos++; } else { new_allele += ctag->snps[i]->col > qtag->len - 1 ? 'N' : qtag->con[ctag->snps[i]->col]; } } } return new_allele; } int populate_hash(map &sample, HashMap &hash_map, vector &hash_map_keys, int min_tag_len) { map::iterator it; vector >::iterator all_it; Locus *tag; char *key; // // Create a hash map out of the set of alleles for each Locus. // for (it = sample.begin(); it != sample.end(); it++) { tag = it->second; for (all_it = tag->strings.begin(); all_it != tag->strings.end(); all_it++) { key = new char[min_tag_len + 1]; strncpy(key, all_it->second.c_str(), min_tag_len); key[min_tag_len] = '\0'; hash_map[key].push_back(make_pair(tag->id, all_it->first)); hash_map_keys.push_back(key); } } //dump_kmer_map(kmer_map); return 0; } int write_matches(string sample_path, map &sample) { map::iterator i; // // Parse the input file names to create the output file // size_t pos_1 = sample_path.find_last_of("/"); string out_file = out_path + sample_path.substr(pos_1 + 1) + ".matches.tsv"; if (in_file_type == FileT::gzsql) out_file += ".gz"; // // Open the output files for writing. // gzFile gz_matches; ofstream matches; if (in_file_type == FileT::gzsql) { gz_matches = gzopen(out_file.c_str(), "wb"); if (!gz_matches) { cerr << "Error: Unable to open gzipped matches file '" << out_file << "': " << strerror(errno) << ".\n"; exit(1); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_matches, libz_buffer_size); #endif } else { matches.open(out_file.c_str()); if (matches.fail()) { cerr << "Error: Unable to open matches file for writing.\n"; exit(1); } } // // Record the version of Stacks used and the date generated as a comment in the catalog. // // Obtain the current date. // stringstream log; time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%F %T", timeinfo); log << "# sstacks version " << VERSION << "; generated on " << date << "\n"; if (in_file_type == FileT::gzsql) gzputs(gz_matches, log.str().c_str()); else matches << log.str(); QLocus *qloc; string type; uint match_depth; stringstream sstr; cerr << "Outputing to file " << out_file << "\n"; for (i = sample.begin(); i != sample.end(); i++) { qloc = i->second; for (uint j = 0; j < qloc->matches.size(); j++) { if (verify_haplotypes == false && search_type == genomic_loc) match_depth = qloc->depth; else match_depth = qloc->alleles.count(qloc->matches[j]->cat_type) > 0 ? qloc->alleles[qloc->matches[j]->cat_type] : qloc->depth; sstr << "0" << "\t" << batch_id << "\t" << qloc->matches[j]->cat_id << "\t" << samp_id << "\t" << qloc->id << "\t" << qloc->matches[j]->cat_type << "\t" << match_depth << "\t" << qloc->lnl << "\t" << qloc->matches[j]->cigar << "\n"; } if (in_file_type == FileT::gzsql) gzputs(gz_matches, sstr.str().c_str()); else matches << sstr.str(); sstr.str(""); } if (in_file_type == FileT::gzsql) gzclose(gz_matches); else matches.close(); return 0; } int parse_command_line(int argc, char* argv[]) { string sample_file; int c; while (1) { static struct option long_options[] = { {"help", no_argument, NULL, 'h'}, {"version", no_argument, NULL, 'v'}, {"genomic_loc", no_argument, NULL, 'g'}, {"verify_hap", no_argument, NULL, 'x'}, {"uniq_haplotypes", no_argument, NULL, 'u'}, {"gapped", no_argument, NULL, 'G'}, {"num_threads", required_argument, NULL, 'p'}, {"batch_id", required_argument, NULL, 'b'}, {"catalog", required_argument, NULL, 'c'}, {"sample_2", required_argument, NULL, 's'}, {"outpath", required_argument, NULL, 'o'}, {0, 0, 0, 0} }; // getopt_long stores the option index here. int option_index = 0; c = getopt_long(argc, argv, "hgGxuvs:c:o:b:p:", long_options, &option_index); // Detect the end of the options. if (c == -1) break; switch (c) { case 'h': help(); break; case 'p': num_threads = atoi(optarg); break; case 'b': batch_id = is_integer(optarg); if (batch_id < 0) { cerr << "Batch ID (-b) must be an integer, e.g. 1, 2, 3\n"; help(); } break; case 's': sample_file = optarg; samples.push(sample_file); break; case 'g': search_type = genomic_loc; break; case 'o': out_path = optarg; break; case 'c': catalog_path = optarg; break; case 'x': verify_haplotypes = false; break; case 'u': require_uniq_haplotypes = true; break; case 'G': gapped_alignments = true; break; case 'v': version(); break; case '?': // getopt_long already printed an error message. help(); break; default: help(); abort(); } } if (catalog_path.length() == 0) { cerr << "You must specify the prefix path to the catalog.\n"; help(); } if (samples.size() == 0) { cerr << "You must specify at least one sample file.\n"; help(); } if (out_path.length() == 0) out_path = "."; if (out_path.at(out_path.length() - 1) != '/') out_path += "/"; return 0; } void version() { std::cerr << "sstacks " << VERSION << "\n\n"; exit(0); } void help() { std::cerr << "sstacks " << VERSION << "\n" << "sstacks -b batch_id -c catalog_file -s sample_file [-s sample_file_2 ...] [-o path] [-p num_threads] [-g] [-x] [-v] [-h]" << "\n" << " p: enable parallel execution with num_threads threads.\n" << " b: MySQL ID of this batch." << "\n" << " c: TSV file from which to load the catalog loci." << "\n" << " s: filename prefix from which to load sample loci." << "\n" << " o: output path to write results." << "\n" << " g: base matching on genomic location, not sequence identity." << "\n" << " x: don't verify haplotype of matching locus." << "\n" << " v: print program version." << "\n" << " h: display this help messsage." << "\n\n" << " Gapped assembly options:\n" << " --gapped: preform gapped alignments between stacks.\n"; exit(0); } stacks-1.44/src/sstacks.h000644 000765 000024 00000006203 12705003145 016113 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2013, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __SSTACKS_H__ #define __SSTACKS_H__ #include "constants.h" #ifdef _OPENMP #include // OpenMP library #endif #include // Process command-line options #include #include #include #include #include using std::pair; using std::make_pair; #include using std::string; #include #include using std::ifstream; using std::ofstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include using std::vector; #include using std::map; #include using std::set; #include using std::queue; #include using std::unordered_map; #ifdef HAVE_SPARSEHASH #include using google::sparse_hash_map; #endif #include "kmers.h" #include "stacks.h" #include "locus.h" #include "GappedAln.h" #include "kmers.h" #include "sql_utilities.h" #include "utils.h" #include "aln_utils.h" #ifdef HAVE_SPARSEHASH typedef sparse_hash_map >, hash_charptr, eqstr> HashMap; #else typedef unordered_map >, hash_charptr, eqstr> HashMap; #endif void help( void ); void version( void ); int parse_command_line(int, char**); int populate_hash(map &, HashMap &, vector &, int); int find_matches_by_sequence(map &, map &); int find_matches_by_genomic_loc(map &, map &); int verify_sequence_match(map &, QLocus *, set &, map > &, uint, unsigned long &, unsigned long &); int search_for_gaps(map &, map &, KmerHashMap &, map > &, double); bool verify_gapped_match(map &, QLocus *, set &, map > &, uint &, uint &, uint &, uint &, uint &); int verify_genomic_loc_match(Locus *, QLocus *, set &, unsigned long &); string generate_query_allele(Locus *, Locus *, allele_type); bool match_alleles(allele_type, allele_type); int generate_query_haplotypes(Locus *, QLocus *, set &); int impute_haplotype(string, vector > &, string &); bool compare_dist(pair, pair); int write_matches(string, map &); #endif // __SSTACKS_H__ stacks-1.44/src/stacks.cc000644 000765 000024 00000004715 12764047443 016112 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // stacks.cc -- routines for the stack-holding containers // // Julian Catchen // jcatchen@uoregon.edu // University of Oregon // // $Id$ // #include "stacks.h" Rem::Rem() { this->id = 0; this->seq = NULL; this->utilized = false; } Rem::Rem(int id, uint seq_id, DNANSeq *seq) { this->id = id; this->utilized = false; this->map.push_back(seq_id); this->seq = new DNANSeq(seq->size(), seq->s); } int Rem::add_id(uint id) { this->map.push_back(id); return 0; } int Rem::add_seq(const DNANSeq *seq) { if (this->seq != NULL) delete this->seq; this->seq = new DNANSeq(seq->size(), seq->s); return 0; } int Rem::add_seq(const char *seq) { if (this->seq != NULL) delete this->seq; this->seq = new DNANSeq(strlen(seq), seq); return 0; } int PStack::add_id(const char *id) { char *f = new char[strlen(id) + 1]; strcpy(f, id); this->map.push_back(f); return 0; } int PStack::add_seq(const char *seq) { if (this->seq != NULL) delete this->seq; this->len = strlen(seq); this->seq = new DNANSeq(this->len, seq); return 0; } int PStack::add_seq(const DNANSeq *seq) { if (this->seq != NULL) delete this->seq; this->seq = new DNANSeq(seq->size(), seq->s); return 0; } int Stack::add_id(uint id) { this->map.push_back(id); return 0; } int Stack::add_seq(const char *seq) { if (this->seq != NULL) delete this->seq; this->seq = new DNANSeq(strlen(seq), seq); return 0; } int Stack::add_seq(const DNANSeq *seq) { if (this->seq != NULL) delete this->seq; this->seq = new DNANSeq(seq->size(), seq->s); return 0; } stacks-1.44/src/stacks.h000644 000765 000024 00000015771 12764047443 015760 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __STACKS_H__ #define __STACKS_H__ #include #include using std::string; #include using std::vector; #include using std::map; #include using std::set; #include #include using std::pair; using std::make_pair; #include using std::ofstream; using std::cerr; #include using std::stringstream; #include "constants.h" #include "DNASeq.h" #include "DNANSeq.h" typedef unsigned int uint; typedef string allele_type; enum snp_type {snp_type_het, snp_type_hom, snp_type_unk}; enum read_type {primary, secondary}; enum strand_type {strand_plus, strand_minus}; enum searcht {sequence, genomic_loc}; class PhyLoc { public: char *chr; uint bp; strand_type strand; void set(const char *chr, uint bp, strand_type strand) { if (this->chr != NULL) delete [] this->chr; this->chr = new char[strlen(chr) + 1]; this->bp = bp; this->strand = strand; strcpy(this->chr, chr); } PhyLoc() { chr = NULL; bp = 0; strand = strand_plus; } PhyLoc(const PhyLoc& other) : bp(other.bp), strand(other.strand) { chr = new char[strlen(other.chr)+1]; strcpy(chr, other.chr); } PhyLoc(const char *chr, uint bp) { this->chr = new char[strlen(chr) + 1]; this->bp = bp; this->strand = strand_plus; strcpy(this->chr, chr); } PhyLoc(const char *chr, uint bp, strand_type strnd) { this->chr = new char[strlen(chr) + 1]; this->bp = bp; this->strand = strnd; strcpy(this->chr, chr); } ~PhyLoc() { if (chr != NULL) delete [] chr; } friend void swap(PhyLoc& p, PhyLoc& q) { char* chr = p.chr; p.chr = q.chr; q.chr = chr; const uint bp = p.bp; p.bp = q.bp; q.bp = bp; const strand_type strand = p.strand; p.strand = q.strand; q.strand = strand; } PhyLoc& operator=(PhyLoc&& other) {swap(*this, other); return *this;} PhyLoc& operator=(const PhyLoc& other) =delete; bool operator==(const PhyLoc& other) const { if (bp == other.bp && strand == other.strand && strcmp(chr, other.chr) == 0) return true; else return false; } bool operator<(const PhyLoc& other) const { const int chrcmp = strcmp(chr, other.chr); if (chrcmp != 0) // Alphanumeric. return chrcmp < 0; else if (bp != other.bp) return bp < other.bp; else // Minus strand first. return strand == strand_minus && other.strand == strand_plus; } }; class SNP { public: snp_type type; // Heterozygous or homozygous uint col; float lratio; char rank_1; char rank_2; char rank_3; char rank_4; SNP() { col = 0; lratio = 0.0; rank_1 = 0; rank_2 = 0; rank_3 = 0; rank_4 = 0; } }; class Gap { public: uint start; uint end; Gap(uint s, uint e) { start = s; end = e; } }; class Aln { public: uint id; uint gap_cnt; double pct_id; string cigar; Aln() { this->id = 0; this->pct_id = 0.0; this->gap_cnt = 0; } Aln(uint id, string cigar, double pct_id, uint gap_cnt) { this->id = id; this->cigar = cigar; this->pct_id = pct_id; this->gap_cnt = gap_cnt; } }; class PStack { public: uint id; uint count; // Number of identical reads forming this stack DNANSeq *seq; // Sequence read uint len; // Read length vector map; // List of sequence read IDs merged into this stack PhyLoc loc; // Physical genome location of this stack. PStack() { id = 0; count = 0; seq = NULL; len = 0; } ~PStack() { delete this->seq; for (unsigned int i = 0; i < this->map.size(); i++) delete [] this->map[i]; } int add_id(const char *); int add_seq(const char *); int add_seq(const DNANSeq *); }; class Stack { public: uint id; DNANSeq *seq; // Sequence read vector map; // List of sequence read IDs merged into this stack Stack() { id = 0; seq = NULL; } ~Stack() { delete this->seq; } uint count() { return this->map.size(); } int add_id(uint); int add_seq(const char *); int add_seq(const DNANSeq *); }; class Rem { public: uint id; vector map; // List of sequence read IDs merged into this stack DNANSeq *seq; // Sequence read bool utilized; Rem(); Rem(int, uint, DNANSeq *); ~Rem() { delete this->seq; } uint count() { return this->map.size(); } int add_id(uint); int add_seq(const char *); int add_seq(const DNANSeq *); }; class CatMatch { public: int batch_id; int cat_id; int sample_id; int tag_id; int depth; double lnl; char *haplotype; char *cigar; CatMatch() { batch_id = 0; cat_id = 0; sample_id = 0; tag_id = 0; depth = 0; lnl = 0.0; haplotype = NULL; cigar = NULL; } ~CatMatch() { delete [] haplotype; delete [] cigar; } }; class ModRes { public: int sample_id; int tag_id; char *model; ModRes(int samp_id, int tag_id, const char *model) { this->sample_id = samp_id; this->tag_id = tag_id; this->model = new char[strlen(model) + 1]; strcpy(this->model, model); } ~ModRes() { delete [] this->model; } }; class SNPRes { public: int sample_id; int tag_id; vector snps; SNPRes(int samp_id, int tag_id) { this->sample_id = samp_id; this->tag_id = tag_id; } ~SNPRes() { for (uint i = 0; i < this->snps.size(); i++) delete this->snps[i]; this->snps.clear(); } }; #endif // __STACKS_H__ stacks-1.44/src/Tsv.h000644 000765 000024 00000003354 12764047443 015236 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // Code to parse the internal (and tempoary) data format. This format is created for // reads that have been aligned to a reference genome. It takes the tab-separated form: // // // // One record per line. // #ifndef __TSV_H__ #define __TSV_H__ #include "input.h" class Tsv: public Input { public: Tsv(const char *path) : Input(path) {}; ~Tsv() {}; Seq *next_seq(); int next_seq(Seq &); }; Seq *Tsv::next_seq() { Seq* s = new Seq(); if(next_seq(*s) != 1) { delete s; s = NULL; } return s; } int Tsv::next_seq(Seq& s) { vector parts; this->fh.getline(this->line, max_len); if (!this->fh.good()) { return 0; } parse_tsv(this->line, parts); string id = parts[0] + "_" + parts[1]; s = Seq(id.c_str(), parts[2].c_str(), parts[3].c_str(), parts[0].c_str(), atoi(parts[1].c_str()), strand_plus); return 1; } #endif // __TSV_H__ stacks-1.44/src/ustacks.cc000644 000765 000024 00000254640 12774470747 016313 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // ustacks -- build denovo stacks // #include "ustacks.h" // // Global variables to hold command-line options. // FileT in_file_type; string in_file; string out_path; int num_threads = 1; int sql_id = 0; bool call_sec_hapl = true; bool set_kmer_len = true; int kmer_len = 0; int max_kmer_len = 19; int min_merge_cov = 3; uint max_subgraph = 3; int dump_graph = 0; int retain_rem_reads = false; int deleverage_stacks = 0; int remove_rep_stacks = 0; int max_utag_dist = 2; int max_rem_dist = -1; bool gapped_alignments = false; double min_match_len = 0.80; double max_gaps = 2.0; int deleverage_trigger; int removal_trigger; // // For use with the multinomial model to call fixed nucleotides. // modelt model_type = snp; double alpha = 0.05; double bound_low = 0.0; double bound_high = 1.0; double p_freq = 0.5; double barcode_err_freq = 0.0; double heterozygote_limit = -3.84; double homozygote_limit = 3.84; int main (int argc, char* argv[]) { parse_command_line(argc, argv); // // Set the max remainder distance to be greater than the max_utag_dist, if it is not // specified on the command line. // if (max_rem_dist == -1) max_rem_dist = max_utag_dist + 2; cerr << "ustacks paramters selected:\n" << " Min depth of coverage to create a stack: " << min_merge_cov << "\n" << " Max distance allowed between stacks: " << max_utag_dist << "\n" << " Max distance allowed to align secondary reads: " << max_rem_dist << "\n" << " Max number of stacks allowed per de novo locus: " << max_subgraph << "\n" << " Deleveraging algorithm: " << (deleverage_stacks ? "enabled" : "disabled") << "\n" << " Removal algorithm: " << (remove_rep_stacks ? "enabled" : "disabled") << "\n" << " Model type: "; switch (model_type) { case snp: cerr << "SNP\n"; break; case fixed: cerr << "Fixed\n"; break; case bounded: cerr << "Bounded; lower epsilon bound: " << bound_low << "; upper bound: " << bound_high << "\n"; break; } cerr << " Alpha significance level for model: " << alpha << "\n" << " Gapped alignments: " << (gapped_alignments ? "enabled" : "disabled") << "\n"; // // Set limits to call het or homozygote according to chi-square distribution with one // degree of freedom: // http://en.wikipedia.org/wiki/Chi-squared_distribution#Table_of_.CF.872_value_vs_p-value // if (alpha == 0.1) { heterozygote_limit = -2.71; homozygote_limit = 2.71; } else if (alpha == 0.05) { heterozygote_limit = -3.84; homozygote_limit = 3.84; } else if (alpha == 0.01) { heterozygote_limit = -6.64; homozygote_limit = 6.64; } else if (alpha == 0.001) { heterozygote_limit = -10.83; homozygote_limit = 10.83; } // // Set the number of OpenMP parallel threads to execute. // #ifdef _OPENMP omp_set_num_threads(num_threads); #endif DNASeqHashMap* radtags = new DNASeqHashMap(); map remainders; set merge_map; map unique; load_radtags(in_file, *radtags); reduce_radtags(*radtags, unique, remainders); delete radtags; // dump_unique_tags(unique); double cov_mean, cov_stdev, cov_max; calc_coverage_distribution(unique, cov_mean, cov_stdev, cov_max); cerr << "Initial coverage mean: " << cov_mean << "; Std Dev: " << cov_stdev << "; Max: " << cov_max << "\n"; calc_triggers(cov_mean, cov_stdev, 1, deleverage_trigger, removal_trigger); cerr << "Deleveraging trigger: " << deleverage_trigger << "; Removal trigger: " << removal_trigger << "\n"; map merged; populate_merged_tags(unique, merged); cerr << merged.size() << " initial stacks were populated; " << remainders.size() << " stacks were set aside as secondary reads.\n"; if (remove_rep_stacks) { cerr << "Calculating distance for removing repetitive stacks.\n"; calc_kmer_distance(merged, 1); cerr << "Removing repetitive stacks.\n"; remove_repetitive_stacks(unique, merged); } calc_coverage_distribution(unique, merged, cov_mean, cov_stdev, cov_max); cerr << "Post-Repeat Removal, coverage depth Mean: " << cov_mean << "; Std Dev: " << cov_stdev << "; Max: " << cov_max << "\n"; cerr << "Calculating distance between stacks...\n"; calc_kmer_distance(merged, max_utag_dist); cerr << "Merging stacks, maximum allowed distance: " << max_utag_dist << " nucleotide(s)\n"; merge_stacks(unique, remainders, merged, merge_map, max_utag_dist); call_consensus(merged, unique, remainders, false); calc_coverage_distribution(unique, merged, cov_mean, cov_stdev, cov_max); cerr << "After merging, coverage depth Mean: " << cov_mean << "; Std Dev: " << cov_stdev << "; Max: " << cov_max << "\n"; //dump_merged_tags(merged); cerr << "Merging remainder radtags\n"; merge_remainders(merged, remainders); calc_coverage_distribution(unique, remainders, merged, cov_mean, cov_stdev, cov_max); cerr << "After remainders merged, coverage depth Mean: " << cov_mean << "; Std Dev: " << cov_stdev << "; Max: " << cov_max << "\n"; if (gapped_alignments) { call_consensus(merged, unique, remainders, false); cerr << "Searching for gaps between merged stacks...\n"; search_for_gaps(merged, min_match_len); merge_gapped_alns(unique, remainders, merged); calc_coverage_distribution(unique, remainders, merged, cov_mean, cov_stdev, cov_max); cerr << "After gapped alignments, coverage depth Mean: " << cov_mean << "; Std Dev: " << cov_stdev << "; Max: " << cov_max << "\n"; } // // Call the final consensus sequence and invoke the SNP model. // cerr << "Calling final consensus sequences, invoking SNP-calling model...\n"; call_consensus(merged, unique, remainders, true); count_raw_reads(unique, remainders, merged); cerr << "Writing loci, SNPs, and alleles to '" << out_path << "'...\n"; write_results(merged, unique, remainders); cerr << "done.\n"; return 0; } int merge_gapped_alns(map &unique, map &rem, map &merged) { map new_merged; map::iterator it; MergedStack *tag_1, *tag_2, *merged_tag; int id = 1; uint merge_cnt = 0; set processed; string cigar_1, cigar_2; vector > cigar; for (it = merged.begin(); it != merged.end(); it++) { if (processed.count(it->first) > 0) continue; tag_1 = it->second; sort(tag_1->alns.begin(), tag_1->alns.end(), rank_alignments); // // No gapped alignments, or no optimal alignment for this stack, or // this stack has already been set aside. // if (tag_1->masked || tag_1->alns.size() == 0) continue; if (tag_1->alns.size() > 1 && tag_1->alns[0].pct_id == tag_1->alns[1].pct_id) continue; // // Found one or more gapped alignments. Make sure the best alignment for each of the aligned pairs // of tags are, reciprocal to each other.. // tag_2 = merged[tag_1->alns[0].id]; sort(tag_2->alns.begin(), tag_2->alns.end(), rank_alignments); if (tag_2->masked || tag_2->alns.size() == 0) continue; if (tag_2->alns.size() > 1 && tag_2->alns[0].pct_id == tag_2->alns[1].pct_id) continue; if (tag_1->id != (int)tag_2->alns[0].id) continue; cigar_1 = invert_cigar(tag_1->alns[0].cigar); cigar_2 = tag_2->alns[0].cigar; if (cigar_1 == cigar_2) { parse_cigar(tag_1->alns[0].cigar.c_str(), cigar); // // Check that the alignment still contains fewer than // max_utag_dist mismatches. // if (dist(tag_1->con, tag_2->con, cigar) > max_utag_dist) continue; // // If the alignment has too many gaps, skip it. // if (tag_1->alns[0].gap_cnt > (max_gaps + 1)) continue; // // If the alignment doesn't span enough of the two sequences, skip it. // if (tag_1->alns[0].pct_id < min_match_len) continue; // // Edit the sequences to accommodate any added gaps. // edit_gapped_seqs(unique, rem, tag_1, cigar); parse_cigar(tag_2->alns[0].cigar.c_str(), cigar); edit_gapped_seqs(unique, rem, tag_2, cigar); // // Merge the tags. // merged_tag = merge_tags(tag_1, tag_2, id); new_merged[id] = merged_tag; id++; // // Record the gaps. // uint pos = 0; for (uint j = 0; j < cigar.size(); j++) { if (cigar[j].first == 'I' || cigar[j].first == 'D') merged_tag->gaps.push_back(Gap(pos, pos + cigar[j].second)); pos += cigar[j].second; } processed.insert(tag_1->id); processed.insert(tag_2->id); merge_cnt++; } } set merge_set; for (it = merged.begin(); it != merged.end(); it++) { if (processed.count(it->first)) continue; tag_1 = it->second; merge_set.insert(tag_1->id); tag_2 = merge_tags(merged, merge_set, id); new_merged[id] = tag_2; merge_set.clear(); id++; } uint new_cnt = new_merged.size(); uint old_cnt = merged.size(); // // Free the memory from the old map of merged tags. // for (it = merged.begin(); it != merged.end(); it++) delete it->second; merged = new_merged; cerr << " " << old_cnt << " stacks merged into " << new_cnt << " stacks; merged " << merge_cnt << " gapped alignments.\n"; return 0; } bool rank_alignments(Aln a, Aln b) { return a.pct_id > b.pct_id; } int edit_gapped_seqs(map &unique, map &rem, MergedStack *tag, vector > &cigar) { int stack_id; Stack *s; Rem *r; char *buf = new char[tag->len + 1]; for (uint i = 0; i < tag->utags.size(); i++) { stack_id = tag->utags[i]; s = unique[stack_id]; buf = s->seq->seq(buf); edit_gaps(cigar, buf); delete s->seq; s->seq = new DNANSeq(tag->len, buf); } for (uint i = 0; i < tag->remtags.size(); i++) { stack_id = tag->remtags[i]; r = rem[stack_id]; buf = r->seq->seq(buf); edit_gaps(cigar, buf); delete r->seq; r->seq = new DNANSeq(tag->len, buf); } delete [] buf; return 0; } int edit_gaps(vector > &cigar, char *seq) { char *buf; uint size = cigar.size(); char op; uint dist, bp, len, buf_len, buf_size, j, k, stop; len = strlen(seq); bp = 0; buf = new char[len + 1]; buf_size = len + 1; for (uint i = 0; i < size; i++) { op = cigar[i].first; dist = cigar[i].second; switch(op) { case 'S': stop = bp + dist; stop = stop > len ? len : stop; while (bp < stop) { seq[bp] = 'N'; bp++; } break; case 'D': // // A deletion has occured in the read relative to the reference genome. // Pad the read with sufficent Ns to match the deletion, shifting the existing // sequence down. Trim the final length to keep the read length consistent. // k = bp >= len ? len : bp; strncpy(buf, seq + k, buf_size - 1); buf[buf_size - 1] = '\0'; buf_len = strlen(buf); stop = bp + dist; stop = stop > len ? len : stop; while (bp < stop) { seq[bp] = 'N'; bp++; } j = bp; k = 0; while (j < len && k < buf_len) { seq[j] = buf[k]; k++; j++; } break; case 'I': case 'M': bp += dist; break; default: break; } } delete [] buf; return 0; } int search_for_gaps(map &merged, double min_match_len) { // // Search for loci that can be merged with a gapped alignment. // KmerHashMap kmer_map; vector kmer_map_keys; MergedStack *tag_1, *tag_2; map::iterator it; // // OpenMP can't parallelize random access iterators, so we convert // our map to a vector of integer keys. // vector keys; for (it = merged.begin(); it != merged.end(); it++) keys.push_back(it->first); // // Calculate the number of k-mers we will generate. If kmer_len == 0, // determine the optimal length for k-mers. // int con_len = strlen(merged[keys[0]]->con); int kmer_len = 19; int num_kmers = con_len - kmer_len + 1; // // Calculate the minimum number of matching k-mers required for a possible sequence match. // int min_hits = (round((double) con_len * min_match_len) - (kmer_len * max_gaps)) - kmer_len + 1; cerr << " Searching with a k-mer length of " << kmer_len << " (" << num_kmers << " k-mers per read); " << min_hits << " k-mer hits required.\n"; populate_kmer_hash(merged, kmer_map, kmer_map_keys, kmer_len); #pragma omp parallel private(tag_1, tag_2) { KmerHashMap::iterator h; vector query_kmers; set uniq_kmers; GappedAln *aln = new GappedAln(con_len); AlignRes a; initialize_kmers(kmer_len, num_kmers, query_kmers); #pragma omp for schedule(dynamic) for (uint i = 0; i < keys.size(); i++) { tag_1 = merged[keys[i]]; // // Don't compute distances for masked tags. // if (tag_1->masked) continue; // // Don't compare tags that are already at or above max_locus_stacks. // if (tag_1->utags.size() >= max_subgraph) continue; generate_kmers_lazily(tag_1->con, kmer_len, num_kmers, query_kmers); uniq_kmers.clear(); for (int j = 0; j < num_kmers; j++) uniq_kmers.insert(query_kmers[j]); map hits; // // Lookup the occurances of each k-mer in the kmer_map // for (set::iterator j = uniq_kmers.begin(); j != uniq_kmers.end(); j++) { h = kmer_map.find(j->c_str()); if (h != kmer_map.end()) for (uint k = 0; k < h->second.size(); k++) hits[h->second[k]]++; } // // Iterate through the list of hits. For each hit that has more than min_hits // check its full length to verify a match. // map::iterator hit_it; for (hit_it = hits.begin(); hit_it != hits.end(); hit_it++) { if (hit_it->second < min_hits) continue; tag_2 = merged[hit_it->first]; // Don't compute distances for masked tags if (tag_2->masked) continue; // Don't compare tag_1 against itself. if (tag_1 == tag_2) continue; // // Don't compare tags that are already at or above max_locus_stacks. // if (tag_2->utags.size() >= max_subgraph) continue; if (aln->align(tag_1->con, tag_2->con)) { a = aln->result(); tag_1->alns.push_back(Aln(tag_2->id, a.cigar, a.pct_id, a.gap_cnt)); } } } // // Free the k-mers we generated for this query. // for (uint j = 0; j < query_kmers.size(); j++) delete [] query_kmers[j]; query_kmers.clear(); delete aln; } free_kmer_hash(kmer_map, kmer_map_keys); return 0; } int merge_remainders(map &merged, map &rem) { map::iterator it; // OpenMP can't parallelize random access iterators, so we convert // our map to a vector of integer keys. vector keys; uint tot = 0; for (it = rem.begin(); it != rem.end(); it++) { keys.push_back(it->first); tot += it->second->count(); } cerr << " " << tot << " remainder sequences left to merge.\n"; if (max_rem_dist <= 0) { cerr << " Matched 0 remainder reads; unable to match " << tot << " remainder reads.\n"; return 0; } // // Calculate the number of k-mers we will generate. If kmer_len == 0, // determine the optimal length for k-mers. // int con_len = strlen(merged.begin()->second->con); if (set_kmer_len) kmer_len = determine_kmer_length(con_len, max_rem_dist); int num_kmers = con_len - kmer_len + 1; // // Calculate the minimum number of matching k-mers required for a possible sequence match. // int min_hits = calc_min_kmer_matches(kmer_len, max_rem_dist, con_len, set_kmer_len ? true : false); cerr << " Distance allowed between stacks: " << max_rem_dist << "; searching with a k-mer length of " << kmer_len << " (" << num_kmers << " k-mers per read); " << min_hits << " k-mer hits required.\n"; KmerHashMap kmer_map; vector kmer_map_keys; populate_kmer_hash(merged, kmer_map, kmer_map_keys, kmer_len); int utilized = 0; #pragma omp parallel private(it) { KmerHashMap::iterator h; vector rem_kmers; char *buf = new char[con_len + 1]; #pragma omp for schedule(dynamic) for (uint j = 0; j < keys.size(); j++) { it = rem.find(keys[j]); Rem *r = it->second; // // Generate the k-mers for this remainder sequence // buf = r->seq->seq(buf); generate_kmers_lazily(buf, kmer_len, num_kmers, rem_kmers); map hits; // // Lookup the occurances of each remainder k-mer in the MergedStack k-mer map // for (int k = 0; k < num_kmers; k++) { h = kmer_map.find(rem_kmers[k]); if (h != kmer_map.end()) for (uint n = 0; n < h->second.size(); n++) hits[h->second[n]]++; } // // Iterate through the list of hits. For each hit that has more than min_hits // check its full length to verify a match. // map dists; map::iterator hit_it; for (hit_it = hits.begin(); hit_it != hits.end(); hit_it++) { if (hit_it->second < min_hits) continue; int d = dist(merged[hit_it->first], buf); // // Store the distance between these two sequences if it is // below the maximum distance // if (d <= max_rem_dist) { dists[hit_it->first] = d; } } // Check to see if there is a uniquely low distance, if so, // merge this remainder tag. If not, discard it, since we // can't locate a single best-fitting Stack to merge it into. map::iterator s; int min_id = -1; int count = 0; int dist = max_rem_dist + 1; for (s = dists.begin(); s != dists.end(); s++) { if ((*s).second < dist) { min_id = (*s).first; count = 1; dist = (*s).second; } else if ((*s).second == dist) { count++; } } // Found a merge partner. if (min_id >= 0 && count == 1) { r->utilized = true; #pragma omp critical { merged[min_id]->remtags.push_back(it->first); utilized += it->second->count(); } } } // // Free the k-mers we generated for this remainder read // for (uint k = 0; k < rem_kmers.size(); k++) delete [] rem_kmers[k]; delete [] buf; } free_kmer_hash(kmer_map, kmer_map_keys); //delete [] buf; cerr << " Matched " << utilized << " remainder reads; unable to match " << tot - utilized << " remainder reads.\n"; return 0; } int call_alleles(MergedStack *mtag, vector &reads, vector &read_types) { int row; int height = reads.size(); string allele; DNANSeq *d; char base; vector::iterator snp; for (row = 0; row < height; row++) { allele.clear(); uint snp_cnt = 0; // // Only call a haplotype from primary reads. // if (!call_sec_hapl && read_types[row] == secondary) continue; for (snp = mtag->snps.begin(); snp != mtag->snps.end(); snp++) { if ((*snp)->type != snp_type_het) continue; snp_cnt++; d = reads[row]; base = (*d)[(*snp)->col]; // // Check to make sure the nucleotide at the location of this SNP is // of one of the two possible states the multinomial model called. // if (base == (*snp)->rank_1 || base == (*snp)->rank_2) allele += base; else break; } if (snp_cnt > 0 && allele.length() == snp_cnt) mtag->alleles[allele]++; } return 0; } int call_consensus(map &merged, map &unique, map &rem, bool invoke_model) { // // OpenMP can't parallelize random access iterators, so we convert // our map to a vector of integer keys. // map::iterator it; vector keys; for (it = merged.begin(); it != merged.end(); it++) keys.push_back(it->first); int i; #pragma omp parallel private(i) { MergedStack *mtag; Stack *utag; Rem *r; #pragma omp for schedule(dynamic) for (i = 0; i < (int) keys.size(); i++) { mtag = merged[keys[i]]; // // Create a two-dimensional array, each row containing one read. For // each unique tag that has been merged together, add the sequence for // that tag into our array as many times as it originally occurred. // vector::iterator j; vector reads; vector read_types; for (j = mtag->utags.begin(); j != mtag->utags.end(); j++) { utag = unique[*j]; for (uint k = 0; k < utag->count(); k++) { reads.push_back(utag->seq); read_types.push_back(primary); } } // For each remainder tag that has been merged into this Stack, add the sequence. for (j = mtag->remtags.begin(); j != mtag->remtags.end(); j++) { r = rem[*j]; for (uint k = 0; k < r->count(); k++) { reads.push_back(r->seq); read_types.push_back(secondary); } } // // Iterate over each column of the array and call the consensus base. // uint row, col; uint length = reads[0]->size(); uint height = reads.size(); string con; map nuc; map::iterator max, n; DNANSeq *d; uint cur_gap = mtag->gaps.size() > 0 ? 0 : 1; for (col = 0; col < length; col++) { // // Don't invoke the model within gaps. // if (cur_gap < mtag->gaps.size() && col == mtag->gaps[cur_gap].start) { do { con += 'N'; SNP *snp = new SNP; snp->type = snp_type_unk; snp->col = col; snp->rank_1 = '-'; snp->rank_2 = '-'; mtag->snps.push_back(snp); col++; } while (col < mtag->gaps[cur_gap].end && col < length); col--; cur_gap++; continue; } nuc['A'] = 0; nuc['G'] = 0; nuc['C'] = 0; nuc['T'] = 0; for (row = 0; row < height; row++) { d = reads[row]; if (nuc.count((*d)[col])) nuc[(*d)[col]]++; } // // Find the base with a plurality of occurances and call it. // max = nuc.end(); for (n = nuc.begin(); n != nuc.end(); n++) { if (max == nuc.end() || n->second > max->second) max = n; } con += max->first; // // Search this column for the presence of a SNP // if (invoke_model) switch(model_type) { case snp: call_multinomial_snp(mtag, col, nuc, true); break; case bounded: call_bounded_multinomial_snp(mtag, col, nuc, true); break; case fixed: call_multinomial_fixed(mtag, col, nuc); break; } } if (invoke_model) { call_alleles(mtag, reads, read_types); if (model_type == fixed) { // // Mask nucleotides that are not fixed. // vector::iterator s; for (s = mtag->snps.begin(); s != mtag->snps.end(); s++) { if ((*s)->type == snp_type_unk) con.replace((*s)->col, 1, "N"); } } } mtag->add_consensus(con.c_str()); } } return 0; } int populate_merged_tags(map &unique, map &merged) { map::iterator i; map::iterator it_new, it_old; Stack *utag; MergedStack *mtag; int k = 0; it_old = merged.begin(); for (i = unique.begin(); i != unique.end(); i++) { utag = (*i).second; mtag = new MergedStack; mtag->id = k; mtag->count = utag->count(); mtag->utags.push_back(utag->id); mtag->add_consensus(utag->seq); // Insert the new MergedStack giving a hint as to which position // to insert it at. it_new = merged.insert(it_old, pair(k, mtag)); it_old = it_new; k++; } return 0; } int merge_stacks(map &unique, map &rem, map &merged, set &merge_map, int round) { map new_merged; map::iterator it, it_old, it_new; MergedStack *tag_1, *tag_2; vector > merge_lists; uint index = 0; int cohort_id = 0; int id = 1; uint delev_cnt = 0; uint blist_cnt = 0; for (it = merged.begin(); it != merged.end(); it++) { tag_1 = it->second; // // This tag may already have been merged by an earlier operation. // if (merge_map.count(tag_1->id) > 0) continue; queue merge_list; pair::iterator,bool> ret; vector >::iterator k; merge_lists.push_back(set()); if (tag_1->masked) { merge_lists[index].insert(tag_1->id); index++; continue; } // // Construct a list of MergedStacks that are within a particular distance // of this tag. // merge_lists[index].insert(tag_1->id); merge_list.push(tag_1->id); while (!merge_list.empty()) { tag_2 = merged[merge_list.front()]; merge_list.pop(); for (k = tag_2->dist.begin(); k != tag_2->dist.end(); k++) { ret = merge_lists[index].insert(k->first); // // If this Tag has not already been added to the merge list (i.e. we were able // to insert it in to our unique_merge_list, which is a set), add it for consideration // later in the loop. // if (ret.second == true) merge_list.push((*k).first); } } // // Record the nodes that have been merged in this round. // set::iterator j; for (j = merge_lists[index].begin(); j != merge_lists[index].end(); j++) merge_map.insert(*j); index++; } #pragma omp parallel private(tag_1, tag_2) { vector merged_tags; #pragma omp for reduction(+:delev_cnt) reduction(+:blist_cnt) for (uint index = 0; index < merge_lists.size(); index++) { // // Deal with the simple case of a single locus that does not need to be merged. // if (merge_lists[index].size() == 1) { tag_1 = merged[*(merge_lists[index].begin())]; tag_2 = merge_tags(merged, merge_lists[index], 0); // // If this tag is masked, keep the old cohort_id. // if (tag_1->masked) { tag_2->cohort_id = tag_1->cohort_id; } else { tag_2->cohort_id = cohort_id; #pragma omp atomic cohort_id++; } merged_tags.push_back(tag_2); continue; } // // Break large loci down by constructing a minimum // spanning tree and severing long distance edges. // if (deleverage_stacks) { vector tags; bool delev; deleverage(unique, rem, merged, merge_lists[index], cohort_id, tags); if (tags.size() == 1) { delev = false; } else { delev_cnt++; delev = true; } for (uint t = 0; t < tags.size(); t++) { //tags[t]->id = id; tags[t]->deleveraged = delev; if (tags[t]->utags.size() > max_subgraph) { tags[t]->masked = true; tags[t]->blacklisted = true; blist_cnt++; } //new_merged.insert(pair(id, tags[t])); merged_tags.push_back(tags[t]); //id++; } #pragma omp atomic cohort_id++; } else { // // If not deleveraging, merge these tags together into a new MergedStack object. // tag_2 = merge_tags(merged, merge_lists[index], 0); tag_2->cohort_id = cohort_id; if (tag_2->utags.size() > max_subgraph) { tag_2->masked = true; tag_2->blacklisted = true; blist_cnt++; } //new_merged.insert(pair(id, tag_2)); merged_tags.push_back(tag_2); #pragma omp atomic cohort_id++; //id++; } } // // Merge the accumulated tags into the new_merged map. // #pragma omp critical { it_old = merged.begin(); for (uint j = 0; j < merged_tags.size(); j++) { merged_tags[j]->id = id; it_new = new_merged.insert(it_old, pair(id, merged_tags[j])); it_old = it_new; id++; } } } uint new_cnt = new_merged.size(); uint old_cnt = merged.size(); // // Free the memory from the old map of merged tags. // for (it = merged.begin(); it != merged.end(); it++) delete it->second; merged = new_merged; cerr << " " << old_cnt << " stacks merged into " << new_cnt << " stacks; deleveraged " << delev_cnt << " stacks; removed " << blist_cnt << " stacks.\n"; return 0; } MergedStack * merge_tags(MergedStack *tag_1, MergedStack *tag_2, int id) { MergedStack *new_tag; new_tag = new MergedStack; new_tag->id = id; new_tag->deleveraged = tag_2->deleveraged || tag_1->deleveraged; new_tag->masked = tag_2->masked || tag_1->masked; new_tag->blacklisted = tag_2->blacklisted || tag_1->blacklisted; new_tag->gappedlumberjack = tag_2->gappedlumberjack || tag_1->gappedlumberjack; new_tag->lumberjackstack = tag_2->lumberjackstack || tag_1->lumberjackstack; for (uint i = 0; i < tag_1->utags.size(); i++) new_tag->utags.push_back(tag_1->utags[i]); for (uint i = 0; i < tag_1->remtags.size(); i++) new_tag->remtags.push_back(tag_1->remtags[i]); new_tag->count = tag_1->count; for (uint i = 0; i < tag_2->utags.size(); i++) new_tag->utags.push_back(tag_2->utags[i]); for (uint i = 0; i < tag_2->remtags.size(); i++) new_tag->remtags.push_back(tag_2->remtags[i]); new_tag->count += tag_2->count; return new_tag; } MergedStack *merge_tags(map &merged, set &merge_list, int id) { set::iterator i; vector::iterator j; MergedStack *tag_1, *tag_2; tag_1 = new MergedStack; tag_1->id = id; for (i = merge_list.begin(); i != merge_list.end(); i++) { tag_2 = merged[(*i)]; tag_1->deleveraged = tag_2->deleveraged ? true : tag_1->deleveraged; tag_1->masked = tag_2->masked ? true : tag_1->masked; tag_1->blacklisted = tag_2->blacklisted ? true : tag_1->blacklisted; tag_1->lumberjackstack = tag_2->lumberjackstack ? true : tag_1->lumberjackstack; for (j = tag_2->utags.begin(); j != tag_2->utags.end(); j++) tag_1->utags.push_back(*j); for (j = tag_2->remtags.begin(); j != tag_2->remtags.end(); j++) tag_1->remtags.push_back(*j); tag_1->count += tag_2->count; } return tag_1; } MergedStack *merge_tags(map &merged, int *merge_list, int merge_list_size, int id) { int i; vector::iterator j; MergedStack *tag_1, *tag_2; tag_1 = new MergedStack; tag_1->id = id; for (i = 0; i < merge_list_size; i++) { tag_2 = merged[merge_list[i]]; tag_1->deleveraged = tag_2->deleveraged ? true : tag_1->deleveraged; tag_1->masked = tag_2->masked ? true : tag_1->masked; tag_1->blacklisted = tag_2->blacklisted ? true : tag_1->blacklisted; tag_1->lumberjackstack = tag_2->lumberjackstack ? true : tag_1->lumberjackstack; for (j = tag_2->utags.begin(); j != tag_2->utags.end(); j++) tag_1->utags.push_back(*j); for (j = tag_2->remtags.begin(); j != tag_2->remtags.end(); j++) tag_1->remtags.push_back(*j); tag_1->count += tag_2->count; } return tag_1; } int remove_repetitive_stacks(map &unique, map &merged) { // // If enabled, check the depth of coverage of each unique tag, and remove // from consideration any tags with depths greater than removal_trigger. These tags // are likely to be multiple repetitive sites that have been merged together. // Because large stacks of unique tags are likely to also generate many one-off // sequencing error reads, remove all seqeunces that are a distance of one from // the RAD-Tag with high depth of coverage. // map::iterator i; vector >::iterator k; map new_merged; MergedStack *tag_1, *tag_2; set already_merged; // // First, iterate through the stacks and populate a list of tags that will be removed // (those above the removal_trigger and those 1 nucleotide away). Sort the list of // stacks so that we process them from largest depth to shortest so the same stacks // are always merged/removed.. // vector > ordered_tags; for (i = merged.begin(); i != merged.end(); i++) { if (i->second->count > removal_trigger) ordered_tags.push_back(make_pair(i->second->id, i->second->count)); } sort(ordered_tags.begin(), ordered_tags.end(), compare_pair_intint); pair::iterator,bool> ret; int id = 0; // // Merge all stacks that are over the removal trigger with their nearest neighbors and // mask them so they are not further processed by the program. // for (uint j = 0; j < ordered_tags.size(); j++) { tag_1 = merged[ordered_tags[j].first]; // // Don't process a tag that has already been merged. // if (already_merged.count(tag_1->id) > 0) continue; // // Construct a list of MergedStacks that are either: // within a distance of 1 nucleotide of this tag, or // are they themselves above the lumberjack stacks limit. // queue merge_queue; set merge_list; merge_queue.push(tag_1->id); merge_list.insert(tag_1->id); already_merged.insert(tag_1->id); while (!merge_queue.empty()) { tag_2 = merged[merge_queue.front()]; merge_queue.pop(); if (tag_2->count < removal_trigger) continue; for (k = tag_2->dist.begin(); k != tag_2->dist.end(); k++) { ret = already_merged.insert(k->first); if (ret.second == true) { merge_queue.push(k->first); merge_list.insert(k->first); } } } // // Merge these tags together into a new MergedStack object. // tag_2 = merge_tags(merged, merge_list, id); tag_2->add_consensus(tag_1->con); tag_2->lumberjackstack = true; tag_2->masked = true; tag_2->blacklisted = true; new_merged.insert(make_pair(id, tag_2)); id++; } // // Move the non-lumberjack stacks, unmodified, into the new merged map. // for (i = merged.begin(); i != merged.end(); i++) { tag_1 = i->second; if (already_merged.count(tag_1->id) > 0) continue; set merge_list; merge_list.insert(tag_1->id); tag_2 = merge_tags(merged, merge_list, id); tag_2->add_consensus(tag_1->con); new_merged.insert(make_pair(id, tag_2)); id++; } cerr << " Removed " << already_merged.size() << " stacks.\n"; // // Free the memory from the old map of merged tags. // map::iterator it; for (it = merged.begin(); it != merged.end(); it++) delete it->second; merged = new_merged; cerr << " " << merged.size() << " stacks remain for merging.\n"; return 0; } int deleverage(map &unique, map &rem, map &merged, set &merge_list, int cohort_id, vector &deleveraged_tags) { set::iterator it; vector >::iterator j; MergedStack *tag_1, *tag_2; uint k, l; // // Create a minimum spanning tree in order to determine the minimum distance // between each node in the list. // MinSpanTree *mst = new MinSpanTree; vector keys; for (it = merge_list.begin(); it != merge_list.end(); it++) { keys.push_back(*it); mst->add_node(*it); tag_1 = merged[*it]; // cerr << " " << *it << " -> " << tag_1->utags[0] << "\n"; } // // Measure the distance between each pair of nodes and add edges to our // minimum spanning tree. // Node *n_1, *n_2; for (k = 0; k < keys.size(); k++) { tag_1 = merged[keys[k]]; n_1 = mst->node(keys[k]); for (l = k+1; l < keys.size(); l++) { tag_2 = merged[keys[l]]; n_2 = mst->node(keys[l]); int d = dist(tag_1, tag_2); n_1->add_edge(mst->node(keys[l]), d); n_2->add_edge(mst->node(keys[k]), d); } } // // Build the minimum spanning tree. // mst->build_tree(); // // Visualize the MST // if (dump_graph) { stringstream gout_file; size_t pos_1 = in_file.find_last_of("/"); size_t pos_2 = in_file.find_last_of("."); gout_file << out_path << in_file.substr(pos_1 + 1, (pos_2 - pos_1 - 1)) << "_" << keys[0] << ".dot"; string vis = mst->vis(true); ofstream gvis(gout_file.str().c_str()); gvis << vis; gvis.close(); } set visited; set dists; queue q; Node *n = mst->head(); q.push(n); while (!q.empty()) { n = q.front(); q.pop(); visited.insert(n->id); for (uint i = 0; i < n->min_adj_list.size(); i++) { if (visited.count(n->min_adj_list[i]->id) == 0) { q.push(n->min_adj_list[i]); // cerr << n->id << " -> " << n->min_adj_list[i]->id << ": "; // // Find the edge distance. // for (uint j = 0; j < n->edges.size(); j++) if (n->edges[j]->child == n->min_adj_list[i]) { // cerr << n->edges[j]->dist << "\n"; dists.insert(n->edges[j]->dist); } } } } // // This set is sorted by definition. Check if there is more than a single // distance separating stacks. // if (dists.size() == 1) { tag_1 = merge_tags(merged, merge_list, 0); deleveraged_tags.push_back(tag_1); delete mst; return 0; } uint min_dist = *(dists.begin()); // // If there is more than a single distance, split the minimum spanning tree // into separate loci, by cutting the tree at the larger distances. // set uniq_merge_list; visited.clear(); n = mst->head(); q.push(n); int id = 0; uniq_merge_list.insert(n->id); while (!q.empty()) { n = q.front(); q.pop(); visited.insert(n->id); for (uint i = 0; i < n->min_adj_list.size(); i++) { if (visited.count(n->min_adj_list[i]->id) == 0) { q.push(n->min_adj_list[i]); for (uint j = 0; j < n->edges.size(); j++) { if (n->edges[j]->child == n->min_adj_list[i]) if (n->edges[j]->dist > min_dist) { // cerr << "Merging the following stacks into a locus:\n"; for (it = uniq_merge_list.begin(); it != uniq_merge_list.end(); it++) { tag_1 = merged[*it]; // cerr << " " << *it << " -> " << tag_1->utags[0] << "\n"; } tag_1 = merge_tags(merged, uniq_merge_list, id); tag_1->cohort_id = cohort_id; deleveraged_tags.push_back(tag_1); uniq_merge_list.clear(); id++; } } uniq_merge_list.insert(n->min_adj_list[i]->id); } } } // cerr << "Merging the following stacks into a locus:\n"; for (it = uniq_merge_list.begin(); it != uniq_merge_list.end(); it++) { tag_1 = merged[*it]; // cerr << " " << *it << " -> " << tag_1->utags[0] << "\n"; } tag_1 = merge_tags(merged, uniq_merge_list, id); tag_1->cohort_id = cohort_id; deleveraged_tags.push_back(tag_1); uniq_merge_list.clear(); delete mst; return 0; } int calc_kmer_distance(map &merged, int utag_dist) { // // Calculate the distance (number of mismatches) between each pair // of Radtags. We expect all radtags to be the same length; // KmerHashMap kmer_map; vector kmer_map_keys; MergedStack *tag_1, *tag_2; map::iterator it; // OpenMP can't parallelize random access iterators, so we convert // our map to a vector of integer keys. vector keys; for (it = merged.begin(); it != merged.end(); it++) keys.push_back(it->first); // // Calculate the number of k-mers we will generate. If kmer_len == 0, // determine the optimal length for k-mers. // int con_len = strlen(merged[keys[0]]->con); if (set_kmer_len) kmer_len = determine_kmer_length(con_len, utag_dist); int num_kmers = con_len - kmer_len + 1; // // Calculate the minimum number of matching k-mers required for a possible sequence match. // int min_hits = calc_min_kmer_matches(kmer_len, utag_dist, con_len, set_kmer_len ? true : false); cerr << " Distance allowed between stacks: " << utag_dist << "; searching with a k-mer length of " << kmer_len << " (" << num_kmers << " k-mers per read); " << min_hits << " k-mer hits required.\n"; populate_kmer_hash(merged, kmer_map, kmer_map_keys, kmer_len); #pragma omp parallel private(tag_1, tag_2) { KmerHashMap::iterator h; vector query_kmers; initialize_kmers(kmer_len, num_kmers, query_kmers); #pragma omp for schedule(dynamic) for (uint i = 0; i < keys.size(); i++) { tag_1 = merged[keys[i]]; // Don't compute distances for masked tags if (tag_1->masked) continue; generate_kmers_lazily(tag_1->con, kmer_len, num_kmers, query_kmers); map hits; int d; // // Lookup the occurances of each k-mer in the kmer_map // for (int j = 0; j < num_kmers; j++) { h = kmer_map.find(query_kmers[j]); if (h != kmer_map.end()) for (uint k = 0; k < h->second.size(); k++) hits[h->second[k]]++; } // // Iterate through the list of hits. For each hit that has more than min_hits // check its full length to verify a match. // map::iterator hit_it; for (hit_it = hits.begin(); hit_it != hits.end(); hit_it++) { if (hit_it->second < min_hits) continue; tag_2 = merged[hit_it->first]; // Don't compute distances for masked tags if (tag_2->masked) continue; // Don't compare tag_1 against itself. if (tag_1 == tag_2) continue; d = dist(tag_1, tag_2); // // Store the distance between these two sequences if it is // below the maximum distance (which governs those // sequences to be merged in the following step of the // algorithm.) // if (d <= utag_dist) tag_1->add_dist(tag_2->id, d); } // Sort the vector of distances. sort(tag_1->dist.begin(), tag_1->dist.end(), compare_dist); } // // Free the k-mers we generated for this query // for (uint j = 0; j < query_kmers.size(); j++) delete [] query_kmers[j]; } free_kmer_hash(kmer_map, kmer_map_keys); return 0; } int calc_distance(map &merged, int utag_dist) { // // Calculate the distance (number of mismatches) between each pair // of Radtags. We expect all radtags to be the same length; // map::iterator it; MergedStack *tag_1, *tag_2; int i, j; // OpenMP can't parallelize random access iterators, so we convert // our map to a vector of integer keys. vector keys; for (it = merged.begin(); it != merged.end(); it++) keys.push_back(it->first); #pragma omp parallel private(i, j, tag_1, tag_2) { #pragma omp for schedule(dynamic) for (i = 0; i < (int) keys.size(); i++) { tag_1 = merged[keys[i]]; // Don't compute distances for masked tags if (tag_1->masked) continue; int d; for (j = 0; j < (int) keys.size(); j++) { tag_2 = merged[keys[j]]; // Don't compute distances for masked tags if (tag_2->masked) continue; // Don't compare tag_1 against itself. if (tag_1 == tag_2) continue; d = dist(tag_1, tag_2); //cerr << " Distance: " << d << "\n"; // // Store the distance between these two sequences if it is // below the maximum distance (which governs those // sequences to be merged in the following step of the // algorithm.) // if (d == utag_dist) { tag_1->add_dist(tag_2->id, d); //cerr << " HIT.\n"; } } // Sort the vector of distances. sort(tag_1->dist.begin(), tag_1->dist.end(), compare_dist); } } return 0; } int reduce_radtags(DNASeqHashMap &radtags, map &unique, map &rem) { DNASeqHashMap::iterator it; Rem *r; Stack *u; int global_id = 1; for (it = radtags.begin(); it != radtags.end(); it++) { if (it->second.count() < min_merge_cov) { // // Don't record this unique RAD-Tag if its coverage is below // the specified cutoff. However, add the reads to the remainder // vector for later processing. // r = new Rem; r->id = global_id; r->add_seq(&it->first); for (uint i = 0; i < it->second.ids.size(); i++) r->add_id(it->second.ids[i]); rem[r->id] = r; global_id++; } else { // // Populate a Stack object for this unique radtag. Create a // map of the IDs for the sequences that have been // collapsed into this radtag. // u = new Stack; u->id = global_id; u->add_seq(&it->first); // Copy the original Fastq IDs from which this unique radtag was built. for (uint i = 0; i < it->second.ids.size(); i++) u->add_id(it->second.ids[i]); unique[u->id] = u; global_id++; } } if (unique.size() == 0) { cerr << "Error: Unable to form any stacks, data appear to be unique.\n"; exit(1); } return 0; } int calc_coverage_distribution(map &unique, double &mean, double &stdev, double &max) { map::iterator i; double m = 0.0; double s = 0.0; double sum = 0.0; uint cnt = 0; double total = 0.0; mean = 0.0; max = 0.0; stdev = 0.0; map depth_dist; map::iterator j; for (i = unique.begin(); i != unique.end(); i++) { cnt = i->second->count(); m += cnt; total++; depth_dist[cnt]++; if (cnt > max) max = cnt; } mean = m / total; // // Calculate the standard deviation // total = 0.0; for (i = unique.begin(); i != unique.end(); i++) { total++; s = i->second->count(); sum += pow((s - mean), 2); } stdev = sqrt(sum / (total - 1)); return 0; } int calc_coverage_distribution(map &unique, map &merged, double &mean, double &stdev, double &max) { map::iterator it; vector::iterator k; Stack *tag; double m = 0.0; double s = 0.0; double sum = 0.0; double cnt = 0.0; mean = 0.0; max = 0.0; stdev = 0.0; for (it = merged.begin(); it != merged.end(); it++) { if (it->second->blacklisted) continue; cnt++; m = 0.0; for (k = it->second->utags.begin(); k != it->second->utags.end(); k++) { tag = unique[*k]; m += tag->count(); } if (m > max) max = m; sum += m; } mean = sum / cnt; // // Calculate the standard deviation // for (it = merged.begin(); it != merged.end(); it++) { s = 0.0; for (k = it->second->utags.begin(); k != it->second->utags.end(); k++) { tag = unique[*k]; s += tag->count(); } sum += pow((s - mean), 2); } stdev = sqrt(sum / (cnt - 1)); return 0; } int calc_coverage_distribution(map &unique, map &rem, map &merged, double &mean, double &stdev, double &max) { map::iterator it; vector::iterator k; Stack *tag; double m = 0.0; double s = 0.0; double sum = 0.0; double cnt = 0.0; mean = 0.0; max = 0.0; stdev = 0.0; for (it = merged.begin(); it != merged.end(); it++) { if (it->second->blacklisted) continue; cnt++; m = 0.0; for (k = it->second->utags.begin(); k != it->second->utags.end(); k++) { tag = unique[*k]; m += tag->count(); } for (uint j = 0; j < it->second->remtags.size(); j++) m += rem[it->second->remtags[j]]->count(); if (m > max) max = m; sum += m; } mean = sum / cnt; // // Calculate the standard deviation // for (it = merged.begin(); it != merged.end(); it++) { s = 0.0; for (k = it->second->utags.begin(); k != it->second->utags.end(); k++) { tag = unique[*k]; s += tag->count(); } for (uint j = 0; j < it->second->remtags.size(); j++) s += rem[it->second->remtags[j]]->count(); sum += pow((s - mean), 2); } stdev = sqrt(sum / (cnt - 1)); return 0; } int count_raw_reads(map &unique, map &rem, map &merged) { map::iterator it; map::iterator sit; vector::iterator k; Stack *tag; long int m = 0; map uniq_ids; map::iterator uit; for (it = merged.begin(); it != merged.end(); it++) { for (k = it->second->utags.begin(); k != it->second->utags.end(); k++) { tag = unique[*k]; m += tag->count(); if (uniq_ids.count(*k) == 0) uniq_ids[*k] = 0; uniq_ids[*k]++; } for (uint j = 0; j < it->second->remtags.size(); j++) m += rem[it->second->remtags[j]]->count(); //m += it->second->remtags.size(); } for (uit = uniq_ids.begin(); uit != uniq_ids.end(); uit++) if (uit->second > 1) cerr << " Unique stack #" << uit->first << " appears in " << uit->second << " merged stacks.\n"; cerr << "Number of utilized reads: " << m << "\n"; //for (sit = unique.begin(); sit != unique.end(); sit++) // if (uniq_ids.count(sit->first) == 0) // cerr << " Stack " << sit->first << ": '" << sit->second->seq << "' unused.\n"; return 0; } int write_results(map &m, map &u, map &r) { map::iterator i; vector::iterator k; vector::iterator s; map::iterator t; MergedStack *tag_1; Stack *tag_2; Rem *rem; stringstream sstr; bool gzip = (in_file_type == FileT::gzfastq || in_file_type == FileT::gzfasta) ? true : false; // // Read in the set of sequencing IDs so they can be included in the output. // vector seq_ids; load_seq_ids(seq_ids); // // Parse the input file name to create the output files // size_t pos_1 = in_file.find_last_of("/"); size_t pos_2 = in_file.find_last_of("."); if (in_file.substr(pos_2) == ".gz") { in_file = in_file.substr(0, pos_2); pos_2 = in_file.find_last_of("."); } string tag_file = out_path + in_file.substr(pos_1 + 1, (pos_2 - pos_1 - 1)) + ".tags.tsv"; string snp_file = out_path + in_file.substr(pos_1 + 1, (pos_2 - pos_1 - 1)) + ".snps.tsv"; string all_file = out_path + in_file.substr(pos_1 + 1, (pos_2 - pos_1 - 1)) + ".alleles.tsv"; string mod_file = out_path + in_file.substr(pos_1 + 1, (pos_2 - pos_1 - 1)) + ".models.tsv"; if (gzip) { tag_file += ".gz"; snp_file += ".gz"; all_file += ".gz"; mod_file += ".gz"; } // // Open the output files for writing. // gzFile gz_tags, gz_snps, gz_alle, gz_mods; ofstream tags, snps, alle, mods; if (gzip) { gz_tags = gzopen(tag_file.c_str(), "wb"); if (!gz_tags) { cerr << "Error: Unable to open gzipped tag file '" << tag_file << "': " << strerror(errno) << ".\n"; exit(1); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_tags, libz_buffer_size); #endif gz_mods = gzopen(mod_file.c_str(), "wb"); if (!gz_mods) { cerr << "Error: Unable to open gzipped model file '" << mod_file << "': " << strerror(errno) << ".\n"; exit(1); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_mods, libz_buffer_size); #endif gz_snps = gzopen(snp_file.c_str(), "wb"); if (!gz_snps) { cerr << "Error: Unable to open gzipped snps file '" << snp_file << "': " << strerror(errno) << ".\n"; exit(1); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_snps, libz_buffer_size); #endif gz_alle = gzopen(all_file.c_str(), "wb"); if (!gz_alle) { cerr << "Error: Unable to open gzipped alleles file '" << all_file << "': " << strerror(errno) << ".\n"; exit(1); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_alle, libz_buffer_size); #endif } else { tags.open(tag_file.c_str()); if (tags.fail()) { cerr << "Error: Unable to open tag file for writing.\n"; exit(1); } mods.open(mod_file.c_str()); if (mods.fail()) { cerr << "Error: Unable to open model file for writing.\n"; exit(1); } snps.open(snp_file.c_str()); if (snps.fail()) { cerr << "Error: Unable to open SNPs file for writing.\n"; exit(1); } alle.open(all_file.c_str()); if (alle.fail()) { cerr << "Error: Unable to open allele file for writing.\n"; exit(1); } } // // Record the version of Stacks used and the date generated as a comment in the catalog. // // Obtain the current date. // stringstream log; time_t rawtime; struct tm *timeinfo; char date[32]; time(&rawtime); timeinfo = localtime(&rawtime); strftime(date, 32, "%F %T", timeinfo); log << "# ustacks version " << VERSION << "; generated on " << date << "\n"; if (gzip) { gzputs(gz_tags, log.str().c_str()); gzputs(gz_mods, log.str().c_str()); gzputs(gz_snps, log.str().c_str()); gzputs(gz_alle, log.str().c_str()); } else { tags << log.str(); mods << log.str(); snps << log.str(); alle << log.str(); } int id; char *buf = new char[m.begin()->second->len + 1]; for (i = m.begin(); i != m.end(); i++) { float total = 0; tag_1 = i->second; // // Calculate the log likelihood of this merged stack. // tag_1->gen_matrix(u, r); tag_1->calc_likelihood(); // First write the consensus sequence sstr << "0" << "\t" << sql_id << "\t" << tag_1->id << "\t" //<< tag_1->cohort_id << "\t" << "" << "\t" // chr << 0 << "\t" // bp << "+" << "\t" // strand << "consensus\t" << "\t" << "\t" << tag_1->con << "\t" << tag_1->deleveraged << "\t" << tag_1->blacklisted << "\t" << tag_1->lumberjackstack << "\t" << tag_1->lnl << "\n"; // // Write a sequence recording the output of the SNP model for each nucleotide. // sstr << "0" << "\t" << sql_id << "\t" << tag_1->id << "\t" //<< "\t" // cohort_id << "\t" // chr << "\t" // bp << "\t" // strand << "model\t" << "\t" << "\t"; for (s = tag_1->snps.begin(); s != tag_1->snps.end(); s++) { switch((*s)->type) { case snp_type_het: sstr << "E"; break; case snp_type_hom: sstr << "O"; break; default: sstr << "U"; break; } } sstr << "\t" << "\t" << "\t" << "\t" << "\n"; if (gzip) gzputs(gz_tags, sstr.str().c_str()); else tags << sstr.str(); if (gzip) gzputs(gz_mods, sstr.str().c_str()); else mods << sstr.str(); sstr.str(""); // // Now write out the components of each unique tag merged into this locus. // id = 0; for (k = tag_1->utags.begin(); k != tag_1->utags.end(); k++) { tag_2 = u[*k]; total += tag_2->count(); for (uint j = 0; j < tag_2->map.size(); j++) { sstr << "0" << "\t" << sql_id << "\t" << tag_1->id << "\t" //<< "\t" // cohort_id << "\t" // chr << "\t" // bp << "\t" // strand << "primary\t" << id << "\t" << seq_ids[tag_2->map[j]] << "\t" << tag_2->seq->seq(buf) << "\t\t\t\t\n"; if (gzip) gzputs(gz_tags, sstr.str().c_str()); else tags << sstr.str(); sstr.str(""); } id++; } // // Write out the remainder tags merged into this unique tag. // for (k = tag_1->remtags.begin(); k != tag_1->remtags.end(); k++) { rem = r[*k]; total += rem->map.size(); for (uint j = 0; j < rem->map.size(); j++) sstr << "0" << "\t" << sql_id << "\t" << tag_1->id << "\t" //<< "\t" // cohort_id << "\t" // chr << "\t" // bp << "\t" // strand << "secondary\t" << "\t" << seq_ids[rem->map[j]] << "\t" << rem->seq->seq(buf) << "\t\t\t\t\n"; if (gzip) gzputs(gz_tags, sstr.str().c_str()); else tags << sstr.str(); sstr.str(""); } // // Write out the model calls for each nucleotide in this locus. // for (s = tag_1->snps.begin(); s != tag_1->snps.end(); s++) { sstr << "0" << "\t" << sql_id << "\t" << tag_1->id << "\t" << (*s)->col << "\t"; switch((*s)->type) { case snp_type_het: sstr << "E\t"; break; case snp_type_hom: sstr << "O\t"; break; default: sstr << "U\t"; break; } sstr << std::fixed << std::setprecision(2) << (*s)->lratio << "\t" << (*s)->rank_1 << "\t" << (*s)->rank_2 << "\t\t\n"; } if (gzip) gzputs(gz_snps, sstr.str().c_str()); else snps << sstr.str(); sstr.str(""); // // Write the expressed alleles seen for the recorded SNPs and // the percentage of tags a particular allele occupies. // for (t = tag_1->alleles.begin(); t != tag_1->alleles.end(); t++) { sstr << "0" << "\t" << sql_id << "\t" << tag_1->id << "\t" << (*t).first << "\t" << (((*t).second/total) * 100) << "\t" << (*t).second << "\n"; } if (gzip) gzputs(gz_alle, sstr.str().c_str()); else alle << sstr.str(); sstr.str(""); } if (gzip) { gzclose(gz_tags); gzclose(gz_mods); gzclose(gz_snps); gzclose(gz_alle); } else { tags.close(); mods.close(); snps.close(); alle.close(); } // // Free sequence IDs. // for (uint i = 0; i < seq_ids.size(); i++) delete [] seq_ids[i]; // // If specified, output reads not utilized in any stacks. // if (retain_rem_reads) { string unused_file = out_path + in_file.substr(pos_1 + 1, (pos_2 - pos_1 - 1)) + ".unused.fa"; gzFile gz_unused; ofstream unused; if (gzip) { unused_file += ".gz"; gz_unused = gzopen(unused_file.c_str(), "wb"); if (!gz_unused) { cerr << "Error: Unable to open gzipped discard file '" << unused_file << "': " << strerror(errno) << ".\n"; exit(1); } #if ZLIB_VERNUM >= 0x1240 gzbuffer(gz_unused, libz_buffer_size); #endif } else { unused.open(unused_file.c_str()); if (unused.fail()) { cerr << "Error: Unable to open discard file for writing.\n"; exit(1); } } map::iterator r_it; for (r_it = r.begin(); r_it != r.end(); r_it++) { if (r_it->second->utilized == false) sstr << ">" << r_it->second->id << "\n" << r_it->second->seq->seq(buf) << "\n"; if (gzip) gzputs(gz_unused, sstr.str().c_str()); else unused << sstr.str(); sstr.str(""); } if (gzip) gzclose(gz_unused); else unused.close(); } delete [] buf; return 0; } int dump_stack_graph(string data_file, map &unique, map &merged, vector &keys, map > &dist_map, map > &cluster_map) { uint s, t; double d, scale, scaled_d; char label[32]; vector colors; std::ofstream data(data_file.c_str()); size_t pos_1 = data_file.find_last_of("/"); size_t pos_2 = data_file.find_last_of("."); string title = data_file.substr(pos_1 + 1, pos_2 - pos_1 - 1); // // Output a list of IDs so we can locate these stacks in the final results. // for (s = 0; s < keys.size(); s++) data << "/* " << keys[s] << ": " << unique[merged[keys[s]]->utags[0]]->map[0] << "; depth: " << merged[keys[s]]->count << " */\n"; // // Output a specification to visualize the stack graph using graphviz: // http://www.graphviz.org/ // data << "graph " << title.c_str() << " {\n" << "rankdir=LR\n" << "size=\"20!\"\n" << "overlap=false\n" << "node [shape=circle style=filled fillcolor=\"#3875d7\" fontname=\"Arial\"];\n" << "edge [fontsize=8.0 fontname=\"Arial\" color=\"#aaaaaa\"];\n"; colors.push_back("red"); colors.push_back("blue"); colors.push_back("green"); colors.push_back("brown"); colors.push_back("purple"); map >::iterator c; set::iterator it; int color_index = 0; string color; // Write out the clusters created by R, prior to writing all the nodes and connections. s = 0; for (c = cluster_map.begin(); c != cluster_map.end(); c++) { data << "subgraph " << s << " {\n" << " edge [penwidth=5 fontsize=12.0 fontcolor=\"black\" color=\"black\"]\n"; if ((*c).second.size() == 1) { color = "white"; data << " node [fillcolor=" << color.c_str() << " fontcolor=\"black\"]\n"; } else { color = colors[color_index % colors.size()]; data << " node [fillcolor=" << color.c_str() << " fontcolor=\"white\"]\n"; color_index++; } for (it = (*c).second.begin(); it != (*c).second.end(); it++) { data << " " << *it << "\n"; } if ((*c).second.size() > 1) { uint j = 0; for (it = (*c).second.begin(); it != (*c).second.end(); it++) { data << *it; if (j < (*c).second.size() - 1) data << " -- "; j++; } } data << "}\n"; s++; } // // Scale the graph to display on a 10 inch canvas. Find the largest edge weight // and scale the edge lengths to fit the canvas. // for (s = 0; s < keys.size(); s++) for (t = s+1; t < keys.size(); t++) scale = dist_map[keys[s]][keys[t]] > scale ? dist_map[keys[s]][keys[t]] : scale; scale = scale / 20; for (s = 0; s < keys.size(); s++) { for (t = s+1; t < keys.size(); t++) { d = dist_map[keys[s]][keys[t]]; scaled_d = d / scale; scaled_d = scaled_d < 0.75 ? 0.75 : scaled_d; sprintf(label, "%.1f", d); data << keys[s] << " -- " << keys[t] << " [len=" << scaled_d << ", label=" << label << "];\n"; } } data << "}\n"; data.close(); return 0; } int dump_unique_tags(map &u) { map::iterator it; vector >::iterator pit; vector::iterator mit; char *c; for (it = u.begin(); it != u.end(); it++) { c = (*it).second->seq->seq(); cerr << "UniqueTag UID: " << (*it).second->id << "\n" << " Seq: " << c << "\n" << " IDs: "; for (uint j = 0; j < it->second->map.size(); j++) cerr << it->second->map[j] << " "; cerr << "\n\n"; delete [] c; } return 0; } int dump_merged_tags(map &m) { map::iterator it; vector >::iterator pit; vector::iterator fit; for (it = m.begin(); it != m.end(); it++) { cerr << "MergedStack ID: " << it->second->id << "\n" << " Consensus: "; if (it->second->con != NULL) cerr << it->second->con << "\n"; else cerr << "\n"; cerr << " IDs: "; for (fit = it->second->utags.begin(); fit != it->second->utags.end(); fit++) cerr << (*fit) << " "; cerr << "\n" << " Distances: "; for (pit = it->second->dist.begin(); pit != it->second->dist.end(); pit++) cerr << (*pit).first << ": " << (*pit).second << ", "; cerr << "\n\n"; } return 0; } int load_radtags(string in_file, DNASeqHashMap &radtags) { Input *fh = NULL; if (in_file_type == FileT::fasta) fh = new Fasta(in_file.c_str()); else if (in_file_type == FileT::fastq) fh = new Fastq(in_file.c_str()); else if (in_file_type == FileT::gzfasta) fh = new GzFasta(in_file.c_str()); else if (in_file_type == FileT::gzfastq) fh = new GzFastq(in_file.c_str()); cerr << "Parsing " << in_file.c_str() << "\n"; long int corrected = 0; long int i = 0; short int seql = 0; short int prev_seql = 0; bool len_mismatch = false; Seq c; c.id = new char[id_len]; c.seq = new char[max_len]; c.qual = new char[max_len]; cerr << "Loading RAD-Tags..."; while ((fh->next_seq(c)) != 0) { if (i % 1000000 == 0 && i>0) cerr << i/1000000 << "M..."; prev_seql = seql; seql = 0; for (char *p = c.seq; *p != '\0'; p++, seql++) switch (*p) { case 'N': case 'n': case '.': *p = 'A'; corrected++; } if (seql != prev_seql && prev_seql > 0) len_mismatch = true; DNASeqHashMap::iterator element = radtags.insert({DNANSeq(seql, c.seq), HVal()}).first; element->second.add_id(i); i++; } cerr << "done\n"; if (i == 0) { cerr << "Error: Unable to load data from '" << in_file.c_str() << "'.\n"; exit(1); } if (len_mismatch) cerr << "Warning: different sequence lengths detected, this will interfere with Stacks algorithms.\n"; cerr << "Loaded " << i << " RAD-Tags.\n" " Inserted " << radtags.size() << " elements into the RAD-Tags hash map.\n" " " << corrected << " reads contained uncalled nucleotides that were modified.\n"; delete fh; return 0; } int load_seq_ids(vector &seq_ids) { Input *fh = NULL; if (in_file_type == FileT::fasta) fh = new Fasta(in_file.c_str()); else if (in_file_type == FileT::fastq) fh = new Fastq(in_file.c_str()); else if (in_file_type == FileT::gzfasta) fh = new GzFasta(in_file.c_str()); else if (in_file_type == FileT::gzfastq) fh = new GzFastq(in_file.c_str()); cerr << " Refetching sequencing IDs from " << in_file.c_str() << "... "; char *id; Seq c; c.id = new char[id_len]; c.seq = new char[max_len]; c.qual = new char[max_len]; while ((fh->next_seq(c)) != 0) { id = new char[strlen(c.id) + 1]; strcpy(id, c.id); seq_ids.push_back(id); } cerr << "read " << seq_ids.size() << " sequence IDs.\n"; delete fh; return 0; } int calc_triggers(double cov_mean, double cov_stdev, double cov_scale, int &deleverage_trigger, int &removal_trigger) { deleverage_trigger = (int) round(cov_mean + cov_stdev * cov_scale); removal_trigger = (int) round(cov_mean + (cov_stdev * 2) * cov_scale); return 0; // // // // Calculate the deleverage trigger. Assume RAD-Tags are selected from // // the sample for sequencing randomly, forming a poisson distribution // // representing the depths of coverage of RAD-Tags in the sample. Calculate // // the trigger value that is larger than the depth of coverage of 99.9999% of stacks. // // // long double lambda = cov_mean; // int k = 0; // long double d = 0.0; // long double e = 0.0; // long double f = 0.0; // long double g = 0.0; // long double h = 0.0; // long double i = 0.0; // do { // e = exp(-1 * lambda); // g = pow(lambda, k); // f = factorial(k); // h = (e * g); // i = h / f; // d += i; // //cerr << "iteration " << k << "; e: " << e << " h: " << h << " g: " << g << " F: " << f << " i: " << i << " D: " << d << "\n"; // k++; // } while (d < 0.999999); // return k - 1; } long double factorial(int i) { long double f = 1; if (i == 0) return 1; do { f = f * i; i--; } while (i > 0); return f; } int parse_command_line(int argc, char* argv[]) { int c; while (1) { static struct option long_options[] = { {"help", no_argument, NULL, 'h'}, {"version", no_argument, NULL, 'v'}, {"infile_type", required_argument, NULL, 't'}, {"file", required_argument, NULL, 'f'}, {"outpath", required_argument, NULL, 'o'}, {"id", required_argument, NULL, 'i'}, {"min_cov", required_argument, NULL, 'm'}, {"max_dist", required_argument, NULL, 'M'}, {"max_sec_dist", required_argument, NULL, 'N'}, {"max_locus_stacks", required_argument, NULL, 'K'}, {"k_len", required_argument, NULL, 'k'}, {"num_threads", required_argument, NULL, 'p'}, {"deleverage", no_argument, NULL, 'd'}, {"remove_rep", no_argument, NULL, 'r'}, {"retain_rem", no_argument, NULL, 'R'}, {"graph", no_argument, NULL, 'g'}, {"sec_hapl", no_argument, NULL, 'H'}, {"gapped", no_argument, NULL, 'G'}, {"max_gaps", required_argument, NULL, 'X'}, {"min_aln_len", required_argument, NULL, 'x'}, {"model_type", required_argument, NULL, 'T'}, {"bc_err_freq", required_argument, NULL, 'e'}, {"bound_low", required_argument, NULL, 'L'}, {"bound_high", required_argument, NULL, 'U'}, {"alpha", required_argument, NULL, 'A'}, {0, 0, 0, 0} }; // getopt_long stores the option index here. int option_index = 0; c = getopt_long(argc, argv, "GhHvdrgRA:L:U:f:o:i:m:e:p:t:M:N:K:k:T:X:x:", long_options, &option_index); // Detect the end of the options. if (c == -1) break; switch (c) { case 'h': help(); break; case 't': if (strcmp(optarg, "tsv") == 0) in_file_type = FileT::tsv; else if (strcmp(optarg, "fasta") == 0) in_file_type = FileT::fasta; else if (strcmp(optarg, "fastq") == 0) in_file_type = FileT::fastq; else if (strcasecmp(optarg, "gzfasta") == 0) in_file_type = FileT::gzfasta; else if (strcasecmp(optarg, "gzfastq") == 0) in_file_type = FileT::gzfastq; else in_file_type = FileT::unknown; break; case 'f': in_file = optarg; break; case 'o': out_path = optarg; break; case 'i': sql_id = is_integer(optarg); if (sql_id < 0) { cerr << "SQL ID (-i) must be an integer, e.g. 1, 2, 3\n"; help(); } break; case 'm': min_merge_cov = is_integer(optarg); break; case 'M': max_utag_dist = is_integer(optarg); break; case 'N': max_rem_dist = is_integer(optarg); break; case 'd': deleverage_stacks++; break; case 'r': remove_rep_stacks++; break; case 'K': max_subgraph = is_integer(optarg); break; case 'k': set_kmer_len = false; kmer_len = is_integer(optarg); break; case 'R': retain_rem_reads = true; break; case 'g': dump_graph++; break; case 'G': gapped_alignments = true; break; case 'X': max_gaps = is_double(optarg); break; case 'x': min_match_len = is_double(optarg); break; case 'T': if (strcmp(optarg, "snp") == 0) { model_type = snp; } else if (strcmp(optarg, "fixed") == 0) { model_type = fixed; } else if (strcmp(optarg, "bounded") == 0) { model_type = bounded; } else { cerr << "Unknown model type specified '" << optarg << "'\n"; help(); } case 'e': barcode_err_freq = is_double(optarg); break; case 'L': bound_low = is_double(optarg); break; case 'U': bound_high = is_double(optarg); break; case 'A': alpha = is_double(optarg); break; case 'H': call_sec_hapl = false; break; case 'p': num_threads = is_integer(optarg); break; case 'v': version(); break; case '?': // getopt_long already printed an error message. help(); break; default: cerr << "Unknown command line option '" << (char) c << "'\n"; help(); abort(); } } if (set_kmer_len == false && (kmer_len < 5 || kmer_len > 31)) { cerr << "Kmer length must be between 5 and 31bp.\n"; help(); } if (alpha != 0.1 && alpha != 0.05 && alpha != 0.01 && alpha != 0.001) { cerr << "SNP model alpha significance level must be either 0.1, 0.05, 0.01, or 0.001.\n"; help(); } if (bound_low != 0 && (bound_low < 0 || bound_low >= 1.0)) { cerr << "SNP model lower bound must be between 0.0 and 1.0.\n"; help(); } if (bound_high != 1 && (bound_high <= 0 || bound_high > 1.0)) { cerr << "SNP model upper bound must be between 0.0 and 1.0.\n"; help(); } if (bound_low > 0 || bound_high < 1.0) { model_type = bounded; } if (in_file.length() == 0 || in_file_type == FileT::unknown) { cerr << "You must specify an input file of a supported type.\n"; help(); } if (out_path.length() == 0) out_path = "."; if (out_path.at(out_path.length() - 1) != '/') out_path += "/"; if (model_type == fixed && barcode_err_freq == 0) { cerr << "You must specify the barcode error frequency.\n"; help(); } return 0; } void version() { std::cerr << "ustacks " << VERSION << "\n\n"; exit(0); } void help() { std::cerr << "ustacks " << VERSION << "\n" << "ustacks -t file_type -f file_path [-d] [-r] [-o path] [-i id] [-m min_cov] [-M max_dist] [-p num_threads] [-R] [-H] [-h]" << "\n" << " t: input file Type. Supported types: fasta, fastq, gzfasta, or gzfastq.\n" << " f: input file path.\n" << " o: output path to write results.\n" << " i: SQL ID to insert into the output to identify this sample.\n" << " m: Minimum depth of coverage required to create a stack (default 3).\n" << " M: Maximum distance (in nucleotides) allowed between stacks (default 2).\n" << " N: Maximum distance allowed to align secondary reads to primary stacks (default: M + 2).\n" << " R: retain unused reads.\n" << " H: disable calling haplotypes from secondary reads.\n" << " p: enable parallel execution with num_threads threads.\n" << " h: display this help messsage.\n\n" << " Stack assembly options:\n" << " r: enable the Removal algorithm, to drop highly-repetitive stacks (and nearby errors) from the algorithm.\n" << " d: enable the Deleveraging algorithm, used for resolving over merged tags.\n" << " --max_locus_stacks : maximum number of stacks at a single de novo locus (default 3).\n" << " --k_len : specify k-mer size for matching between alleles and loci (automatically calculated by default).\n\n" << " Gapped assembly options:\n" << " --gapped: preform gapped alignments between stacks.\n" << " --max_gaps: number of gaps allowed between stacks before merging (default: 2).\n" << " --min_aln_len: minimum length of aligned sequence in a gapped alignment (default: 0.80).\n\n" << " Model options:\n" << " --model_type: either 'snp' (default), 'bounded', or 'fixed'\n" << " For the SNP or Bounded SNP model:\n" << " --alpha : chi square significance level required to call a heterozygote or homozygote, either 0.1, 0.05 (default), 0.01, or 0.001.\n" << " For the Bounded SNP model:\n" << " --bound_low : lower bound for epsilon, the error rate, between 0 and 1.0 (default 0).\n" << " --bound_high : upper bound for epsilon, the error rate, between 0 and 1.0 (default 1).\n" << " For the Fixed model:\n" << " --bc_err_freq : specify the barcode error frequency, between 0 and 1.0.\n"; exit(0); } stacks-1.44/src/ustacks.h000644 000765 000024 00000012453 12764047443 016137 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2016, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __USTACKS_H__ #define __USTACKS_H__ #include "constants.h" #ifdef _OPENMP #include // OpenMP library #endif #include // Process command-line options #include #include #include #include #include using std::pair; using std::make_pair; #include using std::string; #include #include #include using std::ofstream; using std::stringstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include // std::setprecision #include using std::vector; #include using std::map; #include using std::unordered_map; #include using std::queue; #include using std::set; #include #ifdef HAVE_SPARSEHASH #include using google::sparse_hash_map; #endif #include "config.h" #include "kmers.h" #include "utils.h" #include "DNASeq.h" // Class for storing two-bit compressed DNA sequences #include "stacks.h" // Major data structures for holding stacks #include "mstack.h" #include "mst.h" // Minimum spanning tree implementation #include "models.h" // Contains maximum likelihood statistical models. #include "FastaI.h" // Reading input files in FASTA format #include "FastqI.h" // Reading input files in FASTQ format #include "gzFasta.h" // Reading gzipped input files in FASTA format #include "gzFastq.h" // Reading gzipped input files in FASTQ format #include "aln_utils.h" #include "GappedAln.h" class HVal { public: vector ids; int count() { return this->ids.size(); } int add_id(int id) { this->ids.push_back(id); return 0; } }; const int barcode_size = 5; #ifdef HAVE_SPARSEHASH typedef sparse_hash_map DNASeqHashMap; #else typedef unordered_map DNASeqHashMap; #endif void help( void ); void version( void ); int parse_command_line(int, char**); int load_radtags(string, DNASeqHashMap &); int load_seq_ids(vector &); int reduce_radtags(DNASeqHashMap &, map &, map &); int free_radtags_hash(DNASeqHashMap &, vector &); int populate_merged_tags(map &, map &); int merge_stacks(map &, map &, map &, set &, int); int call_consensus(map &, map &, map &, bool); int call_alleles(MergedStack *, vector &, vector &); int merge_remainders(map &, map &); int write_results(map &, map &, map &); // // Match MergedStacks using a k-mer hashing algorithm // int calc_kmer_distance(map &, int); int search_for_gaps(map &, double); int merge_gapped_alns(map &, map &, map &); int edit_gapped_seqs(map &, map &, MergedStack *, vector > &); int edit_gaps(vector > &, char *); int dist(MergedStack *, MergedStack *, vector > &); bool rank_alignments(Aln, Aln); // // Calculate depth of coverage statistics for stacks // int calc_coverage_distribution(map &, double &, double &, double &); int calc_coverage_distribution(map &, map &, double &, double &, double &); int calc_coverage_distribution(map &, map &, map &, double &, double &, double &); int count_raw_reads(map &, map &, map &); // // Dealing with lumberjack (huge) stacks // int calc_triggers(double, double, double, int &, int &); int remove_repetitive_stacks(map &, map &); int deleverage(map &, map &, map &, set &, int, vector &); // // Debugging // int dump_unique_tags(map &); int dump_merged_tags(map &); int dump_stack_graph(string, map &, map &, vector &, map > &, map > &); // // Utilities // MergedStack *merge_tags(MergedStack *, MergedStack *, int); MergedStack *merge_tags(map &, set &, int); MergedStack *merge_tags(map &, int *, int, int); long double factorial(int); // // Deprecated // int calc_distance(map &, int); #endif // __USTACKS_H__ stacks-1.44/src/utils.cc000644 000765 000024 00000011535 12764047443 015760 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // utils.cc -- common routines needed in multiple object files. // // Julian Catchen // jcatchen@uoregon.edu // University of Oregon // #include "utils.h" char reverse(char c) { switch (c) { case 'A': case 'a': return 'T'; break; case 'C': case 'c': return 'G'; break; case 'G': case 'g': return 'C'; break; case 'T': case 't': return 'A'; break; case 'N': case 'n': case '.': return 'N'; break; case '-': default: return '-'; break; } return 'N'; } char * rev_comp(const char *seq) { int len = strlen(seq); int j = 0; char *com = new char[len + 1]; const char *p; for (p = seq + len - 1; p >= seq; p--) { switch (*p) { case 'A': case 'a': com[j] = 'T'; break; case 'C': case 'c': com[j] = 'G'; break; case 'G': case 'g': com[j] = 'C'; break; case 'T': case 't': com[j] = 'A'; break; case 'N': case 'n': case '.': com[j] = 'N'; break; } j++; } com[len] = '\0'; return com; } void reverse_string(char *seq) { int len = strlen(seq); char *p = seq; char *q = seq + len - 1; char tmp; while (q > p) { tmp = *q; *q = *p; *p = tmp; q--; p++; } return; } int is_integer(const char *str) { // // Adapted from the strtol manpage. // char *endptr; // To distinguish success/failure after call errno = 0; long val = strtol(str, &endptr, 10); // // Check for various possible errors // if ((errno == ERANGE && (val == LONG_MAX || val == LONG_MIN)) || (errno != 0 && val == 0)) { return -1; } if (endptr == str || *endptr != '\0') return -1; return (int) val; } double is_double(const char *str) { // // Adapted from the strtol manpage. // char *endptr; // To distinguish success/failure after call errno = 0; double val = strtod(str, &endptr); // // Check for various possible errors // if ((errno == ERANGE && (val == LONG_MAX || val == LONG_MIN)) || (errno != 0 && val == 0)) { return -1; } if (endptr == str || *endptr != '\0') return -1; return val; } double factorial(double n) { double fact = 1; for (double i = n; i > 1; i--) fact *= i; return fact; } double reduced_factorial(double n, double d) { double f = n - d; if (f < 0) return 0; else if (f == 0) return 1; else if (f == 1) return n; f = n; n--; while (n > d) { f *= n; n--; } return f; } double log_factorial(double n) { double fact = 0; for (double i = n; i > 1; i--) fact += log(i); return fact; } double reduced_log_factorial(double n, double d) { double f = n - d; if (f < 0) return 0; else if (f == 0) return 0; else if (f == 1) return log(n); f = log(n); n--; while (n > d) { f += log(n); n--; } return f; } bool compare_pair(pair a, pair b) { return (a.second > b.second); } bool compare_pair_intint(pair a, pair b) { return (a.second > b.second); } bool compare_pair_intdouble(pair a, pair b) { return (a.second < b.second); } bool compare_pair_stringint(pair a, pair b) { return (a.second < b.second); } bool compare_ints(int a, int b) { return (a > b); } bool compare_pair_snp(pair a, pair b) { return (a.second->col < b.second->col); } bool compare_pair_haplotype(pair a, pair b) { return (a.second > b.second); } bool compare_pair_haplotype_rev(pair a, pair b) { return (a.second < b.second); } stacks-1.44/src/utils.h000644 000765 000024 00000004207 12755650230 015612 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __UTILS_H__ #define __UTILS_H__ #include #include #include #include #include using std::cerr; using std::endl; #include using std::pair; using std::make_pair; #include "stacks.h" char reverse(char); char *rev_comp(const char *); void reverse_string(char *); int is_integer(const char *); double is_double(const char *); double factorial(double); double reduced_factorial(double, double); double log_factorial(double); double reduced_log_factorial(double, double); // // Comparison functions for the STL sort routine // bool compare_ints(int, int); bool compare_pair(pair, pair); bool compare_pair_intint(pair, pair); bool compare_pair_intdouble(pair, pair); bool compare_pair_stringint(pair, pair); bool compare_pair_snp(pair, pair); bool compare_pair_haplotype(pair, pair); bool compare_pair_haplotype_rev(pair, pair); // // Comparison classes for STL sets // struct int_increasing { bool operator() (const int& lhs, const int& rhs) const { return lhs < rhs; } }; struct int_decreasing { bool operator() (const int& lhs, const int& rhs) const { return lhs > rhs; } }; #endif // __UTILS_H__ stacks-1.44/src/Vcf.cc000644 000765 000024 00000040556 12774470747 015353 0ustar00catchenstaff000000 000000 #include #include #include #include #include "Vcf.h" using namespace std; const map VcfMeta::predefined = { {"INFO/NS", VcfMeta("INFO","")}, {"INFO/AF", VcfMeta("INFO","")}, {"FORMAT/GT", VcfMeta("FORMAT","")}, {"FORMAT/DP", VcfMeta("FORMAT","")}, {"FORMAT/AD", VcfMeta("FORMAT","")}, {"FORMAT/GL", VcfMeta("FORMAT","")}, }; void VcfHeader::init_meta(const string& fileformat) { add_meta(VcfMeta("fileformat", fileformat)); time_t t; time(&t); char date[9]; strftime(date, 9, "%Y%m%d", localtime(&t)); add_meta(VcfMeta("fileDate", date)); add_meta(VcfMeta("source", string("\"Stacks v") + VERSION + "\"")); } VcfAbstractParser::VcfAbstractParser(const string& path) : path_(path), header_(), line_number_(0), eol_(true), eof_(false), tabs_(), bounds_(), sample_index_(-1) { memset(line_, '\0', Vcf::line_buf_size); } void VcfAbstractParser::read_header() { auto malformed = [this] () { cerr << "Error: Malformed header." << " At line " << line_number_ << " in file '" << path_ << "'.\n"; throw exception(); }; while(true) { getline(line_, Vcf::line_buf_size); ++line_number_; if(eof_) malformed(); if (line_[0] != '#') malformed(); if(line_[1] == '#') { // Meta header line. char* equal = strchr(line_, '='); if(equal == NULL) { // (Notice: Skipping missing '=') tabs_.clear(); tabs_.push_back(line_+strlen(line_)); check_eol(); read_to_eol(); continue; } char* end = equal + strlen(equal); // Check the length of the line (discard lines that don't fit in the buffer). tabs_.clear(); tabs_.push_back(end); check_eol(); if(!eol_) { // (Notice: Skipping long line) read_to_eol(); continue; } string key = string(line_+2, equal); transform(key.begin(), key.end(), key.begin(), ::toupper); header_.add_meta(VcfMeta(key, string(equal+1, end))); continue; } break; } // Final header line. if(strncmp(line_, Vcf::base_fields.c_str(), Vcf::base_fields.length()) != 0) malformed(); // Get tabs. tabs_.clear(); tabs_.push_back(strchr(line_, '\t')); //rem. one tab guaranteed while (tabs_.back()) tabs_.push_back(strchr(tabs_.back()+1, '\t')); tabs_.pop_back(); tabs_.push_back(tabs_.back() + strlen(tabs_.back())); // final '\0' check_eol(); // Check the number of tabs. if(!(tabs_.size() == Vcf::base_fields_no or tabs_.size() >= Vcf::base_fields_no+2)) malformed(); // Check windows line endings. if(*(tabs_.back()-1) == '\r') { tabs_.back() = tabs_.back()-1; *tabs_.back() = '\0'; } // Parse sample names. if(tabs_.size() >= Vcf::base_fields_no+2) { for(size_t i=Vcf::first_sample-1; i < tabs_.size()-2; ++i) header_.add_sample(string(tabs_[i]+1, tabs_[i+1])); while(!eol_){ // Buffer wasn't long enough : parse until eol is found. // Copy the truncated name at the beginning of the buffer. size_t last_field_len = tabs_.back() - *(tabs_.end()-2); *line_ = '\t'; strcpy(line_+1, *(tabs_.end()-2)+1); getline(line_+last_field_len, Vcf::line_buf_size-last_field_len); if(eof_) malformed(); // Get tabs. tabs_.clear(); tabs_.push_back(line_); while (tabs_.back()) tabs_.push_back(strchr(tabs_.back()+1, '\t')); tabs_.pop_back(); tabs_.push_back(tabs_.back() + strlen(tabs_.back())); check_eol(); // Check windows line endings. if(*(tabs_.back()-1) == '\r') { tabs_.back() = tabs_.back()-1; *tabs_.back() = '\0'; } for(size_t i = 0; i= 0x1240 gzbuffer(file_, libz_buffer_size); #endif } #endif // HAVE_LIBZ VcfAbstractParser* Vcf::adaptive_open(const string& path) { VcfAbstractParser* parser = NULL; if (path.length() >= 4 && path.substr(path.length()-4) == ".vcf") { parser = new VcfParser(path); if (parser->fail()) { // Opening failed delete parser; parser = NULL; } #ifdef HAVE_LIBZ } else if (path.length() >= 7 && path.substr(path.length()-7) == ".vcf.gz") { parser = new VcfGzParser(path); if (parser->fail()) { delete parser; parser = NULL; } #endif } else { cerr << "Error: File '" << path << "' : expected '.vcf(.gz)' suffix.\n"; throw exception(); } return parser; } bool VcfAbstractParser::next_record(VcfRecord& record) { getline(line_, Vcf::line_buf_size); if(eof_) return false; ++line_number_; record.clear(); // Get tabs. tabs_.clear(); tabs_.push_back(strchr(line_, '\t')); while(tabs_.back()) tabs_.push_back(strchr(tabs_.back()+1, '\t')); tabs_.pop_back(); if(tabs_.size() > 0) tabs_.push_back(tabs_.back() + strlen(tabs_.back())); else tabs_.push_back(line_ + strlen(line_)); check_eol(); /* Check the number of fields (should be ==8 or >=10 depending on the * presence of samples) : * If the line fits in the buffer but the number of fields is wrong, * raise an error. * If the fixed fields don't fit in the buffer, return a null record. */ if(eol_) { if(tabs_.size() < Vcf::base_fields_no + (header_.samples().empty() ? 0 : 2)) { cerr << "Error: malformed VCF record line (at least " << Vcf::base_fields_no + (header_.samples().empty() ? 0 : 2) << " fields required).\nLine " << line_number_ << " in file '" << path_ << "'.\n"; throw exception (); } } else { if(header_.samples().empty() || tabs_.size() < Vcf::base_fields_no + 2) { cerr << "Warning: In file '" << path_ << "': skipping the very long record at line " << line_number_ << ".\n"; read_to_eol(); return true; } } // Check windows line endings. if(*(tabs_.back()-1) == '\r') { tabs_.back() = tabs_.back()-1; *tabs_.back() = '\0'; } // Separate all substrings (replace \t's with \0's). for(size_t i = 0; ifirst << "=" << i->second; ++i; while(i != r.info.end()) { file_ << ";" << i->first << "=" << i->second; ++i; } } if (not h.samples().empty()) { //FORMAT file_ << "\t"; if (r.format.empty()) { file_ << "."; } else { auto f = r.format.begin(); file_ << *f; ++f; while(f != r.format.end()) { file_ << ":" << *f; ++f; } } //SAMPLES if (r.samples.size() != h.samples().size()) throw exception(); for (const string& s : r.samples) file_ << "\t" << s; } file_ << "\n"; } stacks-1.44/src/Vcf.h000644 000765 000024 00000030573 12750704270 015174 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2010-2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __VCF_H__ #define __VCF_H__ #include "config.h" #include #include #include #include #include #include #include #include #include #include #include #ifdef HAVE_LIBZ #include #endif #include "constants.h" using std::size_t; using std::pair; using std::vector; using std::string; using std::set; using std::map; using std::ifstream; using std::ofstream; using std::cerr; using std::exception; using std::out_of_range; class VcfAbstractParser; class VcfHeader; namespace Vcf { const size_t base_fields_no = 8; // CHROM POS ID REF ALT QUAL FILTER INFO [FORMAT SAMPLE ...]] const string base_fields = "#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO"; const size_t chrom = 0; const size_t pos = 1; const size_t id = 2; const size_t ref = 3; const size_t alt = 4; const size_t qual = 5; const size_t filter = 6; const size_t info = 7; const size_t format = 8; const size_t first_sample = 9; const string base_header = "#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO"; // enum for record types enum class RType { null, expl, // Record with explicitly written alleles, e.g. SNPs, small indels or entirely defined haplotypes invariant, // ALT is empty symbolic, // ALT is e.g. "" breakend // ALT is e.g. "G]17:198982]" }; // Constants for the parser. const size_t line_buf_size = 4096; // Open the given VCF file using VcfParser or VcfGzParser, depending on the // suffix of the file. Return NULL if the opening failed. // (The pointee is dynamically allocated and should be deleted.) VcfAbstractParser* adaptive_open(const string& path); } //namespace Vcf /* * VcfRecord * ========== * Datastructure to store VCF records */ struct VcfRecord { string chrom; // required size_t pos; // required string id; vector alleles; // allele 0 is REF and is required ; case insensitive string qual; vector filter; vector > info; vector format; vector samples; Vcf::RType type; //map allele_indexes; //void refresh_allele_indexes(); VcfRecord() : chrom(), pos(-1), id(), alleles(), qual(), filter(), info(), format(), samples(), type(Vcf::RType::null) {} // Clears all the members. inline void clear(); inline size_t index_of_gt_subfield(const string& key) const; inline string parse_gt_subfield(const string& sample, size_t index) const; // Returns (first allele, second allele), '-1' meaning no data. inline pair parse_genotype(const string& sample) const; inline bool is_snp() const; }; /* * VcfMeta * ========== * Represents one line of VCF metainformation. */ class VcfMeta { string key_; string value_; public: VcfMeta(const string& k, const string& p) : key_(k), value_(p) {} const string& key() const {return key_;} const string& value() const {return value_;} static const map predefined; }; //class Info : public Meta {...}; /* * VcfHeader * ========== * Stores the contents of a VCF header. */ class VcfHeader { vector samples_; vector meta_; map sample_indexes_; // map > meta_indexes; public: VcfHeader() : samples_(), meta_() {} const vector& meta() const {return meta_;} const vector& samples() const {return samples_;} const map& sample_indexes() const {return sample_indexes_;} // Adds the meta lines VERSION, FILEDATE and SOURCE void init_meta(const string& fileformat = "VCFv4.2"); void add_meta(const VcfMeta& m) {meta_.push_back(m);} void add_sample(const string& s) {samples_.push_back(s); sample_indexes_.insert({s, samples_.size()-1});} }; /* * VcfAbstractParser * ========== * * Main class for parsing VCF. The derived non-abstract classes * VcfParser and VcfGzParser only add the file_ attribute and * implement the getline(), eof() and check_eol() methods. * * At present, the parser does not handle : * -- records for which the fixed fields (CHROM to FORMAT) do not * fit in the buffer. Such records usually correspond to large * indels with several alleles. In these cases, the record * passed to next_record() will be empty and have the default * type 'null'. * -- records describing symbolic or breakends alleles. In these * cases the record passed to next_record() will be of type * 'symbolic' or 'breakend', with only the fields CHROM, POS, * ID and REF filled in. */ class VcfAbstractParser { protected: const string path_; VcfHeader header_; size_t line_number_; char line_[Vcf::line_buf_size]; bool eol_; // Set by check_eol(). True if the buffer reaches the end of the currently parsed line. bool eof_; // Set by getline(). True if EOF was reached. vector tabs_; // Vector of pointers to {first_tab, second_tab, ..., \0 } in [line_]. vector bounds_; // Vector of pointers to {leading_tab, first_sep, second_sep, ..., (trailing \t or \0) } in [line_]. size_t sample_index_; // Index of the current (next) sample. virtual void getline(char* ptr, size_t n) =0; virtual void check_eol() =0; // rem. The implementation in gzparser relies on [tabs_] to access the end of the string in [line_]. inline void read_to_eol(); // Reads while [eol_] is false. // Add a sample to [record.samples_] if [samples_to_keep_.at(sample_index_)] is true. inline void add_sample(VcfRecord& record, char* tab1, char* tab2); public: VcfAbstractParser(const string& path); virtual ~VcfAbstractParser() {} // Assess the state of the underlying file. virtual bool fail() =0; // Getters. const string& path() const {return path_;}; const VcfHeader& header() const {return header_;}; size_t line_number() const {return line_number_;} // Parse the header. void read_header(); // Read a record. Returns false on EOF, true otherwise. bool next_record(VcfRecord& record); }; /* * VcfParser * ========= * Implements VcfAbstractParser for plain text files. */ class VcfParser : public VcfAbstractParser { ifstream file_; inline void getline(char* ptr, size_t n); void check_eol() {eol_ = ! file_.fail(); file_.clear();} public: VcfParser(const string& path) : VcfAbstractParser(path), file_(path) {} bool fail() {return file_.fail();} }; #ifdef HAVE_LIBZ /* * VcfGzParser * ========== * Implements VcfAbstractParser for gzipped files. */ class VcfGzParser : public VcfAbstractParser { gzFile file_; inline void getline(char* ptr, size_t n); inline void check_eol(); VcfGzParser(VcfGzParser& p) = delete; // No copy constructor. VcfGzParser& operator=(VcfGzParser& p) = delete; public: VcfGzParser(const string& path); ~VcfGzParser() {if(file_) gzclose(file_);} bool fail() {return file_ == NULL;} }; #endif // HAVE_LIBZ /* * VcfWriter * ========== */ class VcfWriter { private: const string path_; ofstream file_; public: VcfWriter(const string& path) : path_(path), file_(path) {} bool fail() {return file_.fail();} void write_header(const VcfHeader& h); void write_record(const VcfRecord& r, const VcfHeader& h); }; /* * Inline methods. * ========== */ inline void get_bounds(vector& bounds, char* tab1, char* tab2, char sep) { bounds.clear(); bounds.push_back(tab1); do { bounds.push_back(strchr(bounds.back()+1, sep)); // this is last+1 if tab1==tab2 } while (bounds.back()); bounds.pop_back(); bounds.push_back(tab2); } inline void VcfRecord::clear() { pos = -1; chrom.clear(); id.clear(); alleles.clear(); qual.clear(); filter.clear(); info.clear(); format.clear(); samples.clear(); type = Vcf::RType::null; } inline size_t VcfRecord::index_of_gt_subfield(const string& key) const { size_t i = 0; for (const string& f : format) { if (f == key) return i; ++i; } throw out_of_range(key); } inline string VcfRecord::parse_gt_subfield(const string& sample, size_t index) const { string subf; // Skip the first [index] colons. const char* first = sample.c_str(); for(size_t i=0; i VcfRecord::parse_genotype(const string& sample) const { pair genotype = {-1,-1}; if (format.empty() || format[0] != "GT" || sample.empty() || sample[0] == '.') { return genotype; } const char* first = sample.c_str(); const char* slash = strchr(first, '/'); if (slash == NULL) { slash = strchr(first, '|'); if (slash == NULL) { cerr << "Error: Malformed VCF genotype field '" << sample << "', at marker '" << chrom << ":" << pos << "'.\n"; throw exception(); } } if (*(slash+1) == '.') { static bool printed = false; if (not printed) { // Print the warning once. cerr << "Notice: Treating incomplete genotypes (e.g. '1/.') as missing.\n"; printed = true; } return genotype; } const char* colon = strchr(slash, ':'); try { genotype.first = std::stoi(string(first, slash)); genotype.second = std::stoi(colon==NULL ? string(slash+1) : string(slash+1, colon)); if (genotype.first < 0 || genotype.first >= int(alleles.size()) || genotype.second < 0 || genotype.second >= int(alleles.size())) throw exception(); } catch (exception& e) { cerr << "Error: Malformed VCF genotype '" << sample << "', at marker '" << chrom << ":" << pos << "'.\n"; throw e; } return genotype; } inline bool VcfRecord::is_snp() const { if (type != Vcf::RType::expl) return false; for (const string& a : alleles) if (a.length() > 1) return false; return true; } inline void VcfAbstractParser::read_to_eol() { while(!eol_) { getline(line_, Vcf::line_buf_size); if(eof_) { eol_=true; } else { tabs_.clear(); tabs_.push_back(line_+strlen(line_)); check_eol(); } } } inline void VcfAbstractParser::add_sample(VcfRecord& record, char* tab1, char* tab2) { if(tab2 == tab1+1) cerr << "Warning: malformed VCF record line (empty SAMPLE field should be marked by a dot)." << " Line " << line_number_ << " in file " << path_ << "'.\n"; record.samples.push_back(string(tab1+1, tab2)); ++sample_index_; } inline void VcfParser::getline(char* ptr, size_t n) { file_.getline(ptr, n); if(file_.eof()) { eof_=file_.fail(); if(!eof_) cerr << "Notice: File '" << path_ << "' does not end with a newline.\n"; } } #ifdef HAVE_LIBZ inline void VcfGzParser::getline(char* buf, size_t n) { if (gzgets(file_, buf, n) == NULL) eof_ = true; } inline void VcfGzParser::check_eol() { if(*(tabs_.back()-1) == '\n') { // rem. safe, gzgets returns NULL on EOF eol_ = true; tabs_.back() = tabs_.back()-1; *tabs_.back() = '\0'; } else { eol_= false; } } #endif // HAVE_LIBZ #endif // __VCF_H__ stacks-1.44/src/write.cc000644 000765 000024 00000017141 12764047443 015751 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2013-2015, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // write.cc -- common routines for writing FASTA/FASTQ records to a file.. // // Julian Catchen // jcatchen@uoregon.edu // University of Oregon // #include "write.h" int write_fasta(ofstream *fh, Read *href, bool overhang) { char tile[id_len]; sprintf(tile, "%04d", href->tile); int offset = href->inline_bc_len; offset += overhang ? 1 : 0; if (href->fastq_type != generic_fastq) *fh << ">" << href->lane << "_" << tile << "_" << href->x << "_" << href->y << "/" << href->read << "\n" << href->seq + offset << "\n"; else *fh << ">" << href->machine << "/" << href->read << "\n" << href->seq + offset << "\n"; if (fh->fail()) return -1; return 1; } int write_fasta(gzFile *fh, Read *href, bool overhang) { stringstream sstr; char tile[id_len]; sprintf(tile, "%04d", href->tile); int offset = href->inline_bc_len; offset += overhang ? 1 : 0; if (href->fastq_type != generic_fastq) sstr << ">" << href->lane << "_" << tile << "_" << href->x << "_" << href->y << "/" << href->read << "\n" << href->seq + offset << "\n"; else sstr << ">" << href->machine << "/" << href->read << "\n" << href->seq + offset << "\n"; int res = gzputs(*fh, sstr.str().c_str()); return res; } int write_fasta(ofstream *fh, Seq *href) { *fh << ">" << href->id << "\n" << href->seq << "\n"; if (fh->fail()) return -1; return 1; } int write_fasta(gzFile *fh, Seq *href) { stringstream sstr; sstr << ">" << href->id << "\n" << href->seq << "\n"; int res = gzputs(*fh, sstr.str().c_str()); return res; } int write_fastq(ofstream *fh, Read *href, bool overhang) { // // Write the sequence and quality scores in FASTQ format. // char tile[id_len]; sprintf(tile, "%04d", href->tile); int offset = href->inline_bc_len; offset += overhang ? 1 : 0; if (href->fastq_type != generic_fastq) *fh << "@" << href->lane << "_" << tile << "_" << href->x << "_" << href->y << "/" << href->read << "\n" << href->seq + offset << "\n" << "+\n" << href->phred + offset << "\n"; else *fh << "@" << href->machine << "/" << href->read << "\n" << href->seq + offset << "\n" << "+\n" << href->phred + offset << "\n"; if (fh->fail()) return -1; return 1; } int write_fastq(gzFile *fh, Read *href, bool overhang) { // // Write the sequence and quality scores in FASTQ format. // stringstream sstr; char tile[id_len]; sprintf(tile, "%04d", href->tile); int offset = href->inline_bc_len; offset += overhang ? 1 : 0; if (href->fastq_type != generic_fastq) sstr << "@" << href->lane << "_" << tile << "_" << href->x << "_" << href->y << "/" << href->read << "\n" << href->seq + offset << "\n" << "+\n" << href->phred + offset << "\n"; else sstr << "@" << href->machine << "/" << href->read << "\n" << href->seq + offset << "\n" << "+\n" << href->phred + offset << "\n"; int res = gzputs(*fh, sstr.str().c_str()); return res; } int write_fastq(ofstream *fh, Seq *href, int offset) { *fh << "@" << href->id << "\n" << href->seq + offset << "\n" << "+\n" << href->qual + offset << "\n"; if (fh->fail()) return -1; return 1; } int write_fastq(gzFile *fh, Seq *href, int offset) { stringstream sstr; sstr << "@" << href->id << "\n" << href->seq + offset << "\n" << "+\n" << href->qual + offset << "\n"; int res = gzputs(*fh, sstr.str().c_str()); return res; } int write_fasta(ofstream *fh, Seq *href, int offset) { *fh << ">" << href->id << "\n" << href->seq + offset << "\n"; if (fh->fail()) return -1; return 1; } int write_fasta(gzFile *fh, Seq *href, int offset) { stringstream sstr; sstr << ">" << href->id << "\n" << href->seq + offset << "\n"; int res = gzputs(*fh, sstr.str().c_str()); return res; } int write_fastq(ofstream *fh, Seq *href) { *fh << "@" << href->id << "\n" << href->seq << "\n" << "+\n" << href->qual << "\n"; if (fh->fail()) return -1; return 1; } int write_fastq(gzFile *fh, Seq *href) { stringstream sstr; sstr << "@" << href->id << "\n" << href->seq << "\n" << "+\n" << href->qual << "\n"; int res = gzputs(*fh, sstr.str().c_str()); return res; } int write_fastq(ofstream *fh, Seq *href, string msg) { *fh << "@" << href->id << "|" << msg << "\n" << href->seq << "\n" << "+\n" << href->qual << "\n"; if (fh->fail()) return -1; return 1; } int write_fastq(gzFile *fh, Seq *href, string msg) { stringstream sstr; sstr << "@" << href->id << "|" << msg << "\n" << href->seq << "\n" << "+\n" << href->qual << "\n"; int res = gzputs(*fh, sstr.str().c_str()); return res; } int write_fasta(ofstream *fh, Seq *href, string msg) { *fh << ">" << href->id << "|" << msg << "\n" << href->seq << "\n"; if (fh->fail()) return -1; return 1; } int write_fasta(gzFile *fh, Seq *href, string msg) { stringstream sstr; sstr << ">" << href->id << "|" << msg << "\n" << href->seq << "\n"; int res = gzputs(*fh, sstr.str().c_str()); return res; } int write_fasta(ofstream *fh, Seq *href, Read *r) { *fh << ">" << href->id << "\n" << r->seq + r->inline_bc_len << "\n"; if (fh->fail()) return -1; return 1; } int write_fasta(gzFile *fh, Seq *href, Read *r) { stringstream sstr; sstr << ">" << href->id << "\n" << r->seq + r->inline_bc_len << "\n"; int res = gzputs(*fh, sstr.str().c_str()); return res; } int write_fastq(ofstream *fh, Seq *href, Read *r) { *fh << "@" << href->id << "\n" << r->seq + r->inline_bc_len << "\n" << "+\n" << r->phred + r->inline_bc_len << "\n"; if (fh->fail()) return -1; return 1; } int write_fastq(gzFile *fh, Seq *href, Read *r) { stringstream sstr; sstr << "@" << href->id << "\n" << r->seq + r->inline_bc_len << "\n" << "+\n" << r->phred + r->inline_bc_len << "\n"; int res = gzputs(*fh, sstr.str().c_str()); return res; } stacks-1.44/src/write.h000644 000765 000024 00000003576 12764047443 015622 0ustar00catchenstaff000000 000000 // -*-mode:c++; c-style:k&r; c-basic-offset:4;-*- // // Copyright 2013, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // #ifndef __WRITE_H__ #define __WRITE_H__ #include using std::string; #include #include using std::ofstream; using std::cin; using std::cout; using std::cerr; using std::endl; #include #include #include "input.h" #include "clean.h" int write_fastq(ofstream *, Read *, bool); int write_fastq(ofstream *, Seq *); int write_fastq(ofstream *, Seq *, int); int write_fastq(ofstream *, Seq *, string); int write_fasta(ofstream *, Read *, bool); int write_fasta(ofstream *, Seq *); int write_fasta(ofstream *, Seq *, int); int write_fasta(ofstream *, Seq *, string); int write_fastq(ofstream *, Seq *, Read *); int write_fasta(ofstream *, Seq *, Read *); int write_fastq(gzFile *, Read *, bool); int write_fastq(gzFile *, Seq *); int write_fastq(gzFile *, Seq *, int); int write_fastq(gzFile *, Seq *, string); int write_fasta(gzFile *, Read *, bool); int write_fasta(gzFile *, Seq *); int write_fasta(gzFile *, Seq *, int); int write_fasta(gzFile *, Seq *, string); int write_fastq(gzFile *, Seq *, Read *); int write_fasta(gzFile *, Seq *, Read *); #endif // __WRITE_H__ stacks-1.44/sql/catalog_index.sql000644 000765 000024 00000003434 12441417455 017636 0ustar00catchenstaff000000 000000 create table catalog_index ( id int unsigned not null primary key auto_increment, batch_id int unsigned not null, cat_id int unsigned not null, tag_id int unsigned not null, snps int unsigned not null, parents int unsigned not null, progeny int unsigned not null, alleles int unsigned not null, marker enum('aa/bb', 'aa/b-', '-a/bb', 'ab/--', '--/ab', 'aa/ab', 'ab/aa', 'ab/a-', '-a/ab', 'ab/ab', 'ab/ac', 'ab/cd', 'ab/cc', 'cc/ab', 'ab/c-', '-c/ab', ''), uncor_marker enum('aa/bb', 'aa/b-', '-a/bb', 'ab/--', '--/ab', 'aa/ab', 'ab/aa', 'ab/a-', '-a/ab', 'ab/ab', 'ab/ac', 'ab/cd', 'ab/cc', 'cc/ab', 'ab/c-', '-c/ab', ''), valid_progeny int unsigned not null default 0, chisq_pval float not null default 1.0, lnl float not null default 0.0, ratio varchar(512), ests int unsigned not null, pe_radtags int unsigned not null, blast_hits int unsigned not null, geno_cnt int unsigned not null, chr varchar(32), bp int unsigned default 0, type enum('genomic', 'exon', 'intron'), ref_id int unsigned not null, INDEX batch_index (batch_id), INDEX tag_index (tag_id), INDEX snps_index (snps), INDEX parents_index (parents), INDEX progeny_index (progeny), INDEX allele_index (alleles), INDEX marker_index (marker), INDEX valid_index (valid_progeny), INDEX lnl_index (lnl), INDEX chisq_pval_index (chisq_pval), INDEX ests_index (ests), INDEX pe_rad_index (pe_radtags), INDEX hits_index (blast_hits), INDEX geno_index (geno_cnt), INDEX chr_index (chr), INDEX bp_index (bp), INDEX type_index (type) ); stacks-1.44/sql/chr_index.sql000644 000765 000024 00000000305 12335173442 016767 0ustar00catchenstaff000000 000000 create table chr_index ( id int unsigned not null primary key auto_increment, batch_id int unsigned not null, chr varchar(32), max_len int unsigned not null );stacks-1.44/sql/mysql.cnf.dist000644 000765 000024 00000000115 12533677757 017123 0ustar00catchenstaff000000 000000 [client] user=dbuser password=dbpass host=localhost port=3306 local-infile=1 stacks-1.44/sql/stacks.sql000644 000765 000024 00000026163 12704461610 016323 0ustar00catchenstaff000000 000000 SET default_storage_engine=MYISAM; create table batches ( id int unsigned not null primary key auto_increment, date DATE not null, description tinytext, type enum('map', 'population') ); create table samples ( id int unsigned not null primary key auto_increment, batch_id int unsigned not null, sample_id int unsigned not null, type enum('parent', 'progeny', 'sample'), file varchar(128), pop_id varchar(32), group_id varchar(32) ); create table populations ( id int unsigned not null primary key auto_increment, batch_id int unsigned not null, pop_id int unsigned not null, pop_name varchar(128) ); create table sumstats ( batch_id int unsigned not null, tag_id int unsigned not null, chr varchar(32), bp int unsigned default 0, col int unsigned default 0, pop_id varchar(32), p_nuc char, q_nuc char, n int unsigned default 0, p double, obs_het double, obs_hom double, exp_het double, exp_hom double, pi double, pi_s double, pi_s_pval double, fis double, fis_s double, fis_s_pval double, private int unsigned default 0, INDEX batch_id_index (batch_id), INDEX tag_id_index (tag_id) ); create table fst ( batch_id int unsigned not null, tag_id int unsigned not null, pop_id_1 varchar(32), pop_id_2 varchar(32), chr varchar(32), bp int unsigned default 0, col int unsigned default 0, pi_o double, fst double, fishers_p double, odds_ratio double, ci_low double, ci_high double, lod double, fst_c double, fst_s double, amova_fst double, amova_fst_c double, amova_fst_s double, amova_fst_s_pval double, snp_cnt int unsigned default 0, INDEX batch_id_index (batch_id), INDEX tag_id_index (tag_id) ); create table hapstats ( batch_id int unsigned not null, tag_id int unsigned not null, chr varchar(32), bp int unsigned default 0, pop_id varchar(32), n int unsigned default 0, hapcnt double, gene_div double, gene_div_s double, gene_div_pval double, hap_div double, hap_div_s double, hap_div_pval double, haplotypes tinytext, INDEX batch_id_index (batch_id), INDEX tag_id_index (tag_id) ); create table phist ( batch_id int unsigned not null, tag_id int unsigned not null, pop_id_1 varchar(32), pop_id_2 varchar(32), chr varchar(32), bp int unsigned default 0, phist double, phist_s double, phist_pval double, fpst double, fpst_s double, fpst_pval double, dest double, dest_s double, dest_pval double, INDEX batch_id_index (batch_id), INDEX tag_id_index (tag_id) ); create table catalog_tags ( id int unsigned not null primary key auto_increment, batch_id int unsigned not null, tag_id int unsigned not null, chr varchar(32), bp int unsigned default 0, strand enum('+', '-'), relationship enum('consensus', 'primary', 'secondary', 'tertiary'), sub_id int unsigned not null, merge_type tinytext, seq text, INDEX batch_id_index (batch_id), INDEX tag_id_index (tag_id) ); create table catalog_snps ( id int unsigned not null primary key auto_increment, batch_id int unsigned not null, tag_id int unsigned not null, col int unsigned not null, type enum('E', 'O', 'U'), lratio float, rank_1 char(1), rank_2 char(1), rank_3 char(1), rank_4 char(1), INDEX batch_index (batch_id), INDEX tag_index (tag_id) ); create table catalog_alleles ( id int unsigned not null primary key auto_increment, batch_id int unsigned not null, tag_id int unsigned not null, allele varchar(32), read_pct float, read_cnt int unsigned, INDEX batch_index (batch_id), INDEX tag_index (tag_id) ); create table catalog_genotypes ( id int unsigned not null primary key auto_increment, batch_id int unsigned not null, catalog_id int unsigned not null, sample_id int unsigned not null, genotype char(2), INDEX batch_index (batch_id), INDEX catalog_index (catalog_id), INDEX sample_index (sample_id) ); create table genotype_corrections ( id int unsigned not null primary key auto_increment, batch_id int unsigned not null, catalog_id int unsigned not null, sample_id int unsigned not null, genotype char(2), INDEX cat_index (catalog_id), INDEX batch_index (batch_id), INDEX sample_index (sample_id) ); create table catalog_annotations ( id int unsigned not null primary key auto_increment, batch_id int unsigned not null, catalog_id int unsigned not null, external_id varchar(64), INDEX batch_index (batch_id), INDEX catalog_index (catalog_id), INDEX external_index (external_id) ); create table pileup ( id int unsigned not null primary key auto_increment, sample_id int unsigned not null, tag_id int unsigned not null, chr varchar(32), bp int unsigned default 0, relationship enum('consensus', 'model', 'primary', 'secondary'), sub_id int unsigned not null, seq_id varchar(32), seq text, INDEX tag_id_index (tag_id), INDEX sample_id_index (sample_id), INDEX rel_index (relationship) ); create table unique_tags ( id int unsigned not null primary key auto_increment, sample_id int unsigned not null, tag_id int unsigned not null, chr varchar(32), bp int unsigned default 0, strand enum('+', '-'), relationship enum('consensus', 'model', 'primary', 'secondary'), sub_id int unsigned not null, seq_id varchar(32), seq text, deleveraged bool default false, blacklisted bool default false, removed bool default false, lnl float, INDEX tag_id_index (tag_id), INDEX sample_id_index (sample_id), INDEX rel_index (relationship) ); create table snps ( id int unsigned not null primary key auto_increment, sample_id int unsigned not null, tag_id int unsigned not null, col int unsigned not null, type enum('E', 'O', 'U'), lratio float, rank_1 char(1), rank_2 char(1), rank_3 char(1), rank_4 char(1), INDEX samp_index (sample_id), INDEX tag_index (tag_id) ); create table alleles ( id int unsigned not null primary key auto_increment, sample_id int unsigned not null, tag_id int unsigned not null, allele varchar(32), read_pct float, read_cnt int unsigned, INDEX samp_index (sample_id), INDEX tag_index (tag_id) ); create table matches ( id int unsigned not null primary key auto_increment, batch_id int unsigned not null, catalog_id int unsigned not null, sample_id int unsigned not null, tag_id int unsigned not null, allele varchar(256), depth int unsigned not null, lnl float, cigar varchar(256), INDEX batch_id_index (batch_id), INDEX catalog_id_index (catalog_id), INDEX sample_id_index (sample_id), INDEX tag_id_index (tag_id) ); create table markers ( id int unsigned not null primary key auto_increment, batch_id int unsigned not null, catalog_id int unsigned not null, type enum('aa/bb', 'aa/b-', '-a/bb', 'ab/--', '--/ab', 'aa/ab', 'ab/aa', 'ab/a-', '-a/ab', 'ab/ab', 'ab/ac', 'ab/cd', 'ab/cc', 'cc/ab', 'ab/c-', '-c/ab'), progeny int unsigned not null default 0, max_pct float, ratio varchar(512), chisq_pval float, lnl float, geno_map text, uncor_type enum('aa/bb', 'aa/b-', '-a/bb', 'ab/--', '--/ab', 'aa/ab', 'ab/aa', 'ab/a-', '-a/ab', 'ab/ab', 'ab/ac', 'ab/cd', 'ab/cc', 'cc/ab', 'ab/c-', '-c/ab') ); create table sequence ( id int unsigned not null primary key auto_increment, batch_id int unsigned not null, catalog_id int unsigned not null, type enum('pe_radtag', 'est'), seq_id varchar(64), seq text, INDEX catalog_id_index (catalog_id) ); create table sequence_blast ( id int unsigned not null primary key auto_increment, batch_id int unsigned not null default 0, catalog_id int unsigned not null default 0, seq_id int unsigned not null default 0, algorithm enum('blastn', 'blastp', 'blastx', 'tblastn', 'tblastx'), query_id varchar(64), query_len int unsigned not null default 0, hit_id varchar(128), hit_len int unsigned not null default 0, score double, e_value double, percent_ident double, hsp_rank int unsigned not null default 0, aln_len int unsigned not null default 0, aln_homology_str text, query_aln text, query_aln_start int unsigned not null default 0, query_aln_end int unsigned not null default 0, hit_aln text, hit_aln_start int unsigned not null default 0, hit_aln_end int unsigned not null default 0 ); create table chr_index ( id int unsigned not null primary key auto_increment, batch_id int unsigned not null, chr varchar(32), max_len int unsigned not null ); create table ref_radome ( id int unsigned not null primary key, chr varchar(32), bp int unsigned not null, strand enum('1', '-1'), type enum('genomic', 'exon', 'intron'), gene varchar(32), ext_id varchar(32), exon varchar(32), ex_start int unsigned not null, ex_end int unsigned not null, ex_index int unsigned not null, tran_cnt int unsigned not null, dist int unsigned not null, uniq bool default false, tran varchar(32) ); stacks-1.44/sql/tag_index.sql000644 000765 000024 00000001547 12335173442 016777 0ustar00catchenstaff000000 000000 create table tag_index ( id int unsigned not null primary key auto_increment, batch_id int unsigned not null, sample_id int unsigned not null, tag_id int unsigned not null, con_tag_id int unsigned not null, depth int unsigned not null, snps int unsigned not null, catalog_id int unsigned not null, deleveraged bool default false, blacklisted bool default false, removed bool default false, INDEX batch_index (batch_id), INDEX sample_index (sample_id), INDEX tag_index (tag_id), INDEX con_tag_index (con_tag_id), INDEX depth_index (depth), INDEX snps_index (snps), INDEX delv_index (deleveraged), INDEX black_index (blacklisted), INDEX rem_index (removed), INDEX catalog_index (catalog_id) ); stacks-1.44/scripts/convert_stacks.pl000755 000765 000024 00000015222 12441417455 020572 0ustar00catchenstaff000000 000000 #!/usr/bin/perl # # Copyright 2014, Julian Catchen # # This file is part of Stacks. # # Stacks is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation, either version 3 of the License, or # (at your option) any later version. # # Stacks is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with Stacks. If not, see . # use strict; use constant stacks_version => "_VERSION_"; use constant true => 1; use constant false => 0; my $tags_midpt = 11; my $tags_len = 13; my $snps_midpt = 3; my $snps_len = 9; my $debug = 0; my $in_path = ""; my $out_path = ""; my $gzipped = false; parse_command_line(); my (@files, @catalog_files); build_file_list(\@files, \@catalog_files); my ($file, $num_files, $i, $key); $num_files = scalar(@catalog_files); $i = 1; foreach $file (@catalog_files) { printf(STDERR "Converting catalog files, file % 2s of % 2s [%s]\n", $i, $num_files, $file); convert_tags_file($in_path, $out_path, $file . ".catalog"); convert_snps_file($in_path, $out_path, $file . ".catalog"); # # Just copy the unchanged *.alleles.tsv files. # if ($gzipped == true) { `cp $in_path/${file}.catalog.alleles.tsv.gz $out_path/.`; } else { `cp $in_path/${file}.catalog.alleles.tsv $out_path/.`; } $i++; } $num_files = scalar(@files); $i = 1; foreach $file (@files) { printf(STDERR "Converting sample files, file % 2s of % 2s [%s]\n", $i, $num_files, $file); convert_tags_file($in_path, $out_path, $file); convert_snps_file($in_path, $out_path, $file); convert_matches_file($in_path, $out_path, $file); # # Just copy the unchanged *.alleles.tsv files. # if ($gzipped == true) { `cp $in_path/${file}.alleles.tsv.gz $out_path/.`; } else { `cp $in_path/${file}.alleles.tsv $out_path/.`; } $i++; } sub convert_matches_file { my ($in_path, $out_path, $file) = @_; my ($path, $in_fh, $out_fh, $line, @parts); $path = $in_path . "/" . $file . ".matches.tsv"; if ($gzipped) { $path .= ".gz"; open($in_fh, "gunzip -c $path |") or die("Unable to open matches file '$path', $!\n"); } else { open($in_fh, "<$path") or die("Unable to open matches file '$path', $!\n"); } $path = $out_path . "/" . $file . ".matches.tsv"; open($out_fh, ">$path") or die("Unable to open tags output file '$path', $!\n"); while ($line = <$in_fh>) { chomp $line; @parts = split(/\t/, $line); print $out_fh join("\t", @parts), "\t", "0.0\n"; # Missing column } close($in_fh); close($out_fh); `gzip -f $path` if ($gzipped); } sub convert_tags_file { my ($in_path, $out_path, $file) = @_; my ($path, $in_fh, $out_fh, $line, @parts); $path = $in_path . "/" . $file . ".tags.tsv"; if ($gzipped) { $path .= ".gz"; open($in_fh, "gunzip -c $path |") or die("Unable to open tags file '$path', $!\n"); } else { open($in_fh, "<$path") or die("Unable to open tags file '$path', $!\n"); } $path = $out_path . "/" . $file . ".tags.tsv"; open($out_fh, ">$path") or die("Unable to open tags output file '$path', $!\n"); while ($line = <$in_fh>) { chomp $line; if ($parts[6] eq "consensus") { print $out_fh $line, "\t", "0.0\n"; # Missing column } else { print $out_fh $line, "\t", "\n", # Missing column } } close($in_fh); close($out_fh); `gzip -f $path` if ($gzipped); } sub convert_snps_file { my ($in_path, $out_path, $file) = @_; my ($path, $in_fh, $out_fh, $line, @parts); $path = $in_path . "/" . $file . ".snps.tsv"; if ($gzipped) { $path .= ".gz"; open($in_fh, "gunzip -c $path |") or die("Unable to open tags file '$path', $!\n"); } else { open($in_fh, "<$path") or die("Unable to open tags file '$path', $!\n"); } $path = $out_path . "/" . $file . ".snps.tsv"; open($out_fh, ">$path") or die("Unable to open tags output file '$path', $!\n"); while ($line = <$in_fh>) { chomp $line; @parts = split(/\t/, $line); print $out_fh join("\t", @parts[0 .. $snps_midpt]), "\t", "E\t", # Missing column join("\t", @parts[$snps_midpt+1 .. $snps_len-1]), "\n"; } close($in_fh); close($out_fh); `gzip -f $path` if ($gzipped); } sub build_file_list { my ($files, $catalog_files) = @_; my (@wl, @ls, $line, $prefix); @ls = `ls -1 $in_path/*.tags.tsv* 2> /dev/null`; if (scalar(@ls) == 0) { print STDERR "Unable to locate any input files to process within '$in_path'\n"; usage(); } foreach $line (@ls) { chomp $line; if ($line =~ /\.tags\.tsv\.gz$/) { $gzipped = true; ($prefix) = ($line =~ /$in_path\/(.+)\.tags\.tsv\.gz/); } else { ($prefix) = ($line =~ /$in_path\/(.+)\.tags\.tsv/); } next if ($prefix =~ /catalog/); push(@{$files}, $prefix); } @ls = `ls -1 $in_path/*.catalog.tags.tsv* 2> /dev/null`; if (scalar(@ls) == 0) { print STDERR "Unable to locate any catalog input files to process within '$in_path'\n"; usage(); } foreach $line (@ls) { chomp $line; if ($line =~ /\.catalog\.tags\.tsv\.gz$/) { $gzipped = true; ($prefix) = ($line =~ /$in_path\/(.+)\.catalog\.tags\.tsv\.gz/); } else { ($prefix) = ($line =~ /$in_path\/(.+)\.catalog\.tags\.tsv/); } push(@{$catalog_files}, $prefix); } } sub parse_command_line { while (@ARGV) { $_ = shift @ARGV; if ($_ =~ /^-p$/) { $in_path = shift @ARGV; } elsif ($_ =~ /^-o$/) { $out_path = shift @ARGV; } elsif ($_ =~ /^-d$/) { $debug++; } elsif ($_ =~ /^-v$/) { version(); exit(); } elsif ($_ =~ /^-h$/) { usage(); } else { print STDERR "Unknown command line option: '$_'\n"; usage(); } } $in_path = substr($in_path, 0, -1) if (substr($in_path, -1) eq "/"); $out_path = substr($out_path, 0, -1) if (substr($out_path, -1) eq "/"); if ($in_path eq $out_path) { print STDERR "Input and output paths cannot be the same.\n"; usage(); } } sub version { print STDERR "convert_stacks.pl ", stacks_version, "\n"; } sub usage { version(); print STDERR < # # This file is part of Stacks. # # Stacks is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation, either version 3 of the License, or # (at your option) any later version. # # Stacks is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with Stacks. If not, see . # # # Process the data for a genetic map: build stacks in parents and progeny, # create a catalog from the parents, and match progeny against the catatlog. # Call genotypes, and load all data into an MySQL database along the way. # # For the database interactions to work, the 'mysql' program is expected to be # on the path and sufficient permissions set to access the specified database. # use strict; use POSIX; use File::Temp qw/ mktemp /; use File::Spec; use constant stacks_version => "_VERSION_"; use constant true => 1; use constant false => 0; my $dry_run = false; my $sql = true; my $create_db = false; my $overw_db = false; my $gapped_alns = false; my $mysql_config = "_PKGDATADIR_" . "sql/mysql.cnf"; my $mysql_tables = "_PKGDATADIR_" . "sql/stacks.sql"; my $exe_path = "_BINDIR_"; my $out_path = ""; my $popmap_path = ""; my $sample_path = ""; my $db = ""; my $data_type = "map"; my $min_cov = 0; my $min_rcov = 0; my $batch_id = -1; my $sample_id = 1; my $desc = ""; # Database description of this dataset my $date = ""; # Date relevent to this data, formatted for SQL: 2009-05-31 my $gzip = false; my @parents; my @progeny; my @samples; my (@_ustacks, @_cstacks, @_sstacks, @_genotypes, @_populations); my $cmd_str = $0 . " " . join(" ", @ARGV); parse_command_line(); my $cnf = (-e $ENV{"HOME"} . "/.my.cnf") ? $ENV{"HOME"} . "/.my.cnf" : $mysql_config; # # Check for the existence of the necessary pipeline programs # die ("Unable to find '" . $exe_path . "ustacks'.\n") if (!-e $exe_path . "ustacks" || !-x $exe_path . "ustacks"); die ("Unable to find '" . $exe_path . "cstacks'.\n") if (!-e $exe_path . "cstacks" || !-x $exe_path . "cstacks"); die ("Unable to find '" . $exe_path . "sstacks'.\n") if (!-e $exe_path . "sstacks" || !-x $exe_path . "sstacks"); die ("Unable to find '" . $exe_path . "genotypes'.\n") if (!-e $exe_path . "genotypes" || !-x $exe_path . "genotypes"); die ("Unable to find '" . $exe_path . "populations'.\n") if (!-e $exe_path . "populations" || !-x $exe_path . "populations"); die ("Unable to find '" . $exe_path . "index_radtags.pl'.\n") if (!-e $exe_path . "index_radtags.pl" || !-x $exe_path . "index_radtags.pl"); my ($log, $log_fh, $sample); my (@sample_list, %pop_ids, %pops, %grp_ids, %grps, %sample_ids); parse_population_map(\@sample_list, \%pop_ids, \%pops, \%grp_ids, \%grps); initialize_samples(\@parents, \@progeny, \@samples, \@sample_list, \%pop_ids, \%grp_ids); # # Open the log file # $log = "$out_path/denovo_map.log"; open($log_fh, ">$log") or die("Unable to open log file '$log'; $!\n"); print $log_fh "denovo_map.pl version ", stacks_version, " started at ", strftime("%Y-%m-%d %H:%M:%S", (localtime(time))), "\n", $cmd_str, "\n\n"; initialize_database($log_fh, \@parents, \@progeny, \@samples, \%sample_ids) if ($sql == true); execute_stacks($log_fh, $sample_id, \@parents, \@progeny, \@samples, \%sample_ids); load_sql_data($log_fh, \%pops, \@parents, \@progeny, \@samples) if ($sql == true); print $log_fh "\ndenovo_map.pl completed at ", strftime("%Y-%m-%d %H:%M:%S", (localtime(time))), "\n"; close($log_fh); sub execute_stacks { my ($log_fh, $sample_id, $parents, $progeny, $samples, $sample_ids) = @_; my (@results, @depths_of_cov); my ($pop_cnt, $sample, $num_files, $i, $cmd, $pipe_fh, $path, $cat_file); my $minc = $min_cov > 0 ? "-m $min_cov" : ""; my $minrc = $min_rcov > 0 ? "-m $min_rcov" : $minc; $i = 1; $num_files = scalar(@{$parents}) + scalar(@{$progeny}) + scalar(@{$samples}); # # Assemble RAD loci in each individual. # foreach $sample (@parents, @progeny, @samples) { printf("Identifying unique stacks; file % 3s of % 3s [%s]\n", $i, $num_files, $sample->{'file'}); printf($log_fh "Identifying unique stacks; file % 3s of % 3s [%s]\n", $i, $num_files, $sample->{'file'}); if (scalar(keys %{$sample_ids}) > 0) { $sample_id = $sample_ids->{$sample->{'file'}}; } $path = $sample->{'path'} . $sample->{'file'} . "." . $sample->{'suffix'}; if ($sample->{'type'} eq "sample") { $cmd = $exe_path . "ustacks -t $sample->{'fmt'} -f $path -o $out_path -i $sample_id -r $minc " . join(" ", @_ustacks) . " 2>&1"; } elsif ($sample->{'type'} eq "parent") { $cmd = $exe_path . "ustacks -t $sample->{'fmt'} -f $path -o $out_path -i $sample_id -r $minc " . join(" ", @_ustacks) . " 2>&1"; } elsif ($sample->{'type'} eq "progeny") { $cmd = $exe_path . "ustacks -t $sample->{'fmt'} -f $path -o $out_path -i $sample_id -r $minrc " . join(" ", @_ustacks) . " 2>&1"; } print STDERR " $cmd\n"; print $log_fh "$cmd\n"; if ($dry_run == false) { @results = `$cmd`; # # Pull the depth of coverage from ustacks. # my $depth; if ($gapped_alns) { my @lines = grep(/^After gapped alignments, coverage depth Mean/, @results); ($depth) = ($lines[0] =~ /^After gapped alignments, coverage depth Mean: (\d+\.?\d*); Std Dev: .+; Max: .+$/); } else { my @lines = grep(/^After remainders merged, coverage depth Mean/, @results); ($depth) = ($lines[0] =~ /^After remainders merged, coverage depth Mean: (\d+\.?\d*); Std Dev: .+; Max: .+$/); } push(@depths_of_cov, [$sample->{'file'}, $depth]); } write_results(\@results, $log_fh); $i++; $sample_id++; } write_depths_of_cov(\@depths_of_cov, $log_fh); # # Generate catalog of RAD loci. # print STDERR "Generating catalog...\n"; my $file_paths = ""; foreach $sample (@parents, @samples) { $file_paths .= "-s $out_path/$sample->{'file'} "; } $cmd = $exe_path . "cstacks -b $batch_id -o $out_path $file_paths " . join(" ", @_cstacks) . " 2>&1"; print STDERR " $cmd\n"; print $log_fh "$cmd\n\n"; if ($dry_run == false) { open($pipe_fh, "$cmd |"); while (<$pipe_fh>) { print $log_fh $_; if ($_ =~ /failed/i) { print STDERR "Catalog construction failed.\n"; exit(1); } } close($pipe_fh); } # # Match parents, progeny, or samples to the catalog. # $file_paths = ""; print STDERR "Matching samples to the catalog...\n"; foreach $sample (@parents, @progeny, @samples) { $file_paths .= "-s $out_path/$sample->{'file'} "; } $cat_file = "batch_" . $batch_id; $cmd = $exe_path . "sstacks -b $batch_id -c $out_path/$cat_file -o $out_path $file_paths " . join(" ", @_sstacks) . " 2>&1"; print STDERR " $cmd\n"; print $log_fh "$cmd\n\n"; if ($dry_run == false) { open($pipe_fh, "$cmd |"); while (<$pipe_fh>) { print $log_fh $_; } close($pipe_fh); } if ($data_type eq "map") { # # Generate a set of observed haplotypes and a set of markers and generic genotypes # printf(STDERR "Generating genotypes...\n"); $cmd = $exe_path . "genotypes -b $batch_id -P $out_path -r 1 -c -s " . join(" ", @_genotypes) . " 2>&1"; print STDERR "$cmd\n"; print $log_fh "$cmd\n"; if ($dry_run == 0) { open($pipe_fh, "$cmd |"); while (<$pipe_fh>) { print $log_fh $_; } close($pipe_fh); } } else { printf(STDERR "Calculating population-level summary statistics\n"); $cmd = $exe_path . "populations -b $batch_id -P $out_path -s " . join(" ", @_populations) . " 2>&1"; print STDERR "$cmd\n"; print $log_fh "$cmd\n"; if ($dry_run == 0) { open($pipe_fh, "$cmd |"); while (<$pipe_fh>) { print $log_fh $_; } close($pipe_fh); } } } sub parse_population_map { my ($sample_list, $pop_ids, $pops, $grp_ids, $grps) = @_; my ($fh, @parts, $line, $sample); return if (length($popmap_path) == 0); open($fh, "<$popmap_path") or die("Unable to open population map, '$popmap_path', $!\n"); while ($line = <$fh>) { chomp $line; next if ($line =~ /^\s*#/); @parts = split(/\t/, $line); if (scalar(@parts) > 3) { die("Unable to parse population map, '$popmap_path' (map should contain no more than three columns).\n"); } push(@{$sample_list}, $parts[0]); $pop_ids->{$parts[0]} = $parts[1]; $pops->{$parts[1]}++; if (scalar(@parts) > 2) { $grp_ids->{$parts[0]} = $parts[2]; $grps->{$parts[2]}++; } } if (scalar(keys %{$grps}) == 0) { $grps->{"1"}++; foreach $sample (@{$sample_list}) { $grp_ids->{$sample} = "1"; } } print STDERR "Parsed population map: ", scalar(@{$sample_list}), " files in ", scalar(keys %{$pops}); scalar(keys %{$pops}) == 1 ? print STDERR " population" : print STDERR " populations"; print STDERR " and ", scalar(keys %{$grps}); scalar(keys %{$grps}) == 1 ? print STDERR " group.\n" : print STDERR " groups.\n"; close($fh); } sub initialize_samples { my ($parents, $progeny, $samples, $sample_list, $pop_ids, $grp_ids) = @_; my ($local_gzip, $file, $prefix, $suffix, $path, $found, $i); if (scalar(@{$sample_list}) > 0 && scalar(@{$samples}) == 0) { my @suffixes = ("fq", "fastq", "fq.gz", "fastq.gz", "fa", "fasta", "fa.gz", "fasta.gz"); my @fmts = ("fastq", "fastq", "gzfastq", "gzfastq", "fasta", "fasta", "gzfasta", "gzfasta"); # # If a population map was specified and no samples were provided on the command line. # foreach $sample (@{$sample_list}) { $found = false; for ($i = 0; $i < scalar(@suffixes); $i++) { $path = $sample_path . $sample . "." . $suffixes[$i]; if (-e $path) { if ($i == 2 || $i == 3 || $i == 6 || $i == 7) { $gzip = true; } push(@{$samples}, {'path' => $sample_path, 'file' => $sample, 'suffix' => $suffixes[$i], 'type' => "sample", 'fmt' => $fmts[$i]}); $found = true; last; } } if ($found == false) { die("Unable to find sample '$sample' in directory '$sample_path' as specified in the population map, '$popmap_path'.\n"); } } } else { # # Process any samples that were specified on the command line. # foreach $sample (@{$parents}, @{$progeny}, @{$samples}) { $local_gzip = false; ($prefix, $suffix) = ($sample->{'path'} =~ /^(.+)\.(.+)$/); if ($suffix eq "gz") { $gzip = true; $local_gzip = true; ($prefix, $suffix) = ($prefix =~ /^(.+)\.(.+)$/); } $sample->{'suffix'} = $suffix; $sample->{'suffix'} .= ".gz" if ($local_gzip == true); if ($prefix =~ /^.*\/.+$/) { ($path, $file) = ($prefix =~ /^(.*\/)(.+)$/); } else { $file = $prefix; $path = ""; } $sample->{'path'} = $path; $sample->{'file'} = $file; if ($local_gzip == true) { if ($suffix =~ /^fa$/ || $suffix =~ /^fasta$/) { $sample->{'fmt'} = "gzfasta"; } elsif ($suffix =~ /^fq$/ || $suffix =~ /^fastq$/) { $sample->{'fmt'} = "gzfastq"; } else { die("Unknown input file type for file '" . $sample->{'path'} . "'.\n"); } } else { if ($suffix =~ /^fa$/ || $suffix =~ /^fasta$/) { $sample->{'fmt'} = "fasta"; } elsif ($suffix =~ /^fq$/ || $suffix =~ /^fastq$/) { $sample->{'fmt'} = "fastq"; } else { die("Unknown input file type for file '" . $sample->{'path'} . "'.\n"); } } $path = $sample->{'path'} . $sample->{'file'} . "." . $sample->{'suffix'}; if (!-e $path) { die("Unable to locate sample file '$path'\n"); } } foreach $sample (@{$parents}) { $sample->{'type'} = "parent"; } foreach $sample (@{$progeny}) { $sample->{'type'} = "progeny"; } foreach $sample (@{$samples}) { $sample->{'type'} = "sample"; } } # # If a population map was specified, make sure all samples in the list were found (and vice versa) and assign popualtion IDs. # if (scalar(@{$sample_list}) > 0) { my %sample_hash; foreach $sample (@{$samples}) { $sample_hash{$sample->{'file'}}++; if (!defined($pop_ids->{$sample->{'file'}})) { die("Unable to find an entry for '" . $sample->{'file'} . "' in the population map, '$popmap_path'.\n"); } else { $sample->{'pop_id'} = $pop_ids->{$sample->{'file'}}; } if (!defined($grp_ids->{$sample->{'file'}})) { die("Unable to find an entry for '" . $sample->{'file'} . "' in the population map, '$popmap_path'.\n"); } else { $sample->{'grp_id'} = $grp_ids->{$sample->{'file'}}; } } foreach $sample (@{$sample_list}) { if (!defined($sample_hash{$sample})) { die("Unable to find a file corresponding to the population map entry '" . $sample . "' in the population map, '$popmap_path'.\n"); } } } else { foreach $sample (@{$parents}, @{$progeny}, @{$samples}) { $sample->{'pop_id'} = "1"; $sample->{'grp_id'} = "1"; $pop_ids->{$sample->{'file'}} = $sample->{'pop_id'}; $grp_ids->{$sample->{'file'}} = $sample->{'grp_id'}; } } # # Check that no duplicate files were specified. # my (%files, $file); foreach $file (@{$parents}, @{$progeny}, @{$samples}) { $files{$file}++; } foreach $file (keys %files) { if ($files{$file} > 1) { die("A duplicate file was specified which may create undefined results, '$file'\n"); } } print STDERR "Found ", scalar(@{$parents}), " parental file(s).\n" if (scalar(@{$parents}) > 0); print STDERR "Found ", scalar(@{$progeny}), " progeny file(s).\n" if (scalar(@{$progeny}) > 0); print STDERR "Found ", scalar(@{$samples}), " sample file(s).\n" if (scalar(@{$samples}) > 0); } sub initialize_database { my ($log_fh, $parents, $progeny, $samples, $sample_ids) = @_; my (@results, $sample_id, $sample); print $log_fh "Initializing the database...\n"; # # Create the database. # if ($create_db) { # # Check that the database doesn't already exist. # if ($dry_run == false) { @results = `mysql --defaults-file=$cnf -N -B -e "SHOW DATABASES LIKE '$db'"`; if (scalar(@results) > 0 && $overw_db == false) { die("Unable to create database '$db', it already exists.\n"); } } if ($overw_db == true) { `mysql --defaults-file=$cnf -N -B -e "DROP DATABASE IF EXISTS $db"` if ($dry_run == false); print $log_fh "mysql --defaults-file=$cnf -N -B -e \"DROP DATABASE IF EXISTS $db\"\n"; } `mysql --defaults-file=$cnf -e "CREATE DATABASE $db"` if ($dry_run == false); print $log_fh "mysql --defaults-file=$cnf $db -e \"CREATE DATABASE $db\"\n"; `mysql --defaults-file=$cnf $db < $mysql_tables` if ($dry_run == false); print $log_fh "mysql --defaults-file=$cnf $db < $mysql_tables\n"; } # # Set the SQL Batch ID for this set of loci, along with description and date of # sequencing. Insert this batch data into the database. # `mysql --defaults-file=$cnf $db -e "INSERT INTO batches SET id=$batch_id, description='$desc', date='$date', type='$data_type'"` if ($dry_run == false); print $log_fh "mysql --defaults-file=$cnf $db -e \"INSERT INTO batches SET id=$batch_id, description='$desc', date='$date', type='$data_type'\"\n"; print $log_fh "Loading sample data into the MySQL database...\n"; my $i = 1; foreach $sample (@{$parents}, @{$progeny}, @{$samples}) { if ($dry_run == false) { `mysql --defaults-file=$cnf $db -e "INSERT INTO samples SET sample_id=$i, batch_id=$batch_id, type='$sample->{'type'}', file='$sample->{'file'}', pop_id='$sample->{'pop_id'}', group_id='$sample->{'grp_id'}'"`; @results = `mysql --defaults-file=$cnf $db -N -B -e "SELECT id FROM samples WHERE sample_id=$i AND batch_id=$batch_id AND type='$sample->{'type'}' AND file='$sample->{'file'}'"`; chomp $results[0]; $sample_id = $results[0]; # # Save the sample ID to use when running ustacks. # $sample_ids->{$sample->{'file'}} = $sample_id; } print $log_fh "mysql --defaults-file=$cnf $db -e \"INSERT INTO samples SET sample_id=$i, batch_id=$batch_id, type='$sample->{'type'}', file='$sample->{'file'}', pop_id='$sample->{'pop_id'}', group_id='$sample->{'grp_id'}'\"\n"; $i++; } print $log_fh "\n"; } sub load_sql_data { my ($log_fh, $pops, $parents, $progeny, $samples) = @_; my ($pop_cnt, $sample, $num_files, $i, $file); print STDERR "\nComputation is complete, loading results to the database '$db'.\n"; my $pop_cnt = scalar(keys %{$pops}); $i = 1; $num_files = scalar(@{$parents}) + scalar(@{$progeny}) + scalar(@{$samples}); foreach $sample (@{$parents}, @{$progeny}, @{$samples}) { printf(STDERR "Loading ustacks output to $db; file % 3s of % 3s [%s]...", $i, $num_files, $sample->{'file'}); if ($gzip == true) { $file = "$out_path/$sample->{'file'}" . ".tags.tsv.gz"; import_gzsql_file($log_fh, $file, "unique_tags", 1); $file = "$out_path/$sample->{'file'}" . ".snps.tsv.gz"; import_gzsql_file($log_fh, $file, "snps", 1); $file = "$out_path/$sample->{'file'}" . ".alleles.tsv.gz"; import_gzsql_file($log_fh, $file, "alleles", 1); } else { $file = "$out_path/$sample->{'file'}" . ".tags.tsv"; import_sql_file($log_fh, $file, "unique_tags", 1); $file = "$out_path/$sample->{'file'}" . ".snps.tsv"; import_sql_file($log_fh, $file, "snps", 1); $file = "$out_path/$sample->{'file'}" . ".alleles.tsv"; import_sql_file($log_fh, $file, "alleles", 1); } print STDERR "done.\n"; $i++; } print STDERR "Importing catalog to $db..."; my $cat_file = "batch_" . $batch_id; if ($gzip == true) { $file = "$out_path/$cat_file" . ".catalog.tags.tsv.gz"; import_gzsql_file($log_fh, $file, "catalog_tags", 1); $file = "$out_path/$cat_file" . ".catalog.snps.tsv.gz"; import_gzsql_file($log_fh, $file, "catalog_snps", 1); $file = "$out_path/$cat_file" . ".catalog.alleles.tsv.gz"; import_gzsql_file($log_fh, $file, "catalog_alleles", 1); } else { $file = "$out_path/$cat_file" . ".catalog.tags.tsv"; import_sql_file($log_fh, $file, "catalog_tags", 1); $file = "$out_path/$cat_file" . ".catalog.snps.tsv"; import_sql_file($log_fh, $file, "catalog_snps", 1); $file = "$out_path/$cat_file" . ".catalog.alleles.tsv"; import_sql_file($log_fh, $file, "catalog_alleles", 1); } print STDERR "done.\n"; # # Load the sstacks results to the database if requested. # $i = 1; foreach $sample (@{$parents}, @{$progeny}, @{$samples}) { printf(STDERR "Loading sstacks output to $db; file % 3s of % 3s [%s]...", $i, $num_files, $sample->{'file'}); if ($gzip == true) { $file = "$out_path/" . $sample->{'file'} . ".matches.tsv.gz"; import_gzsql_file($log_fh, $file, "matches", 1); } else { $file = "$out_path/" . $sample->{'file'} . ".matches.tsv"; import_sql_file($log_fh, $file, "matches", 1); } print STDERR "done.\n"; $i++; } if ($data_type eq "map") { # # Load the outputs generated by genotypes to the database. # $file = "$out_path/batch_" . $batch_id . ".markers.tsv"; import_sql_file($log_fh, $file, "markers", 1); $file = "$out_path/batch_" . $batch_id . ".genotypes_1.txt"; import_sql_file($log_fh, $file, "catalog_genotypes", 1); } else { # # Load the outputs generated by populations to the database. # $file = "$out_path/batch_" . $batch_id . ".markers.tsv"; import_sql_file($log_fh, $file, "markers", 1); $file = "$out_path/batch_" . $batch_id . ".sumstats.tsv"; import_sql_file($log_fh, $file, "sumstats", $pop_cnt+1); $file = "$out_path/batch_" . $batch_id . ".hapstats.tsv"; import_sql_file($log_fh, $file, "hapstats", $pop_cnt+1); # # Import the Fst files. # my $fst_cnt = 0; my (@keys, $m, $n); @keys = sort keys %pops; for ($m = 0; $m < scalar(@keys); $m++) { for ($n = 0; $n < scalar(@keys); $n++) { $file = "$out_path/batch_" . $batch_id . ".fst_" . $keys[$m] . "-" . $keys[$n] . ".tsv"; if (-e $file) { import_sql_file($log_fh, $file, "fst", 1); $fst_cnt++; } } } print STDERR "Imported $fst_cnt Fst file(s).\n"; # # Import the Phi_st files. # $fst_cnt = 0; for ($m = 0; $m < scalar(@keys); $m++) { for ($n = 0; $n < scalar(@keys); $n++) { $file = "$out_path/batch_" . $batch_id . ".phistats_" . $keys[$m] . "-" . $keys[$n] . ".tsv"; if (-e $file) { import_sql_file($log_fh, $file, "phist", 3); $fst_cnt++; } } } print STDERR "Imported $fst_cnt Haplotype Fst file(s).\n"; } print $log_fh "\n"; # # Index the radtags database # my ($cmd, @results); print STDERR "Indexing the database...\n"; $cmd = $exe_path . "index_radtags.pl -D $db -t -c 2>&1"; print STDERR "$cmd\n"; print $log_fh "$cmd\n"; @results = `$cmd` if ($dry_run == false); print $log_fh @results; } sub write_results { my ($results, $log_fh) = @_; my $line; foreach $line (@{$results}) { if ($line =~ /\r/) { $line =~ s/^.+\r(.*\n)$/\1/; } print $log_fh $line; } } sub write_depths_of_cov { my ($depths, $log_fh) = @_; print STDERR "\nDepths of Coverage for Processed Samples:\n"; print $log_fh "\nDepths of Coverage for Processed Samples:\n"; foreach $a (@{$depths}) { print STDERR $a->[0], ": ", $a->[1], "x\n"; print $log_fh $a->[0], ": ", $a->[1], "x\n"; } print STDERR "\n"; print $log_fh "\n"; } sub import_sql_file { my ($log_fh, $file, $table, $skip_lines) = @_; my (@results, $ignore); $ignore = "IGNORE $skip_lines LINES" if ($skip_lines > 0); @results = `mysql --defaults-file=$cnf $db -e "LOAD DATA LOCAL INFILE '$file' INTO TABLE $table $ignore"` if ($sql == true && $dry_run == false); if ($sql == true) { print $log_fh "mysql --defaults-file=$cnf $db -e \"LOAD DATA LOCAL INFILE '$file' INTO TABLE $table $ignore\"\n", @results; } } sub import_gzsql_file { my ($log_fh, $file, $table, $skip_lines) = @_; my (@results, $ignore); $ignore = "IGNORE $skip_lines LINES" if ($skip_lines > 0); # # Get a temporary file name and create a named pipe. # my $tmpdir = File::Spec->tmpdir(); my $named_pipe = mktemp($tmpdir . "/denovo_map_XXXXXX"); if ($sql == true && $dry_run == false) { mkfifo($named_pipe, 0700) || die("Unable to create named pipe for loading gzipped data: $named_pipe, $!"); print $log_fh "Streaming $file into named pipe $named_pipe.\n"; } # # Dump our gzipped data onto the named pipe. # system("gunzip -c $file > $named_pipe &") if ($sql == true && $dry_run == false); @results = `mysql --defaults-file=$cnf $db -e "LOAD DATA LOCAL INFILE '$named_pipe' INTO TABLE $table $ignore"` if ($sql == true && $dry_run == false); if ($sql == true) { print $log_fh "mysql --defaults-file=$cnf $db -e \"LOAD DATA LOCAL INFILE '$named_pipe' INTO TABLE $table $ignore\"\n", @results; } # # Remove the pipe. # unlink($named_pipe) if ($sql == true && $dry_run == false); } sub parse_command_line { my ($arg); while (@ARGV) { $_ = shift @ARGV; if ($_ =~ /^-v$/) { version(); exit(); } elsif ($_ =~ /^-h$/) { usage(); } elsif ($_ =~ /^-p$/) { push(@parents, { 'path' => shift @ARGV }); } elsif ($_ =~ /^-r$/) { push(@progeny, { 'path' => shift @ARGV }); } elsif ($_ =~ /^-s$/) { push(@samples, { 'path' => shift @ARGV }); } elsif ($_ =~ /^-d$/) { $dry_run = true; } elsif ($_ =~ /^-o$/) { $out_path = shift @ARGV; } elsif ($_ =~ /^-D$/) { $desc = shift @ARGV; } elsif ($_ =~ /^-e$/) { $exe_path = shift @ARGV; } elsif ($_ =~ /^-b$/) { $batch_id = shift @ARGV; } elsif ($_ =~ /^-i$/) { $sample_id = shift @ARGV; } elsif ($_ =~ /^-a$/) { $date = shift @ARGV; } elsif ($_ =~ /^-S$/) { $sql = false; } elsif ($_ =~ /^-B$/) { $db = shift @ARGV; } elsif ($_ =~ /^-m$/) { $min_cov = shift @ARGV; } elsif ($_ =~ /^-P$/) { $min_rcov = shift @ARGV; } elsif ($_ =~ /^--samples$/) { $sample_path = shift @ARGV; } elsif ($_ =~ /^-O$/) { $popmap_path = shift @ARGV; push(@_populations, "-M " . $popmap_path); } elsif ($_ =~ /^--gapped$/) { $gapped_alns = true; push(@_ustacks, "--gapped "); push(@_cstacks, "--gapped "); push(@_sstacks, "--gapped "); } elsif ($_ =~ /^--create_db$/) { $create_db = true; } elsif ($_ =~ /^--overw_db$/) { $overw_db = true; $create_db = true; } elsif ($_ =~ /^-A$/) { $arg = shift @ARGV; push(@_genotypes, "-t " . $arg); $arg = lc($arg); if ($arg ne "gen" && $arg ne "cp" && $arg ne "f2" && $arg ne "bc1" && $arg ne "dh") { print STDERR "Unknown genetic mapping cross specified: '$arg'\n"; usage(); } } elsif ($_ =~ /^-t$/) { push(@_ustacks, "-d "); } elsif ($_ =~ /^-T$/) { $arg = shift @ARGV; push(@_ustacks, "-p " . $arg); push(@_cstacks, "-p " . $arg); push(@_sstacks, "-p " . $arg); push(@_populations, "-t " . $arg); } elsif ($_ =~ /^-M$/) { push(@_ustacks, "-M " . shift @ARGV); } elsif ($_ =~ /^-N$/) { push(@_ustacks, "-N " . shift @ARGV); } elsif ($_ =~ /^-n$/) { push(@_cstacks, "-n " . shift @ARGV); } elsif ($_ =~ /^-H$/) { push(@_ustacks, "-H "); } elsif ($_ =~ /^--bound_low$/) { push(@_ustacks, "--bound_low " . shift @ARGV); push(@_ustacks, "--model_type bounded"); } elsif ($_ =~ /^--bound_high$/) { push(@_ustacks, "--bound_high " . shift @ARGV); push(@_ustacks, "--model_type bounded"); } elsif ($_ =~ /^--alpha$/) { push(@_ustacks, "--alpha " . shift @ARGV); } elsif ($_ =~ /^-X$/) { # # Pass an arbitrary command-line option to a pipeline program. # # Command line option must be of the form '-X "program:option"' # $arg = shift @ARGV; my ($prog, $opt) = ($arg =~ /^(\w+):(.+)$/); if ($prog eq "ustacks") { push(@_ustacks, $opt); } elsif ($prog eq "cstacks") { push(@_cstacks, $opt); } elsif ($prog eq "sstacks") { push(@_sstacks, $opt); } elsif ($prog eq "genotypes") { push(@_genotypes, $opt); } elsif ($prog eq "populations") { push(@_populations, $opt); } else { print STDERR "Unknown pipeline program, '$arg'\n"; usage(); } } else { print STDERR "Unknown command line option: '$_'\n"; usage(); } } $exe_path = $exe_path . "/" if (substr($exe_path, -1) ne "/"); $out_path = substr($out_path, 0, -1) if (substr($out_path, -1) eq "/"); if ($batch_id !~ /^\d+$/ || $batch_id < 0) { print STDERR "You must specify a batch ID and it must be an integer (e.g. 1, 2, 3).\n"; usage(); } if ($sql == true && length($date) == 0) { $date = strftime("%Y-%m-%d", (localtime(time))); } if (scalar(@parents) > 0 && scalar(@samples) > 0) { print STDERR "You must specify either parent or sample files, but not both.\n"; usage(); } if (scalar(@parents) == 0 && scalar(@samples) == 0 && length($popmap_path) == 0) { print STDERR "You must specify at least one parent or sample file.\n"; usage(); } if (length($sample_path) > 0) { $sample_path .= "/" if (substr($sample_path, -1) ne "/"); } if (scalar(@samples) > 0 || length($popmap_path) > 0) { $data_type = "population"; } else { $data_type = "map"; } } sub version { print STDERR "denovo_map.pl ", stacks_version, "\n"; } sub usage { version(); print STDERR <: specify a path to the directory of samples (samples will be read from population map). Stack assembly options: m: specify a minimum number of identical, raw reads required to create a stack. M: specify the number of mismatches allowed between loci when processing a single individual (default 2). n: specify the number of mismatches allowed between loci when building the catalog (default 1). --gapped: perform gapped assemblies in ustacks, cstacks, and sstacks (default: off). Advanced (rarely used) options: P: specify a minimum number of identical, raw reads required to create a stack in 'progeny' individuals. N: specify the number of mismatches allowed when aligning secondary reads to primary stacks (default M+2). t: remove, or break up, highly repetitive RAD-Tags in the ustacks program. H: disable calling haplotypes from secondary reads. Database options: B: specify a database to load data into. D: a description of this batch to be stored in the database. S: disable recording SQL data in the database. i: starting sample_id, this is determined automatically if database interaction is enabled. --create_db: create the database specified by '-B' and populate the tables. --overw_db: delete the database before creating a new copy of it (turns on --create_db). SNP Model Options (these options are passed on to ustacks): --bound_low : lower bound for epsilon, the error rate, between 0 and 1.0 (default 0). --bound_high : upper bound for epsilon, the error rate, between 0 and 1.0 (default 1). --alpha : chi square significance level required to call a heterozygote or homozygote, either 0.1, 0.05 (default), 0.01, or 0.001. Arbitrary command line options: -X "program:option": pass a command line option to one of the pipeline components, e.g.'-X "ustacks:--max_locus_stacks 4"'. EOQ exit(0); } stacks-1.44/scripts/exec_velvet.pl000755 000765 000024 00000021334 12441417455 020054 0ustar00catchenstaff000000 000000 #!/usr/bin/env perl # # Copyright 2011, Julian Catchen # # This file is part of Stacks. # # Stacks is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation, either version 3 of the License, or # (at your option) any later version. # # Stacks is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with Stacks. If not, see . # use strict; use constant stacks_version => "_VERSION_"; my $debug = 0; my $paired = 0; my $amos = 0; my $ins_len_dist = 0; my $single_path = ""; my $paired_path = ""; my $sanger_path = ""; my $out_path = "."; my $white_list = ""; my $hash_len = 27; my $insert_len = 0; my $exp_cov = 0; my $cov_cut = 0.0; my $min_len = 0; my $read_trk = 0; my $clean = 1; my $collate = 0; my $exe_path = ""; my $velveth = "velveth"; my $velvetg = "velvetg"; parse_command_line(); if (length($exe_path) > 0) { $velveth = $exe_path . "/" . $velveth; $velvetg = $exe_path . "/" . $velvetg; } # # Test that we can execute the velvet programs # die ("Unable to find '" . $velveth . "'.\n") if (!-e $velveth || !-x $velveth); die ("Unable to find '" . $velvetg . "'.\n") if (!-e $velvetg || !-x $velvetg); my (@locus_files, $num_files, $file, $input_file, $output_file, $hres_file, $gres_file, $collate_fh); build_file_list(\@locus_files); $num_files = scalar(@locus_files); if ($collate) { open($collate_fh, ">$out_path/collated.fa") or die("Unable to open collate file, $!\n"); } my ($sing_data, $pair_data, $sang_data, $ins, $cov, $afg, $cut, $min, $read, $cln); $ins = $paired > 0 ? "-ins_length $insert_len" : "-ins_length auto"; $cov = $paired > 0 ? "-exp_cov $exp_cov" : "-exp_cov auto"; $cut = $cov_cut > 0 ? "-cov_cutoff $cov_cut" : "-cov_cutoff auto"; $read = $read_trk > 0 ? "-read_trkg yes" : ""; $cln = $clean > 0 ? "-very_clean yes" : ""; #$min = $min_len > 0 ? "-min_contig_lgth $min_len" : "" # # Write out the parameters for this assembly # open(PARAM, "> $out_path/velvet_parameters.txt") or die("Unable to open parameter file: $!\n"); print PARAM "Single-end Path: ", $single_path, "\n", "Paired-end Path: ", $paired_path, "\n", "Sanger Path: ", $sanger_path, "\n", "Hash Length: ", $hash_len, "\n", "Insert Length: ", $ins, "\n", "Coverage: ", $cov, "\n", "Coverage Cutoff: ", $cut, "\n", "Miniumum contig length: ", $min, "\n", "Read tracking: ", $read, "\n", "Very Clean: ", $cln, "\n"; close(PARAM); my $i = 1; foreach $file (@locus_files) { ($file) = ($file =~ /(.+)\.fa/); $output_file = $out_path . "/" . $file; $hres_file = $out_path . "/" . $file . "-h.output"; $gres_file = $out_path . "/" . $file . "-g.output"; $sing_data = length($single_path) > 0 ? '-short ' . $single_path . "/" . $file . ".fa" : ""; $pair_data = length($paired_path) > 0 ? '-shortPaired ' . $paired_path . "/" . $file . ".fa" : ""; $sang_data = length($sanger_path) > 0 ? '-long ' . $sanger_path . "/" . $file . ".fa" : ""; print STDERR "Assembling locus '$file'; run $i of $num_files \r"; # Execute velveth to build hash table, then velvetg to assemble print STDERR "$velveth $output_file $hash_len -fasta $sing_data $pair_data $sang_data &> $hres_file\n" if ($debug); `$velveth $output_file $hash_len -fasta $sing_data $pair_data $sang_data &> $hres_file`; print STDERR "$velvetg $output_file $ins $cov $cut $min $read $cln &> $gres_file\n" if ($debug); `$velvetg $output_file $ins $cov $cut $min $read $cln &> $gres_file`; collate_and_clean($out_path, $file, $collate_fh) if ($collate); $i++; } close($collate_fh) if ($collate); sub collate_and_clean { my ($out_path, $file, $collate_fh) = @_; my (@seqs, $seq); parse_fasta("$out_path/$file/contigs.fa", \@seqs); foreach $seq (@seqs) { next if (length($seq->{'seq'}) < $min_len); $seq->{'id'} = $file . "|" . $seq->{'id'}; print_fasta($collate_fh, $seq); } `rm $out_path/$file-g.output`; `rm $out_path/$file-h.output`; `rm -r $out_path/$file`; } sub parse_fasta { my ($file, $seqs) = @_; my ($fh, $line, $buf, $id, $seq); open($fh, "<$file") or die("Unable to open Velvet output file: $file, $!\n"); while ($line = <$fh>) { chomp $line; if (substr($line, 0, 1) eq ">") { if (length($buf) > 0) { $seq = {}; $seq->{'id'} = $id; $seq->{'seq'} = $buf; push(@{$seqs}, $seq); $buf = ""; } $id = substr($line, 1); } else { $buf .= $line; } } if (length($buf) > 0 && length($id) > 0) { $seq = {}; $seq->{'id'} = $id; $seq->{'seq'} = $buf; push(@{$seqs}, $seq); } close($fh); } sub print_fasta { my ($fh, $seq) = @_; my ($s); print $fh ">", $seq->{'id'}, "\n"; $s = $seq->{'seq'}; while (length($s) > 60) { print $fh substr($s, 0, 60), "\n"; $s = substr($s, 60); } print $fh $s, "\n" if (length($s) > 0); } sub build_file_list { my ($files) = @_; my (@ls, $line, $file, $path); # Load a white list of files to process if it is supplied. my @wl; if (length($white_list) > 0) { load_white_list(\@wl); } $path = length($paired_path) > 0 ? $paired_path : $single_path; @ls = `ls -1 $path/`; foreach $line (@ls) { chomp $line; next if (length($line) == 0); next if ($line !~ /.+\.fa$/ && $line !~ /.+\.fasta$/); ($file) = ($line =~ /^(.+\.fas?t?a?)/); if (scalar(@wl) > 0) { next if (!grep(/^$file$/, @wl)); } push(@{$files}, $file); } } sub load_white_list { my ($wl) = @_; open(WHITE, "<" . $white_list) or die("Unable to open white list file '$white_list': $!\n"); my $line = ""; while ($line = ) { chomp $line; next if (length($line) == 0); push(@{$wl}, $line); } close(WHITE); } sub parse_command_line { while (@ARGV) { $_ = shift @ARGV; if ($_ =~ /^-s$/) { $single_path = shift @ARGV; } elsif ($_ =~ /^-p$/) { $paired_path = shift @ARGV; } elsif ($_ =~ /^-l$/) { $sanger_path = shift @ARGV; } elsif ($_ =~ /^-o$/) { $out_path = shift @ARGV; } elsif ($_ =~ /^-c$/) { $collate++; } elsif ($_ =~ /^-W$/) { $white_list = shift @ARGV; } elsif ($_ =~ /^-I$/) { $insert_len = shift @ARGV; } elsif ($_ =~ /^-C$/) { $exp_cov = shift @ARGV; } elsif ($_ =~ /^-T$/) { $cov_cut = shift @ARGV; } elsif ($_ =~ /^-R$/) { $read_trk = shift @ARGV; } elsif ($_ =~ /^-M$/) { $min_len = shift @ARGV; } elsif ($_ =~ /^-H$/) { $hash_len = shift @ARGV; } elsif ($_ =~ /^-P$/) { $paired++; } elsif ($_ =~ /^-L$/) { $clean = 0; } elsif ($_ =~ /^-e$/) { $exe_path = shift @ARGV; } elsif ($_ =~ /^-v$/) { version(); exit(); } elsif ($_ =~ /^-d$/) { $debug++; } elsif ($_ =~ /^-h$/) { usage(); } else { print STDERR "Unknown command line option '$_'\n"; usage(); } } $single_path = substr($single_path, 0, -1) if (substr($single_path, -1) eq "/"); $paired_path = substr($paired_path, 0, -1) if (substr($paired_path, -1) eq "/"); $out_path = substr($out_path, 0, -1) if (substr($out_path, -1) eq "/"); $exe_path = substr($exe_path, 0, -1) if (substr($exe_path, -1) eq "/"); } sub version { print STDERR "exec_velvet.pl ", stacks_version, "\n"; } sub usage { version(); print STDERR < # # This file is part of Stacks. # # Stacks is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation, either version 3 of the License, or # (at your option) any later version. # # Stacks is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with Stacks. If not, see . # # # Read in a set of filtering paramters, query a Stacks pipeline database based on # those filters and write the results into a compact tab-separated values or excel file. # use strict; use DBI; #use Excel::Writer::XLSX; use Spreadsheet::WriteExcel; use constant stacks_version => "_VERSION_"; my $mysql_config = "_PKGDATADIR_" . "sql/mysql.cnf"; my $out_file = ""; my $type = "tsv"; my $batch_id = 0; my $data_type = "haplo"; my $map_type = "gen"; my $all_depth = 0; my $allele_depth_lim = 1; my $locus_depth_lim = 0; my $locus_lnl_lim = -10000.0; my $man_cor = 0; my $db = ""; my $translate_genotypes = {'dh' => \&trans_dh_map, 'cp' => \&trans_cp_map, 'bc1' => \&trans_bc1_map, 'f2' => \&trans_f2_map, 'gen' => \&trans_gen_map}; my @valid_filters = ("cata", "alle_l", "alle_u", "snps_l", "snps_u", "pare_l", "pare_u", "prog", "vprog", "mark", "est", "pe", "blast", "gcnt", "chisq_l", "chisq_u", "lnl_l", "lnl_u", "ref", "loc"); parse_command_line(); my (%sth, %loci, %samples, %depths, %filters); prepare_sql_handles(\%sth, \%filters); populate(\%sth, \%loci, \%samples, \%depths, \%filters); apply_corrected_genotypes(\%sth, \%loci) if ($man_cor > 0); if ($data_type eq "haplo") { write_observed_haplotypes(\%loci, \%samples); } elsif ($data_type eq "geno") { write_genotypes(\%loci, \%samples, \%depths); } print "Success\n"; sub populate { my ($sth, $loci, $samples, $depths, $filters) = @_; my (%delev); my ($row, $snp_row, $all_row, $gen_row, $locus); my @params; # # Cache all the stacks that were deleveraged. # $sth->{'delev'}->execute($batch_id); while ($row = $sth->{'delev'}->fetchrow_hashref()) { $delev{$row->{'sample_id'} . "_" . $row->{'tag_id'}}++; } # # Pull list of samples for this batch # $sth->{'samp'}->execute($batch_id); while ($row = $sth->{'samp'}->fetchrow_hashref()) { $samples->{$row->{'file'}} = $row->{'id'}; } prepare_filter_parameters(\@params, $filters); # # Fetch the results and populate the array of groups. # $sth->{'tag'}->execute(@params); while ($row = $sth->{'tag'}->fetchrow_hashref()) { $locus = {}; $locus->{'id'} = $row->{'tag_id'}; $locus->{'annotation'} = defined($row->{'external_id'}) ? $row->{'external_id'} : ""; $locus->{'chr'} = $row->{'chr'}; $locus->{'bp'} = $row->{'bp'}; $locus->{'delev'} = 0; $locus->{'marker'} = $row->{'marker'}; $locus->{'seq'} = $row->{'seq'}; $locus->{'alleles'} = ""; $locus->{'snps'} = ""; $locus->{'gtypes'} = {}; $locus->{'num_alleles'} = $row->{'alleles'}; $locus->{'num_snps'} = $row->{'snps'}; $locus->{'num_parents'} = $row->{'parents'}; $locus->{'num_progeny'} = $row->{'progeny'}; $locus->{'valid_progeny'} = $row->{'valid_progeny'}; $locus->{'num_ests'} = $row->{'ests'}; $locus->{'num_pe_tags'} = $row->{'pe_radtags'}; $locus->{'num_blast'} = $row->{'blast_hits'}; $locus->{'gcnt'} = $row->{'geno_cnt'}; $loci->{$row->{'tag_id'}} = $locus; } if ($data_type eq "haplo") { # # Add observed haplotypes # $sth->{'mat'}->execute($batch_id, $allele_depth_lim); while ($gen_row = $sth->{'mat'}->fetchrow_hashref()) { next if (!defined($loci->{$gen_row->{'catalog_id'}})); $locus = $loci->{$gen_row->{'catalog_id'}}; if (!defined($locus->{'gtypes'}->{$gen_row->{'file'}})) { $locus->{'gtypes'}->{$gen_row->{'file'}} = []; } push(@{$locus->{'gtypes'}->{$gen_row->{'file'}}}, {'file' => $gen_row->{'file'}, 'allele' => $gen_row->{'allele'}, 'tag_id' => $gen_row->{'tag_id'}, 'depth' => $gen_row->{'depth'}, 'lnl' => $gen_row->{'lnl'}}); # # Check if this particular sample was deleveraged # if (defined($delev{$gen_row->{'id'} . "_" . $gen_row->{'tag_id'}}) && $delev{$gen_row->{'id'} . "_" . $gen_row->{'tag_id'}} >= 1) { $locus->{'delev'}++; } } } elsif ($data_type eq "geno") { # # Add genotypes # $sth->{'gtypes'}->execute($batch_id); while ($gen_row = $sth->{'gtypes'}->fetchrow_hashref()) { next if (!defined($loci->{$gen_row->{'catalog_id'}})); $locus = $loci->{$gen_row->{'catalog_id'}}; if (!defined($locus->{'gtypes'}->{$gen_row->{'file'}})) { $locus->{'gtypes'}->{$gen_row->{'file'}} = []; } push(@{$locus->{'gtypes'}->{$gen_row->{'file'}}}, {'file' => $gen_row->{'file'}, 'gtype' => $gen_row->{'genotype'}}); } } # # Fetch SNPs and Alleles # $sth->{'snp'}->execute($batch_id); while ($snp_row = $sth->{'snp'}->fetchrow_hashref()) { next if (!defined($loci->{$snp_row->{'tag_id'}})); $loci->{$snp_row->{'tag_id'}}->{'snps'} .= $snp_row->{'col'} . "," . $snp_row->{'rank_1'} . ">" . $snp_row->{'rank_2'} . ";"; } $sth->{'allele'}->execute($batch_id); while ($all_row = $sth->{'allele'}->fetchrow_hashref()) { next if (!defined($loci->{$all_row->{'tag_id'}})); $loci->{$all_row->{'tag_id'}}->{'alleles'} .= $all_row->{'allele'} . ";"; } # # If exporting genotypes and a locus depth limit was specified, fetch locus depths. # if ($data_type eq "geno" && $locus_depth_lim > 0) { $sth->{'depths'}->execute($batch_id); while ($row = $sth->{'depths'}->fetchrow_hashref()) { next if (!defined($loci->{$row->{'catalog_id'}})); if (!defined($depths->{$row->{'catalog_id'}})) { $depths->{$row->{'catalog_id'}} = {}; } $depths->{$row->{'catalog_id'}}->{$row->{'file'}} += $row->{'depth'}; } } } sub apply_corrected_genotypes { my ($sth, $loci) = @_; my (%corrections, $locus, $key, $row, $sample); #print STDERR "Applying manually corrected genotypes to export data...\n"; # # Fetch the manual corrections from the database. # $sth->{'corr'}->execute($batch_id) or die("Unable to select results from $db.\n"); while ($row = $sth->{'corr'}->fetchrow_hashref()) { if (!defined($corrections{$row->{'catalog_id'}})) { $corrections{$row->{'catalog_id'}} = {}; } $corrections{$row->{'catalog_id'}}->{$row->{'file'}} = $row->{'genotype'}; } foreach $key (keys %{$loci}) { $locus = $loci->{$key}; next if (!defined($corrections{$locus->{'id'}})); foreach $sample (keys %{$corrections{$locus->{'id'}}}) { @{$locus->{'gtypes'}->{$sample}} = (); push(@{$locus->{'gtypes'}->{$sample}}, {'file' => $sample, 'gtype' => $corrections{$locus->{'id'}}->{$sample}}); } } } sub prepare_filter_parameters { my ($params, $filters) = @_; my ($filter); push(@{$params}, $batch_id); foreach $filter (keys %{$filters}) { if ($filter eq "snps") { push(@{$params}, $filters->{'snps_l'}); push(@{$params}, $filters->{'snps_u'}); } elsif ($filter eq "alle") { push(@{$params}, $filters->{'alle_l'}); push(@{$params}, $filters->{'alle_u'}); } elsif ($filter eq "pare") { push(@{$params}, $filters->{'pare_l'}); push(@{$params}, $filters->{'pare_u'}); } elsif ($filter eq "lnl") { push(@{$params}, $filters->{'lnl_l'}); push(@{$params}, $filters->{'lnl_u'}); } elsif ($filter eq "chisq") { push(@{$params}, $filters->{'chisq_l'}); push(@{$params}, $filters->{'chisq_u'}); } elsif ($filter eq "prog") { push(@{$params}, $filters->{'prog'}); } elsif ($filter eq "vprog") { push(@{$params}, $filters->{'vprog'}); } elsif ($filter eq "cata") { push(@{$params}, $filters->{'cata'}); } elsif ($filter eq "gcnt") { push(@{$params}, $filters->{'gcnt'}); } elsif ($filter eq "est") { push(@{$params}, 0); } elsif ($filter eq "pe") { push(@{$params}, 0); } elsif ($filter eq "blast") { push(@{$params}, 0); } elsif ($filter eq "ref") { push(@{$params}, $filters->{'ref'}); } elsif ($filter eq "loc") { push(@{$params}, $filters->{'chr'}); push(@{$params}, $filters->{'sbp'} * 1000000); push(@{$params}, $filters->{'ebp'} * 1000000); } elsif ($filter eq "mark") { if ($filters->{'mark'} eq "Any") { push(@{$params}, "%/%"); } else { push(@{$params}, $filters->{'mark'}); } } } } sub apply_query_filters { my ($filters) = @_; my ($query, $filter) = ""; my %sql_filters = ("cata" => "(catalog_index.tag_id = ?)", "alle" => "(alleles >= ? AND alleles <= ?)", "snps" => "(snps >= ? AND snps <= ?)", "pare" => "(parents >= ? AND parents <= ?)", "prog" => "(progeny >= ?)", "vprog" => "(valid_progeny >= ?)", "lnl" => "(lnl >= ? AND lnl <= ?)", "mark" => "(marker LIKE ?)", "est" => "(ests > ?)", "pe" => "(pe_radtags > ?)", "blast" => "(blast_hits > ?)", "gcnt" => "(geno_cnt >= ?)", "chisq" => "(chisq_pval >= ? AND chisq_pval <= ?)", "ref" => "(catalog_index.type = ?)", "loc" => "(catalog_index.chr = ? && catalog_index.bp >= ? && catalog_index.bp <= ?)"); if (scalar(keys %{$filters}) > 0) { foreach $filter (keys %{$filters}) { next if (!defined($sql_filters{$filter})); $query .= " AND "; $query .= $sql_filters{$filter}; } } return $query; } sub write_observed_haplotypes { my ($loci, $samples, $filters) = @_; my ($workbook, $worksheet); my ($out_fh, $str, $cat_id, $id, $locus, $gtypes, $types, $tot_depth); if ($type eq "xls") { $workbook = Spreadsheet::WriteExcel->new($out_file) or die("Unable to initiate excel spreadsheet.\n"); $worksheet = $workbook->add_worksheet() or die("Unable to add a worksheet to our excel spreadsheet.\n"); } else { open($out_fh, ">$out_file") or die("Unable to open output file '$out_file'\n"); } # # Order the samples by sample ID # my @ordered_sam = sort {$samples->{$a} <=> $samples->{$b}} keys %{$samples}; # # Print the heading out for the spreadsheet # my $i = 0; $str = "# " if ($type ne "xls"); $str .= "Catalog ID\t" . "Annotation\t" . "Chr\t" . "BP\t" . "Consensus Sequence\t" . "Num Parents\t" . "Num Progeny\t" . "Num SNPs\t" . "SNPs\t" . "Num Alleles\t" . "Alleles\t" . "Deleveraged\t"; foreach $id (@ordered_sam) { $str .= $id . "\t"; } $str = substr($str, 0, -1); $str .= "\n"; $type eq "xls" ? write_excel($worksheet, $i, $str) : print $out_fh $str; $i++; foreach $cat_id (keys %{$loci}) { $locus = $loci->{$cat_id}; $locus->{'snps'} = substr($locus->{'snps'}, 0, -1) if (length($locus->{'snps'}) > 0); $locus->{'alleles'} = substr($locus->{'alleles'}, 0, -1) if (length($locus->{'alleles'}) > 0); $str = $cat_id . "\t" . $locus->{'annotation'} . "\t" . $locus->{'chr'} . "\t" . $locus->{'bp'} . "\t" . $locus->{'seq'} . "\t" . $locus->{'num_parents'} . "\t" . $locus->{'num_progeny'} . "\t" . $locus->{'num_snps'} . "\t" . $locus->{'snps'} . "\t" . $locus->{'num_alleles'} . "\t" . $locus->{'alleles'} . "\t" . $locus->{'delev'} . "\t"; foreach $id (@ordered_sam) { $types = $locus->{'gtypes'}->{$id}; if (!defined($types)) { $str .= "\t"; next; } # # Check total locus depth. # $tot_depth = 0; foreach $type (@{$types}) { $tot_depth += $type->{'depth'}; } if ($tot_depth < $locus_depth_lim) { $str .= "\t"; next; } if ($types->[0]->{'lnl'} < $locus_lnl_lim) { $str .= "\t"; next; } foreach $type (@{$types}) { $str .= $all_depth ? $type->{'depth'} : $type->{'allele'}; $str .= "/"; } $str = substr($str, 0, -1); $str .= "\t"; } $str = substr($str, 0, -1); $str .= "\n"; $type eq "xls" ? write_excel($worksheet, $i, $str) : print $out_fh $str; $i++; } $str = "\n"; $type eq "xls" ? write_excel($worksheet, $i, $str) : print $out_fh $str; $i++; foreach $id (@ordered_sam) { $str = "\t" . $samples->{$id} . "\t" . $id . "\n"; $type eq "xls" ? write_excel($worksheet, $i, $str) : print $out_fh $str; $i++; } $type eq "xls" ? $workbook->close() : close($out_fh); } sub write_genotypes { my ($loci, $samples, $depths) = @_; my ($workbook, $worksheet); my ($out_fh, $str, $cat_id, $id, $locus, $gtypes, $types); if ($type eq "xls") { $workbook = Spreadsheet::WriteExcel->new($out_file) or die("Unable to initiate excel spreadsheet.\n"); $worksheet = $workbook->add_worksheet() or die("Unable to add a worksheet to our excel spreadsheet.\n"); } else { open($out_fh, ">$out_file") or die("Unable to open output file '$out_file'\n"); } # # Order the samples by sample ID # my @ordered_sam = sort {$samples->{$a} <=> $samples->{$b}} keys %{$samples}; # # Print the heading out for the spreadsheet # my $i = 0; $str = "Catalog ID\t" . "Annotation\t" . "Chr\t" . "BP\t" . "Marker\t"; foreach $id (@ordered_sam) { $str .= $id . "\t"; } $str = substr($str, 0, -1); $str .= "\n"; $type eq "xls" ? write_excel($worksheet, $i, $str) : print $out_fh $str; $i++; my ($trans_marker); foreach $cat_id (keys %{$loci}) { $locus = $loci->{$cat_id}; $trans_marker = translate_marker($map_type, $locus->{'marker'}); $str = $cat_id . "\t" . $locus->{'annotation'} . "\t" . $locus->{'chr'} . "\t" . $locus->{'bp'} . "\t" . $trans_marker . "\t"; foreach $id (@ordered_sam) { $types = $locus->{'gtypes'}->{$id}; # # Check that there is a genotype for this sample. # if (!defined($types)) { $str .= "\t"; next; } # # Check the depth of coverage for this locus, if requested. # if ($locus_depth_lim > 0) { if (!defined($depths->{$cat_id}->{$id}) || $depths->{$cat_id}->{$id} < $locus_depth_lim) { $str .= "\t"; next; } } my $trans_gtype = $translate_genotypes->{$map_type}->($trans_marker, $types->[0]->{'gtype'}); $str .= $trans_gtype . "\t"; } $str = substr($str, 0, -1); $str .= "\n"; $type eq "xls" ? write_excel($worksheet, $i, $str) : print $out_fh $str; $i++; } $str = "\n"; $type eq "xls" ? write_excel($worksheet, $i, $str) : print $out_fh $str; $i++; foreach $id (@ordered_sam) { $str = $samples->{$id} . "\t" . $id . "\n"; $type eq "xls" ? write_excel($worksheet, $i, $str) : print $out_fh $str; $i++; } $type eq "xls" ? $workbook->close() : close($out_fh); } sub write_excel { my ($worksheet, $i, $str) = @_; chomp $str; my @row = split(/\t/, $str); my $j = 0; foreach my $r (@row) { $worksheet->write($i, $j, $r); $j++; } } sub translate_marker { my ($map_type, $in_marker) = @_; my %dictionary; return $in_marker if ($map_type eq "gen"); $dictionary{"dh"} = {}; $dictionary{"cp"} = {}; $dictionary{"f2"} = {}; $dictionary{"bc1"} = {}; $dictionary{"dh"}->{"ab/--"} = "abx--"; $dictionary{"dh"}->{"--/ab"} = "--xab"; $dictionary{"cp"}->{"ab/--"} = "lmx--"; $dictionary{"cp"}->{"--/ab"} = "--xnp"; $dictionary{"cp"}->{"ab/aa"} = "lmxll"; $dictionary{"cp"}->{"aa/ab"} = "nnxnp"; $dictionary{"cp"}->{"ab/ab"} = "hkxhk"; $dictionary{"cp"}->{"ab/ac"} = "efxeg"; $dictionary{"cp"}->{"ab/cd"} = "abxcd"; $dictionary{"f2"}->{"aa/bb"} = "aaxbb"; $dictionary{"f2"}->{"ab/cd"} = "abxcd"; $dictionary{"f2"}->{"ab/aa"} = "abxaa"; $dictionary{"f2"}->{"aa/ab"} = "aaxab"; $dictionary{"f2"}->{"ab/cc"} = "abxcc"; $dictionary{"f2"}->{"cc/ab"} = "ccxab"; $dictionary{"bc1"}->{"aa/bb"} = "aaxbb"; $dictionary{"bc1"}->{"bb/aa"} = "bbxaa"; $dictionary{"bc1"}->{"ab/cc"} = "abxcc"; $dictionary{"bc1"}->{"cc/ab"} = "ccxab"; return defined($dictionary{$map_type}->{$in_marker}) ? $dictionary{$map_type}->{$in_marker} : ""; } sub trans_bc1_map { my ($marker, $in_gtype) = @_; my (%types, %dictionary); $dictionary{"aaxbb"} = {}; $dictionary{"bbxaa"} = {}; $dictionary{"abxcc"} = {}; $dictionary{"ccxab"} = {}; $dictionary{"aaxbb"}->{"--"} = "-"; $dictionary{"aaxbb"}->{"aa"} = "b"; $dictionary{"aaxbb"}->{"ab"} = "h"; $dictionary{"aaxbb"}->{"bb"} = "h"; $dictionary{"bbxaa"}->{"--"} = "-"; $dictionary{"bbxaa"}->{"aa"} = "h"; $dictionary{"bbxaa"}->{"ab"} = "h"; $dictionary{"bbxaa"}->{"bb"} = "a"; $dictionary{"abxcc"}->{"--"} = "-"; $dictionary{"abxcc"}->{"ac"} = "h"; $dictionary{"abxcc"}->{"bc"} = "h"; $dictionary{"abxcc"}->{"ab"} = "b"; $dictionary{"abxcc"}->{"aa"} = "b"; $dictionary{"abxcc"}->{"bb"} = "b"; $dictionary{"ccxab"}->{"--"} = "-"; $dictionary{"ccxab"}->{"ac"} = "h"; $dictionary{"ccxab"}->{"bc"} = "h"; $dictionary{"ccxab"}->{"ab"} = "a"; $dictionary{"ccxab"}->{"aa"} = "a"; $dictionary{"ccxab"}->{"bb"} = "a"; my $out_gtype = defined($dictionary{$marker}->{lc($in_gtype)}) ? $dictionary{$marker}->{lc($in_gtype)} : "-"; if (lc($in_gtype) ne $in_gtype) { return uc($out_gtype); } else { return $out_gtype; } } sub trans_dh_map { my ($marker, $in_gtype) = @_; my (%types, %dictionary); $dictionary{"abx--"} = {}; $dictionary{"--xab"} = {}; $dictionary{"abx--"}->{"aa"} = "a"; $dictionary{"abx--"}->{"bb"} = "b"; $dictionary{"abx--"}->{"--"} = "-"; $dictionary{"--xab"}->{"aa"} = "a"; $dictionary{"--xab"}->{"bb"} = "b"; $dictionary{"--xab"}->{"--"} = "-"; my $out_gtype = defined($dictionary{$marker}->{lc($in_gtype)}) ? $dictionary{$marker}->{lc($in_gtype)} : "-"; if (lc($in_gtype) ne $in_gtype) { return uc($out_gtype); } else { return $out_gtype; } } sub trans_f2_map { my ($marker, $in_gtype) = @_; my (%types, %dictionary); $dictionary{"aaxbb"} = {}; $dictionary{"abxcd"} = {}; $dictionary{"abxaa"} = {}; $dictionary{"aaxab"} = {}; $dictionary{"abxcc"} = {}; $dictionary{"ccxab"} = {}; $dictionary{"aaxbb"}->{"aa"} = "a"; $dictionary{"aaxbb"}->{"ab"} = "h"; $dictionary{"aaxbb"}->{"bb"} = "b"; $dictionary{"aaxbb"}->{"--"} = "-"; $dictionary{"abxcd"}->{"aa"} = "a"; $dictionary{"abxcd"}->{"ab"} = "a"; $dictionary{"abxcd"}->{"bb"} = "a"; $dictionary{"abxcd"}->{"cc"} = "b"; $dictionary{"abxcd"}->{"cd"} = "b"; $dictionary{"abxcd"}->{"dd"} = "b"; $dictionary{"abxcd"}->{"ac"} = "h"; $dictionary{"abxcd"}->{"ad"} = "h"; $dictionary{"abxcd"}->{"bc"} = "h"; $dictionary{"abxcd"}->{"bd"} = "h"; $dictionary{"abxcd"}->{"--"} = "-"; $dictionary{"abxaa"}->{"aa"} = "-"; $dictionary{"abxaa"}->{"ab"} = "-"; $dictionary{"abxaa"}->{"bb"} = "a"; $dictionary{"abxaa"}->{"--"} = "-"; $dictionary{"aaxab"}->{"aa"} = "-"; $dictionary{"aaxab"}->{"ab"} = "-"; $dictionary{"aaxab"}->{"bb"} = "b"; $dictionary{"aaxab"}->{"--"} = "-"; $dictionary{"abxcc"}->{"a"} = "a"; $dictionary{"abxcc"}->{"ab"} = "a"; $dictionary{"abxcc"}->{"bb"} = "a"; $dictionary{"abxcc"}->{"cc"} = "b"; $dictionary{"abxcc"}->{"ac"} = "-"; $dictionary{"abxcc"}->{"bc"} = "-"; $dictionary{"abxcc"}->{"--"} = "-"; $dictionary{"ccxab"}->{"aa"} = "b"; $dictionary{"ccxab"}->{"ab"} = "b"; $dictionary{"ccxab"}->{"bb"} = "b"; $dictionary{"ccxab"}->{"cc"} = "a"; $dictionary{"ccxab"}->{"ac"} = "-"; $dictionary{"ccxab"}->{"bc"} = "-"; $dictionary{"ccxab"}->{"--"} = "-"; my $out_gtype = defined($dictionary{$marker}->{lc($in_gtype)}) ? $dictionary{$marker}->{lc($in_gtype)} : "-"; if (lc($in_gtype) ne $in_gtype) { return uc($out_gtype); } else { return $out_gtype; } } sub trans_cp_map { my ($marker, $in_gtype) = @_; my (%types, %dictionary); $dictionary{"lmx--"} = {}; $dictionary{"--xnp"} = {}; $dictionary{"lmxll"} = {}; $dictionary{"nnxnp"} = {}; $dictionary{"hkxhk"} = {}; $dictionary{"efxeg"} = {}; $dictionary{"abxcd"} = {}; $dictionary{"lmx--"}->{"--"} = "--"; $dictionary{"lmx--"}->{"aa"} = "ll"; $dictionary{"lmx--"}->{"bb"} = "lm"; $dictionary{"lmx--"}->{"ab"} = "lm"; $dictionary{"--xnp"}->{"--"} = "--"; $dictionary{"--xnp"}->{"aa"} = "nn"; $dictionary{"--xnp"}->{"bb"} = "np"; $dictionary{"--xnp"}->{"ab"} = "np"; $dictionary{"lmxll"}->{"--"} = "--"; $dictionary{"lmxll"}->{"aa"} = "ll"; $dictionary{"lmxll"}->{"ab"} = "lm"; $dictionary{"nnxnp"}->{"--"} = "--"; $dictionary{"nnxnp"}->{"aa"} = "nn"; $dictionary{"nnxnp"}->{"ab"} = "np"; $dictionary{"hkxhk"}->{"--"} = "--"; $dictionary{"hkxhk"}->{"ab"} = "hk"; $dictionary{"hkxhk"}->{"aa"} = "hh"; $dictionary{"hkxhk"}->{"bb"} = "kk"; $dictionary{"efxeg"}->{"--"} = "--"; $dictionary{"efxeg"}->{"ab"} = "ef"; $dictionary{"efxeg"}->{"ac"} = "eg"; $dictionary{"efxeg"}->{"bc"} = "fg"; $dictionary{"efxeg"}->{"aa"} = "ee"; $dictionary{"abxcd"}->{"--"} = "--"; $dictionary{"abxcd"}->{"ac"} = "ac"; $dictionary{"abxcd"}->{"ad"} = "ad"; $dictionary{"abxcd"}->{"bc"} = "bc"; $dictionary{"abxcd"}->{"bd"} = "bd"; my $out_gtype = defined($dictionary{$marker}->{lc($in_gtype)}) ? $dictionary{$marker}->{lc($in_gtype)} : "-"; if (lc($in_gtype) ne $in_gtype) { return uc($out_gtype); } else { return $out_gtype; } } sub trans_gen_map { my ($marker, $in_gtype) = @_; return $in_gtype; } sub prepare_sql_handles { my ($sth, $filters) = @_; # # Connect to the database, check for the existence of a MySQL config file in the home # directory first, otherwise use the stacks-distributed one. # my $cnf = (defined($ENV{"HOME"}) && -e $ENV{"HOME"} . "/.my.cnf") ? $ENV{"HOME"} . "/.my.cnf" : $mysql_config; $sth->{'dbh'} = DBI->connect("DBI:mysql:$db:mysql_read_default_file=$cnf") or die("Unable to connect to the $db MySQL Database!\n" . $DBI::errstr); my $query; $query = "SELECT sample_id, tag_id FROM tag_index " . "WHERE batch_id=? AND deleveraged=true"; $sth->{'delev'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT id, file FROM samples " . "WHERE batch_id=? ORDER BY id"; $sth->{'samp'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT tag_id, allele FROM catalog_alleles " . "WHERE batch_id=?"; $sth->{'allele'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT tag_id, col, rank_1, rank_2 FROM catalog_snps " . "WHERE batch_id=?"; $sth->{'snp'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT catalog_id, samples.id as id, samples.sample_id, samples.type, file, tag_id, allele, depth, lnl " . "FROM matches " . "JOIN samples ON (matches.sample_id=samples.id) " . "WHERE matches.batch_id=? AND matches.depth>?"; $sth->{'mat'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT catalog_id, file, depth " . "FROM matches " . "JOIN samples ON (matches.sample_id=samples.id) " . "WHERE matches.batch_id=?"; $sth->{'depths'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT catalog_id, samples.id as id, samples.sample_id, samples.type, file, genotype " . "FROM catalog_genotypes " . "JOIN samples ON (catalog_genotypes.sample_id=samples.id) " . "WHERE catalog_genotypes.batch_id=?"; $sth->{'gtypes'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT gc.catalog_id, gc.sample_id, gc.genotype, file " . "FROM genotype_corrections as gc " . "JOIN samples ON (gc.sample_id=samples.id) " . "WHERE gc.batch_id=?"; $sth->{'corr'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT catalog_index.tag_id as tag_id, catalog_index.chr, catalog_index.bp, " . "snps, alleles, parents, progeny, valid_progeny, " . "seq, marker, chisq_pval, lnl, ratio, ests, pe_radtags, blast_hits, geno_cnt, external_id " . "FROM catalog_index " . "JOIN catalog_tags ON (catalog_index.cat_id=catalog_tags.id) " . "LEFT JOIN catalog_annotations ON " . "(" . "catalog_index.batch_id=catalog_annotations.batch_id AND " . "catalog_index.tag_id=catalog_annotations.catalog_id" . ") " . "WHERE catalog_index.batch_id=?"; $query .= apply_query_filters($filters); $sth->{'tag'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); } sub close_sql_handles { my ($sth) = @_; my $key; foreach $key (keys %{$sth}) { next if ($key =~ /dbh/); $sth->{$key}->finish(); } foreach $key (keys %{$sth}) { next if ($key !~ /dbh/); $sth->{$key}->disconnect(); } } sub parse_command_line { my ($filter, $name, $value); while (@ARGV) { $_ = shift @ARGV; if ($_ =~ /^-f$/) { $out_file = shift @ARGV; } elsif ($_ =~ /^-o$/) { $type = shift @ARGV; } elsif ($_ =~ /^-a$/) { $data_type = lc(shift @ARGV); } elsif ($_ =~ /^-b$/) { $batch_id = shift @ARGV; } elsif ($_ =~ /^-D$/) { $db = shift @ARGV; } elsif ($_ =~ /^-m$/) { $map_type = lc(shift @ARGV); } elsif ($_ =~ /^-A$/) { $allele_depth_lim = shift @ARGV; } elsif ($_ =~ /^-L$/) { $locus_depth_lim = shift @ARGV; } elsif ($_ =~ /^-I$/) { $locus_lnl_lim = shift @ARGV; } elsif ($_ =~ /^-d$/) { $all_depth++; } elsif ($_ =~ /^-c$/) { $man_cor++; } elsif ($_ =~ /^-v$/) { version(); exit(); } elsif ($_ =~ /^-h$/) { usage(); } elsif ($_ =~ /^-F$/) { $filter = shift @ARGV; ($name, $value) = split(/=/, $filter); if (length($name) == 0 || length($value) == 0) { print STDERR "Error parsing filter '$filter'\n"; usage(); } $filters{$name} = $value; } else { print STDERR "Unknown command line option: '$_'\n"; usage(); } } if (defined($filters{'chr'})) { $filters{'loc'} = 1; $filters{'sbp'} = 0 if (!defined($filters{'sbp'})); $filters{'ebp'} = 500000000 if (!defined($filters{'ebp'})); } if (defined($filters{'snps_l'}) || defined($filters{'snps_u'})) { $filters{'snps'} = 1; $filters{'snps_l'} = 1 if (!defined($filters{'snps_l'})); $filters{'snps_u'} = 100 if (!defined($filters{'snps_u'})); } if (defined($filters{'alle_l'}) || defined($filters{'alle_u'})) { $filters{'alle'} = 1; $filters{'alle_l'} = 1 if (!defined($filters{'alle_l'})); $filters{'alle_u'} = 100 if (!defined($filters{'alle_u'})); } if (defined($filters{'pare_l'}) || defined($filters{'pare_u'})) { $filters{'pare'} = 1; $filters{'pare_l'} = 1 if (!defined($filters{'pare_l'})); $filters{'pare_u'} = 1000 if (!defined($filters{'pare_u'})); } if (defined($filters{'lnl_l'}) || defined($filters{'lnl_u'})) { $filters{'lnl'} = 1; $filters{'lnl_l'} = -500 if (!defined($filters{'lnl_l'})); $filters{'lnl_u'} = 0 if (!defined($filters{'lnl_u'})); } if ($out_file eq "") { print STDERR "You must specify the file to write data to!\n"; usage(); } if ($type ne "tsv" && $type ne "xls") { print STDERR "Unknown output file type specified '$type'.\n"; usage(); } if ($data_type ne "haplo" && $data_type ne "geno") { print STDERR "Unknown data type specified, 'haplo' and 'geno' are currently accepted.\n"; usage(); } if ($data_type eq "geno" && $map_type ne "bc1" && $map_type ne "dh" && $map_type ne "f2" && $map_type ne "cp" && $map_type ne "gen") { print STDERR "Unknown map type specified, 'bc1', 'dh', 'f2', 'cp', and 'gen' are currently accepted.\n"; usage(); } if ($data_type ne "geno" && $man_cor > 0) { print STDERR "You can only specify manual corrections when exporting genotypes.\n"; usage(); } if ($data_type ne "haplo" && $all_depth > 0) { print STDERR "You must use a data type of 'haplo' to export allele depths.\n"; usage(); } } sub version { print STDERR "export_sql.pl ", stacks_version, "\n"; } sub usage { version(); my $filt; my $i = 1; foreach my $f (@valid_filters) { $filt .= $f . ", "; $filt .= "\n " if ($i % 10 == 0); $i++; } $filt = substr($filt, 0, -11); print STDERR "export_sql.pl -D db -b batch_id -a type -f file -o tsv|xls [-m type -c] [-F filter=value ...] [-L lim] [-d] [-h]\n", " D: database to export from.\n", " b: batch ID of the dataset to export.\n", " a: type of data to export, either 'geno' or 'haplo', for genotypes or observed haplotypes.\n", " f: file to output data.\n", " o: type of data to export: 'tsv' or 'xls'.\n", " d: output depths of alleles instead of the allele values (must use 'haplo' data type).\n", " m: map type. If genotypes are to be exported, specify the map type.\n", " c: include manual corrections if exporting genotypes.\n", " Filters that are applied to select among the catalog loci:\n", " F: one or more filters in the format name=value.\n", " Supported filters: \n", " $filt\n\n", " Filters to be applied to individual sets of haplotype calls (for those selected catalog loci):\n", " A: specify an minimum allele depth limit.\n", " L: specify a minimum locus depth limit.\n", " I: specify a minimum locus log likelihood limit.\n", " h: display this help message.\n\n"; exit(0); } stacks-1.44/scripts/extract_interpop_chars.pl000755 000765 000024 00000017251 12441417455 022320 0ustar00catchenstaff000000 000000 #!/usr/bin/env perl # # Copyright 2010-2013, Julian Catchen # # This file is part of Stacks. # # Stacks is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation, either version 3 of the License, or # (at your option) any later version. # # Stacks is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with Stacks. If not, see . # use strict; use Bio::SeqIO; use constant true => 1; use constant false => 0; my $debug = 0; my $in_file = ""; my $barcode_list = ""; my $out_file = ""; my $pop_limit = 10; parse_command_line(); my (@barcodes, %barcode_key, %ids, %characters, %polymorph_level, %site_dist); load_barcode_list(\@barcodes, \%barcode_key); my ($id, $barcode); # # Initialize the characters hash. # foreach $barcode (@barcodes) { $characters{$barcode} = ""; $polymorph_level{$barcode} = {}; $polymorph_level{$barcode}->{'poly_loci'} = 0; $polymorph_level{$barcode}->{'loci'} = 0; $polymorph_level{$barcode}->{'poly'} = 0; $polymorph_level{$barcode}->{'sites'} = 0; } # # Obtain a list of unique tags to process # parse_interspecies_tags($in_file, \%ids); my $i = 1; my $num_ids = keys %ids; foreach $id (keys %ids) { print STDERR "Processing tag $i of $num_ids. \r" if ($i % 1000 == 0); process_tag($ids{$id}, \@barcodes, \%characters, \%polymorph_level, \%site_dist); $i++; } print_results(\%characters, \%barcode_key, \%polymorph_level, \%site_dist); print STDERR "\n"; sub process_tag { my ($tag, $barcodes, $characters, $poly_level, $dist) = @_; my (%substack, $bc, $len, $col); return if (check_tag_counts($tag) == false); # if (scalar(keys %{$tag->{'count'}}) >= 10) { # print STDERR "Keeping tag ", $tag->{'id'}, " with tags from ", scalar(keys %{$tag->{'count'}}), " populations.\n"; # } # # Make sure enough populations have this marker before recording it. # return if (scalar(keys %{$tag->{'seqs'}}) < $pop_limit); foreach $bc (@{$barcodes}) { $substack{$bc} = []; } foreach $bc (keys %{$tag->{'seqs'}}) { my $aref; # # Store the individual reads as a two-dimensional array. # @{$aref} = split(//, $tag->{'seqs'}->{$bc}); $substack{$bc} = $aref; # # Tally the number of non-homozygous sites and the total # number of sites at each locus. # $len = scalar(@{$aref}); $poly_level->{$bc}->{'loci'}++; $poly_level->{$bc}->{'sites'} += scalar(@{$aref}); my $poly = 0; foreach $col (0..$len - 1) { if ($aref->[$col] eq "N") { $poly_level->{$bc}->{'poly'}++; $poly++; } } if ($poly > 0) { $poly_level->{$bc}->{'poly_loci'}++; } } $len = 0; foreach $bc (@{$barcodes}) { $len = scalar(@{$substack{$bc}}); last if ($len > 0); } foreach $col (0..$len - 1) { if (homozygous(\%substack, $col) == false) { record_character(\%substack, $col, $characters, $dist); } } } sub check_tag_counts { my ($tag) = @_; # # Check to make sure that there is only a single tag from # each individual sample. # my ($bc); foreach $bc (keys %{$tag->{'count'}}) { return false if ($tag->{'count'}->{$bc} > 1); } return true; } sub homozygous { my ($substack, $col) = @_; my ($bc, %nuc); foreach $bc (keys %{$substack}) { if (scalar(@{$substack->{$bc}}) > 0) { $nuc{$substack->{$bc}->[$col]}++; } } my @keys = sort {$nuc{$b} <=> $nuc{$a}} keys(%nuc); return true if (scalar(@keys) == 1); return true if (scalar(@keys) == 2 && ($keys[0] eq "N" || $keys[1] eq "N")); return false; } sub record_character { my ($stack, $col, $characters, $dist) = @_; my ($bc, $cnt); $cnt = 0; foreach $bc (keys %{$stack}) { if (scalar(@{$stack->{$bc}}) == 0) { $characters->{$bc} .= "N"; } else { $characters->{$bc} .= $stack->{$bc}->[$col]; $cnt++ if ($stack->{$bc}->[$col] ne "N"); } } $dist->{$cnt}++; } sub print_results { my ($characters, $key, $poly_level, $dist) = @_; my ($out, $seq, $log_fh, $barcode); $out = Bio::SeqIO->new(-file => ">$out_file", -format => "fasta"); foreach $barcode (sort keys %{$characters}) { $seq = Bio::Seq->new('-seq' => $characters->{$barcode}, '-display_id' => $key->{$barcode}, '-alphabet' => 'dna'); print STDERR "Writing sequence for population '", $key->{$barcode}, " / ", $barcode, "' with a length ", $seq->length(), "\n"; $out->write_seq($seq); } print STDERR "Pop\tTotal Loci\tNon-hom Loci\tTotal Sites\tNon-homozygous sites\n"; foreach $barcode (sort keys %{$poly_level}) { print STDERR $key->{$barcode}, "\t", $poly_level->{$barcode}->{'loci'}, "\t", $poly_level->{$barcode}->{'poly_loci'}, "\t", $poly_level->{$barcode}->{'sites'}, "\t", $poly_level->{$barcode}->{'poly'}, "\n"; } print STDERR "Number of Populations\tSites\n"; foreach $barcode (sort {$b <=> $a} keys %{$dist}) { print STDERR $barcode, "\t", $dist->{$barcode}, "\n"; } } sub parse_interspecies_tags { my ($in_path, $ids) = @_; my (@parts, $line, $id, $barcode, $catalog_id); open(IN, "<$in_path") or die("Unable to open input file '$in_path'; $!\n"); while ($line = ) { chomp $line; @parts = split(/\t/, $line); next if ($parts[5] eq "consensus"); $id = $parts[2]; ($barcode, $catalog_id) = ($parts[7] =~ /^(\d+)\_(\d+)/); #($barcode, $catalog_id) = ($parts[4] =~ /^(\d+)\_(\d+)/); #print STDERR "Barcode_key: $barcode, $barcode_key{$barcode}\n"; #next if ($barcode_key{$barcode} eq "CB2"); if (!defined($ids->{$id})) { $ids->{$id} = {}; $ids->{$id}->{'id'} = $parts[2]; $ids->{$id}->{'seqs'} = {}; $ids->{$id}->{'count'} = {}; $ids->{$id}->{'batch_id'} = $parts[1]; } $ids->{$id}->{'seqs'}->{$barcode} = $parts[8]; #$ids->{$id}->{'seqs'}->{$barcode} = $parts[5]; $ids->{$id}->{'count'}->{$barcode}++; #print STDERR "Adding '$parts[8]' with barcode $barcode to $id\n"; } } sub load_barcode_list { my ($bl, $bk) = @_; open(BC, "<" . $barcode_list) or die("Unable to open barcodes file '$barcode_list': $!\n"); my ($line, $pop_id, $sample_id); while ($line = ) { chomp $line; next if (length($line) == 0 || substr($line, 0, 1) eq "#"); ($pop_id, $sample_id) = ($line =~ /^(\w+)\t(\d+)$/); $bk->{$sample_id} = $pop_id; push(@{$bl}, $sample_id); } close(BC); if (scalar(@{$bl}) == 0) { print STDERR "Unable to load any barcodes from '$barcode_list'\n"; usage(); } } sub parse_command_line { while (@ARGV) { $_ = shift @ARGV; if ($_ =~ /^-d$/) { $debug++; } elsif ($_ =~ /^-f$/) { $in_file = shift @ARGV; } elsif ($_ =~ /^-b$/) { $barcode_list = shift @ARGV; } elsif ($_ =~ /^-o$/) { $out_file = shift @ARGV; } elsif ($_ =~ /^-p$/) { $pop_limit = shift @ARGV; } elsif ($_ =~ /^-h$/) { usage(); } else { print STDERR "Unknown command line options received: $_\n"; usage(); } } } sub usage { print << "EOQ"; extract-interpop-chars.pl -f in_file -b barcodes [-o path] [-p limit] [-d] [-h] f: input file containing interspecies tags. b: list of barcodes to process (one barcode per population). o: output file. p: minimum number of populations required before recording a marker. h: display this help message. d: turn on debug output. EOQ exit(0); } stacks-1.44/scripts/index_radtags.pl000755 000765 000024 00000047636 12533677757 020413 0ustar00catchenstaff000000 000000 #!/usr/bin/env perl # # Copyright 2010, Julian Catchen # # This file is part of Stacks. # # Stacks is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation, either version 3 of the License, or # (at your option) any later version. # # Stacks is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with Stacks. If not, see . # # # Written by Julian Catchen # use strict; use File::Temp qw/tempfile/; use DBI; use constant stacks_version => "_VERSION_"; my $mysql_config = "_PKGDATADIR_" . "sql/mysql.cnf"; my $sql_path = "_PKGDATADIR_" . "sql/"; my $sql_tag_table = ""; my $sql_cat_table = ""; my $sql_chr_table = ""; my $db = ""; my $debug = 0; my $catalog_index = 0; my $tag_index = 0; my $radome_index = 0; parse_command_line(); my $cnf = (-e $ENV{"HOME"} . "/.my.cnf") ? $ENV{"HOME"} . "/.my.cnf" : $mysql_config; # # Make sure the SQL definition files are available # if ($catalog_index && !-e $sql_cat_table) { print STDERR "Unable to locate catalog SQL definition.\n"; usage(); } if ($tag_index && !-e $sql_tag_table) { print STDERR "Unable to locate tag_index SQL definition.\n"; usage(); } # # Connect to the database and prepare our queries. # my (%sth); prepare_sql_handles(\%sth); if ($catalog_index) { gen_cat_index(\%sth); } if ($tag_index) { gen_tag_index(\%sth); } close_sql_handles(\%sth); sub gen_cat_index { my ($sth) = @_; die ("Unable to find SQL file: '$sql_cat_table'\n") if (!-e $sql_tag_table); print STDERR "Generating catalog tag index\n"; my ($fh, $catalog_file) = tempfile("catalog_index_XXXXXXXX", UNLINK => 1, TMPDIR => 1); my ($row, $tag, $count, $par_cnt, $pro_cnt, $allele_cnt, $marker, $uncor_marker, $valid_pro, $chisq_pval, $lnl, $ratio, $ests, $pe_radtags, $blast_hits, $geno_cnt, $ref_type, $ref_id, $bp); my (%snps, %markers, %genotypes, %seqs, %hits, %parents, %progeny, %alleles, %chrs, %radome); print STDERR " Fetching catalog SNPs..."; fetch_catalog_snps(\%sth, \%snps); print STDERR "done.\n"; print STDERR " Fetching markers..."; fetch_markers(\%sth, \%markers); print STDERR "done.\n"; print STDERR " Fetching genotypes..."; fetch_genotypes(\%sth, \%genotypes); print STDERR "done.\n"; print STDERR " Fetching associated sequences..."; sequence_matches(\%sth, \%seqs, \%hits); print STDERR "done.\n"; print STDERR " Fetching catalog matches..."; catalog_matches(\%sth, \%parents, \%progeny, \%alleles); print STDERR "done.\n"; $ref_type = ""; $ref_id = 0; if ($radome_index) { print STDERR " Fetching reference RAD data..."; radome_ref(\%sth, \%radome); print STDERR "done.\n"; } print STDERR " Assembling catalog tags at the database..."; $sth->{'cat_tags'}->execute() or die("Unable to select results from $db.\n"); print STDERR "done.\n"; #my $num_rows = $sth->{'cat_tags'}->rows(); #my $i = 1; print STDERR " Processing catalog tags\n"; while ($row = $sth->{'cat_tags'}->fetchrow_hashref()) { # # Determine the number of SNPs contained within this RAD-Tag # if (defined($snps{$row->{'batch_id'}}->{$row->{'tag_id'}})) { $count = $snps{$row->{'batch_id'}}->{$row->{'tag_id'}}; } else { $count = 0; } # # Determine the number of parental samples that match this catalog RAD-Tag # if (defined($parents{$row->{'batch_id'}}->{$row->{'tag_id'}})) { $par_cnt = scalar(keys %{$parents{$row->{'batch_id'}}->{$row->{'tag_id'}}}); } else { $par_cnt = 0; } # # Determine the number of progeny samples that match this catalog RAD-Tag # if (defined($progeny{$row->{'batch_id'}}->{$row->{'tag_id'}})) { $pro_cnt = scalar(keys %{$progeny{$row->{'batch_id'}}->{$row->{'tag_id'}}}); } else { $pro_cnt = 0; } # # Determine the number of genotypes associated with this RAD-Tag # if (defined($genotypes{$row->{'batch_id'}}->{$row->{'tag_id'}})) { $geno_cnt = $genotypes{$row->{'batch_id'}}->{$row->{'tag_id'}}; } else { $geno_cnt = 0; } # # Determine the number of alleles for this catalog RAD-Tag # if (defined($alleles{$row->{'batch_id'}}->{$row->{'tag_id'}})) { $allele_cnt = scalar(keys %{$alleles{$row->{'batch_id'}}->{$row->{'tag_id'}}}); } else { $allele_cnt = 0; } # # Annotate the RAD site # if ($radome_index) { my $key = $row->{'chr'} . "|" . $row->{'bp'} . "|" . $row->{'strand'}; if (defined($radome{$key})) { $ref_type = $radome{$key}->{'type'}; $ref_id = $radome{$key}->{'id'}; } else { # # Check for a read aligned on the other strand. # $bp = $row->{'bp'} - 4; $key = $row->{'chr'} . "|" . $bp . "|" . $row->{'strand'}; if (defined($radome{$key})) { $ref_type = $radome{$key}->{'type'}; $ref_id = $radome{$key}->{'id'} } else { $ref_type = "genomic"; $ref_id = 0; } } } # # Determine if there are any sequences associated with this marker # $ests = 0; $pe_radtags = 0; $blast_hits = 0; if (defined($seqs{$row->{'batch_id'}}->{$row->{'tag_id'}}->{'est'})) { $ests = $seqs{$row->{'batch_id'}}->{$row->{'tag_id'}}->{'est'}; } if (defined($seqs{$row->{'batch_id'}}->{$row->{'tag_id'}}->{'pe_radtag'})) { $pe_radtags = $seqs{$row->{'batch_id'}}->{$row->{'tag_id'}}->{'pe_radtag'}; } if (defined($hits{$row->{'batch_id'}}->{$row->{'tag_id'}})) { $blast_hits = $hits{$row->{'batch_id'}}->{$row->{'tag_id'}}; } # # Does this RAD-Tag have a mappable marker? # if (defined($markers{$row->{'batch_id'}}->{$row->{'tag_id'}})) { $tag = $markers{$row->{'batch_id'}}->{$row->{'tag_id'}}; $marker = $tag->{'marker'}; $uncor_marker = $tag->{'uncor_marker'}; $chisq_pval = $tag->{'chisq_pval'}; $valid_pro = $tag->{'valid_pro'}; $ratio = $tag->{'ratio'}; $lnl = $tag->{'lnl'}; } else { $marker = ""; $uncor_marker = ""; $valid_pro = 0; $chisq_pval = 1.0; $ratio = ""; $lnl = 0.0; } # # Record the chromosomes present in this dataset (if any). # if (length($row->{'chr'}) > 0) { $chrs{$row->{'batch_id'}}->{$row->{'chr'}} = $row->{'bp'} > $chrs{$row->{'batch_id'}}->{$row->{'chr'}} ? $row->{'bp'} : $chrs{$row->{'batch_id'}}->{$row->{'chr'}}; } print $fh "0\t", $row->{'batch_id'}, "\t", $row->{'id'}, "\t", $row->{'tag_id'}, "\t", $count, "\t", $par_cnt, "\t", $pro_cnt, "\t", $allele_cnt, "\t", $marker, "\t", $uncor_marker, "\t", $valid_pro, "\t", $chisq_pval, "\t", $lnl, "\t", $ratio, "\t", $ests, "\t", $pe_radtags, "\t", $blast_hits, "\t", $geno_cnt, "\t", $row->{'chr'}, "\t", $row->{'bp'}, "\t", $ref_type, "\t", $ref_id, "\n"; } close($fh); `mysql --defaults-file=$cnf $db -e "DROP TABLE IF EXISTS catalog_index"`; `mysql --defaults-file=$cnf $db < $sql_cat_table`; import_sql_file($catalog_file, 'catalog_index'); if (scalar(keys %chrs) > 0) { my ($batch_id, $chr, $max); my ($fh, $chr_file) = tempfile("chr_index_XXXXXXXX", UNLINK => 1, TMPDIR => 1); foreach $batch_id (sort keys %chrs) { foreach $chr (sort keys %{$chrs{$batch_id}}) { # # Round up the maximum chromosome length to the nearest megabase. # $max = int($chrs{$batch_id}->{$chr} / 1000000); $max += $chrs{$batch_id}->{$chr} % 1000000 > 0 ? 1 : 0; print $fh "0\t", $batch_id, "\t", $chr, "\t", $max, "\n"; } } `mysql --defaults-file=$cnf $db -e "DROP TABLE IF EXISTS chr_index"`; `mysql --defaults-file=$cnf $db < $sql_chr_table`; import_sql_file($chr_file, 'chr_index'); close($fh); } } sub fetch_catalog_snps { my ($sth, $snps) = @_; my ($row); $sth->{'cat_snps'}->execute() or die("Unable to select results from $db.\n"); while ($row = $sth->{'cat_snps'}->fetchrow_hashref()) { if (!defined($snps->{$row->{'batch_id'}})) { $snps->{$row->{'batch_id'}} = {}; } $snps->{$row->{'batch_id'}}->{$row->{'tag_id'}}++; } } sub fetch_genotypes { my ($sth, $genotypes) = @_; my ($row); $sth->{'cat_geno'}->execute() or die("Unable to select results from $db.\n"); while ($row = $sth->{'cat_geno'}->fetchrow_hashref()) { if (!defined($genotypes->{$row->{'batch_id'}})) { $genotypes->{$row->{'batch_id'}} = {}; } if ($row->{'genotype'} ne "-" && $row->{'genotype'} ne "--") { $genotypes->{$row->{'batch_id'}}->{$row->{'tag_id'}}++; } } } sub fetch_markers { my ($sth, $markers) = @_; my ($row, $tag); $sth->{'marker'}->execute() or die("Unable to select results from $db.\n"); while ($row = $sth->{'marker'}->fetchrow_hashref()) { $tag = {}; $tag->{'marker'} = $row->{'type'}; $tag->{'uncor_marker'} = $row->{'uncor_type'}; $tag->{'chisq_pval'} = $row->{'chisq_pval'}; $tag->{'valid_pro'} = $row->{'progeny'}; $tag->{'ratio'} = $row->{'ratio'}; $tag->{'lnl'} = $row->{'lnl'}; if (!defined($markers->{$row->{'batch_id'}})) { $markers->{$row->{'batch_id'}} = {}; } $markers->{$row->{'batch_id'}}->{$row->{'catalog_id'}} = $tag; } } sub radome_ref { my ($sth, $radome) = @_; my ($row, $key, $strand); $sth->{'radome'}->execute() or die("Unable to select results from $db.\n"); while ($row = $sth->{'radome'}->fetchrow_hashref()) { $strand = $row->{'strand'} == 1 ? "+" : "-"; $key = $row->{'chr'} . "|" . $row->{'bp'} . "|" . $strand; if (!defined($radome->{$key}) || $row->{'type'} eq "exon") { $radome->{$key} = {'type' => $row->{'type'}, 'id' => $row->{'id'}}; } } } sub catalog_matches { my ($sth, $parents, $progeny, $alleles) = @_; my ($row, $key); $sth->{'cat_matches'}->execute() or die("Unable to select results from $db.\n"); while ($row = $sth->{'cat_matches'}->fetchrow_hashref()) { $key = $row->{'sample_id'} . "_" . $row->{'tag_id'}; if ($row->{'type'} eq "parent" || $row->{'type'} eq "sample") { if (!defined($parents->{$row->{'batch_id'}})) { $parents->{$row->{'batch_id'}} = {}; } if (!defined($parents->{$row->{'batch_id'}}->{$row->{'catalog_id'}})) { $parents->{$row->{'batch_id'}}->{$row->{'catalog_id'}} = {}; } $parents->{$row->{'batch_id'}}->{$row->{'catalog_id'}}->{$key}++; } elsif ($row->{'type'} eq "progeny") { if (!defined($progeny->{$row->{'batch_id'}})) { $progeny->{$row->{'batch_id'}} = {}; } if (!defined($progeny->{$row->{'batch_id'}}->{$row->{'catalog_id'}})) { $progeny->{$row->{'batch_id'}}->{$row->{'catalog_id'}} = {}; } $progeny->{$row->{'batch_id'}}->{$row->{'catalog_id'}}->{$key}++; } if (!defined($alleles->{$row->{'batch_id'}})) { $alleles->{$row->{'batch_id'}} = {}; } if (!defined($alleles->{$row->{'batch_id'}}->{$row->{'catalog_id'}})) { $alleles->{$row->{'batch_id'}}->{$row->{'catalog_id'}} = {}; } $alleles->{$row->{'batch_id'}}->{$row->{'catalog_id'}}->{$row->{'allele'}}++; } } sub sequence_matches { my ($sth, $seqs, $hits) = @_; my ($row); $sth->{'cat_seqs'}->execute() or die("Unable to select results from $db.\n"); while ($row = $sth->{'cat_seqs'}->fetchrow_hashref()) { if (!defined($seqs->{$row->{'batch_id'}})) { $seqs->{$row->{'batch_id'}} = {}; } if (!defined($seqs->{$row->{'batch_id'}}->{$row->{'catalog_id'}})) { $seqs->{$row->{'batch_id'}}->{$row->{'catalog_id'}} = {}; } $seqs->{$row->{'batch_id'}}->{$row->{'catalog_id'}}->{$row->{'type'}}++; } $sth->{'cat_hits'}->execute() or die("Unable to select results from $db.\n"); while ($row = $sth->{'cat_hits'}->fetchrow_hashref()) { if (!defined($hits->{$row->{'batch_id'}})) { $hits->{$row->{'batch_id'}} = {}; } $hits->{$row->{'batch_id'}}->{$row->{'catalog_id'}}++; } } sub gen_tag_index { my ($sth) = @_; die ("Unable to find SQL file: '$sql_tag_table'\n") if (!-e $sql_tag_table); my ($fh, $tag_file) = tempfile("tag_index_XXXXXXXX", UNLINK => 1, TMPDIR => 1); print STDERR "Generating unique tag index...\n"; my ($sample_row, $tags_row, $row, $catalog_id, $i, $num_samples); $sth->{'sample'}->execute() or die("Unable to select results from $db.\n"); $num_samples = $sth->{'sample'}->rows(); $i = 1; while ($sample_row = $sth->{'sample'}->fetchrow_hashref()) { print STDERR "Processing sample $i of $num_samples \r"; my (%depth, %snps, %cats); $sth->{'tags'}->execute($sample_row->{'batch_id'}, $sample_row->{'sample_id'}) or die("Unable to select results from $db.\n"); fetch_depth_counts($sth, \%depth, $sample_row->{'sample_id'}); fetch_snp_counts($sth, \%snps, $sample_row->{'sample_id'}); fetch_catalog_ids($sth, \%cats, $sample_row->{'batch_id'}, $sample_row->{'sample_id'}); while ($tags_row = $sth->{'tags'}->fetchrow_hashref()) { print $fh "0\t", $sample_row->{'batch_id'}, "\t", $sample_row->{'sample_id'}, "\t", $tags_row->{'tag_id'}, "\t", $tags_row->{'id'}, "\t", $depth{$tags_row->{'tag_id'}}, "\t", defined($snps{$tags_row->{'tag_id'}}) ? $snps{$tags_row->{'tag_id'}} : 0, "\t", $cats{$tags_row->{'tag_id'}}, "\t", $tags_row->{'deleveraged'}, "\t", $tags_row->{'blacklisted'}, "\t", $tags_row->{'removed'}, "\n"; } $i++; } print STDERR "\n"; close($fh); `mysql --defaults-file=$cnf $db -e "DROP TABLE IF EXISTS tag_index"`; `mysql --defaults-file=$cnf $db < $sql_tag_table`; import_sql_file($tag_file, 'tag_index'); } sub fetch_depth_counts { my ($sth, $depths, $sample_id) = @_; my ($row); # # Determine the depth of coverage for the RAD-Tags in this sample # $sth->{'depth'}->execute($sample_id) or die("Unable to select results from $db.\n"); while ($row = $sth->{'depth'}->fetchrow_hashref()) { $depths->{$row->{'tag_id'}}++; } } sub fetch_snp_counts { my ($sth, $snps, $sample_id) = @_; my ($row); # # Determine the number of SNPs contained within each RAD-Tag in this sample. # $sth->{'snps'}->execute($sample_id) or die("Unable to select results from $db.\n"); while ($row = $sth->{'snps'}->fetchrow_hashref()) { $snps->{$row->{'tag_id'}}++; } } sub fetch_catalog_ids { my ($sth, $cats, $batch_id, $sample_id) = @_; my ($row); # # Determine the catalog ID that corresponds to the RAD-Tags in this sample # $sth->{'match'}->execute($batch_id, $sample_id) or die("Unable to select results from $db.\n"); while ($row = $sth->{'match'}->fetchrow_hashref()) { $cats->{$row->{'tag_id'}} = $row->{'catalog_id'}; } } sub import_sql_file { my ($file, $table) = @_; my (@results); @results = `mysql --defaults-file=$cnf $db -e "LOAD DATA LOCAL INFILE '$file' INTO TABLE $table"`; print STDERR "mysql --defaults-file=$cnf $db -e \"LOAD DATA LOCAL INFILE '$file' INTO TABLE $table\"\n", @results; } sub prepare_sql_handles { my ($sth, $outg) = @_; # # Connect to the database, check for the existence of a MySQL config file in the home # directory first, otherwise use the stacks-distributed one. # $sth->{'dbh'} = DBI->connect("DBI:mysql:$db:mysql_read_default_file=$cnf") or die("Unable to connect to the $db MySQL Database!\n" . $DBI::errstr); my $query; $query = "SELECT batch_id, id as sample_id, type FROM samples"; $sth->{'sample'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT unique_tags.id as id, tag_id, seq, deleveraged, blacklisted, removed FROM unique_tags " . "JOIN samples ON (unique_tags.sample_id=samples.id) " . "WHERE relationship='consensus' AND samples.batch_id=? AND unique_tags.sample_id=?"; $sth->{'tags'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT tag_id FROM unique_tags " . "WHERE relationship!='consensus' AND relationship != 'model' AND unique_tags.sample_id=?"; $sth->{'depth'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT tag_id FROM snps " . "JOIN samples ON (snps.sample_id=samples.id) " . "WHERE snps.type='E' AND samples.id=?"; $sth->{'snps'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT tag_id, catalog_id FROM matches " . "WHERE batch_id=? AND sample_id=?"; $sth->{'match'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT catalog_tags.id as id, tag_id, batch_id, chr, bp, strand, seq FROM catalog_tags " . "JOIN batches ON (catalog_tags.batch_id=batches.id) WHERE relationship='consensus'"; $sth->{'cat_tags'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT batch_id, tag_id FROM catalog_snps WHERE type='E'"; $sth->{'cat_snps'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT batch_id, catalog_id as tag_id, sample_id, genotype FROM catalog_genotypes"; $sth->{'cat_geno'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT batch_id, catalog_id, type, uncor_type, progeny, chisq_pval, ratio, lnl FROM markers"; $sth->{'marker'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT samples.batch_id, catalog_id, tag_id, matches.sample_id, allele, type FROM matches " . "JOIN samples ON (samples.id=matches.sample_id)"; $sth->{'cat_matches'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); # "JOIN samples ON (samples.sample_id=matches.sample_id AND samples.batch_id=matches.batch_id)"; $query = "SELECT batch_id, catalog_id, type FROM sequence"; $sth->{'cat_seqs'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT batch_id, catalog_id FROM sequence_blast"; $sth->{'cat_hits'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); $query = "SELECT id, chr, bp, strand, type FROM ref_radome"; $sth->{'radome'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); } sub close_sql_handles { my ($sth) = @_; my $key; foreach $key (keys %{$sth}) { next if ($key =~ /dbh/); $sth->{$key}->finish(); } foreach $key (keys %{$sth}) { next if ($key !~ /dbh/); $sth->{$key}->disconnect(); } } sub parse_command_line { while (@ARGV) { $_ = shift @ARGV; if ($_ =~ /^-d$/) { $debug++; } elsif ($_ =~ /^-D$/) { $db = shift @ARGV; } elsif ($_ =~ /^-s$/) { $sql_path = shift @ARGV; } elsif ($_ =~ /^-c$/) { $catalog_index++; } elsif ($_ =~ /^-t$/) { $tag_index++; } elsif ($_ =~ /^-r$/) { $radome_index++; } elsif ($_ =~ /^-v$/) { version(); exit(); } elsif ($_ =~ /^-h$/) { usage(); } else { print STDERR "Unknown command line options received: $_\n"; usage(); } } if (length($db) == 0) { print STDERR "You must specify a database to index.\n"; usage(); } $sql_path .= "/" if (substr($sql_path, -1, 1) ne "/"); $sql_tag_table = $sql_path . "tag_index.sql"; $sql_cat_table = $sql_path . "catalog_index.sql"; $sql_chr_table = $sql_path . "chr_index.sql"; } sub version { print STDERR "index_radtags.pl ", stacks_version, "\n"; } sub usage { version(); print << "EOQ"; index_radtags.pl -D db [-c] [-t] [-s path] [-d] [-h] D: radtag database to examine. c: generate a catalog index. t: generate a unique tags index. s: path to SQL definition files for catalog/tag index tables (if not in default, installed location). h: display this help message. d: turn on debug output. EOQ exit(0); } stacks-1.44/scripts/integrate_alignments.py000755 000765 000024 00000025062 12710427076 021764 0ustar00catchenstaff000000 000000 #!/usr/bin/env python import optparse import sys import os import gzip import re # # Global configuration variables. # path = "" aln_path = "" out_path = "" batch_id = -1 def parse_command_line(): global out_path global aln_path global path global batch_id p = optparse.OptionParser() # # Add options. # p.add_option("-o", action="store", dest="out_path", help="write modified Stacks files to this output path.") p.add_option("-a", action="store", dest="aln_path", help="SAM file containing catalog locus alignments.") p.add_option("-p", action="store", dest="path", help="path to Stacks directory.") p.add_option("-b", action="store", dest="batch_id", help="Stacks batch ID.") # # Parse the command line # (opts, args) = p.parse_args() if opts.out_path != None: out_path = opts.out_path if opts.aln_path != None: aln_path = opts.aln_path if opts.path != None: path = opts.path if opts.batch_id != None: batch_id = int(opts.batch_id) if len(out_path) == 0 or os.path.exists(out_path) == False: print >> sys.stderr, "You must specify a valid path to write files to." p.print_help() sys.exit() if out_path.endswith("/") == False: out_path += "/" if len(aln_path) == 0 or os.path.exists(aln_path) == False: print >> sys.stderr, "You must specify a valid path to a SAM file containing catalog locus alignments." p.print_help() sys.exit() if len(path) == 0 or os.path.exists(path) == False: print >> sys.stderr, "You must specify a valid path to Stacks input files." p.print_help() sys.exit() if batch_id < 0: pritn >> sys.stderr, "You must specify the batch ID that was supplied to Stacks." p.print_help() sys.exit() if path.endswith("/") == False: path += "/" def find_stacks_files(path, files): try: entries = os.listdir(path) for entry in entries: pos = entry.find(".matches.tsv.gz") if (pos == -1): pos = entry.find(".matches.tsv") if (pos != -1): files.append(entry[0:pos]) print >> sys.stderr, "Found", len(files), "Stacks samples." except: print >> sys.stderr, "Unable to read files from Stacks directory, '", path, "'" def parse_catalog_alignments(aln_path, alns): fh = open(aln_path, "r") for line in fh: line = line.strip("\n") if len(line) == 0 or line[0] == "#" or line[0] == "@": continue parts = line.split("\t") locus = int(parts[0]) chr = parts[2] bp = int(parts[3]) flag = int(parts[1]) # # Check which strand the read is aligned to. # if flag & 16 > 0: alns[locus] = (chr, bp, "-"); else: alns[locus] = (chr, bp, "+"); fh.close() print >> sys.stderr, "Loaded", len(alns), "catalog locus alignments from '", aln_path, "'." def convert_sample(path, file, out_path, alns): matches = {} # # Open the matches file and load the matches to the catalog. # p = path + file + ".matches.tsv.gz" if os.path.exists(p): gzipped = True; fh = gzip.open(p, 'rb') else: gzipped = False; fh = open(path + file + ".matches.tsv", "r") for line in fh: line = line.strip("\n") if len(line) == 0 or line[0] == "#": continue parts = line.split("\t") cat_locus = int(parts[2]) sample_locus = int(parts[4]) matches[sample_locus] = cat_locus fh.close() # # Open the tags file and rewrite it with the alignment coordinates. # if gzipped: fh = gzip.open(path + file + ".tags.tsv.gz", "rb") else: fh = open(path + file + ".tags.tsv", "r") out_fh = open(out_path + file + ".tags.tsv", "w") alns_written = {} for line in fh: if line[0] == "#": out_fh.write(line) continue if len(line) == 0: continue parts = line.split("\t") sample_locus = int(parts[2]) read_type = parts[6] if read_type == "consensus": if sample_locus not in matches: continue; cat_locus = matches[sample_locus] if cat_locus not in alns: continue; (chr, bp, strand) = alns[cat_locus] if sample_locus in alns_written: alns_written[sample_locus] += 1 else: alns_written[sample_locus] = 1; buf = "\t".join(parts[0:3]) + "\t" + chr + "\t" + str(bp) + "\t" + strand + "\t" + "\t".join(parts[6:]) out_fh.write(buf) elif sample_locus in alns_written: out_fh.write(line) fh.close() out_fh.close() # # Open the SNPs, Alleles, and Matches files and rewrite those that had alignment coordinates. # if gzipped: fh = gzip.open(path + file + ".snps.tsv.gz", "rb") else: fh = open(path + file + ".snps.tsv", "r") out_fh = open(out_path + file + ".snps.tsv", "w") for line in fh: if line[0] == "#": out_fh.write(line) continue if len(line) == 0: continue parts = line.split("\t") sample_locus = int(parts[2]) if sample_locus in alns_written: out_fh.write(line) fh.close() out_fh.close() if gzipped: fh = gzip.open(path + file + ".alleles.tsv.gz", "rb") else: fh = open(path + file + ".alleles.tsv", "r") out_fh = open(out_path + file + ".alleles.tsv", "w") for line in fh: if line[0] == "#": out_fh.write(line) continue if len(line) == 0: continue parts = line.split("\t") sample_locus = int(parts[2]) if sample_locus in alns_written: out_fh.write(line) fh.close() out_fh.close() if gzipped: fh = gzip.open(path + file + ".matches.tsv.gz", "rb") else: fh = open(path + file + ".matches.tsv", "r") out_fh = open(out_path + file + ".matches.tsv", "w") for line in fh: if line[0] == "#": out_fh.write(line) continue if len(line) == 0: continue parts = line.split("\t") sample_locus = int(parts[2]) if sample_locus in alns_written: out_fh.write(line) fh.close() out_fh.close() # # If it exists, open the model file and rewrite it with the alignment coordinates. # if gzipped: if os.path.exists(path + file + ".models.tsv.gz") == False: return len(alns_written) elif os.path.exists(path + file + ".models.tsv") == False: return len(alns_written) if gzipped: fh = gzip.open(path + file + ".models.tsv.gz", "rb") else: fh = open(path + file + ".models.tsv", "r") out_fh = open(out_path + file + ".models.tsv", "w") for line in fh: if line[0] == "#": out_fh.write(line) continue if len(line) == 0: continue parts = line.split("\t") sample_locus = int(parts[2]) read_type = parts[6] if sample_locus in alns_written: if read_type == "consensus": (chr, bp, strand) = alns[cat_locus] buf = "\t".join(parts[0:3]) + "\t" + chr + "\t" + str(bp) + "\t" + strand + "\t" + "\t".join(parts[6:]) out_fh.write(buf) else: out_fh.write(line) fh.close() out_fh.close() return len(alns_written) def convert_catalog(path, batch_id, out_path, alns): # # Open the tags file and rewrite it with the alignment coordinates. # p = path + "batch_" + str(batch_id) + ".catalog.tags.tsv.gz" if os.path.exists(p): gzipped = True; fh = gzip.open(path + "batch_" + str(batch_id) + ".catalog.tags.tsv.gz", "rb") else: gzipped = False; fh = open(path + "batch_" + str(batch_id) + ".catalog.tags.tsv", "r") out_fh = open(out_path + "batch_" + str(batch_id) + ".catalog.tags.tsv", "w") alns_written = {} for line in fh: if line[0] == "#": out_fh.write(line) continue if len(line) == 0: continue parts = line.split("\t") cat_locus = int(parts[2]) if cat_locus not in alns: continue; (chr, bp, strand) = alns[cat_locus] if cat_locus in alns_written: alns_written[cat_locus] += 1 else: alns_written[cat_locus] = 1; buf = "\t".join(parts[0:3]) + "\t" + chr + "\t" + str(bp) + "\t" + strand + "\t" + "\t".join(parts[6:]) out_fh.write(buf) fh.close() out_fh.close() if gzipped: fh = gzip.open(path + "batch_" + str(batch_id) + ".catalog.snps.tsv.gz", "rb") else: fh = open(path + "batch_" + str(batch_id) + ".catalog.snps.tsv", "r") out_fh = open(out_path + "batch_" + str(batch_id) + ".catalog.snps.tsv", "w") for line in fh: if line[0] == "#": out_fh.write(line) continue if len(line) == 0: continue parts = line.split("\t") cat_locus = int(parts[2]) if cat_locus in alns_written: out_fh.write(line) fh.close() out_fh.close() if gzipped: fh = gzip.open(path + "batch_" + str(batch_id) + ".catalog.alleles.tsv.gz", "rb") else: fh = open(path + "batch_" + str(batch_id) + ".catalog.alleles.tsv", "r") out_fh = open(out_path + "batch_" + str(batch_id) + ".catalog.alleles.tsv", "w") for line in fh: if line[0] == "#": out_fh.write(line) continue if len(line) == 0: continue parts = line.split("\t") cat_locus = int(parts[2]) if cat_locus in alns_written: out_fh.write(line) fh.close() out_fh.close() return len(alns_written) parse_command_line() files = [] alns = {} find_stacks_files(path, files) parse_catalog_alignments(aln_path, alns) i = 1 for file in files: print >> sys.stderr, "Processing file", str(i), "of", len(files), "['" + file + "']" cnt = convert_sample(path, file, out_path, alns) print >> sys.stderr, " Added alignments for", cnt, "loci." i += 1 # # Now process the catalog. # print >> sys.stderr, "Processing the catalog" cnt = convert_catalog(path, batch_id, out_path, alns) print >> sys.stderr, " Added alignments for", cnt, "catalog loci." stacks-1.44/scripts/load_radtags.pl000755 000765 000024 00000035362 12533677757 020214 0ustar00catchenstaff000000 000000 #!/usr/bin/env perl # # Copyright 2011-2014, Julian Catchen # # This file is part of Stacks. # # Stacks is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation, either version 3 of the License, or # (at your option) any later version. # # Stacks is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with Stacks. If not, see . # # # Load a set of output files from the Stacks pipeline into a Stacks MySQL database. # # By Julian Catchen # use strict; use POSIX; use File::Temp qw/ mktemp /; use File::Spec; use constant stacks_version => "_VERSION_"; my $mysql_config = "_PKGDATADIR_" . "sql/mysql.cnf"; my $dry_run = 0; my $db = ""; my $in_path = "."; my $sample_id = 0; my $desc = ""; my $date = ""; my $batch_id = 0; my $batch = 0; my $catalog = 0; my $stacks_type = ""; my $popmap_path = ""; my $ignore_tags = 0; my $white_list = ""; parse_command_line(); my $cnf = (-e $ENV{"HOME"} . "/.my.cnf") ? $ENV{"HOME"} . "/.my.cnf" : $mysql_config; if (length($date) == 0) { $date = strftime("%Y-%m-%d", (localtime(time))); } my (@results, @files, @catalog, @parent_ids, @pop_ids, %pops, %sample_ids); build_file_list(\@files, \@catalog); extract_parental_ids(scalar(@files), \@catalog, \@parent_ids); extract_sample_ids(\@files, \%sample_ids); parse_population_map(\@files, \%sample_ids, \@pop_ids, \%pops); print STDERR "Stacks pipeline type: '", $stacks_type, "'\n", scalar(@files), " files to process: ", join(", ", @files), "\n", scalar(@catalog), " catalog files to process: ", join(", ", @catalog), "\n"; if ($stacks_type eq "map") { print STDERR scalar(@parent_ids), " parent IDs identified: ", join(", ", @parent_ids), "\n"; } if ($batch) { if (!$dry_run) { @results = `mysql --defaults-file=$cnf $db -e "INSERT INTO batches SET id=$batch_id, description='$desc', date='$date', type='$stacks_type'"`; } print STDERR "mysql --defaults-file=$cnf $db ", "-e \"INSERT INTO batches SET id=$batch_id, description='$desc', date='$date', type='$stacks_type'\"\n", @results; } my ($file, $f, $i, $cnt, $type, $pop_id); # # Import the catalog # if ($catalog) { foreach $file (@catalog) { $f = $in_path . "/$file" . ".catalog.tags.tsv"; if (-e $f) { import_sql_file($f, "catalog_tags", 1); } elsif (-e $f . ".gz") { $f = $in_path . "/$file" . ".catalog.tags.tsv.gz"; import_gzsql_file($f, "catalog_tags", 1); } $f = $in_path . "/$file" . ".catalog.snps.tsv"; if (-e $f) { import_sql_file($f, "catalog_snps", 1); } elsif (-e $f . ".gz") { $f = $in_path . "/$file" . ".catalog.snps.tsv.gz"; import_gzsql_file($f, "catalog_snps", 1); } $f = $in_path . "/$file" . ".catalog.alleles.tsv"; if (-e $f) { import_sql_file($f, "catalog_alleles", 1); } elsif (-e $f . ".gz") { $f = $in_path . "/$file" . ".catalog.alleles.tsv.gz"; import_gzsql_file($f, "catalog_alleles", 1); } } } if ($stacks_type eq "map") { $f = "$in_path/batch_" . $batch_id . ".markers.tsv"; import_sql_file($f, "markers", 1); $f = "$in_path/batch_" . $batch_id . ".genotypes_1.txt"; import_sql_file($f, "catalog_genotypes", 1); } elsif ($stacks_type eq "population") { $f = "$in_path/batch_" . $batch_id . ".markers.tsv"; import_sql_file($f, "markers", 1); $f = "$in_path/batch_" . $batch_id . ".sumstats.tsv"; import_sql_file($f, "sumstats", scalar(keys %pops) + 1); $f = "$in_path/batch_" . $batch_id . ".hapstats.tsv"; import_sql_file($f, "hapstats", scalar(keys %pops) + 1); # # Import the Fst files. # my $fst_cnt = 0; my (@keys, $m, $n); @keys = sort keys %pops; for ($m = 0; $m < scalar(@keys); $m++) { for ($n = 0; $n < scalar(@keys); $n++) { $f = "$in_path/batch_" . $batch_id . ".fst_" . $keys[$m] . "-" . $keys[$n] . ".tsv"; if (-e $file) { import_sql_file($f, "fst", 1); $fst_cnt++; } } } print STDERR "Imported $fst_cnt SNP Fst file(s).\n"; # # Import the Phi_st files. # $fst_cnt = 0; for ($m = 0; $m < scalar(@keys); $m++) { for ($n = 0; $n < scalar(@keys); $n++) { $f = "$in_path/batch_" . $batch_id . ".phistats_" . $keys[$m] . "-" . $keys[$n] . ".tsv"; if (-e $file) { import_sql_file($f, "phist", 3); $fst_cnt++; } } } print STDERR "Imported $fst_cnt Haplotype Fst file(s).\n"; } $i = 1; $cnt = scalar(@files); foreach $file (sort {$sample_ids{$a} <=> $sample_ids{$b}} @files) { print STDERR "Processing sample $i of $cnt\n"; $f = $in_path . "/$file" . ".matches.tsv"; if (-e $f) { import_sql_file($f, "matches", 1); } elsif (-e $f . ".gz") { $f = $in_path . "/$file" . ".matches.tsv.gz"; import_gzsql_file($f, "matches", 1); } $i++; } $i = 1; foreach $file (sort {$sample_ids{$a} <=> $sample_ids{$b}} @files) { print STDERR "Processing sample $i of $cnt\n"; # # Pull out the sample ID and insert it into the database # $sample_id = $sample_ids{$file}; if ($stacks_type eq "map") { $type = (grep(/^$sample_id$/, @parent_ids) > 0) ? 'parent' : 'progeny'; } else { $type = "sample"; } $pop_id = shift(@pop_ids); if (!$dry_run) { @results = `mysql --defaults-file=$cnf $db -e "INSERT INTO samples SET id=$sample_id, sample_id=$sample_id, batch_id=$batch_id, type='$type', file='$file', pop_id='$pop_id'"`; } print STDERR "mysql --defaults-file=$cnf $db ", "-e \"INSERT INTO samples SET id=$sample_id, sample_id=$sample_id, batch_id=$batch_id, type='$type', file='$file', pop_id='$pop_id'\"\n", @results; $f = $in_path . "/$file" . ".tags.tsv"; if (-e $f) { import_sql_file($f, "unique_tags", 1) if ($ignore_tags == 0); } elsif (-e $f . ".gz") { $f = $in_path . "/$file" . ".tags.tsv.gz"; import_gzsql_file($f, "unique_tags", 1) if ($ignore_tags == 0); } $f = $in_path . "/$file" . ".snps.tsv"; if (-e $f) { import_sql_file($f, "snps", 1); } elsif (-e $f . ".gz") { $f = $in_path . "/$file" . ".snps.tsv.gz"; import_gzsql_file($f, "snps", 1); } $f = $in_path . "/$file" . ".alleles.tsv"; if (-e $f) { import_sql_file($f, "alleles", 1); } elsif (-e $f . ".gz") { $f = $in_path . "/$file" . ".alleles.tsv.gz"; import_gzsql_file($f, "alleles", 1); } $i++; } print STDERR "\nDon't forget to index your Stacks database -- run index_radtags.pl\n\n"; sub parse_population_map { my ($samples, $sample_ids, $pop_ids, $pops) = @_; my ($fh, @parts, $line, %ids, $file, $path); if (length($popmap_path) == 0) { foreach $path (@{$samples}) { push(@{$pop_ids}, 1); $pops->{1}++; } return; } open($fh, "<$popmap_path") or die("Unable to open population map, '$popmap_path', $!\n"); while ($line = <$fh>) { chomp $line; @parts = split(/\t/, $line); if (scalar(@parts) > 3) { die("Unable to parse population map, '$popmap_path' (map should contain no more than three columns).\n"); } $ids{$parts[0]} = $parts[1]; } foreach $file (sort {$sample_ids->{$a} <=> $sample_ids->{$b}} @{$samples}) { if (!defined($ids{$file})) { die("Unable to find '$file' in the population map, '$popmap_path'.\n"); } push(@{$pop_ids}, $ids{$file}); $pops->{$ids{$file}}++; } print STDERR "Parsed population map: ", scalar(@{$samples}), " files in ", scalar(keys %pops), " populations.\n"; close($fh); } sub extract_parental_ids { my ($sample_cnt, $catalog, $parental_ids) = @_; my ($fh, $prefix, $path, $line, @parts, $tag_id, @tag_ids, $id, $tag, %ids); print STDERR "Scanning catalog for sample IDs..."; foreach $prefix (@catalog) { $path = $in_path . "/" . $prefix . ".catalog.tags.tsv"; if (-e $path) { open($fh, "<$path") or die("Unable to open catalog file: '$path', $!\n"); } elsif (-e $path . ".gz") { open($fh, "gunzip -c " . $path . ".gz |") or die("Unable to open catalog file: '$path', $!\n"); } while ($line = <$fh>) { chomp $line; @parts = split(/\t/, $line); @tag_ids = split(/,/, $parts[8]); foreach $tag_id (@tag_ids) { ($id, $tag) = split(/_/, $tag_id); $ids{$id}++; } } close($fh); } @{$parental_ids} = keys %ids; # # Determine the type of pipeline run: either a 'map' or a 'population' type. # If all samples are parental, i.e. in the catalog, then this is a population type # otherwise, it is a map type. # if (length($stacks_type) == 0) { $stacks_type = (scalar(@{$parental_ids}) == $sample_cnt) ? "population" : "map"; } print STDERR "done.\n"; } sub extract_sample_ids { my ($files, $sample_ids) = @_; my ($file, $f, $line, @results, @parts); print STDERR "Collecting sample IDs from Stacks output files..."; foreach $file (@{$files}) { $f = $in_path . "/$file" . ".tags.tsv"; if (-e $f) { @results = `head -n 2 $f | tail -n 1`; } elsif (-e $f . ".gz") { $f = $in_path . "/$file" . ".tags.tsv.gz"; @results = `gunzip -c $f | head -n 2 | tail -n 1`; } else { die("Unable to find file $f\n"); } chomp $results[0]; @parts = split(/\t/, $results[0]); # # Sample ID is expected to be the first column in the *.tags.tsv file. # $sample_ids->{$file} = $parts[1]; } print STDERR "done.\n"; } sub import_sql_file { my ($file, $table, $skip_lines) = @_; my (@results, $ignore); if (!-e $file) { print STDERR "File '$file' does not exist.\n"; return; } $ignore = " IGNORE $skip_lines LINES" if ($skip_lines > 0); if (!$dry_run) { @results = `mysql --defaults-file=$cnf $db -e "LOAD DATA LOCAL INFILE '$file' INTO TABLE $table$ignore"`; } print STDERR "mysql --defaults-file=$cnf $db ", "-e \"LOAD DATA LOCAL INFILE '$file' INTO TABLE $table$ignore\"\n", @results; } sub import_gzsql_file { my ($file, $table, $skip_lines) = @_; my (@results, $ignore); if (!-e $file) { print STDERR "File '$file' does not exist.\n"; return; } $ignore = "IGNORE $skip_lines LINES" if ($skip_lines > 0); # # Get a temporary file name and create a named pipe. # my $tmpdir = File::Spec->tmpdir(); my $named_pipe = mktemp($tmpdir . "/denovo_map_XXXXXX"); if ($dry_run == 0) { mkfifo($named_pipe, 0700) || die("Unable to create named pipe for loading gzipped data: $named_pipe, $!"); } print STDERR "Streaming $file into named pipe $named_pipe.\n"; # # Dump our gzipped data onto the named pipe. # system("gunzip -c $file > $named_pipe &") if ($dry_run == 0); @results = `mysql --defaults-file=$cnf $db -e "LOAD DATA LOCAL INFILE '$named_pipe' INTO TABLE $table $ignore"` if ($dry_run == 0); print STDERR "mysql --defaults-file=$cnf $db -e \"LOAD DATA LOCAL INFILE '$named_pipe' INTO TABLE $table $ignore\"\n", @results; # # Remove the pipe. # unlink($named_pipe) if ($dry_run == 0); } sub build_file_list { my ($files, $catalog_files) = @_; my (@wl, @ls, $line, $prefix); # Load a white list of files to process if it is supplied. if (length($white_list) > 0) { load_white_list(\@wl); } @ls = `ls -1 $in_path/*.tags.tsv* 2> /dev/null`; if (scalar(@ls) == 0) { print STDERR "Unable to locate any input files to process within '$in_path'\n"; usage(); } foreach $line (@ls) { chomp $line; if ($line =~ /\.tags\.tsv\.gz$/) { ($prefix) = ($line =~ /$in_path\/(.+)\.tags\.tsv\.gz/); } else { ($prefix) = ($line =~ /$in_path\/(.+)\.tags\.tsv/); } next if ($prefix =~ /catalog/); if (scalar(@wl) > 0) { next if (!grep(/^$prefix$/, @wl)); } push(@{$files}, $prefix); } if ($catalog > 0) { @ls = `ls -1 $in_path/*.catalog.tags.tsv* 2> /dev/null`; if (scalar(@ls) == 0) { print STDERR "Unable to locate any catalog input files to process within '$in_path'\n"; usage(); } foreach $line (@ls) { chomp $line; if ($line =~ /\.catalog\.tags\.tsv\.gz$/) { ($prefix) = ($line =~ /$in_path\/(.+)\.catalog\.tags\.tsv\.gz/); } else { ($prefix) = ($line =~ /$in_path\/(.+)\.catalog\.tags\.tsv/); } if (scalar(@wl) > 0) { next if (!grep(/^$prefix$/, @wl)); } push(@{$catalog_files}, $prefix); } } } sub load_white_list { my ($wl) = @_; open(WHITE, "<" . $white_list) or die("Unable to open white list file '$white_list': $!\n"); my $line = ""; while ($line = ) { chomp $line; next if (length($line) == 0); push(@{$wl}, $line); } close(WHITE); } sub parse_command_line { while (@ARGV) { $_ = shift @ARGV; if ($_ =~ /^-p$/) { $in_path = shift @ARGV; } elsif ($_ =~ /^-D$/) { $db = shift @ARGV; } elsif ($_ =~ /^-c$/) { $catalog++; } elsif ($_ =~ /^-B$/) { $batch++; } elsif ($_ =~ /^-b$/) { $batch_id = shift @ARGV; } elsif ($_ =~ /^-e$/) { $desc = shift @ARGV; } elsif ($_ =~ /^-a$/) { $date = shift @ARGV; } elsif ($_ =~ /^-W$/) { $white_list = shift @ARGV; } elsif ($_ =~ /^-t$/) { $stacks_type = shift @ARGV; } elsif ($_ =~ /^-M$/) { $popmap_path = shift @ARGV; } elsif ($_ =~ /^-U$/) { $ignore_tags++; } elsif ($_ =~ /^-d$/) { $dry_run++; } elsif ($_ =~ /^-v$/) { version(); exit(); } elsif ($_ =~ /^-h$/) { usage(); } else { print STDERR "Unknown command line option: '$_'\n"; usage(); } } $in_path = substr($in_path, 0, -1) if (substr($in_path, -1) eq "/"); if (length($db) == 0) { print STDERR "You must specify a database.\n"; usage(); } } sub version { print STDERR "load_radtags.pl ", stacks_version, "\n"; } sub usage { version(); print STDERR < # # This file is part of Stacks. # # Stacks is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation, either version 3 of the License, or # (at your option) any later version. # # Stacks is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with Stacks. If not, see . # use strict; use DBI; use constant stacks_version => "_VERSION_"; my $debug = 0; my $mysql_config = "_PKGDATADIR_" . "sql/mysql.cnf"; my $db = ""; my $batch_id = 0; my $in_path = ""; my $seq_path = ""; my $seq_type = ""; parse_command_line(); my (%sth); prepare_sql_handles(\%sth); if ($seq_type eq "est") { load_ests(\%sth); } elsif ($seq_type eq "pe_radtag") { load_pe_radtags(\%sth); } else { print STDERR "Unknown sequence type specified '$seq_type'\n"; } sub load_pe_radtags { my ($sth) = @_; # # Open the file containing sequences and load them into the database. # my ($fh, $line, $id, $buf, $catalog_id, $seq_id, $count); open($fh, "<$seq_path") or die("Unable to open FASTA input file: '$seq_path', $!\n"); $count = 0; while ($line = <$fh>) { chomp $line; if (substr($line, 0, 1) eq ">") { if (length($buf) > 0) { ($catalog_id, $seq_id) = split(/\|/, $id); $sth->{'load'}->execute($batch_id, $catalog_id, $seq_type, $seq_id, $buf) or die("Unable to insert results into $db.\n"); $count++; $buf = ""; } $id = substr($line, 1); } else { $buf .= $line; } } if (length($buf) > 0 && length($id) > 0) { ($catalog_id, $seq_id) = split(/\|/, $id); $sth->{'load'}->execute($batch_id, $catalog_id, $seq_type, $seq_id, $buf) or die("Unable to insert results into $db.\n"); $count++; } print STDERR "Loaded $count RAD-Tag paired-end contigs into the database.\n"; close($fh); close_sql_handles(\%sth); } sub load_ests { my ($sth) = @_; # # Load a TSV file of BLAST hits linking sequences to markers in the catalog. # my %hits; load_blast_hits(\%hits); my ($fh, $line, $id, $buf, $catalog_id, $count); # # Open the file containing sequences. For each sequence that has a BLAST # hit to a marker in the catalog, load the sequence into the database. # open($fh, "<$seq_path") or die("Unable to open FASTA input file: '$seq_path', $!\n"); $count = 0; while ($line = <$fh>) { chomp $line; if (substr($line, 0, 1) eq ">") { if (length($buf) > 0) { if (defined($hits{$id})) { foreach $catalog_id (@{$hits{$id}}) { $sth->{'load'}->execute($batch_id, $catalog_id, $seq_type, $id, $buf) or die("Unable to insert results into $db.\n"); $count++; } } $buf = ""; } $id = substr($line, 1); } else { $buf .= $line; } } if (length($buf) > 0 && length($id) > 0) { if (defined($hits{$id})) { foreach $catalog_id (@{$hits{$id}}) { $sth->{'load'}->execute($batch_id, $catalog_id, $seq_type, $id, $buf) or die("Unable to insert results into $db.\n"); $count++; } } } print STDERR "Loaded $count EST sequences into the database.\n"; close($fh); close_sql_handles(\%sth); } sub load_blast_hits { my ($hits) = @_; my (@parts, $line); open(HITS, "<" . $in_path) or die("Unable to open BLAST hits file '$in_path' $!\n"); # # First line is a header. # $line = ; while ($line = ) { chomp $line; @parts = split(/\t/, $line); if (!defined($hits->{$parts[7]})) { $hits->{$parts[7]} = []; } if (!grep(/^$parts[2]$/, @{$hits->{$parts[7]}})) { push(@{$hits->{$parts[7]}}, $parts[2]); } } print STDERR "Loaded ", scalar(keys %{$hits}), " distinct BLAST hits.\n"; close(HITS); } sub prepare_sql_handles { my ($sth) = @_; my $cnf = (defined($ENV{"HOME"}) && -e $ENV{"HOME"} . "/.my.cnf") ? $ENV{"HOME"} . "/.my.cnf" : $mysql_config; $sth->{'dbh'} = DBI->connect("DBI:mysql:$db:mysql_read_default_file=$cnf") or die("Unable to connect to the $db MySQL Database!\n" . $DBI::errstr); my $query; $query = "INSERT INTO sequence SET batch_id=?, catalog_id=?, type=?, seq_id=?, seq=?"; $sth->{'load'} = $sth->{'dbh'}->prepare($query) or die($sth->{'dbh'}->errstr()); } sub close_sql_handles { my ($sth) = @_; my $key; foreach $key (keys %{$sth}) { next if ($key =~ /dbh/); $sth->{$key}->finish(); } foreach $key (keys %{$sth}) { next if ($key !~ /dbh/); $sth->{$key}->disconnect(); } } sub parse_command_line { while (@ARGV) { $_ = shift @ARGV; if ($_ =~ /^-p$/) { $in_path = shift @ARGV; } elsif ($_ =~ /^-f$/) { $seq_path = shift @ARGV; } elsif ($_ =~ /^-b$/) { $batch_id = shift @ARGV; } elsif ($_ =~ /^-D$/) { $db = shift @ARGV; } elsif ($_ =~ /^-t$/) { $seq_type = shift @ARGV; } elsif ($_ =~ /^-d$/) { $debug++; } elsif ($_ =~ /^-h$/) { usage(); } else { print STDERR "Unknown command line option: '$_'\n"; usage(); } } if (length($seq_type) == 0) { print STDERR "You must specify the sequence type.\n"; usage(); } if ($batch_id == 0) { print STDERR "You must specify the batch ID.\n"; usage(); } } sub version { print STDERR "load_sequences.pl ", stacks_version, "\n"; } sub usage { version(); print STDERR < # # This file is part of Stacks. # # Stacks is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation, either version 3 of the License, or # (at your option) any later version. # # Stacks is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with Stacks. If not, see . # # # Process the data for a genetic map: build stacks in parents and progeny, # create a catalog from the parents, and match progeny against the catatlog. # Call genotypes, and load all data into an MySQL database along the way. # # For the database interactions to work, the 'mysql' program is expected to be # on the path and sufficient permissions set to access the specified database. # use strict; use POSIX; use File::Temp qw/ mktemp /; use File::Spec; use constant stacks_version => "_VERSION_"; use constant true => 1; use constant false => 0; my $dry_run = false; my $sql = true; my $create_db = false; my $overw_db = false; my $mysql_config = "_PKGDATADIR_" . "sql/mysql.cnf"; my $mysql_tables = "_PKGDATADIR_" . "sql/stacks.sql"; my $exe_path = "_BINDIR_"; my $out_path = ""; my $popmap_path = ""; my $sample_path = ""; my $db = ""; my $data_type = "map"; my $min_cov = 0; my $min_rcov = 0; my $batch_id = -1; my $sample_id = 1; my $desc = ""; # Database description of this dataset my $date = ""; # Date relevent to this data, formatted for SQL: 2009-05-31 my $gzip = false; my @parents; my @progeny; my @samples; my (@_pstacks, @_cstacks, @_sstacks, @_genotypes, @_populations); my $cmd_str = $0 . " " . join(" ", @ARGV); parse_command_line(); my $cnf = (-e $ENV{"HOME"} . "/.my.cnf") ? $ENV{"HOME"} . "/.my.cnf" : $mysql_config; # # Check for the existence of the necessary pipeline programs # die ("Unable to find '" . $exe_path . "pstacks'.\n") if (!-e $exe_path . "pstacks" || !-x $exe_path . "pstacks"); die ("Unable to find '" . $exe_path . "cstacks'.\n") if (!-e $exe_path . "cstacks" || !-x $exe_path . "cstacks"); die ("Unable to find '" . $exe_path . "sstacks'.\n") if (!-e $exe_path . "sstacks" || !-x $exe_path . "sstacks"); die ("Unable to find '" . $exe_path . "genotypes'.\n") if (!-e $exe_path . "genotypes" || !-x $exe_path . "genotypes"); die ("Unable to find '" . $exe_path . "populations'.\n") if (!-e $exe_path . "populations" || !-x $exe_path . "populations"); die ("Unable to find '" . $exe_path . "index_radtags.pl'.\n") if (!-e $exe_path . "index_radtags.pl" || !-x $exe_path . "index_radtags.pl"); my ($log, $log_fh, $sample); my (@sample_list, %pop_ids, %pops, %grp_ids, %grps, %sample_ids); parse_population_map(\@sample_list, \%pop_ids, \%pops, \%grp_ids, \%grps); initialize_samples(\@parents, \@progeny, \@samples, \@sample_list, \%pop_ids, \%grp_ids); # # Open the log file # $log = "$out_path/ref_map.log"; open($log_fh, ">$log") or die("Unable to open log file '$log'; $!\n"); print $log_fh "ref_map.pl version ", stacks_version, " started at ", strftime("%Y-%m-%d %H:%M:%S", (localtime(time))), "\n", $cmd_str, "\n\n"; initialize_database($log_fh, \@parents, \@progeny, \@samples, \%sample_ids) if ($sql == true); execute_stacks($log_fh, $sample_id, \@parents, \@progeny, \@samples, \%sample_ids); load_sql_data($log_fh, \%pops, \@parents, \@progeny, \@samples) if ($sql == true); print $log_fh "\nref_map.pl completed at ", strftime("%Y-%m-%d %H:%M:%S", (localtime(time))), "\n"; close($log_fh); sub execute_stacks { my ($log_fh, $sample_id, $parents, $progeny, $samples, $sample_ids) = @_; my (@results, @depths_of_cov); my ($pop_cnt, $sample, $num_files, $i, $cmd, $pipe_fh, $path, $cat_file); my $minc = $min_cov > 0 ? "-m $min_cov" : ""; my $minrc = $min_rcov > 0 ? "-m $min_rcov" : $minc; $i = 1; $num_files = scalar(@{$parents}) + scalar(@{$progeny}) + scalar(@{$samples}); # # Assemble RAD loci in each individual. # foreach $sample (@parents, @progeny, @samples) { printf("Identifying unique stacks; file % 3s of % 3s [%s]\n", $i, $num_files, $sample->{'file'}); printf($log_fh "Identifying unique stacks; file % 3s of % 3s [%s]\n", $i, $num_files, $sample->{'file'}); if (scalar(keys %{$sample_ids}) > 0) { $sample_id = $sample_ids->{$sample->{'file'}}; } $path = $sample->{'path'} . $sample->{'file'} . "." . $sample->{'suffix'}; if ($sample->{'type'} eq "sample") { $cmd = $exe_path . "pstacks -t $sample->{'fmt'} -f $path -o $out_path -i $sample_id $minc " . join(" ", @_pstacks) . " 2>&1"; } elsif ($sample->{'type'} eq "parent") { $cmd = $exe_path . "pstacks -t $sample->{'fmt'} -f $path -o $out_path -i $sample_id $minc " . join(" ", @_pstacks) . " 2>&1"; } elsif ($sample->{'type'} eq "progeny") { $cmd = $exe_path . "pstacks -t $sample->{'fmt'} -f $path -o $out_path -i $sample_id $minrc " . join(" ", @_pstacks) . " 2>&1"; } print STDERR " $cmd\n"; print $log_fh "$cmd\n"; if ($dry_run == false) { @results = `$cmd`; # # Pull the depth of coverage from pstacks. # my @lines = grep(/Mean coverage depth is/, @results); my ($depth) = ($lines[0] =~ /^ Mean coverage depth is (\d+\.?\d*); Std Dev: .+; Max: .+$/); push(@depths_of_cov, [$sample->{'file'}, $depth]); } write_results(\@results, $log_fh); $i++; $sample_id++; } write_depths_of_cov(\@depths_of_cov, $log_fh); # # Generate catalog of RAD loci. # print STDERR "Generating catalog...\n"; my $file_paths = ""; foreach $sample (@parents, @samples) { $file_paths .= "-s $out_path/$sample->{'file'} "; } $cmd = $exe_path . "cstacks -g -b $batch_id -o $out_path $file_paths " . join(" ", @_cstacks) . " 2>&1"; print STDERR " $cmd\n"; print $log_fh "$cmd\n\n"; if ($dry_run == false) { open($pipe_fh, "$cmd |"); while (<$pipe_fh>) { print $log_fh $_; if ($_ =~ /failed/i) { print STDERR "Catalog construction failed.\n"; exit(1); } } close($pipe_fh); } # # Match parents, progeny, or samples to the catalog. # $file_paths = ""; print STDERR "Matching samples to the catalog...\n"; foreach $sample (@parents, @progeny, @samples) { $file_paths .= "-s $out_path/$sample->{'file'} "; } $cat_file = "batch_" . $batch_id; $cmd = $exe_path . "sstacks -g -b $batch_id -c $out_path/$cat_file -o $out_path $file_paths " . join(" ", @_sstacks) . " 2>&1"; print STDERR " $cmd\n"; print $log_fh "$cmd\n\n"; if ($dry_run == false) { open($pipe_fh, "$cmd |"); while (<$pipe_fh>) { print $log_fh $_; } close($pipe_fh); } if ($data_type eq "map") { # # Generate a set of observed haplotypes and a set of markers and generic genotypes # printf(STDERR "Generating genotypes...\n"); $cmd = $exe_path . "genotypes -b $batch_id -P $out_path -r 1 -c -s " . join(" ", @_genotypes) . " 2>&1"; print STDERR "$cmd\n"; print $log_fh "$cmd\n"; if ($dry_run == 0) { open($pipe_fh, "$cmd |"); while (<$pipe_fh>) { print $log_fh $_; } close($pipe_fh); } } else { printf(STDERR "Calculating population-level summary statistics\n"); $cmd = $exe_path . "populations -b $batch_id -P $out_path -s " . join(" ", @_populations) . " 2>&1"; print STDERR "$cmd\n"; print $log_fh "$cmd\n"; if ($dry_run == 0) { open($pipe_fh, "$cmd |"); while (<$pipe_fh>) { print $log_fh $_; } close($pipe_fh); } } } sub parse_population_map { my ($sample_list, $pop_ids, $pops, $grp_ids, $grps) = @_; my ($fh, @parts, $line, $sample); return if (length($popmap_path) == 0); open($fh, "<$popmap_path") or die("Unable to open population map, '$popmap_path', $!\n"); while ($line = <$fh>) { chomp $line; next if ($line =~ /^\s*#/); @parts = split(/\t/, $line); if (scalar(@parts) > 3) { die("Unable to parse population map, '$popmap_path' (map should contain no more than three columns).\n"); } push(@{$sample_list}, $parts[0]); $pop_ids->{$parts[0]} = $parts[1]; $pops->{$parts[1]}++; if (scalar(@parts) > 2) { $grp_ids->{$parts[0]} = $parts[2]; $grps->{$parts[2]}++; } } if (scalar(keys %{$grps}) == 0) { $grps->{"1"}++; foreach $sample (@{$sample_list}) { $grp_ids->{$sample} = "1"; } } print STDERR "Parsed population map: ", scalar(@{$sample_list}), " files in ", scalar(keys %{$pops}); scalar(keys %{$pops}) == 1 ? print STDERR " population" : print STDERR " populations"; print STDERR " and ", scalar(keys %{$grps}); scalar(keys %{$grps}) == 1 ? print STDERR " group.\n" : print STDERR " groups.\n"; close($fh); } sub initialize_samples { my ($parents, $progeny, $samples, $sample_list, $pop_ids, $grp_ids) = @_; my ($local_gzip, $file, $prefix, $suffix, $path, $found, $i); if (scalar(@{$sample_list}) > 0 && scalar(@{$samples}) == 0) { my @suffixes = ("sam", "bam", "map", "bowtie"); my @fmts = ("sam", "bam", "map", "bowtie"); # # If a population map was specified and no samples were provided on the command line. # foreach $sample (@{$sample_list}) { $found = false; for ($i = 0; $i < scalar(@suffixes); $i++) { $path = $sample_path . $sample . "." . $suffixes[$i]; if (-e $path) { if ($i == 1) { $gzip = true; } push(@{$samples}, {'path' => $sample_path, 'file' => $sample, 'suffix' => $suffixes[$i], 'type' => "sample", 'fmt' => $fmts[$i]}); $found = true; last; } } if ($found == false) { die("Unable to find sample '$sample' in directory '$sample_path' as specified in the population map, '$popmap_path'.\n"); } } } else { # # Process any samples were specified on the command line. # foreach $sample (@{$parents}, @{$progeny}, @{$samples}) { ($prefix, $suffix) = ($sample->{'path'} =~ /^(.+)\.(.+)$/); $sample->{'suffix'} = $suffix; if ($prefix =~ /^.*\/.+$/) { ($path, $file) = ($prefix =~ /^(.*\/)(.+)$/); } else { $file = $prefix; $path = ""; } $sample->{'path'} = $path; $sample->{'file'} = $file; if ($suffix =~ /^bam$/) { $sample->{'fmt'} = "bam"; $gzip = true; } elsif ($suffix =~ /^sam$/) { $sample->{'fmt'} = "sam"; } elsif ($suffix =~ /^map$/) { $sample->{'fmt'} = "map"; } elsif ($suffix =~ /^bowtie$/) { $sample->{'fmt'} = "bowtie"; } else { die("Unknown input file type for file '" . $sample->{'path'} . "'.\n"); } $path = $sample->{'path'} . $sample->{'file'} . "." . $sample->{'suffix'}; if (!-e $path) { die("Unable to locate sample file '$path'\n"); } } foreach $sample (@{$parents}) { $sample->{'type'} = "parent"; } foreach $sample (@{$progeny}) { $sample->{'type'} = "progeny"; } foreach $sample (@{$samples}) { $sample->{'type'} = "sample"; } } # # If a population map was specified, make sure all samples in the list were found (and vice versa) and assign popualtion IDs. # if (scalar(@{$sample_list}) > 0) { my %sample_hash; foreach $sample (@{$samples}) { $sample_hash{$sample->{'file'}}++; if (!defined($pop_ids->{$sample->{'file'}})) { die("Unable to find an entry for '" . $sample->{'file'} . "' in the population map, '$popmap_path'.\n"); } else { $sample->{'pop_id'} = $pop_ids->{$sample->{'file'}}; } if (!defined($grp_ids->{$sample->{'file'}})) { die("Unable to find an entry for '" . $sample->{'file'} . "' in the population map, '$popmap_path'.\n"); } else { $sample->{'grp_id'} = $grp_ids->{$sample->{'file'}}; } } foreach $sample (@{$sample_list}) { if (!defined($sample_hash{$sample})) { die("Unable to find a file corresponding to the population map entry '" . $sample . "' in the population map, '$popmap_path'.\n"); } } } else { foreach $sample (@{$parents}, @{$progeny}, @{$samples}) { $sample->{'pop_id'} = "1"; $sample->{'grp_id'} = "1"; $pop_ids->{$sample->{'file'}} = $sample->{'pop_id'}; $grp_ids->{$sample->{'file'}} = $sample->{'grp_id'}; } } # # Check that no duplicate files were specified. # my (%files, $file); foreach $file (@{$parents}, @{$progeny}, @{$samples}) { $files{$file}++; } foreach $file (keys %files) { if ($files{$file} > 1) { die("A duplicate file was specified which may create undefined results, '$file'\n"); } } print STDERR "Found ", scalar(@{$parents}), " parental file(s).\n" if (scalar(@{$parents}) > 0); print STDERR "Found ", scalar(@{$progeny}), " progeny file(s).\n" if (scalar(@{$progeny}) > 0); print STDERR "Found ", scalar(@{$samples}), " sample file(s).\n" if (scalar(@{$samples}) > 0); } sub initialize_database { my ($log_fh, $parents, $progeny, $samples, $sample_ids) = @_; my (@results, $sample_id, $sample); print $log_fh "Initializing the database...\n"; # # Create the database. # if ($create_db) { # # Check that the database doesn't already exist. # if ($dry_run == false) { @results = `mysql --defaults-file=$cnf -N -B -e "SHOW DATABASES LIKE '$db'"`; if (scalar(@results) > 0 && $overw_db == false) { die("Unable to create database '$db', it already exists.\n"); } } if ($overw_db == true) { `mysql --defaults-file=$cnf -N -B -e "DROP DATABASE IF EXISTS $db"` if ($dry_run == false); print $log_fh "mysql --defaults-file=$cnf -N -B -e \"DROP DATABASE IF EXISTS '$db'\"\n"; } `mysql --defaults-file=$cnf -e "CREATE DATABASE $db"` if ($dry_run == false); print $log_fh "mysql --defaults-file=$cnf $db -e \"CREATE DATABASE $db\"\n"; `mysql --defaults-file=$cnf $db < $mysql_tables` if ($dry_run == false); print $log_fh "mysql --defaults-file=$cnf $db < $mysql_tables\n"; } # # Set the SQL Batch ID for this set of loci, along with description and date of # sequencing. Insert this batch data into the database. # `mysql --defaults-file=$cnf $db -e "INSERT INTO batches SET id=$batch_id, description='$desc', date='$date', type='$data_type'"` if ($dry_run == false); print $log_fh "mysql --defaults-file=$cnf $db -e \"INSERT INTO batches SET id=$batch_id, description='$desc', date='$date', type='$data_type'\"\n"; print $log_fh "Loading sample data into the MySQL database...\n"; my $i = 1; foreach $sample (@{$parents}, @{$progeny}, @{$samples}) { if ($dry_run == false) { `mysql --defaults-file=$cnf $db -e "INSERT INTO samples SET sample_id=$i, batch_id=$batch_id, type='$sample->{'type'}', file='$sample->{'file'}', pop_id='$sample->{'pop_id'}', group_id='$sample->{'grp_id'}'"`; @results = `mysql --defaults-file=$cnf $db -N -B -e "SELECT id FROM samples WHERE sample_id=$i AND batch_id=$batch_id AND type='$sample->{'type'}' AND file='$sample->{'file'}'"`; chomp $results[0]; $sample_id = $results[0]; # # Save the sample ID to use when running pstacks. # $sample_ids->{$sample->{'file'}} = $sample_id; } print $log_fh "mysql --defaults-file=$cnf $db -e \"INSERT INTO samples SET sample_id=$i, batch_id=$batch_id, type='$sample->{'type'}', file='$sample->{'file'}', pop_id='$sample->{'pop_id'}', group_id='$sample->{'grp_id'}'\"\n"; $i++; } print $log_fh "\n"; } sub load_sql_data { my ($log_fh, $pops, $parents, $progeny, $samples) = @_; my ($pop_cnt, $sample, $num_files, $i, $file); print STDERR "\nComputation is complete, loading results to the database '$db'.\n"; my $pop_cnt = scalar(keys %{$pops}); $i = 1; $num_files = scalar(@{$parents}) + scalar(@{$progeny}) + scalar(@{$samples}); foreach $sample (@{$parents}, @{$progeny}, @{$samples}) { printf(STDERR "Loading pstacks output to $db; file % 3s of % 3s [%s]...", $i, $num_files, $sample->{'file'}); if ($gzip == true) { $file = "$out_path/$sample->{'file'}" . ".tags.tsv.gz"; import_gzsql_file($log_fh, $file, "unique_tags", 1); $file = "$out_path/$sample->{'file'}" . ".snps.tsv.gz"; import_gzsql_file($log_fh, $file, "snps", 1); $file = "$out_path/$sample->{'file'}" . ".alleles.tsv.gz"; import_gzsql_file($log_fh, $file, "alleles", 1); } else { $file = "$out_path/$sample->{'file'}" . ".tags.tsv"; import_sql_file($log_fh, $file, "unique_tags", 1); $file = "$out_path/$sample->{'file'}" . ".snps.tsv"; import_sql_file($log_fh, $file, "snps", 1); $file = "$out_path/$sample->{'file'}" . ".alleles.tsv"; import_sql_file($log_fh, $file, "alleles", 1); } print STDERR "done.\n"; $i++; } print STDERR "Importing catalog to $db..."; my $cat_file = "batch_" . $batch_id; if ($gzip == true) { $file = "$out_path/$cat_file" . ".catalog.tags.tsv.gz"; import_gzsql_file($log_fh, $file, "catalog_tags", 1); $file = "$out_path/$cat_file" . ".catalog.snps.tsv.gz"; import_gzsql_file($log_fh, $file, "catalog_snps", 1); $file = "$out_path/$cat_file" . ".catalog.alleles.tsv.gz"; import_gzsql_file($log_fh, $file, "catalog_alleles", 1); } else { $file = "$out_path/$cat_file" . ".catalog.tags.tsv"; import_sql_file($log_fh, $file, "catalog_tags", 1); $file = "$out_path/$cat_file" . ".catalog.snps.tsv"; import_sql_file($log_fh, $file, "catalog_snps", 1); $file = "$out_path/$cat_file" . ".catalog.alleles.tsv"; import_sql_file($log_fh, $file, "catalog_alleles", 1); } print STDERR "done.\n"; # # Load the sstacks results to the database if requested. # $i = 1; foreach $sample (@{$parents}, @{$progeny}, @{$samples}) { printf(STDERR "Loading sstacks output to $db; file % 3s of % 3s [%s]...", $i, $num_files, $sample->{'file'}); if ($gzip == true) { $file = "$out_path/" . $sample->{'file'} . ".matches.tsv.gz"; import_gzsql_file($log_fh, $file, "matches", 1); } else { $file = "$out_path/" . $sample->{'file'} . ".matches.tsv"; import_sql_file($log_fh, $file, "matches", 1); } print STDERR "done.\n"; $i++; } if ($data_type eq "map") { # # Load the outputs generated by genotypes to the database. # $file = "$out_path/batch_" . $batch_id . ".markers.tsv"; import_sql_file($log_fh, $file, "markers", 1); $file = "$out_path/batch_" . $batch_id . ".genotypes_1.txt"; import_sql_file($log_fh, $file, "catalog_genotypes", 1); } else { # # Load the outputs generated by populations to the database. # $file = "$out_path/batch_" . $batch_id . ".markers.tsv"; import_sql_file($log_fh, $file, "markers", 1); $file = "$out_path/batch_" . $batch_id . ".sumstats.tsv"; import_sql_file($log_fh, $file, "sumstats", $pop_cnt+1); $file = "$out_path/batch_" . $batch_id . ".hapstats.tsv"; import_sql_file($log_fh, $file, "hapstats", $pop_cnt+1); # # Import the Fst files. # my $fst_cnt = 0; my (@keys, $m, $n); @keys = sort keys %pops; for ($m = 0; $m < scalar(@keys); $m++) { for ($n = 0; $n < scalar(@keys); $n++) { $file = "$out_path/batch_" . $batch_id . ".fst_" . $keys[$m] . "-" . $keys[$n] . ".tsv"; if (-e $file) { import_sql_file($log_fh, $file, "fst", 1); $fst_cnt++; } } } print STDERR "Imported $fst_cnt Fst file(s).\n"; # # Import the Phi_st files. # $fst_cnt = 0; for ($m = 0; $m < scalar(@keys); $m++) { for ($n = 0; $n < scalar(@keys); $n++) { $file = "$out_path/batch_" . $batch_id . ".phistats_" . $keys[$m] . "-" . $keys[$n] . ".tsv"; if (-e $file) { import_sql_file($log_fh, $file, "phist", 3); $fst_cnt++; } } } print STDERR "Imported $fst_cnt Haplotype Fst file(s).\n"; } print $log_fh "\n"; # # Index the radtags database # my ($cmd, @results); print STDERR "Indexing the database...\n"; $cmd = $exe_path . "index_radtags.pl -D $db -t -c 2>&1"; print STDERR "$cmd\n"; print $log_fh "$cmd\n"; @results = `$cmd` if ($dry_run == false); print $log_fh @results; } sub write_results { my ($results, $log_fh) = @_; my $line; foreach $line (@{$results}) { if ($line =~ /\r/) { $line =~ s/^.+\r(.*\n)$/\1/; } print $log_fh $line; } } sub write_depths_of_cov { my ($depths, $log_fh) = @_; print STDERR "\nDepths of Coverage for Processed Samples:\n"; print $log_fh "\nDepths of Coverage for Processed Samples:\n"; foreach $a (@{$depths}) { print STDERR $a->[0], ": ", $a->[1], "x\n"; print $log_fh $a->[0], ": ", $a->[1], "x\n"; } print STDERR "\n"; print $log_fh "\n"; } sub import_sql_file { my ($log_fh, $file, $table, $skip_lines) = @_; my (@results, $ignore); $ignore = "IGNORE $skip_lines LINES" if ($skip_lines > 0); @results = `mysql --defaults-file=$cnf $db -e "LOAD DATA LOCAL INFILE '$file' INTO TABLE $table $ignore"` if ($sql == true && $dry_run == false); if ($sql == true) { print $log_fh "mysql --defaults-file=$cnf $db -e \"LOAD DATA LOCAL INFILE '$file' INTO TABLE $table $ignore\"\n", @results; } } sub import_gzsql_file { my ($log_fh, $file, $table, $skip_lines) = @_; my (@results, $ignore); $ignore = "IGNORE $skip_lines LINES" if ($skip_lines > 0); # # Get a temporary file name and create a named pipe. # my $tmpdir = File::Spec->tmpdir(); my $named_pipe = mktemp($tmpdir . "/denovo_map_XXXXXX"); if ($sql == true && $dry_run == false) { mkfifo($named_pipe, 0700) || die("Unable to create named pipe for loading gzipped data: $named_pipe, $!"); print $log_fh "Streaming $file into named pipe $named_pipe.\n"; } # # Dump our gzipped data onto the named pipe. # system("gunzip -c $file > $named_pipe &") if ($sql == true && $dry_run == false); @results = `mysql --defaults-file=$cnf $db -e "LOAD DATA LOCAL INFILE '$named_pipe' INTO TABLE $table $ignore"` if ($sql == true && $dry_run == false); if ($sql == 1) { print $log_fh "mysql --defaults-file=$cnf $db -e \"LOAD DATA LOCAL INFILE '$named_pipe' INTO TABLE $table $ignore\"\n", @results; } # # Remove the pipe. # unlink($named_pipe) if ($sql == true && $dry_run == false); } sub parse_command_line { my ($arg); while (@ARGV) { $_ = shift @ARGV; if ($_ =~ /^-v$/) { version(); exit(); } elsif ($_ =~ /^-h$/) { usage(); } elsif ($_ =~ /^-p$/) { push(@parents, { 'path' => shift @ARGV }); } elsif ($_ =~ /^-r$/) { push(@progeny, { 'path' => shift @ARGV }); } elsif ($_ =~ /^-s$/) { push(@samples, { 'path' => shift @ARGV }); } elsif ($_ =~ /^-d$/) { $dry_run = true; } elsif ($_ =~ /^-o$/) { $out_path = shift @ARGV; } elsif ($_ =~ /^-D$/) { $desc = shift @ARGV; } elsif ($_ =~ /^-e$/) { $exe_path = shift @ARGV; } elsif ($_ =~ /^-b$/) { $batch_id = shift @ARGV; } elsif ($_ =~ /^-i$/) { $sample_id = shift @ARGV; } elsif ($_ =~ /^-a$/) { $date = shift @ARGV; } elsif ($_ =~ /^-S$/) { $sql = false; } elsif ($_ =~ /^-B$/) { $db = shift @ARGV; } elsif ($_ =~ /^-m$/) { $min_cov = shift @ARGV; } elsif ($_ =~ /^-P$/) { $min_rcov = shift @ARGV; } elsif ($_ =~ /^--samples$/) { $sample_path = shift @ARGV; } elsif ($_ =~ /^-O$/) { $popmap_path = shift @ARGV; push(@_populations, "-M " . $popmap_path); } elsif ($_ =~ /^--create_db$/) { $create_db = true; } elsif ($_ =~ /^--overw_db$/) { $overw_db = true; $create_db = true; } elsif ($_ =~ /^-A$/) { $arg = shift @ARGV; push(@_genotypes, "-t " . $arg); $arg = lc($arg); if ($arg ne "gen" && $arg ne "cp" && $arg ne "f2" && $arg ne "bc1" && $arg ne "dh") { print STDERR "Unknown genetic mapping cross specified: '$arg'\n"; usage(); } } elsif ($_ =~ /^-T$/) { $arg = shift @ARGV; push(@_pstacks, "-p " . $arg); push(@_cstacks, "-p " . $arg); push(@_sstacks, "-p " . $arg); push(@_populations, "-t " . $arg); } elsif ($_ =~ /^--bound_low$/) { push(@_pstacks, "--bound_low " . shift @ARGV); push(@_pstacks, "--model_type bounded"); } elsif ($_ =~ /^--bound_high$/) { push(@_pstacks, "--bound_high " . shift @ARGV); push(@_pstacks, "--model_type bounded"); } elsif ($_ =~ /^--alpha$/) { push(@_pstacks, "--alpha " . shift @ARGV); } elsif ($_ =~ /^-X$/) { # # Pass an arbitrary command-line option to a pipeline program. # # Command line option must be of the form '-X "program:option"' # $arg = shift @ARGV; my ($prog, $opt) = ($arg =~ /^(\w+):(.+)$/); if ($prog eq "pstacks") { push(@_pstacks, $opt); } elsif ($prog eq "cstacks") { push(@_cstacks, $opt); } elsif ($prog eq "sstacks") { push(@_sstacks, $opt); } elsif ($prog eq "genotypes") { push(@_genotypes, $opt); } elsif ($prog eq "populations") { push(@_populations, $opt); } else { print STDERR "Unknown pipeline program, '$arg'\n"; usage(); } } else { print STDERR "Unknown command line option: '$_'\n"; usage(); } } $exe_path = $exe_path . "/" if (substr($exe_path, -1) ne "/"); $out_path = substr($out_path, 0, -1) if (substr($out_path, -1) eq "/"); if ($batch_id !~ /^\d+$/ || $batch_id < 0) { print STDERR "You must specify a batch ID and it must be an integer (e.g. 1, 2, 3).\n"; usage(); } if ($sql == true && length($date) == 0) { $date = strftime("%Y-%m-%d", (localtime(time))); } if (scalar(@parents) > 0 && scalar(@samples) > 0) { print STDERR "You must specify either parent or sample files, but not both.\n"; usage(); } if (scalar(@parents) == 0 && scalar(@samples) == 0 && length($popmap_path) == 0) { print STDERR "You must specify at least one parent or sample file.\n"; usage(); } if (length($sample_path) > 0) { $sample_path .= "/" if (substr($sample_path, -1) ne "/"); } if (scalar(@samples) > 0 || length($popmap_path) > 0) { $data_type = "population"; } else { $data_type = "map"; } } sub version { print STDERR "ref_map.pl ", stacks_version, "\n"; } sub usage { version(); print STDERR <: specify a path to the directory of samples (samples will be read from population map). Stack assembly options: m: specify a minimum number of identical, raw reads required to create a stack (default 3). P: specify a minimum number of identical, raw reads required to create a stack in 'progeny' individuals. Database options: B: specify a database to load data into. D: a description of this batch to be stored in the database. S: disable recording SQL data in the database. i: starting sample_id, this is determined automatically if database interaction is enabled. --create_db: create the database specified by '-B' and populate the tables. --overw_db: delete the database before creating a new copy of it (turns on --create_db). SNP Model Options (these options are passed on to pstacks): --bound_low : lower bound for epsilon, the error rate, between 0 and 1.0 (default 0). --bound_high : upper bound for epsilon, the error rate, between 0 and 1.0 (default 1). --alpha : chi square significance level required to call a heterozygote or homozygote, either 0.1, 0.05 (default), 0.01, or 0.001. Arbitrary command line options: -X "program:option": pass a command line option to one of the pipeline components, e.g.'-X "sstacks:-x"'. EOQ exit(0); } stacks-1.44/scripts/sort_read_pairs.pl000755 000765 000024 00000035110 12533677757 020737 0ustar00catchenstaff000000 000000 #!/usr/bin/env perl # # Copyright 2011-2015, Julian Catchen # # This file is part of Stacks. # # Stacks is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation, either version 3 of the License, or # (at your option) any later version. # # Stacks is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with Stacks. If not, see . # # # Sort paired-end sequences according to the stacks the non-paired-end # was found in. # # By Julian Catchen # use strict; use constant stacks_version => "_VERSION_"; use constant true => 1; use constant false => 0; my $debug = 0; my $white_list = ""; my $cat_white_list = ""; my $in_path = ""; my $out_path = ""; my $samp_path = ""; my $out_type = "fasta"; my $in_type = "fastq"; my $gzipped = false; parse_command_line(); my (@files, %matches, %stacks, %reads, %marker_wl); build_file_list(\@files); my ($file, $num_files, $i, $key); if (length($cat_white_list) > 0) { load_white_list($cat_white_list, \%marker_wl); print STDERR "Loaded ", scalar(keys %marker_wl), " catalog IDs from '$cat_white_list'\n"; } $num_files = scalar(@files); $i = 1; foreach $file (@files) { printf(STDERR "Loading catalog matches, file % 2s of % 2s [%s]\n", $i, $num_files, $file->{'prefix'}); # # Load the sstacks file, listing matches between the stacks and the catalog # load_matches($in_path, $file, \%matches, \%marker_wl); $i++; } # # Determine which catalog loci have more than a single match from each sample and blacklist them. # print STDERR "Identifying catalog loci that have more than one match from a single sample..."; my %multiple_matches; check_mult_catalog_matches(\%matches, \%multiple_matches); print STDERR "done\n", " These loci will be excluded when collating paired-end reads;\n", " A list of them has been recorded: $out_path/sort_read_pairs.log\n", scalar(keys %matches), " total catalog loci; ", scalar(keys %multiple_matches), " will be excluded, processing ", scalar(keys %matches) - scalar(keys %multiple_matches), " loci.\n"; # # Check if files already exist, if so, exit to prevent adding data to existing files. # my ($cat_id, $path); foreach $cat_id (keys %matches) { # # Check that this catalog ID only has a single match from each sample. # next if (defined($multiple_matches{$cat_id})); $path = $out_path . "/" . $cat_id; $path .= $out_type eq "fasta" ? ".fa" : ".fq"; if (-e $path) { die("Error: output files already exist. This program will append data to files if\n" . "they already exist. Please delete these files and re-execute sort_read_pairs.pl.\n"); } } $i = 1; foreach $file (@files) { printf(STDERR "Processing file % 2s of % 2s [%s]\n", $i, $num_files, $file->{'prefix'}); # # Load the ustacks tag file for each sample, containing the read-to-stack mappings # $stacks{$file->{'prefix'}} = {}; print STDERR " Loading tag file..."; load_stacks($in_path, $file, $stacks{$file->{'prefix'}}); print STDERR "done.\n"; # # Map the read-pairs to the stack/catalog match they correspond to. # $reads{$file->{'prefix'}} = {}; print STDERR " Loading sample file..."; $in_type eq "fastq" ? process_fastq_read_pairs($samp_path, $file, \%stacks, $reads{$file->{'prefix'}}) : process_fasta_read_pairs($samp_path, $file, \%stacks, $reads{$file->{'prefix'}}); print STDERR "done.\n"; print STDERR " Printing results..."; print_results($out_path, \%matches, \%stacks, \%reads, \%multiple_matches); print STDERR "done.\n"; # # Clean up memory usage. # print STDERR " Clearing memory..."; undef(%{$stacks{$file->{'prefix'}}}); undef(%{$reads{$file->{'prefix'}}}); print STDERR " done.\n"; $i++; } sub load_matches { my ($in_path, $in_file, $matches, $marker_wl) = @_; my ($file, $in_fh, $line, @parts, $key); if ($gzipped == true) { $file = $in_path . "/" . $in_file->{'prefix'} . $in_file->{'suffix'} . ".matches.tsv.gz"; open($in_fh, "gunzip -c $file |") or die("Unable to open catalog matches file '$file', $!\n"); } else { $file = $in_path . "/" . $in_file->{'prefix'} . $in_file->{'suffix'} . ".matches.tsv"; open($in_fh, "<$file") or die("Unable to open catalog matches file '$file', $!\n"); } while ($line = <$in_fh>) { chomp $line; @parts = split(/\t/, $line); if (length($cat_white_list) > 0) { next if (!defined($marker_wl->{$parts[2]})); } if (!defined($matches->{$parts[2]})) { $matches->{$parts[2]} = {}; } # # Index by catalog_ID -> sample_ID|stack_ID # $key = $in_file->{'prefix'} . "|" . $parts[4]; $matches->{$parts[2]}->{$key}++; } close($in_fh); } sub load_stacks { my ($in_path, $in_file, $stacks) = @_; my ($file, $in_fh, $line, @parts); if ($gzipped == true) { $file = $in_path . "/" . $in_file->{'prefix'} . $in_file->{'suffix'} . ".tags.tsv.gz"; open($in_fh, "gunzip -c $file |") or die("Unable to open '$file', $!\n"); } else { $file = $in_path . "/" . $in_file->{'prefix'} . $in_file->{'suffix'} . ".tags.tsv"; open($in_fh, "<$file") or die("Unable to open '$file', $!\n"); } while ($line = <$in_fh>) { chomp $line; @parts = split(/\t/, $line); next if ($parts[6] eq "consensus" || $parts[6] eq "model"); # # Index by sequence ID -> stack ID # $stacks->{substr($parts[8], 0, -2)} = $parts[2]; } close($in_fh); } sub process_fastq_read_pairs { my ($in_path, $in_file, $stacks, $reads) = @_; my ($file, $in_fh, $line, $seq, $qual, $key, $read_id); if ($in_file->{'suffix'} eq ".1") { if ($gzipped == true) { $file = $in_path . "/" . $in_file->{'prefix'} . ".2.fq.gz"; open($in_fh, "gunzip -c $file |") or die("Unable to open paired-end input file '$file'\n"); } else { $file = $in_path . "/" . $in_file->{'prefix'} . ".2.fq"; open($in_fh, "<$file") or die("Unable to open paired-end input file '$file'\n"); } } else { if ($gzipped == true) { $file = $in_path . "/" . $in_file->{'prefix'} . ".fq_2.gz"; open($in_fh, "gunzip -c $file |") or die("Unable to open paired-end input file '$file'\n"); } else { $file = $in_path . "/" . $in_file->{'prefix'} . ".fq_2"; open($in_fh, "<$file") or die("Unable to open paired-end input file '$file'\n"); } } while ($line = <$in_fh>) { next if (substr($line, 0, 1) ne "@"); chomp $line; $read_id = substr($line, 1, -2); $seq = <$in_fh>; chomp $seq; # # Read the repeated ID and the quality scores. # <$in_fh>; $qual = <$in_fh>; chomp $qual; $key = $stacks->{$in_file->{'prefix'}}->{$read_id}; next if (!defined($key)); if (!defined($reads->{$key})) { $reads->{$key} = []; } push(@{$reads->{$key}}, {'id' => $read_id, 'seq' => $seq, 'qual' => $qual}); } } sub process_fasta_read_pairs { my ($in_path, $in_file, $stacks, $reads) = @_; my ($file, $in_fh, $line, $seq, $qual, $key, $read_id); if ($in_file->{'suffix'} eq ".1") { if ($gzipped == true) { $file = $in_path . "/" . $in_file->{'prefix'} . ".2.fa.gz"; open($in_fh, "gunzip -c $file |") or die("Unable to open paired-end input file '$file'\n"); } else { $file = $in_path . "/" . $in_file->{'prefix'} . ".2.fa"; open($in_fh, "<$file") or die("Unable to open paired-end input file '$file'\n"); } } else { if ($gzipped == true) { $file = $in_path . "/" . $in_file->{'prefix'} . ".fa_2.gz"; open($in_fh, "gunzip -c $file |") or die("Unable to open paired-end input file '$file'\n"); } else { $file = $in_path . "/" . $in_file->{'prefix'} . ".fa_2"; open($in_fh, "<$file") or die("Unable to open paired-end input file '$file'\n"); } } while ($line = <$in_fh>) { next if (substr($line, 0, 1) ne ">"); chomp $line; $read_id = substr($line, 1, -2); $seq = <$in_fh>; chomp $seq; $key = $stacks->{$in_file->{'prefix'}}->{$read_id}; next if (!defined($key)); if (!defined($reads->{$key})) { $reads->{$key} = []; } push(@{$reads->{$key}}, {'id' => $read_id, 'seq' => $seq, 'qual' => ""}); } } sub print_results { my ($out_path, $matches, $stacks, $reads, $multiple_matches) = @_; my ($path, $cat_id, $sample, $stack_id, $read, $out_fh, $i, @keys, $count, $key, $mult_hits); # # If a catalog ID matches stacks from multiple samples, print them out together. # foreach $cat_id (keys %{$matches}) { # # Check that this catalog ID only has a single match from each sample. # next if (defined($multiple_matches->{$cat_id})); $path = $out_path . "/" . $cat_id; $path .= $out_type eq "fasta" ? ".fa" : ".fq"; open($out_fh, ">>$path") or die("Unable to open $path; '$!'\n"); foreach $key (keys %{$matches->{$cat_id}}) { ($sample, $stack_id) = split(/\|/, $key); foreach $read (@{$reads->{$sample}->{$stack_id}}) { if ($out_type eq "fasta") { print $out_fh ">", $cat_id, "|", $sample, "|", $stack_id, "|", $read->{'id'}, "\n", $read->{'seq'}, "\n"; } else { print $out_fh "@", $cat_id, "|", $sample, "|", $stack_id, "|", $read->{'id'}, "\n", $read->{'seq'}, "\n", "+\n", $read->{'qual'}, "\n"; } } } close($out_fh); } } sub check_mult_catalog_matches { my ($matches, $multiple_matches) = @_; my ($fh, $key, $sample, $stack_id); # # Find catalog loci that have more than a single match from one or more samples # and log those loci that will be excluded. # open($fh, ">$out_path/sort_read_pairs.log") or die("Unable to open log file, $!\n"); print $fh "# The catalog loci listed below have more than a single match from one or more individuals, indicating undermerged or repetitive loci.\n", "# CatalogLocus Sample1:Locus1,Locus2;Sample2:Locus1,Locus2\n"; foreach $cat_id (keys %{$matches}) { my %samples; foreach $key (keys %{$matches->{$cat_id}}) { ($sample, $stack_id) = split(/\|/, $key); push(@{$samples{$sample}}, $stack_id); } my $mult_hits = 0; my $str = ""; foreach $sample (keys %samples) { if (scalar(@{$samples{$sample}}) > 1) { $mult_hits++; $str .= $sample . ":" . join(",", @{$samples{$sample}}) . "; "; } } if ($mult_hits > 0) { print $fh $cat_id, "\t", substr($str, 0, -1), "\n"; $multiple_matches{$cat_id}++; } } close($fh); } sub count_reads { my ($catalog, $reads) = @_; my ($count, $key, $sample, $stack_id); $count = 0; foreach $key (keys %{$catalog}) { ($sample, $stack_id) = split(/\|/, $key); if (defined($reads->{$sample}->{$stack_id})) { $count += scalar(@{$reads->{$sample}->{$stack_id}}); } } return $count; } sub build_file_list { my ($files) = @_; my (@ls, $line, $file, $prefix, $suffix); # Load a white list of files to process if it is supplied. my %wl; if (length($white_list) > 0) { load_white_list($white_list, \%wl); print STDERR "Loaded ", scalar(keys %wl), " filenames from '$white_list'\n"; } @ls = glob("$in_path/*.tags.tsv"); if (scalar @ls == 0) { @ls = glob("$in_path/*.tags.tsv.gz"); $gzipped = true if (scalar @ls > 0); } foreach $line (@ls) { chomp $line; next if (length($line) == 0); next if ($line =~ /batch_\d+\.catalog/); ($file) = ($line =~ /$in_path\/(.+)\.tags\.tsv\.?g?z?$/); if (length($white_list) > 0) { next if (!defined($wl{$file})); } if ($file =~ /\.1$/) { ($prefix, $suffix) = ($file =~ /^(.+)(\.1)$/); } else { $prefix = $file; $suffix = ""; } push(@{$files}, {'prefix' => $prefix, 'suffix' => $suffix}); } } sub load_white_list { my ($list, $wl) = @_; open(WHITE, "<" . $list) or die("Unable to open white list file '$white_list': $!\n"); my $line = ""; while ($line = ) { chomp $line; next if (length($line) == 0); next if ($line =~ /^\s*#/); $wl->{$line}++; } close(WHITE); } sub parse_command_line { while (@ARGV) { $_ = shift @ARGV; if ($_ =~ /^-p$/) { $in_path = shift @ARGV; } elsif ($_ =~ /^-o$/) { $out_path = shift @ARGV; } elsif ($_ =~ /^-s$/) { $samp_path = shift @ARGV; } elsif ($_ =~ /^-t$/) { $out_type = shift @ARGV; } elsif ($_ =~ /^-i$/) { $in_type = shift @ARGV; } elsif ($_ =~ /^-W$/) { $white_list = shift @ARGV; } elsif ($_ =~ /^-w$/) { $cat_white_list = shift @ARGV; } elsif ($_ =~ /^-d$/) { $debug++; } elsif ($_ =~ /^-v$/) { version(); exit(); } elsif ($_ =~ /^-h$/) { usage(); } else { print STDERR "Unknown command line option: '$_'\n"; usage(); } } if ($out_type ne "fasta" && $out_type ne "fastq") { print STDERR "Output type must be either 'fasta' or 'fastq'.\n"; usage(); } if ($in_type ne "fasta" && $in_type ne "fastq") { print STDERR "Input type must be either 'fasta' or 'fastq'.\n"; usage(); } if (length($in_path) == 0) { print STDERR "You must specify a path to the Stacks output files.\n"; usage(); } if (length($out_path) == 0) { print STDERR "You must specify a path to write the collated output files.\n"; usage(); } if (length($samp_path) == 0) { print STDERR "You must specify a path to the paired-end reads.\n"; usage(); } $in_path = substr($in_path, 0, -1) if (substr($in_path, -1) eq "/"); $out_path = substr($out_path, 0, -1) if (substr($out_path, -1) eq "/"); $samp_path = substr($samp_path, 0, -1) if (substr($samp_path, -1) eq "/"); } sub version { print STDERR "sort_read_pairs.pl ", stacks_version, "\n"; } sub usage { version(); print STDERR < # # This file is part of Stacks. # # Stacks is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation, either version 3 of the License, or # (at your option) any later version. # # Stacks is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with Stacks. If not, see . # # # Execute the Stacks Pipeline exporter script, wait for it to finish, send # a notification email to the person who submitted the export. # # Written by Julian Catchen # use strict; use File::Temp qw/tempfile/; use Net::SMTP; use constant stacks_version => "_VERSION_"; # # Configuration: # exe_path: Path to the export executable. # output_path: Path to web-accessible directory to output the export data # url: URL to reach the directory specified by output_path # local_host: Name of localhost to present to SMTP server # smtp_host: Name of SMTP server through which to send mail # from: email address to use in the 'From' field of the message # my $exe_path = "_BINDIR_" . "export_sql.pl"; my $output_path = "_PKGDATADIR_" . "php/export/"; my $url = "http://stackshost.edu/stacks/export/"; my $local_host = "localhost"; my $smtp_host = "localhost"; my $from = "stacks\@stackshost.edu"; my $debug = 0; my $db = ""; my $batch_id = 0; my $out_file = ""; # Generated file name for output data my $out_type = "tsv"; # Output format my $email = ""; # Email address to send message to my $data_type = "haplo"; # Type of export: observed haplotypes or a genetic map. my $map_type = "geno"; # If exporting a genetic map, specify map type. my $depth_lim = 1; # Minimum stack depth limit my $man_cor = 0; # Include manual corrections in the export my $filter_str = ""; # Comma separated list of filters parse_command_line(); my (@filters, @cmd_opts, $filter); # # Generate file name # my $template = "stacks_export_XXXXXXXX"; my $suffix = $out_type eq "xls" ? ".xls" : ".tsv"; my (undef, $out_file) = tempfile($template, OPEN => 0, DIR => $output_path, SUFFIX => $suffix); # # Prepare the command line parameters # push(@cmd_opts, "-D $db", "-b $batch_id", "-f $out_file", "-o $out_type", "-a $data_type"); if ($data_type eq "geno") { push(@cmd_opts, "-m $map_type"); } if ($data_type eq "geno" && $man_cor > 0) { push(@cmd_opts, "-c"); } if ($data_type eq "haplo" && $depth_lim > 1) { push(@cmd_opts, "-L $depth_lim"); } if (length($filter_str) > 0) { @filters = split(/,/, $filter_str); foreach $filter (@filters) { push(@cmd_opts, "-F $filter"); } } my $cmd = join(" ", @cmd_opts); $cmd = $exe_path . " " . $cmd; print STDERR "CMD: $cmd\n" if ($debug); # # Execute the exporter program # my @results = `$cmd`; #my @results = `echo Success`; # # Check the results, we expext a one line result: either 'Success' or 'Failure' # chomp $results[0]; if ($results[0] eq "Success") { send_email('success', $out_file); } else { send_email('failure', $out_file); } sub send_email { my ($result) = @_; my $smtp = Net::SMTP->new($smtp_host, 'Hello' => $local_host, 'Timeout' => 60); $smtp->mail($from); $smtp->recipient($email); my $msg .= "From: $from\r\n" . "To: $email\r\n" . "Subject: Stacks pipeline export complete\r\n" . "\r\n"; if ($result eq "success") { # # Trim the path off the output file # my ($f) = ($out_file =~ /.*\/(stacks_export_\w{8}\.\w{3})$/); $msg .= "Your data has been exported and can be downloaded from: " . $url . $f . "\r\n"; } else { $msg .= "There has been an error exporting your data, please contact the system administrator.\r\n"; } $smtp->data($msg); $smtp->quit(); } sub parse_command_line { while (@ARGV) { $_ = shift @ARGV; if ($_ =~ /^-d$/) { $debug++; } elsif ($_ =~ /^-D$/) { $db = shift @ARGV; } elsif ($_ =~ /^-b$/) { $batch_id = shift @ARGV; } elsif ($_ =~ /^-e$/) { $email = shift @ARGV; } elsif ($_ =~ /^-a$/) { $data_type = lc(shift @ARGV); } elsif ($_ =~ /^-m$/) { $map_type = lc(shift @ARGV); } elsif ($_ =~ /^-L$/) { $depth_lim = shift @ARGV; } elsif ($_ =~ /^-c$/) { $man_cor++; } elsif ($_ =~ /^-t$/) { $out_type = shift @ARGV; } elsif ($_ =~ /^-F$/) { $filter_str = shift @ARGV; } elsif ($_ =~ /^-v$/) { version(); exit(); } elsif ($_ =~ /^-h$/) { usage(); } else { print STDERR "Unknown command line options received: $_\n"; usage(); } } if (length($db) == 0) { print STDERR "You must specify a database to index.\n"; usage(); } if (length($email) == 0) { print STDERR "You must specify an email for notification.\n"; usage(); } if ($out_type ne "tsv" && $out_type ne "xls") { print STDERR "Output type can only be 'tsv' or 'xls'.\n"; usage(); } if ($batch_id !~ /^\d{1,4}$/) { print STDERR "Batch ID must be a numeric value.\n"; usage(); } } sub version { print STDERR "stacks_export_notify.pl ", stacks_version, "\n"; } sub usage { version(); print << "EOQ"; stacks_export_notify.pl -e email -D db -b batch_id [-t type] [-F filters] [-d] [-h] e: email to use for notification. D: radtag database to examine. b: batch_id of data set to export. a: type of data to export, either 'gen' or 'haplo', for genotypes or observed haplotypes. t: output type, either 'tsv' or 'xls'. F: comma separated list of filters to apply to the data. L: if exporting observed haplotypes, specify a stack depth limit. m: map type. If genotypes are to be exported, specify the map type. c: include manual corrections if exporting genotypes. h: display this help message. d: turn on debug output. EOQ exit(0); } stacks-1.44/php/ajax.js000644 000765 000024 00000001341 12533677757 015577 0ustar00catchenstaff000000 000000 var req; function loadXMLDoc(url) { req = false; // branch for native XMLHttpRequest object if(window.XMLHttpRequest && !(window.ActiveXObject)) { try { req = new XMLHttpRequest(); } catch(e) { req = false; } // branch for IE/Windows ActiveX version } else if(window.ActiveXObject) { try { req = new ActiveXObject("Msxml2.XMLHTTP"); } catch(e) { try { req = new ActiveXObject("Microsoft.XMLHTTP"); } catch(e) { req = false; } } } if(req) { req.onreadystatechange = processReqChange; req.open("GET", url, true); req.send(""); } } stacks-1.44/php/annotate.js000644 000765 000024 00000016374 12335173442 016457 0ustar00catchenstaff000000 000000 // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // var http_req; function toggle_annotation(id) { var div_obj = document.getElementById(id + "_div"); var a_obj = document.getElementById(id + "_ann"); var frm_obj = document.getElementById(id + "_frm"); // Turn ON the anchor, turn OFF the form if (a_obj.style.display == "none") { a_obj.style.display = ""; div_obj.style.display = "none"; if (a_obj.innerHTML.length > 0) a_obj.innerHTML = "annotate"; } // Turn OFF the anchor, turn ON the form else { a_obj.style.display = "none"; div_obj.style.display = ""; if (a_obj.innerHTML != "annotate") frm_obj.ext_id.value = a_obj.innerHTML; } } function annotate_marker(id) { // // Fetch the marker annotation // var form_obj = document.getElementById(id + "_frm"); var url = form_obj.url.value + "&" + "ext_id=" + form_obj.ext_id.value; // // Prepare and send XMLHttpRequest Object. // http_req = false; try { http_req = new XMLHttpRequest(); } catch(e) { http_req = false; } if (http_req) { http_req.onreadystatechange = process_annotation; http_req.open("GET", url, true); http_req.send(""); } toggle_annotation(id); } function process_annotation() { // // Possible readyState values: // 0 = uninitialized // 1 = loading // 2 = loaded // 3 = interactive // 4 = complete // if (http_req.readyState == 4) { // Check that the status is "OK" if (http_req.status == 200) { var xml_doc = http_req.responseXML; var tag_obj = xml_doc.getElementsByTagName("marker_id"); var txt_obj = xml_doc.getElementsByTagName("text"); var tag_id = tag_obj[0].childNodes[0].nodeValue; var txt; if (txt_obj[0].childNodes.length > 0) txt = txt_obj[0].childNodes[0].nodeValue; else txt = "annotate"; var a_obj = document.getElementById(tag_id + "_ann"); a_obj.innerHTML = txt; } else { alert("There was a problem retrieving the XML data:\n" + http_req.statusText); } } } function toggle_correction(id) { var div_obj = document.getElementById(id + "_div"); var sel_obj = document.getElementById(id + "_sel"); var s_obj = document.getElementById(id); if (div_obj.style.display == "none") { div_obj.style.display = ""; sel_obj.style.display = "none"; } else { div_obj.style.display = "none"; sel_obj.style.display = ""; s_obj.focus(); } } function cancel_correction(id) { var div_obj = document.getElementById(id + "_div"); var sel_obj = document.getElementById(id + "_sel"); div_obj.style.display = ""; sel_obj.style.display = "none"; } function correct_genotype(id, url) { // // Fetch the marker annotation // var sel_obj = document.getElementById(id); url = url + "&" + "gtype=" + sel_obj.options[sel_obj.selectedIndex].text; // // Prepare and send XMLHttpRequest Object. // http_req = false; try { http_req = new XMLHttpRequest(); } catch(e) { http_req = false; } if (http_req) { http_req.onreadystatechange = process_correction; http_req.open("GET", url, true); http_req.send(""); } toggle_correction(id); } function process_correction() { // // Possible readyState values: // 0 = uninitialized // 1 = loading // 2 = loaded // 3 = interactive // 4 = complete // if (http_req.readyState == 4) { // Check that the status is "OK" if (http_req.status == 200) { var xml_doc = http_req.responseXML; var tag_obj = xml_doc.getElementsByTagName("div_id"); var div_id = tag_obj[0].childNodes[0].nodeValue; var cor_obj = xml_doc.getElementsByTagName("corrected"); var cor = cor_obj[0].childNodes[0].nodeValue; var txt_obj = xml_doc.getElementsByTagName("gtype"); var gtype = txt_obj[0].childNodes[0].nodeValue; var div_obj = document.getElementById(div_id + "_div"); var txt; if (cor == "true") txt = "" + gtype + ""; else txt = gtype; div_obj.innerHTML = "" + txt + ""; } else { alert("There was a problem retrieving the XML data:\n" + http_req.statusText); } } } function toggle_population(id) { var div_obj = document.getElementById(id + "_div"); var a_obj = document.getElementById(id + "_pop"); var frm_obj = document.getElementById(id + "_frm"); // Turn ON the anchor, turn OFF the form if (a_obj.style.display == "none") { a_obj.style.display = ""; div_obj.style.display = "none"; } // Turn OFF the anchor, turn ON the form else { a_obj.style.display = "none"; div_obj.style.display = ""; frm_obj.pop_name.value = a_obj.innerHTML; } } function annotate_population(id) { // // Fetch the marker annotation // var form_obj = document.getElementById(id + "_frm"); var url = form_obj.url.value + "&" + "pop_name=" + escape(form_obj.pop_name.value); // // Prepare and send XMLHttpRequest Object. // http_req = false; try { http_req = new XMLHttpRequest(); } catch(e) { http_req = false; } if (http_req) { http_req.onreadystatechange = process_pop_annotation; http_req.open("GET", url, true); http_req.send(""); } toggle_population(id); } function process_pop_annotation() { // // Possible readyState values: // 0 = uninitialized // 1 = loading // 2 = loaded // 3 = interactive // 4 = complete // if (http_req.readyState == 4) { // Check that the status is "OK" if (http_req.status == 200) { var xml_doc = http_req.responseXML; var id_obj = xml_doc.getElementsByTagName("pop_id"); var name_obj = xml_doc.getElementsByTagName("text"); var pop_id = id_obj[0].childNodes[0].nodeValue; var txt; if (name_obj[0].childNodes.length > 0) txt = name_obj[0].childNodes[0].nodeValue; else txt = "Population " + pop_id; var a_obj = document.getElementById(pop_id + "_pop"); a_obj.innerHTML = txt; } else { alert("There was a problem retrieving the XML data:\n" + http_req.statusText); } } } stacks-1.44/php/annotate_marker.php000644 000765 000024 00000007570 12663652222 020173 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); $database = isset($_GET['db']) ? $_GET['db'] : ""; $tag_id = isset($_GET['tag_id']) ? $_GET['tag_id'] : 0; $batch_id = isset($_GET['batch_id']) ? $_GET['batch_id'] : 0; $ext_id = isset($_GET['ext_id']) ? $_GET['ext_id'] : ""; // Connect to the database if (!isset($db)) $db = db_connect($database); // Save these variables for automatic URL formation later on. $display = array(); $display['db'] = $database; $display['tag_id'] = $tag_id; $display['batch_id'] = $batch_id; $display['ext_id'] = $ext_id; // // Prepare some SQL queries // $query = "SELECT id, external_id " . "FROM catalog_annotations " . "WHERE batch_id=? and catalog_id=?"; if (!($db['sel_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "UPDATE catalog_annotations SET external_id=? WHERE id=?"; if (!($db['upd_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "INSERT INTO catalog_annotations SET batch_id=?, catalog_id=?, external_id=?"; if (!($db['ins_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "DELETE FROM catalog_annotations WHERE id=?"; if (!($db['del_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); // // Fetch any existing annotation for this marker // if (!$db['sel_sth']->bind_param("ii", $display['batch_id'], $display['tag_id'])) write_db_error($db['sel_sth'], __FILE__, __LINE__); if (!$db['sel_sth']->execute()) write_db_error($db['sel_sth'], __FILE__, __LINE__); $res = $db['sel_sth']->get_result(); $external_id = ""; $sql_id = 0; if ($row = $res->fetch_assoc()) { $external_id = $row['external_id']; $sql_id = $row['id']; } if ($external_id != $ext_id) { // // Is this annotation being reset to the original value? If so, delete the corrected record. // if (strlen($external_id) > 0 && strlen($ext_id) == 0) { if (!$db['del_sth']->bind_param("i", $sql_id)) write_db_error($db['del_sth'], __FILE__, __LINE__); if (!$db['del_sth']->execute()) write_db_error($db['del_sth'], __FILE__, __LINE__); // // Are we changing an existing annotation? // } else if (strlen($external_id) > 0 && strlen($ext_id) > 0) { if (!$db['upd_sth']->bind_param("si", $ext_id, $sql_id)) write_db_error($db['upd_sth'], __FILE__, __LINE__); if (!$db['upd_sth']->execute()) write_db_error($db['upd_sth'], __FILE__, __LINE__); // // Otherwise, add a new annotation. // } else if (strlen($ext_id) > 0) { if (!$db['ins_sth']->bind_param("iis", $display['batch_id'], $display['tag_id'], $ext_id)) write_db_error($db['ins_sth'], __FILE__, __LINE__); if (!$db['ins_sth']->execute()) write_db_error($db['ins_sth'], __FILE__, __LINE__); } } header("Content-type: text/xml"); $xml_output = "\n" . "\n" . "$ext_id\n" . "$tag_id\n" . "\n"; echo $xml_output; ?> stacks-1.44/php/catalog.php000644 000765 000024 00000134274 12676533172 016443 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); $batch_id = isset($_GET['id']) ? $_GET['id'] : 0; $database = isset($_GET['db']) ? $_GET['db'] : ""; $page = isset($_GET['p']) ? $_GET['p'] : 1; $per_page = isset($_GET['pp']) ? $_GET['pp'] : 10; // Connect to the database $db = db_connect($database); // Save these variables for automatic URL formation later on. $display = array(); $display['id'] = $batch_id; $display['db'] = $database; $display['p'] = $page; $display['pp'] = $per_page; $display['filter_type'] = array(); // // Process the filtering parameters // $param = array(); process_filter($display); prepare_filter_parameters($display, $param); // // Prepare some SQL queries // $query = "SELECT batches.id as id, date, description, type FROM batches " . "WHERE batches.id=?"; if (!($db['batch_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT COUNT(tag_id) as count FROM catalog_index " . "WHERE batch_id=?"; $query .= apply_query_filters($display); if (!($db['count_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT alleles as count FROM catalog_index " . "WHERE batch_id=? AND tag_id=?"; if (!($db['allele_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT col, rank_2 FROM catalog_snps " . "JOIN batches ON (catalog_snps.batch_id=batches.id) " . "WHERE batch_id=? AND tag_id=? ORDER BY col"; if (!($db['snp_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT chr, max_len FROM chr_index " . "WHERE batch_id=?"; if (!($db['chrs_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT max(ests) as ests, max(pe_radtags) as pe_radtags, max(blast_hits) as blast_hits " . "FROM catalog_index WHERE batch_id=?"; if (!($db['seq_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); if (!$db['seq_sth']->bind_param("i", $batch_id)) write_db_error($db['seq_sth'], __FILE__, __LINE__); if (!$db['seq_sth']->execute()) write_db_error($db['seq_sth'], __FILE__, __LINE__); $res = $db['seq_sth']->get_result(); $row = $res->fetch_assoc(); $cols = array(); if ($row['ests'] == 0 && $row['pe_radtags'] == 0 && $row['blast_hits'] == 0) $cols['seq'] = false; else $cols['seq'] = true; $query = "SELECT count(id) as cnt FROM catalog_genotypes WHERE batch_id=?"; if (!($db['gcnt_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); if (!$db['gcnt_sth']->bind_param("i", $batch_id)) write_db_error($db['gcnt_sth'], __FILE__, __LINE__); if (!$db['gcnt_sth']->execute()) write_db_error($db['gcnt_sth'], __FILE__, __LINE__); $res = $db['gcnt_sth']->get_result(); $row = $res->fetch_assoc(); if ($row['cnt'] > 0) $cols['gcnt'] = true; else $cols['gcnt'] = false; // // Pull information about this batch // if (!$db['batch_sth']->bind_param("i", $batch_id)) write_db_error($db['batch_sth'], __FILE__, __LINE__); if (!$db['batch_sth']->execute()) write_db_error($db['batch_sth'], __FILE__, __LINE__); $res = $db['batch_sth']->get_result(); $row = $res->fetch_assoc(); $batch = array(); $batch['id'] = $row['id']; $batch['desc'] = $row['description']; $batch['date'] = $row['date']; $batch['type'] = $row['type']; $page_title = "RAD-Tag Catalog Viewer"; write_header($page_title, $batch); echo <<< EOQ

Batch #$batch[id] [$batch[date]; $batch[desc]]

EOQ; if ($batch['type'] == "population") write_pop_filter($cols); else write_map_filter($cols); // // How many columns will we print // $num_cols = 9; foreach ($cols as $col) if ($col == false) $num_cols--; // // Generate Excel export URL // $excel_export = generate_url("export_batch.php", false); echo <<< EOQ
EOQ; } else { echo <<< EOQ EOQ; } if ($cols['seq'] == true) print " \n"; print "\n"; $query = "SELECT catalog_index.tag_id as tag_id, alleles, parents, progeny, valid_progeny, " . "seq, marker, uncor_marker, chisq_pval, lnl, ratio, ests, pe_radtags, blast_hits, external_id, geno_cnt, " . "catalog_index.chr, catalog_index.bp, catalog_tags.strand, catalog_index.type, gene, ext_id, ex_start, ex_end, ex_index " . "FROM catalog_index " . "JOIN catalog_tags ON (catalog_index.cat_id=catalog_tags.id) " . "LEFT JOIN catalog_annotations ON (catalog_index.batch_id=catalog_annotations.batch_id AND catalog_index.tag_id=catalog_annotations.catalog_id) " . "LEFT JOIN ref_radome ON (catalog_index.ref_id=ref_radome.id) " . "WHERE catalog_index.batch_id=?"; $query .= apply_query_filters($display); $query .= " LIMIT " . ($start_group - 1) . ", " . $display['pp']; if (!($db['tag_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); array_unshift($param, $db['tag_sth']); call_user_func_array("mysqli_stmt_bind_param", $param); array_shift($param); if (!$db['tag_sth']->execute()) write_db_error($db['tag_sth'], __FILE__, __LINE__); $res = $db['tag_sth']->get_result(); $tag_id = 0; $db['snp_sth']->bind_param("ii", $batch_id, $tag_id); while ($row = $res->fetch_assoc()) { $tag_id = $row['tag_id']; // // Query the database to find how many SNPs were found in this sample. // $snps = array(); if (!$db['snp_sth']->execute()) write_db_error($db['snp_sth'], __FILE__, __LINE__); $snp_res = $db['snp_sth']->get_result(); while ($snp_row = $snp_res->fetch_assoc()) { array_push($snps, array('col' => $snp_row['col'], 'rank' => $snp_row['rank_2'])); } $url = "$root_path/pop_view.php?db=$database&batch_id=$batch_id&type=$batch[type]&tag_id=$tag_id"; $annotation = strlen($row['external_id']) > 0 ? $row['external_id'] : "annotate"; echo <<< EOQ EOQ; if (count($snps) == 0) print " \n"; else print " \n"; $s = print_snps($row['tag_id'], $row['seq'], $row['seq'], $snps, true); $ratio = explode(";", $row['ratio']); $ratio_parsed = ""; $i = 0; foreach ($ratio as $r) { if (strlen($r) == 0) continue; preg_match("/([a-z]+):(\d+)\((\d+\.?\d*%)\)/", $r, $matches); for ($j = 0; $j < strlen($matches[1]); $j++) { $color = $color_map[$matches[1][$j]]; $ratio_parsed .= "" . $matches[1][$j] . ""; } $ratio_parsed .= ": $matches[2]"; if ($matches[3] > 0) $ratio_parsed .= " ($matches[3])"; $ratio_parsed .= "
"; $i++; } $url = "$root_path/sequence_blast.php?db=$database&batch_id=$batch_id&tag_id=$row[tag_id]"; if ($row['blast_hits'] > 0 || $row['pe_radtags'] > 0 || $row['ests'] > 0) { $blast_hits_str = "
" . "" . "blast hits: $row[blast_hits]"; } else { $blast_hits_str = "blast hits: $row[blast_hits]"; } if (strlen($row['chr']) > 0) { print "
\n"; } else { print "\n"; } if ($batch['type'] == "map") { print " \n"; if (strlen($row['uncor_marker']) > 0 && $row['marker'] != $row['uncor_marker']) print " \n"; else print " \n"; } else { print " \n"; } echo <<< EOQ EOQ; if ($batch['type'] == "map") print "\n"; if ($cols['seq'] == true) echo <<< EOQ EOQ; echo <<< EOQ EOQ; } print "\n" . "
EOQ; // // Figure out how many results there are (including filtering) // and write out the proper pagination links // array_unshift($param, $db['count_sth']); call_user_func_array("mysqli_stmt_bind_param", $param); array_shift($param); if (!$db['count_sth']->execute()) write_db_error($db['count_sth'], __FILE__, __LINE__); $res = $db['count_sth']->get_result(); $row = $res->fetch_assoc(); $pagination_count = $row['count']; $start_group = 0; $end_group = 0; write_pagination($pagination_count, $start_group, $end_group, "catalog.php"); if ($batch['type'] == "map") { echo <<< EOQ
Id SNP Consensus Matches Marker Ratio ChiSq P-value
Id SNP Consensus Matches RatioSequence
$annotation
NoYes [" . count($snps) . "nuc]\n" . "
$s
\n" . "
Chr: $row[chr], " . print_bp($row['bp']) . ", $row[strand]\n"; if ($row['type'] == "exon") { if (strlen($row['ext_id']) == 0) $gene = $row['gene']; else $gene = $row['ext_id']; print ", Gene: " . $gene . "\n"; } print ", LnL: $row[lnl]
$s
\n" . "
LnL: $row[lnl]
\n" . "		$row[parents]" . " / " . "$row[progeny]" . " / " . "$row[valid_progeny]" . " / " . "$row[geno_cnt]$row[marker]*$row[marker]$row[parents] $ratio_parsed $row[chisq_pval]
ests: $row[ests]pe: $row[pe_radtags]
$blast_hits_str
\n"; write_pagination($pagination_count, $start_group, $end_group, "catalog.php"); echo <<< EOQ
EOQ; write_footer(); function generate_hidden_form_vars($var) { global $root_path, $display; $vars = ""; foreach ($display as $key => $d) { if (strstr($key, $var)) continue; if (is_array($d)) { foreach ($d as $e) { $vars .= " \n"; } } else { $vars .= " \n"; } } return $vars; } function generate_per_page_select($name, $per_page) { $pages = array("10", "50", "100", "all"); $ctl = " \n"; return $ctl; } function generate_url($destination, $prefix) { global $root_path, $display; if ($prefix) $url = "href=\"" . $root_path . "/" . $destination . "?"; else $url = $root_path . "/" . $destination . "?"; foreach ($display as $key => $d) { if (is_array($d)) { foreach ($d as $e) $url .= "{$key}[]=$e&"; } else { $url .= "$key=$d&"; } } // Remove the hanging '&' $url = substr($url, 0, -1); if ($prefix) $url .= "\""; return $url; } function generate_page_list($page, $num_pages, $destination) { global $display; $page_list = ""; if ($page <= 4) { for ($i = 1; $i < $page; $i++) { $display['p'] = $i; $p = generate_url($destination, true); $page_list .= "$i\n"; } } else { $display['p'] = 1; $p = generate_url($destination, true); $page_list .= "1 ...\n"; foreach (array($page - 3, $page - 2, $page - 1) as $i) { $display['p'] = $i; $p = generate_url($destination, true); $page_list .= "$i\n"; } } $page_list .= " $page\n"; if ($page <= $num_pages - 4) { for ($i = $page + 1; $i <= $page + 3; $i++) { $display['p'] = $i; $p = generate_url($destination, true); $page_list .= "$i\n"; } $display['p'] = $num_pages; $p = generate_url($destination, true); $page_list .= "... $num_pages\n"; } else { for ($i = $page + 1; $i <= $num_pages; $i++) { $display['p'] = $i; $p = generate_url($destination, true); $page_list .= "$i\n"; } } $display['p'] = $page; $page_list = "\n" . $page_list . "\n"; return $page_list; } function write_pagination($num_tags, &$start_gene, &$end_gene, $destination) { global $img_path, $root_path, $display; $cur_page = $display['p']; $page = $display['p']; $per_page = $display['pp']; if ($per_page == "all") $per_page = $num_tags; // // First figure out the total number of pages. If there are // additional genes left over, add an extra page. // $num_pages = floor($num_tags / $per_page); $num_pages += $num_tags % $per_page >= 1 ? 1 : 0; if ($num_pages > 0 && $page > $num_pages) { $page = $num_pages; $cur_page = $num_pages; } // Determine the start and end gene numbers $start_gene = 1 + (($page - 1) * $per_page); $end_gene = $start_gene + $per_page > $num_tags ? $num_tags : ($start_gene + $per_page - 1); // Generate the URLs for our links $display['p'] -= 1; $prev_page = generate_url($destination, true); $display['p'] += 2; $next_page = generate_url($destination, true); $display['p'] = $cur_page; print "\n" . "\n" . "\n"; $page_list = ""; if ($num_pages > 1) $page_list = generate_page_list($page, $num_pages, $destination); print $page_list; $hidden_vars = generate_hidden_form_vars("pp"); $per_page_ctl = generate_per_page_select("pp", $display['pp']); echo <<< EOQ
\n"; if ($page == 1) { if ($num_pages == 1) { echo <<< EOQ No Previous Page $page No Next Page EOQ; } else { echo <<< EOQ No Previous Page $page View Next Page EOQ; } } else if ($page == $num_pages) { echo <<< EOQ View Previous Page $page No Next Page EOQ; } else { echo <<< EOQ View Previous Page $page View Next Page EOQ; } print " ($num_tags tags)\n" . "
$hidden_vars tags per page   $per_page_ctl
EOQ; } function write_map_filter($cols) { global $img_path, $root_path, $display; $max_chr_len = 0; $hidden_vars = generate_hidden_form_vars("filter"); $chrs = fetch_chrs($max_chr_len); $filters = array("cata" => array(), "alle" => array(), "snps" => array(), "pare" => array(), "prog" => array(), "vprog" => array(), "mark" => array(), "gcnt" => array(), "chisq" => array(), "loc" => array(), "ref" => array(), "est" => array(), "pe" => array(), "blast" => array()); $fall = isset($display['filter_alle_l']) ? $display['filter_alle_l'] : ""; $falu = isset($display['filter_alle_u']) ? $display['filter_alle_u'] : "100"; $fsnl = isset($display['filter_snps_l']) ? $display['filter_snps_l'] : ""; $fsnu = isset($display['filter_snps_u']) ? $display['filter_snps_u'] : "100"; $fpal = isset($display['filter_pare_l']) ? $display['filter_pare_l'] : ""; $fpau = isset($display['filter_pare_u']) ? $display['filter_pare_u'] : "2"; $fpr = isset($display['filter_prog']) ? $display['filter_prog'] : ""; $fvp = isset($display['filter_vprog']) ? $display['filter_vprog'] : ""; $fgc = isset($display['filter_gcnt']) ? $display['filter_gcnt'] : ""; $csql = isset($display['filter_chisq_l']) ? $display['filter_chisq_l'] : "0.0"; $csqu = isset($display['filter_chisq_u']) ? $display['filter_chisq_u'] : "1.0"; $fma = isset($display['filter_mark']) ? $display['filter_mark'] : ""; $ref = isset($display['filter_ref']) ? $display['filter_ref'] : ""; $fch = isset($display['filter_chr']) ? $display['filter_chr'] : ""; $fsb = isset($display['filter_sbp']) ? $display['filter_sbp'] : 0; $feb = isset($display['filter_ebp']) ? $display['filter_ebp'] : $max_chr_len; $r = range(1, 9); $r = array_merge($r, range(10, 100, 5)); array_push($r, 1000); $alle_l_ctl = generate_element_select("filter_alle_l", $r, $fall, ""); $alle_u_ctl = generate_element_select("filter_alle_u", $r, $falu, ""); $snps_l_ctl = generate_element_select("filter_snps_l", $r, $fsnl, ""); $snps_u_ctl = generate_element_select("filter_snps_u", $r, $fsnu, ""); $r = range(1, 9); $r = array_merge($r, range(10, 500, 10)); array_push($r, 1000, 2000, 10000); $pare_l_ctl = generate_element_select("filter_pare_l", $r, $fpal, ""); $pare_u_ctl = generate_element_select("filter_pare_u", $r, $fpau, ""); $prog_ctl = generate_element_select("filter_prog", $r, $fpr, ""); $vprog_ctl = generate_element_select("filter_vprog", $r, $fvp, ""); $gcnt_ctl = generate_element_select("filter_gcnt", $r, $fgc, ""); $csql_ctl = generate_element_select("filter_chisq_l", array(0.0, 0.0005, 0.001, 0.005, 0.01, 0.05, 0.1, 0.15, 0.2, 0.25, 0.5, 1.0), $csql, ""); $csqu_ctl = generate_element_select("filter_chisq_u", array(0.0, 0.0005, 0.001, 0.005, 0.01, 0.05, 0.1, 0.15, 0.2, 0.25, 0.5, 1.0), $csqu, ""); $chr_ctl = generate_element_select("filter_chr", $chrs, $fch, ""); $sbp_ctl = generate_element_select("filter_sbp", range(0, $max_chr_len), $fsb, ""); $ebp_ctl = generate_element_select("filter_ebp", range(0, $max_chr_len), $feb, ""); $ref_ctl = generate_element_select("filter_ref", array("exon", "intron", "genomic"), $ref, ""); $mark_ctl = generate_element_select("filter_mark", array('Any', 'aa/bb', 'ab/--', '--/ab', 'aa/ab', 'ab/aa', 'ab/ab', 'ab/ac', 'ab/cd', 'ab/cc', 'cc/ab'), $fma, ""); if (isset($display['filter_type'])) { foreach ($filters as $key => $f) if (in_array($key, $display['filter_type'])) { $filters[$key]['sel'] = "checked=\"checked\""; $filters[$key]['tr'] = "class=\"active_filter\""; } else { $filters[$key]['sel'] = ""; $filters[$key]['tr'] = ""; } } else { $filters['none']['sel'] = "checked=\"checked\""; } echo <<< EOQ

Filter Results By

$hidden_vars EOQ; if (count($chrs) > 0) { echo <<< EOQ EOQ; } $cat_id_filter = isset($display['filter_cata']) ? $display['filter_cata'] : ""; echo <<< EOQ
Location: $chr_ctl
Start: $sbp_ctl Mb
End: $ebp_ctl Mb
Type: $ref_ctl
EOQ; if ($cols['seq'] == true) echo <<< EOQ EOQ; echo <<< EOQ
Catalog ID:
Alleles: $alle_l_ctl $alle_u_ctl
SNPs: $snps_l_ctl $snps_u_ctl
   
Parental matches: $pare_l_ctl $pare_u_ctl
Progeny matches: $prog_ctl
Segregation distortion: $csql_ctl $csqu_ctl
Mappable progeny: $vprog_ctl
Mappable markers: $mark_ctl
Genotypes: $gcnt_ctl
Contains ESTs Contains Paired-end RAD-Tags Contains BLAST Hits
EOQ; } function write_pop_filter($cols) { global $img_path, $root_path, $display; $max_chr_len = 0; $hidden_vars = generate_hidden_form_vars("filter"); $chrs = fetch_chrs($max_chr_len); $filters = array("cata" => array(), "alle" => array(), "snps" => array(), "pare" => array(), "gcnt" => array(), "lnl" => array(), "loc" => array(), "ref" => array(), "est" => array(), "pe" => array(), "blast" => array()); $fall = isset($display['filter_alle_l']) ? $display['filter_alle_l'] : ""; $falu = isset($display['filter_alle_u']) ? $display['filter_alle_u'] : "100"; $fsnl = isset($display['filter_snps_l']) ? $display['filter_snps_l'] : ""; $fsnu = isset($display['filter_snps_u']) ? $display['filter_snps_u'] : "100"; $fpal = isset($display['filter_pare_l']) ? $display['filter_pare_l'] : ""; $fpau = isset($display['filter_pare_u']) ? $display['filter_pare_u'] : "1000"; $ref = isset($display['filter_ref']) ? $display['filter_ref'] : ""; $fch = isset($display['filter_chr']) ? $display['filter_chr'] : ""; $fsb = isset($display['filter_sbp']) ? $display['filter_sbp'] : 0; $feb = isset($display['filter_ebp']) ? $display['filter_ebp'] : $max_chr_len; $flnl = isset($display['filter_lnl_l']) ? $display['filter_lnl_l'] : -500; $flnu = isset($display['filter_lnl_u']) ? $display['filter_lnl_u'] : 0; $r = range(1, 9); $r = array_merge($r, range(10, 100, 5)); array_push($r, 1000); $alle_l_ctl = generate_element_select("filter_alle_l", $r, $fall, ""); $alle_u_ctl = generate_element_select("filter_alle_u", $r, $falu, ""); $snps_l_ctl = generate_element_select("filter_snps_l", $r, $fsnl, ""); $snps_u_ctl = generate_element_select("filter_snps_u", $r, $fsnu, ""); $r = range(1, 9); $r = array_merge($r, range(10, 500, 10)); array_push($r, 1000, 2000, 10000); $pare_l_ctl = generate_element_select("filter_pare_l", $r, $fpal, ""); $pare_u_ctl = generate_element_select("filter_pare_u", $r, $fpau, ""); $chr_ctl = generate_element_select("filter_chr", $chrs, $fch, ""); $sbp_ctl = generate_element_select("filter_sbp", range(0, $max_chr_len), $fsb, ""); $ebp_ctl = generate_element_select("filter_ebp", range(0, $max_chr_len), $feb, ""); $ref_ctl = generate_element_select("filter_ref", array("exon", "intron", "genomic"), $ref, ""); $r = range(0, 9); $r = array_merge($r, range(10, 100, 5)); $r = array_merge($r, range(200, 500, 100)); for ($i = 0; $i < count($r); $i++) $r[$i] = $r[$i] * -1; $lnl_l_ctl = generate_element_select("filter_lnl_l", $r, $flnl, ""); $lnl_u_ctl = generate_element_select("filter_lnl_u", $r, $flnu, ""); if (isset($display['filter_type'])) { foreach ($filters as $key => $f) if (in_array($key, $display['filter_type'])) { $filters[$key]['sel'] = "checked=\"checked\""; $filters[$key]['tr'] = "class=\"active_filter\""; } else { $filters[$key]['sel'] = ""; $filters[$key]['tr'] = ""; } } else { $filters['none']['sel'] = "checked=\"checked\""; } echo <<< EOQ

Filter Results By

$hidden_vars EOQ; if (count($chrs) > 0) { echo <<< EOQ EOQ; } $cat_id_filter = isset($display['filter_cata']) ? $display['filter_cata'] : ""; echo <<< EOQ
Location: $chr_ctl
Start: $sbp_ctl Mb
End: $ebp_ctl Mb
Type: $ref_ctl
EOQ; if ($cols['seq'] == true) echo <<< EOQ EOQ; echo <<< EOQ
Catalog ID:
Alleles: $alle_l_ctl $alle_u_ctl
SNPs: $snps_l_ctl $snps_u_ctl
   
Matching samples: $pare_l_ctl $pare_u_ctl
LnL: $lnl_l_ctl $lnl_u_ctl
   
EOQ; if ($cols['gcnt'] == true) { echo <<< EOQ EOQ; } else { echo <<< EOQ EOQ; } echo <<< EOQ
   
   
$gcnt_ctl
   
Contains ESTs Contains Paired-end RAD-Tags Contains BLAST Hits
EOQ; } function process_filter(&$display_params) { if (!isset($_GET['filter_type'])) return; foreach ($_GET['filter_type'] as $filter) { array_push($display_params['filter_type'], $filter); if ($filter == "alle") { $display_params['filter_alle_l'] = $_GET['filter_alle_l']; $display_params['filter_alle_u'] = $_GET['filter_alle_u']; } else if ($filter == "snps") { $display_params['filter_snps_l'] = $_GET['filter_snps_l']; $display_params['filter_snps_u'] = $_GET['filter_snps_u']; } else if ($filter == "pare") { $display_params['filter_pare_l'] = $_GET['filter_pare_l']; $display_params['filter_pare_u'] = $_GET['filter_pare_u']; } else if ($filter == "lnl") { $display_params['filter_lnl_l'] = $_GET['filter_lnl_l']; $display_params['filter_lnl_u'] = $_GET['filter_lnl_u']; } else if ($filter == "prog") { $display_params['filter_prog'] = $_GET['filter_prog']; } else if ($filter == "vprog") { $display_params['filter_vprog'] = $_GET['filter_vprog']; } else if ($filter == "cata") { $display_params['filter_cata'] = $_GET['filter_cata']; } else if ($filter == "mark") { $display_params['filter_mark'] = ($_GET['filter_mark'] == "Any") ? "%/%" : $_GET['filter_mark']; } else if ($filter == "gcnt") { $display_params['filter_gcnt'] = $_GET['filter_gcnt']; } else if ($filter == "chisq") { $display_params['filter_chisq_l'] = $_GET['filter_chisq_l']; $display_params['filter_chisq_u'] = $_GET['filter_chisq_u']; } else if ($filter == "ref") { $display_params['filter_ref'] = $_GET['filter_ref']; } else if ($filter == "loc") { $display_params['filter_chr'] = $_GET['filter_chr']; $display_params['filter_sbp'] = $_GET['filter_sbp'] * 1000000; $display_params['filter_ebp'] = $_GET['filter_ebp'] * 1000000; } } } function prepare_filter_parameters(&$display_params, &$param) { $filters = $display_params['filter_type']; $param[] = &$display_params['id']; $typestr = "i"; if (!isset($filters)) { array_unshift($param, $typestr); return; } foreach ($filters as $filter) { if ($filter == "snps") { $param[] = &$display_params['filter_snps_l']; $param[] = &$display_params['filter_snps_u']; $typestr .= "ii"; } else if ($filter == "alle") { $param[] = &$display_params['filter_alle_l']; $param[] = &$display_params['filter_alle_u']; $typestr .= "ii"; } else if ($filter == "pare") { $param[] = &$display_params['filter_pare_l']; $param[] = &$display_params['filter_pare_u']; $typestr .= "ii"; } else if ($filter == "lnl") { $param[] = &$display_params['filter_lnl_l']; $param[] = &$display_params['filter_lnl_u']; $typestr .= "ii"; } else if ($filter == "prog") { $param[] = &$display_params['filter_prog']; $typestr .= "i"; } else if ($filter == "vprog") { $param[] = &$display_params['filter_vprog']; $typestr .= "i"; } else if ($filter == "cata") { $param[] = &$display_params['filter_cata']; $typestr .= "i"; } else if ($filter == "gcnt") { $param[] = &$display_params['filter_gcnt']; $typestr .= "i"; } else if ($filter == "chisq") { $param[] = &$display_params['filter_chisq_l']; $param[] = &$display_params['filter_chisq_u']; $typestr .= "dd"; } else if ($filter == "ref") { $param[] = &$display_params['filter_ref']; $typestr .= "s"; } else if ($filter == "loc") { $param[] = &$display_params['filter_chr']; $param[] = &$display_params['filter_sbp']; $param[] = &$display_params['filter_ebp']; $typestr .= "sii"; } else if ($filter == "mark") { $param[] = &$display_params['filter_mark']; $typestr .= "s"; } } array_unshift($param, $typestr); } function apply_query_filters($display_params) { $order = 0; $query = ""; $sql_filters = array("cata" => "(catalog_index.tag_id = ?)", "alle" => "(alleles >= ? AND alleles <= ?)", "snps" => "(snps >= ? AND snps <= ?)", "pare" => "(parents >= ? AND parents <= ?)", "prog" => "(progeny >= ?)", "vprog" => "(valid_progeny >= ?)", "lnl" => "(lnl >= ? AND lnl <= ?)", "mark" => "(marker LIKE ?)", "est" => "(ests > 0)", "pe" => "(pe_radtags > 0)", "blast" => "(blast_hits > 0)", "gcnt" => "(geno_cnt >= ?)", "chisq" => "(chisq_pval >= ? AND chisq_pval <= ?)", "ref" => "(catalog_index.type = ?)", "loc" => "(catalog_index.chr = ? && catalog_index.bp >= ? && catalog_index.bp <= ?)"); $filters = $display_params['filter_type']; if (count($filters) > 0) { $query = " AND "; while (count($filters) > 0) { $filter = array_shift($filters); $query .= $sql_filters[$filter]; $query .= count($filters) > 0 ? " AND " : ""; if ($filter == "loc") $order++; } if ($order) $query .= " ORDER BY chr, bp"; } return $query; } function fetch_chrs(&$max_len) { global $db, $batch_id; $max_len = 0; $chrs = array(); if (!$db['chrs_sth']->bind_param("i", $batch_id)) write_db_error($db['chrs_sth'], __FILE__, __LINE__); if (!$db['chrs_sth']->execute()) write_db_error($db['chrs_sth'], __FILE__, __LINE__); $res = $db['chrs_sth']->get_result(); while ($row = $res->fetch_assoc()) { if ($row['max_len'] > $max_len) $max_len = $row['max_len']; array_push($chrs, $row['chr']); } return $chrs; } ?> stacks-1.44/php/catalog_genotypes.php000644 000765 000024 00000014514 12663652222 020524 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); if (isset($_GET['db'])) $database = $_GET['db']; else if (isset($_POST['db'])) $database = $_POST['db']; if (isset($_GET['tag_id'])) $tag_id = $_GET['tag_id']; else if (isset($_POST['tag_id'])) $tag_id = $_POST['tag_id']; else $tag_id = 0; if (isset($_GET['batch_id'])) $batch_id = $_GET['batch_id']; else if (isset($_POST['batch_id'])) $batch_id = $_POST['batch_id']; else $batch_id = 0; // Connect to the database $db = db_connect($database); // Save these variables for automatic URL formation later on. $display = array(); $display['db'] = $database; $display['tag_id'] = $tag_id; $display['batch_id'] = $batch_id; // // Prepare the possible select lists we will want to construct // $marker_types = array('ab/--' => array('aa', 'bb', '-'), '--/ab' => array('aa', 'bb', '-'), 'ab/aa' => array('aa', 'ab', '-'), 'aa/ab' => array('aa', 'ab', '-'), 'ab/ab' => array('aa', 'ab', 'bb', '-'), 'ab/ac' => array('aa', 'ab', 'ac', 'bc', '-'), 'ab/cd' => array('aa', 'bb', 'cc', 'dd', 'ac', 'ad', 'bc', 'bd', '-'), 'aa/bb' => array('aa', 'bb', 'ab', '-'), 'ab/cc' => array('aa', 'bb', 'ab', 'ac', 'bc', 'cc', '-'), 'cc/ab' => array('aa', 'bb', 'ab', 'ac', 'bc', 'cc', '-')); // // Prepare some SQL queries // $query = "SELECT count(samples.id) as count FROM samples WHERE batch_id=?"; if (!($db['samp_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT marker, catalog_genotypes.sample_id, file, " . "catalog_genotypes.genotype, genotype_corrections.genotype as corrected " . "FROM catalog_genotypes " . "LEFT JOIN genotype_corrections ON " . "(genotype_corrections.catalog_id=catalog_genotypes.catalog_id AND " . "genotype_corrections.sample_id=catalog_genotypes.sample_id AND " . "genotype_corrections.batch_id=catalog_genotypes.batch_id) " . "JOIN samples ON (catalog_genotypes.sample_id=samples.id) " . "JOIN catalog_index ON (catalog_genotypes.catalog_id=catalog_index.tag_id AND " . "catalog_genotypes.batch_id=catalog_index.batch_id) " . "WHERE catalog_genotypes.batch_id=? and catalog_genotypes.catalog_id=? " . "ORDER BY catalog_genotypes.sample_id"; if (!($db['geno_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $page_title = "Catalog Genotype Viewer"; write_compact_header($page_title); // // Get number of samples so we can determine how many rows to display // in the genotype table. // if (!$db['samp_sth']->bind_param("i", $batch_id)) write_db_error($db['samp_sth'], __FILE__, __LINE__); if (!$db['samp_sth']->execute()) write_db_error($db['samp_sth'], __FILE__, __LINE__); $res = $db['samp_sth']->get_result(); $row = $res->fetch_assoc(); $num_samples = $row['count']; $num_cols = 10; $num_rows = ceil($num_samples / $num_cols); $gtypes = array(); if (!$db['geno_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['geno_sth'], __FILE__, __LINE__); if (!$db['geno_sth']->execute()) write_db_error($db['geno_sth'], __FILE__, __LINE__); $res = $db['geno_sth']->get_result(); if ($res->num_rows == 0) { print "

" . "This marker has no genotypes, probably because this tag does not have enough mappable progeny.

\n"; write_compact_footer(); return; } while ($row = $res->fetch_assoc()) { $gtypes[$row['sample_id']] = array('file' => $row['file'], 'genotype' => $row['genotype'], 'corrected' => $row['corrected'], 'marker' => $row['marker']); } print "
\n" . "\n" . "\n" . "\n" . "\n" . "\n" . "\n"; $i = 0; foreach ($gtypes as $sample_id => $sample) { $i++; $id = "gtype_" . $batch_id . "_" . $tag_id . "_" . $sample_id; if (strlen($sample['corrected']) > 0) { $sel = generate_element_select($id, $marker_types[$sample['marker']], strtolower($sample['corrected']), ""); $genotype = "$sample[corrected]"; } else { $sel = generate_element_select($id, $marker_types[$sample['marker']], strtolower($sample['genotype']), ""); $genotype = $sample['genotype']; } print " \n"; if ($i % $num_cols == 0) print "\n" . "\n"; } while ($i % $num_cols != 0) { print " \n"; $i++; } echo <<< EOQ
" . "" . ucfirst(str_replace("_", " ", $sample['file'])) . "
" . "" . $genotype . "\n" . "
\n" . $sel . "
" . "
EOQ; write_compact_footer(); ?> stacks-1.44/php/CatalogClass.php000644 000765 000024 00000024267 12665337341 017367 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("Locus.php"); class Catalog { var $db; // Hash of database statement handles. var $display; // An array of form variables necessary to write URLs. var $params; // An array of parameters to fetch our results from the database. var $queries; // An array of SQL queries. var $loci; // Array of groups objects. var $num_loci; // Total number of groups, after filtering function Catalog($db, $batch_id, $display_params) { $this->db = $db; $this->batch = $batch_id; $this->display = $display_params; $this->queries = array(); $this->params = array(); $this->loci = array(); $this->num_loci = 0; $this->prepare_queries(); } function prepare_queries() { $query = "SELECT allele FROM catalog_alleles " . "WHERE batch_id=? AND tag_id=?"; if (!($this->db['allele_sth'] = $this->db['dbh']->prepare($query))) write_db_error($this->db['dbh'], __FILE__, __LINE__); $query = "SELECT col, rank_1, rank_2 FROM catalog_snps " . "WHERE batch_id=? AND tag_id=? ORDER BY col"; if (!($this->db['snp_sth'] = $this->db['dbh']->prepare($query))) write_db_error($this->db['dbh'], __FILE__, __LINE__); $query = "SELECT samples.id, samples.sample_id, samples.type, file, tag_id, allele " . "FROM matches " . "JOIN samples ON (matches.sample_id=samples.id) " . "WHERE matches.batch_id=? AND catalog_id=? ORDER BY samples.id"; if (!($this->db['mat_sth'] = $this->db['dbh']->prepare($query))) write_db_error($this->db['dbh'], __FILE__, __LINE__); $query = "SELECT COUNT(tag_id) as count FROM catalog_index " . "WHERE batch_id=?"; $query .= $this->apply_query_filters(); $this->queries['tag_count'] = $query; $query = "SELECT catalog_index.tag_id as tag_id, chr, bp, snps, alleles, parents, progeny, valid_progeny, " . "seq, marker, max_pct, ratio, ests, pe_radtags, blast_hits, external_id " . "FROM catalog_index " . "JOIN catalog_tags ON (catalog_index.cat_id=catalog_tags.id) " . "LEFT JOIN catalog_annotations ON " . "(" . "catalog_index.batch_id=catalog_annotations.batch_id AND " . "catalog_index.tag_id=catalog_annotations.catalog_id" . ") " . "WHERE catalog_index.batch_id=?"; $query .= $this->apply_query_filters(); $this->queries['tag'] = $query; } function prepare_filter_parameters() { $this->params[] = &$this->batch; $typestr = "i"; $filters = $this->display['filter_type']; if (!isset($filters)) { array_unshift($param, $typestr); return; } foreach ($filters as $filter) { if ($filter == "snps") { $this->params[] = &$this->display['filter_snps_l']; $this->params[] = &$this->display['filter_snps_u']; $typestr .= "ii"; } else if ($filter == "alle") { $this->params[] = &$this->display['filter_alle_l']; $this->params[] = &$this->display['filter_alle_u']; $typestr .= "ii"; } else if ($filter == "pare") { $this->params[] = &$this->display['filter_pare_l']; $this->params[] = &$this->display['filter_pare_u']; $typestr .= "ii"; } else if ($filter == "prog") { $this->params[] = &$this->display['filter_prog']; $typestr .= "i"; } else if ($filter == "vprog") { $this->params[] = &$this->display['filter_vprog']; $typestr .= "i"; } else if ($filter == "lnl") { $this->params[] = &$this->display['filter_lnl_l']; $this->params[] = &$this->display['filter_lnl_u']; $typestr .= "ii"; } else if ($filter == "cata") { $this->params[] = &$this->display['filter_cata']; $typestr .= "i"; } else if ($filter == "gcnt") { $this->params[] = &$this->display['filter_gcnt']; $typestr .= "i"; } else if ($filter == "ref") { $this->params[] = &$this->display['filter_ref']; $typestr .= "i"; } else if ($filter == "loc") { $this->params[] = &$this->display['filter_chr']; $this->params[] = &$this->display['filter_sbp']; $this->params[] = &$this->display['filter_ebp']; $typestr .= "sii"; } else if ($filter == "mark") { $this->params[] = &$this->display['filter_mark']; } } array_unshift($this->params, $typestr); } function apply_query_filters() { $sql_filters = array("cata" => "(catalog_index.tag_id = ?)", "alle" => "(alleles >= ? AND alleles <= ?)", "snps" => "(snps >= ? AND snps <= ?)", "pare" => "(parents >= ? AND parents <= ?)", "prog" => "(progeny >= ?)", "vprog" => "(valid_progeny >= ?)", "lnl" => "(lnl >= ? AND lnl <= ?)", "mark" => "(marker LIKE ?)", "est" => "(ests > 0)", "pe" => "(pe_radtags > 0)", "blast" => "(blast_hits > 0)", "gcnt" => "(geno_cnt >= ?)", "ref" => "(catalog_index.type = ?)", "loc" => "(catalog_index.chr = ? && catalog_index.bp >= ? && catalog_index.bp <= ?)"); $filters = $this->display['filter_type']; if (count($filters) > 0) { $query = " AND "; while (count($filters) > 0) { $filter = array_shift($filters); $query .= $sql_filters[$filter]; $query .= count($filters) > 0 ? " AND " : ""; } } return $query; } function &loci() { return $this->loci; } function &locus($id) { return $this->loci[$id]; } function num_loci() { return $this->num_loci; } function determine_count() { if (!($this->db['tag_count_sth'] = $this->db['dbh']->prepare($this->queries['tag_count']))) write_db_error($this->db['dbh'], __FILE__, __LINE__); $this->prepare_filter_parameters(); array_unshift($this->params, $this->db['tag_count_sth']); call_user_func_array("mysqli_stmt_bind_param", $this->params); array_shift($this->params); if (!$this->db['tag_count_sth']->execute()) write_db_error($this->db['tag_count_sth'], __FILE__, __LINE__); $res = $this->db['tag_count_sth']->get_result(); $row = $res->fetch_assoc(); $this->num_loci = $row['count']; } function populate($start_group, $num_groups) { // // We only want to load genes between $start_gene and $end_gene. // $this->queries['tag'] .= " LIMIT " . $start_group . ", " . $num_groups; if (!($this->db['tag_sth'] = $this->db['dbh']->prepare($this->queries['tag']))) write_db_error($this->db['dbh'], __FILE__, __LINE__); $this->prepare_filter_parameters(); // // Fetch the results and populate the array of groups. // array_unshift($this->params, $this->db['tag_sth']); call_user_func_array("mysqli_stmt_bind_param", $this->params); array_shift($this->params); if (!$this->db['tag_sth']->execute()) write_db_error($this->db['tag_sth'], __FILE__, __LINE__); $res = $this->db['tag_sth']->get_result(); while ($row = $res->fetch_assoc()) { $locus = new Locus(); $locus->id = $row['tag_id']; $locus->annotation = $row['external_id']; $locus->chr = $row['chr']; $locus->bp = $row['bp']; $locus->marker = $row['marker']; $locus->seq = $row['seq']; $locus->num_alleles = $row['alleles']; $locus->num_snps = $row['snps']; $locus->num_parents = $row['parents']; $locus->num_progeny = $row['progeny']; $locus->valid_progeny = $row['valid_progeny']; $locus->num_ests = $row['ests']; $locus->num_pe_tags = $row['pe_radtags']; $locus->num_blast = $row['blast_hits']; // // Fetch SNPs and Alleles // if (!$this->db['snp_sth']->bind_param("ii", $this->batch, $locus->id)) write_db_error($db['snp_sth'], __FILE__, __LINE__); if (!$this->db['snp_sth']->execute()) write_db_error($db['snp_sth'], __FILE__, __LINE__); $snp_res = $this->db['snp_sth']->get_result(); while ($snp_row = $snp_res->fetch_assoc()) { $locus->snps .= $snp_row['col'] . "," . $snp_row['rank_1'] . ">" . $snp_row['rank_2'] . ";"; } $locus->snps = substr($locus->snps, 0, -1); if (!$this->db['all_sth']->bind_param("ii", $this->batch, $locus->id)) write_db_error($db['all_sth'], __FILE__, __LINE__); if (!$this->db['all_sth']->execute()) write_db_error($db['all_sth'], __FILE__, __LINE__); $all_res = $this->db['all_sth']->get_result(); while ($all_row = $all_res->fetch_assoc()) { $locus->alleles .= $all_row['allele'] . ";"; } $locus->alleles = substr($locus->alleles, 0, -1); // // Add genotypes // if (!$this->db['mat_sth']->bind_param("ii", $this->batch, $locus->id)) write_db_error($db['mat_sth'], __FILE__, __LINE__); if (!$this->db['mat_sth']->execute()) write_db_error($db['mat_sth'], __FILE__, __LINE__); $gen_res = $this->db['mat_sth']->get_result(); while ($gen_row = $gen_res->fetch_assoc()) $locus->add_genotype($gen_row['id'], $gen_row['file'], $gen_row['allele']); $this->loci[$row['tag_id']] = $locus; } } } stacks-1.44/php/constants.php.dist000644 000765 000024 00000005762 12533677757 020020 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // // // Credentials to access Stacks MySQL databases // $db_user = "dbuser"; $db_pass = "dbpass"; $db_host = "localhost"; // // Path to the MySQL client program // $mysql_bin = "/usr/bin/mysql"; // // File system location of PHP/HTML files // $system_path = "_PKGDATADIR_" . "php/"; // // Location of perl script to run the exporting jobs. // $export_cmd = "_BINDIR_" . "stacks_export_notify.pl"; // // Name to print in page header // $site_title = "Stacks Analysis Pipeline"; // // URL path to root of PHP/HTML files // $root_path = "/stacks"; $img_path = "/stacks/images"; // // Length to wrap sequences at. // $display_len = 80; // // Colors for printing alleles/haplotypes // $colors = array("#008000", "#c00000", "#ffc600", "#29356c", "#860000", "#dc6200", "#4b398e", "#008f56", "#bf1e25", "#4cb8ff"); $color_size = count($colors); // // Color map version 2; for printing alleles/haplotypes. // $color_map = array( 'a' => "#0074D9", // Blue 'b' => "#FF4136", // Red 'c' => "#008000", // Dark Green 'd' => "#FF851B", // Orange 'e' => "#001f3f", // Navy 'f' => "#85144b", // Maroon 'g' => "#F012BE", // Fuchsia 'h' => "#39CCCC", // Teal 'i' => "#3D9970", // Olive 'j' => "#01FF70", // Lime 'k' => "#FFDC00", // Yellow 'l' => "#B10DC9", // Purple 'm' => "#111111", // Black 'n' => "#7FDBFF", // Aqua 'o' => "#AAAAAA", // Gray 'p' => "#DDDDDD", // Silver 'q' => "#2ECC40", // Green 'r' => "#c00000", 's' => "#ffc600", 't' => "#29356c", 'u' => "#860000", 'v' => "#dc6200", 'w' => "#4b398e", 'x' => "#008f56", 'y' => "#bf1e25", 'z' => "#4cb8ff", 'A' => "#26081C", 'B' => "#C44900", 'C' => "#036016", 'D' => "#936426", 'E' => "#E87EA1", 'F' => "#692E38", 'G' => "#19423F", 'H' => "#625F64", 'I' => "#432534", 'J' => "#323845", 'K' => "#758E4F", 'L' => "#B287A3", 'M' => "#C0F8D1", 'N' => "#03440C", 'O' => "#8EDCE6", 'P' => "#D5DCF9", 'Q' => "#A7B0CA", 'R' => "#0CBABA", 'S' => "#725E54", 'T' => "#EE2677", 'U' => "#84A98C", 'V' => "#FCCA46", 'W' => "#A1C181", 'X' => "#443627", 'Y' => "#FE7F2D", 'Z' => "#2EBE62" ); $color_map_size = count($color_map); ?> stacks-1.44/php/correct_genotype.php000644 000765 000024 00000011620 12663652222 020363 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); $database = isset($_GET['db']) ? $_GET['db'] : ""; $tag_id = isset($_GET['tag_id']) ? $_GET['tag_id'] : 0; $batch_id = isset($_GET['batch_id']) ? $_GET['batch_id'] : 0; $sample_id = isset($_GET['sample_id']) ? $_GET['sample_id'] : 0; $gtype = isset($_GET['gtype']) ? $_GET['gtype'] : 0; // Connect to the database $db = db_connect($database); // // Prepare some SQL queries // $query = "UPDATE genotype_corrections SET genotype=? WHERE id=?"; if (!($db['upd_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "INSERT INTO genotype_corrections SET batch_id=?, catalog_id=?, sample_id=?, genotype=?"; if (!($db['ins_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "DELETE FROM genotype_corrections WHERE id=?"; if (!($db['del_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT catalog_genotypes.id as id, catalog_genotypes.sample_id, catalog_genotypes.genotype, " . "genotype_corrections.genotype as correction, genotype_corrections.id as cid " . "FROM catalog_genotypes " . "LEFT JOIN genotype_corrections ON " . "(genotype_corrections.catalog_id=catalog_genotypes.catalog_id AND " . "genotype_corrections.sample_id=catalog_genotypes.sample_id AND " . "genotype_corrections.batch_id=catalog_genotypes.batch_id) " . "WHERE catalog_genotypes.batch_id=? AND " . "catalog_genotypes.catalog_id=? AND " . "catalog_genotypes.sample_id=?"; if (!($db['geno_sth'] = $db['dbh']->prepare($query))) write_db_error($db['geno_sth'], __FILE__, __LINE__); // // Fetch the existing genotypes from the database // if (!$db['geno_sth']->bind_param("iii", $batch_id, $tag_id, $sample_id)) write_db_error($db['geno_sth'], __FILE__, __LINE__); if (!$db['geno_sth']->execute()) write_db_error($db['geno_sth'], __FILE__, __LINE__); $res = $db['geno_sth']->get_result(); if ($row = $res->fetch_assoc()) { $sample = array('id' => $row['id'], 'genotype' => strtolower($row['genotype']), 'corrected' => $row['correction'], 'corrected_id' => $row['cid']); } else { return; } $corrected = "false"; if ($gtype == "clr") { if (!$db['del_sth']->bind_param("i", $sample['corrected_id'])) write_db_error($db['del_sth'], __FILE__, __LINE__); if (!$db['del_sth']->execute()) write_db_error($db['del_sth'], __FILE__, __LINE__); $corrected = "false"; $gtype = $sample['genotype']; // // Is this genotype being reset to the original value? If so, delete the corrected record. // } else if ($gtype == $sample['genotype'] && strlen($sample['corrected_id']) > 0) { if (!$db['del_sth']->bind_param("i", $sample['corrected_id'])) write_db_error($db['del_sth'], __FILE__, __LINE__); if (!$db['del_sth']->execute()) write_db_error($db['del_sth'], __FILE__, __LINE__); $corrected = "false"; // // Is the corrected value for this genotype being changed? If so, update the corrected record. // } else if ($gtype != $sample['genotype'] && strlen($sample['corrected_id']) > 0) { if (!$db['upd_sth']->bind_param("si", strtoupper($gtype), $sample['corrected_id'])) write_db_error($db['upd_sth'], __FILE__, __LINE__); if (!$db['upd_sth']->execute()) write_db_error($db['upd_sth'], __FILE__, __LINE__); $corrected = "true"; // // Otherwise, add a new correction. // } else if ($gtype != $sample['genotype']) { if (!$db['ins_sth']->bind_param("iiis", $batch_id, $tag_id, $sample_id, strtoupper($gtype))) write_db_error($db['ins_sth'], __FILE__, __LINE__); if (!$db['ins_sth']->execute()) write_db_error($db['ins_sth'], __FILE__, __LINE__); $corrected = "true"; } $id = "gtype_" . $batch_id . "_" . $tag_id . "_" . $sample_id; if ($corrected == "true") $gtype = strtoupper($gtype); header("Content-type: text/xml"); $xml_output = "\n" . "\n" . "$corrected\n" . "$gtype\n" . "$id\n" . "\n"; echo $xml_output; ?> stacks-1.44/php/correct_genotypes.php000644 000765 000024 00000015323 12663652222 020552 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); $database = isset($_POST['db']) ? $_POST['db'] : ""; $tag_id = isset($_POST['tag_id']) ? $_POST['tag_id'] : 0; $batch_id = isset($_POST['batch_id']) ? $_POST['batch_id'] : 0; $op = isset($_POST['op']) ? $_POST['op'] : "display"; // Connect to the database if (!isset($db)) $db = db_connect($database); // Save these variables for automatic URL formation later on. $display = array(); $display['db'] = $database; $display['tag_id'] = $tag_id; $display['batch_id'] = $batch_id; $diplsay['op'] = $op; // // Prepare some SQL queries // $query = "SELECT catalog_genotypes.id as id, catalog_genotypes.sample_id, catalog_genotypes.genotype, " . "genotype_corrections.genotype as correction, genotype_corrections.id as cid, file " . "FROM catalog_genotypes " . "LEFT JOIN genotype_corrections ON " . "(genotype_corrections.catalog_id=catalog_genotypes.catalog_id AND " . "genotype_corrections.sample_id=catalog_genotypes.sample_id AND " . "genotype_corrections.batch_id=catalog_genotypes.batch_id) " . "JOIN samples ON (catalog_genotypes.sample_id=samples.id) " . "WHERE catalog_genotypes.batch_id=? and catalog_genotypes.catalog_id=? " . "ORDER BY catalog_genotypes.sample_id"; if (!($db['geno_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "UPDATE genotype_corrections SET genotype=? WHERE id=?"; if (!($db['upd_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "INSERT INTO genotype_corrections SET batch_id=?, catalog_id=?, sample_id=?, genotype=?"; if (!($db['ins_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "DELETE FROM genotype_corrections WHERE id=?"; if (!($db['del_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT genotype_corrections.id FROM genotype_corrections " . "JOIN catalog_genotypes ON " . "(genotype_corrections.catalog_id=catalog_genotypes.catalog_id AND " . "genotype_corrections.sample_id=catalog_genotypes.sample_id AND " . "genotype_corrections.batch_id=catalog_genotypes.batch_id) " . "WHERE genotype_corrections.batch_id=? AND genotype_corrections.catalog_id=?"; if (!($db['res_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); if ($op == "reset") { reset_marker($display); } else if ($op == "correct") { correct_marker($display); } include_once("catalog_genotypes.php"); function reset_marker($display) { global $db; if (!$db['res_sth']->bind_param("ii", $display['batch_id'], $display['tag_id'])) write_db_error($db['res_sth'], __FILE__, __LINE__); if (!$db['res_sth']->execute()) write_db_error($db['res_sth'], __FILE__, __LINE__); $res = $db['res_sth']->get_result(); $row_id = 0; if (!$db['del_sth']->bind_param("i", $row_id)) write_db_error($db['del_sth'], __FILE__, __LINE__); while ($row = $res->fetch_assoc()) { $row_id = $row['id']; if (!$db['del_sth']->execute()) write_db_error($db['del_sth'], __FILE__, __LINE__); } } function correct_marker($display) { global $db; $gtypes = array(); $form_gtypes = array(); // // Fetch the existing genotypes from the database // if (!$db['geno_sth']->bind_param("ii", $display['batch_id'], $display['tag_id'])) write_db_error($db['geno_sth'], __FILE__, __LINE__); if (!$db['geno_sth']->execute()) write_db_error($db['geno_sth'], __FILE__, __LINE__); $res = $db['geno_sth']->get_result(); while ($row = $res->fetch_assoc()) { $gtypes[$row['sample_id']] = array('id' => $row['id'], 'file' => $row['file'], 'genotype' => strtolower($row['genotype']), 'corrected' => $row['correction'], 'corrected_id' => $row['cid']); } // // Fetch the corrected genotypes from the submitted form // foreach ($_POST as $key => $value) { if (substr($key, 0, 5) != "gtype") continue; // ID should look like: 'gtype_batchid_catalogid_sampleid' $parts = explode("_", $key); $form_gtypes[$parts[3]] = strtolower($value); //print "Assigning $value to $parts[3]
\n"; } foreach ($form_gtypes as $sample_id => $sample) { //print "LOOKING at sample ID: $sample_id: $sample, original value: " . $gtypes[$sample_id]['genotype'] . "
\n"; // // Is this genotype being reset to the original value? If so, delete the corrected record. // if ($sample == $gtypes[$sample_id]['genotype'] && strlen($gtypes[$sample_id]['corrected_id']) > 0) { if (!$db['del_sth']->bind_param("i", $gtypes[$sample_id]['corrected_id']) write_db_error($db['del_sth'], __FILE__, __LINE__); if (!$db['del_sth']->execute()) write_db_error($db['del_sth'], __FILE__, __LINE__); // // Is the corrected value for this genotype being changed? If so, update the corrected record. // } else if ($sample != $gtypes[$sample_id]['genotype'] && strlen($gtypes[$sample_id]['corrected_id']) > 0) { if (!$db['upd_sth']->bind_param("si", strtoupper($sample), $gtypes[$sample_id]['corrected_id']) write_db_error($db['upd_sth'], __FILE__, __LINE__); if (!$db['upd_sth']->execute()) write_db_error($db['upd_sth'], __FILE__, __LINE__); // // Otherwise, add a new correction. // } else if ($sample != $gtypes[$sample_id]['genotype']) { if (!$db['ins_sth']->bind_param("iiis", $display['batch_id'], $display['tag_id'], $sample_id, strtoupper($sample)) write_db_error($db['ins_sth'], __FILE__, __LINE__); if (!$db['ins_sth']->execute()) write_db_error($db['ins_sth'], __FILE__, __LINE__); } } } ?> stacks-1.44/php/db_functions.php000644 000765 000024 00000002703 12663356440 017472 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // function db_connect($database) { global $db_user, $db_pass, $db_host; $dbh = new mysqli($db_host, $db_user, $db_pass, $database); if ($dbh->connect_errno) { die("File: " . __FILE__ . " (line " . __LINE__ . ") Failed to connect to MySQL: (" . $dbh->connect_errno . ") " . $dbh->connect_error); } // // The $db array will hold the database handle and // common, prepared SQL statements. // $db = array(); $db['dbh'] = $dbh; $db['name'] = $database; return $db; } function write_db_error($sth, $file, $line) { $error_str = "File: $file (line $line)
\n " . $sth->errno . ": " . $sth->error . "
\n"; die($error_str); } ?> stacks-1.44/php/export.js000644 000765 000024 00000010302 12335173442 016150 0ustar00catchenstaff000000 000000 // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // var http_req; function toggle_export_popup(id) { var div_obj = document.getElementById(id); if (div_obj.style.display == "none") { div_obj.style.display = ""; } else { div_obj.style.display = "none"; } } function close_export_popup(id) { var div_obj = document.getElementById(id); div_obj.style.display = "none"; } function export_data(id) { var form_obj = document.getElementById(id + "_frm"); var otype, dtype; for (i = 0; i < form_obj.otype.length; i++) if (form_obj.otype[i].checked) otype = form_obj.otype[i].value; for (i = 0; i < form_obj.dtype.length; i++) if (form_obj.dtype[i].checked) dtype = form_obj.dtype[i].value; var url = form_obj.url.value + "&" + "email=" + form_obj.email.value + "&" + "dtype=" + dtype + "&" + "dlim=" + form_obj.dlim.value + "&" + "mtype=" + form_obj.mtype.value + "&" + "mcor=" + form_obj.mcor.value + "&" + "otype=" + otype; //alert(url); // // Prepare and send XMLHttpRequest Object. // http_req = false; try { http_req = new XMLHttpRequest(); } catch(e) { http_req = false; } if (http_req) { http_req.onreadystatechange = process_export; http_req.open("GET", url, true); http_req.send(""); } var txt_obj = document.getElementById(id + "_txt"); while (txt_obj.childNodes.length > 0) txt_obj.removeChild(txt_obj.firstChild); } function gen_export_result(loci, email) { var div_obj = document.getElementById("export_popup_txt"); var res_p = document.createElement("p"); res_p.innerHTML = "Exporting " + loci + " loci. " + "E-mail will be sent to " + email + "" + " when it is complete."; var app_obj = div_obj.appendChild(res_p); var res_p = document.createElement("p"); res_p.innerHTML = "close"; app_obj.appendChild(res_p); } function process_export() { // // Possible readyState values: // 0 = uninitialized // 1 = loading // 2 = loaded // 3 = interactive // 4 = complete // if (http_req.readyState == 4) { // Check that the status is "OK" if (http_req.status == 200) { var xml_doc = http_req.responseXML; var obj = xml_doc.getElementsByTagName("loci"); var loci = obj[0].childNodes[0].nodeValue; obj = xml_doc.getElementsByTagName("email"); var email = obj[0].childNodes[0].nodeValue; obj = xml_doc.getElementsByTagName("msg"); var msg = obj[0].childNodes[0].nodeValue; //alert(msg); gen_export_result(loci, email); } else { alert("There was a problem retrieving the XML data:\n" + http_req.statusText); } } } function toggle_vis(form_id, name) { var form_obj = document.getElementById(form_id); var g_obj = document.getElementById('gopts'); var h_obj = document.getElementById('hopts'); for(i = 0; i < form_obj.elements.length; i++) if (form_obj.elements[i].name == name && form_obj.elements[i].value == "geno") if (form_obj.elements[i].checked == true) { g_obj.style.display = ""; h_obj.style.display = "none"; } else { g_obj.style.display = "none"; h_obj.style.display = ""; } } stacks-1.44/php/export_batch.php000644 000765 000024 00000012315 12665337341 017500 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); require_once("CatalogClass.php"); $database = isset($_GET['db']) ? $_GET['db'] : ""; $batch_id = isset($_GET['id']) ? $_GET['id'] : 0; $email = isset($_GET['email']) ? $_GET['email'] : ""; $data_type = isset($_GET['dtype']) ? $_GET['dtype'] : "haplo"; $map_type = isset($_GET['mtype']) ? $_GET['mtype'] : "gen"; $depth_lim = isset($_GET['dlim']) ? $_GET['dlim'] : "1"; $man_cor = isset($_GET['mcor']) ? $_GET['mcor'] : "0"; $ex_type = isset($_GET['otype']) ? $_GET['otype'] : "tsv"; // Connect to the database $db = db_connect($database); // Save these variables for automatic URL formation later on. $display = array(); $display['id'] = $batch_id; $display['db'] = $database; $display['filter_type'] = array(); $filters = array(); process_filter($display, $filters); // // Create a new catalog object to hold filtered loci // $catalog = new Catalog($db, $batch_id, $display); $catalog->determine_count(); $loci_cnt = $catalog->num_loci(); $mc = ($man_cor > 0) ? "-c" : ""; if ($data_type == "haplo") { $dt = "-a haplo"; $dl = "-L $depth_lim"; } else if ($data_type == "geno") { $dt = "-a geno -m $map_type $mc"; $dl = ""; } $cmd = $export_cmd . " -D $database -b $batch_id $dt $dl -e $email -t $ex_type -F " . implode(",", $filters); header("Content-type: text/xml"); $xml_output = "\n" . "\n" . "" . number_format($loci_cnt) . "\n" . "$email\n" . "$cmd\n" . "\n"; echo $xml_output; // // Execute the email notification program, which will run the export and email // the submitter upon its conclusion. // // export_cmd is defined in the constants file. // exec($cmd . " >/dev/null &"); function process_filter(&$display_params, &$filters) { if (!isset($_GET['filter_type'])) return; foreach ($_GET['filter_type'] as $filter) { array_push($display_params['filter_type'], $filter); if ($filter == "alle") { $display_params['filter_alle_l'] = $_GET['filter_alle_l']; array_push($filters, "alle_l=" . $_GET['filter_alle_l']); $display_params['filter_alle_u'] = $_GET['filter_alle_u']; array_push($filters, "alle_u=" . $_GET['filter_alle_u']); } else if ($filter == "snps") { $display_params['filter_snps_l'] = $_GET['filter_snps_l']; array_push($filters, "snps_l=" . $_GET['filter_snps_l']); $display_params['filter_snps_u'] = $_GET['filter_snps_u']; array_push($filters, "snps_u=" . $_GET['filter_snps_u']); } else if ($filter == "pare") { $display_params['filter_pare_l'] = $_GET['filter_pare_l']; array_push($filters, "pare_l=" . $_GET['filter_pare_l']); $display_params['filter_pare_u'] = $_GET['filter_pare_u']; array_push($filters, "pare_u=" . $_GET['filter_pare_u']); } else if ($filter == "prog") { $display_params['filter_prog'] = $_GET['filter_prog']; array_push($filters, "prog=" . $_GET['filter_prog']); } else if ($filter == "vprog") { $display_params['filter_vprog'] = $_GET['filter_vprog']; array_push($filters, "vprog=" . $_GET['filter_vprog']); } else if ($filter == "lnl") { $display_params['filter_lnl_l'] = $_GET['filter_lnl_l']; array_push($filters, "lnl_l=" . $_GET['filter_lnl_l']); $display_params['filter_lnl_u'] = $_GET['filter_lnl_u']; array_push($filters, "lnl_u=" . $_GET['filter_lnl_l']); } else if ($filter == "cata") { $display_params['filter_cata'] = $_GET['filter_cata']; array_push($filters, "cata=" . $_GET['filter_cata']); } else if ($filter == "mark") { $display_params['filter_mark'] = ($_GET['filter_mark'] == "Any") ? "%/%" : $_GET['filter_mark']; array_push($filters, "mark=" . $_GET['filter_mark']); } else if ($filter == "gcnt") { $display_params['filter_gcnt'] = $_GET['filter_gcnt']; array_push($filters, "gcnt=" . $_GET['filter_gcnt']); } else if ($filter == "ref") { $display_params['filter_ref'] = $_GET['filter_ref']; array_push($filters, "ref=" . $_GET['filter_ref']); } else if ($filter == "loc") { $display_params['filter_chr'] = $_GET['filter_chr']; array_push($filters, "chr=" . $_GET['filter_chr']); $display_params['filter_sbp'] = $_GET['filter_sbp'] * 1000000; array_push($filters, "sbp=" . $_GET['filter_sbp']); $display_params['filter_ebp'] = $_GET['filter_ebp'] * 1000000; array_push($filters, "ebp=" . $_GET['filter_ebp']); } } } ?> stacks-1.44/php/fst_view.php000644 000765 000024 00000010427 12663652222 016642 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); $database = isset($_GET['db']) ? $_GET['db'] : ""; $tag_id = isset($_GET['tag_id']) ? $_GET['tag_id'] : 0; $batch_id = isset($_GET['batch_id']) ? $_GET['batch_id'] : 0; $batch_type = isset($_GET['type']) ? $_GET['type'] : "map"; // Connect to the database $db = db_connect($database); // Save these variables for automatic URL formation later on. $display = array(); $display['db'] = $database; $display['tag_id'] = $tag_id; $display['batch_id'] = $batch_id; // // Prepare some SQL queries // $query = "SELECT pop_id, pop_name FROM populations " . "WHERE batch_id=?"; if (!($db['pop_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT col, pop_id_1, pop_id_2, pi_o, amova_fst_c as fst, fishers_p, lod FROM fst " . "WHERE batch_id=? AND tag_id=?"; if (!($db['fst_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); // // Fetch population names if available. // $pop_names = array(); if ($batch_type == "population") { if (!$db['pop_sth']->bind_param("i", $batch_id)) write_db_error($db['pop_sth'], __FILE__, __LINE__); if (!$db['pop_sth']->execute()) write_db_error($db['pop_sth'], __FILE__, __LINE__); $res = $db['pop_sth']->get_result(); while ($row = $res->fetch_assoc()) $pop_names[$row['pop_id']] = $row['pop_name']; } if (!$db['fst_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['fst_sth'], __FILE__, __LINE__); if (!$db['fst_sth']->execute()) write_db_error($db['fst_sth'], __FILE__, __LINE__); $res = $db['fst_sth']->get_result(); $stats = array(); $pops = array(); while ($row = $res->fetch_assoc()) { $a = array('col' => $row['col'], 'pid_1' => $row['pop_id_1'], 'pid_2' => $row['pop_id_2'], 'pi_o' => $row['pi_o'], 'fishers_p' => $row['fishers_p'], 'lod' => $row['lod'], 'fst' => $row['fst']); if (!isset($stats[$row['col']])) $stats[$row['col']] = array(); array_push($stats[$row['col']], $a); $pops[$row['pop_id_1']] = $row['pop_id_1']; $pops[$row['pop_id_2']] = $row['pop_id_2']; } ksort($stats); ksort($pops); // // Assign population IDs for any missing population names. // foreach ($pops as $pop_id) if (!isset($pop_names[$pop_id])) $pop_names[$pop_id] = $pop_id; ksort($pop_names); $json_str = "{" . "\"path\": \"$root_path\"," . "\"batch_id\": \"$batch_id\"," . "\"db\": \"$database\"," . "\"id\": \"$tag_id\"," . "\"type\": \"$batch_type\","; $json_str .= "\"popkey\": {"; // // Print the population key. // foreach ($pop_names as $pop_id => $population) { $json_str .= "\"$pop_id\": \"$population\","; } $json_str = substr($json_str, 0, -1); $json_str .= "}," . "\"columns\": {"; foreach ($stats as $col => $stat) { $json_str .= "\"$col\": ["; foreach ($stat as $s) { $fst = $s['fst'] != 0 ? sprintf("%.3f", $s['fst']) : "0"; $lod = $s['lod'] != 0 ? sprintf("%.3f", $s['lod']) : "0"; $pio = $s['pi_o'] != 0 ? sprintf("%.3f", $s['pi_o']) : "0"; $p = $s['fishers_p'] != 0 ? sprintf("%.3f", $s['fishers_p']) : "0"; $json_str .= "{" . "\"pid_1\": \"$s[pid_1]\"," . "\"pid_2\": \"$s[pid_2]\"," . "\"pi_o\": \"$pio\"," . "\"p\": \"$p\"," . "\"lod\": \"$lod\"," . "\"fst\": \"$fst\"" . "},"; } if (count($stat) > 0) $json_str = substr($json_str, 0, -1); $json_str .= "],"; } if (count($stats) > 0) $json_str = substr($json_str, 0, -1); $json_str .= "}}"; echo $json_str; ?> stacks-1.44/php/hapstat_view.php000644 000765 000024 00000007165 12663632701 017517 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); $database = isset($_GET['db']) ? $_GET['db'] : ""; $tag_id = isset($_GET['tag_id']) ? $_GET['tag_id'] : 0; $batch_id = isset($_GET['batch_id']) ? $_GET['batch_id'] : 0; $batch_type = isset($_GET['type']) ? $_GET['type'] : "map"; // Connect to the database $db = db_connect($database); // Save these variables for automatic URL formation later on. $display = array(); $display['db'] = $database; $display['tag_id'] = $tag_id; $display['batch_id'] = $batch_id; // // Prepare some SQL queries // $query = "SELECT pop_id, pop_name FROM populations " . "WHERE batch_id=?"; if (!($db['pop_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT pop_id, bp, n, hapcnt, gene_div, hap_div FROM hapstats " . "WHERE batch_id=? AND tag_id=?"; if (!($db['stats_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); // // Fetch population names if available. // $pop_names = array(); if ($batch_type == "population") { if (!$db['pop_sth']->bind_param("i", $batch_id)) write_db_error($db['pop_sth'], __FILE__, __LINE__); if (!$db['pop_sth']->execute()) write_db_error($db['pop_sth'], __FILE__, __LINE__); $res = $db['pop_sth']->get_result(); while ($row = $res->fetch_assoc()) $pop_names[$row['pop_id']] = $row['pop_name']; } if (!$db['stats_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['stats_sth'], __FILE__, __LINE__); if (!$db['stats_sth']->execute()) write_db_error($db['stats_sth'], __FILE__, __LINE__); $res = $db['stats_sth']->get_result(); $stats = array(); while ($row = $res->fetch_assoc()) { $a = array('bp' => $row['bp'], 'n' => $row['n'], 'hapcnt' => $row['hapcnt'], 'gene_div' => $row['gene_div'], 'hap_div' => $row['hap_div'], 'pop_id' => $row['pop_id']); $stats[$row['pop_id']] = $a; } ksort($stats); $json_str = "{" . "\"path\": \"$root_path\"," . "\"batch_id\": \"$batch_id\"," . "\"db\": \"$database\"," . "\"id\": \"$tag_id\"," . "\"type\": \"$batch_type\","; $json_str .= "\"hapstats\": ["; foreach ($stats as $pop_id => $stat) if (!isset($pop_names[$pop_id])) $pop_names[$pop_id] = $pop_id; $rows = 0; foreach ($stats as $pop_id => $s) { $gdiv = $s['gene_div'] > 0 ? sprintf("%.3f", $s['gene_div']) : $s['gene_div']; $hdiv = $s['hap_div'] > 0 ? sprintf("%.3f", $s['hap_div']) : $s['hap_div']; $json_str .= "{" . "\"pop_id\": \"" . $pop_names[$pop_id] . "\"," . "\"bp\": \"$s[bp]\"," . "\"n\": \"$s[n]\"," . "\"hapcnt\": \"$s[hapcnt]\"," . "\"genediv\": \"$gdiv\"," . "\"hapdiv\": \"$hdiv\"" . "},"; $rows++; } if ($rows > 0) $json_str = substr($json_str, 0, -1); $json_str .= "]}"; echo $json_str; ?> stacks-1.44/php/header.php000644 000765 000024 00000001517 12663356440 016247 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("constants.php"); require_once("db_functions.php"); require_once("stacks_functions.php"); ?> stacks-1.44/php/images/000755 000765 000024 00000000000 13001012452 015522 5ustar00catchenstaff000000 000000 stacks-1.44/php/index.php000644 000765 000024 00000006710 12663632701 016123 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); $database = isset($_GET['db']) ? $_GET['db'] : ""; if (strlen($database) == 0) write_database_list($database); else write_database($database); function write_database_list($database) { global $db_user, $db_pass, $db_host, $root_path, $img_path, $mysql_bin; $databases = array(); exec("$mysql_bin --user=$db_user --password=$db_pass -h $db_host -N -B -e \"SHOW DATABASES LIKE '%_radtags'\"", $databases); $page_title = "Stacks Databases"; write_header($page_title); echo <<< EOQ

Stacks Databases

EOQ; foreach ($databases as $dbase) { print "\n" . " \n" . "\n"; } echo <<< EOQ
Database
$dbase
EOQ; write_footer(); } function write_database($database) { global $root_path, $img_path; // // Connect to the database // $db = db_connect($database); // // Prepare some SQL queries // $query = "SELECT id, date, description, type FROM batches"; if (!($db['batch_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $page_title = "RAD-Tag Analyses"; write_header($page_title); echo <<< EOQ

RAD-Tag Samples

EOQ; if (!$db['batch_sth']->execute()) write_db_error($db['batch_sth'], __FILE__, __LINE__); $res = $db['batch_sth']->get_result(); while ($row = $res->fetch_assoc()) { print "\n" . " \n" . " \n" . " \n" . " \n" . " \n" . " \n" . "\n"; } echo <<< EOQ
    Batch ID Date Type Description
CatalogSamples" . $row['id'] . "" . $row['date'] . "" . $row['type'] . "" . $row['description'] . "
EOQ; write_footer(); } ?> stacks-1.44/php/last_modified.php000644 000765 000024 00000000035 12663356440 017614 0ustar00catchenstaff000000 000000 Tue Feb 23 15:53:56 PST 2016 stacks-1.44/php/Locus.php000644 000765 000024 00000004200 12335173442 016067 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // class Locus { var $id; var $seq; var $annotation; var $chr; var $bp; var $snps; var $alleles; var $genotypes; // Array of samples that possess this locus var $num_parents; var $num_progeny; var $num_snps; var $num_alleles; var $num_ests; var $num_pe_tags; var $num_blast; var $valid_progeny; var $marker; function Locus() { $this->genotypes = array(); $this->annotation = ""; $this->chr = ""; $this->bp = 0; $this->marker = ""; $this->snps = ""; $this->alleles = ""; $this->num_parents = 0; $this->num_progeny = 0; $this->valid_progeny = 0; $this->num_alleles = 0; $this->num_snps = 0; $this->num_ests = 0; $this->num_pe_tags = 0; $this->num_blast = 0; } function &genotypes() { return $this->genotypes; } function &genotype($id) { if (isset($this->genotypes[$id])) return $this->genotypes[$id]; else return NULL; } function add_genotype($id, $file, $allele) { $a = array('file' => $file, 'allele' => $allele, 'tag_id' => $id); if (!isset($this->genotypes[$file])) $this->genotypes[$file] = array(); array_push($this->genotypes[$file], $a); } } ?> stacks-1.44/php/phist_view.php000644 000765 000024 00000007400 12663652222 017172 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); $database = isset($_GET['db']) ? $_GET['db'] : ""; $tag_id = isset($_GET['tag_id']) ? $_GET['tag_id'] : 0; $batch_id = isset($_GET['batch_id']) ? $_GET['batch_id'] : 0; $batch_type = isset($_GET['type']) ? $_GET['type'] : "map"; // Connect to the database $db = db_connect($database); // Save these variables for automatic URL formation later on. $display = array(); $display['db'] = $database; $display['tag_id'] = $tag_id; $display['batch_id'] = $batch_id; // // Prepare some SQL queries // $query = "SELECT pop_id, pop_name FROM populations " . "WHERE batch_id=?"; if (!($db['pop_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT pop_id_1, pop_id_2, phist, fpst FROM phist " . "WHERE batch_id=? AND tag_id=?"; if (!($db['fst_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); // // Fetch population names if available. // $pop_names = array(); if ($batch_type == "population") { if (!$db['pop_sth']->bind_param("i", $batch_id)) write_db_error($db['pop_sth'], __FILE__, __LINE__); if (!$db['pop_sth']->execute()) write_db_error($db['pop_sth'], __FILE__, __LINE__); $res = $db['pop_sth']->get_result(); while ($row = $res->fetch_assoc()) $pop_names[$row['pop_id']] = $row['pop_name']; } if (!$db['fst_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['fst_sth'], __FILE__, __LINE__); if (!$db['fst_sth']->execute()) write_db_error($db['fst_sth'], __FILE__, __LINE__); $res = $db['fst_sth']->get_result(); $stats = array(); $pops = array(); while ($row = $res->fetch_assoc()) { $a = array('pid_1' => $row['pop_id_1'], 'pid_2' => $row['pop_id_2'], 'phist' => $row['phist'], 'fpst' => $row['fpst']); array_push($stats, $a); $pops[$row['pop_id_1']] = $row['pop_id_1']; $pops[$row['pop_id_2']] = $row['pop_id_2']; } ksort($stats); ksort($pops); // // Assign population IDs for any missing population names. // foreach ($pops as $pop_id) if (!isset($pop_names[$pop_id])) $pop_names[$pop_id] = $pop_id; ksort($pop_names); $json_str = "{" . "\"path\": \"$root_path\"," . "\"batch_id\": \"$batch_id\"," . "\"db\": \"$database\"," . "\"id\": \"$tag_id\"," . "\"type\": \"$batch_type\","; $json_str .= "\"popkey\": {"; // // Print the population key. // foreach ($pop_names as $pop_id => $population) { $json_str .= "\"$pop_id\": \"$population\","; } $json_str = substr($json_str, 0, -1); $json_str .= "}," . "\"phist\": ["; foreach ($stats as $s) { $phist = $s['phist'] != 0 ? sprintf("%.3f", $s['phist']) : "0"; $fpst = $s['fpst'] != 0 ? sprintf("%.3f", $s['fpst']) : "0"; $json_str .= "{" . "\"pid_1\": \"$s[pid_1]\"," . "\"pid_2\": \"$s[pid_2]\"," . "\"phist\": \"$phist\"," . "\"fstp\": \"$fpst\"" . "},"; } if (count($stats) > 0) $json_str = substr($json_str, 0, -1); $json_str .= "]}"; echo $json_str; ?> stacks-1.44/php/pop_view.php000644 000765 000024 00000026777 12663632701 016663 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); $database = isset($_GET['db']) ? $_GET['db'] : ""; $tag_id = isset($_GET['tag_id']) ? $_GET['tag_id'] : 0; $batch_id = isset($_GET['batch_id']) ? $_GET['batch_id'] : 0; $batch_type = isset($_GET['type']) ? $_GET['type'] : "map"; // Connect to the database $db = db_connect($database); // Save these variables for automatic URL formation later on. $display = array(); $display['db'] = $database; $display['tag_id'] = $tag_id; $display['batch_id'] = $batch_id; // // Prepare some SQL queries // $query = "SELECT samples.id, samples.sample_id, samples.type, file, tag_id, allele, depth, lnl, pop_id " . "FROM matches " . "JOIN samples ON (matches.sample_id=samples.id) " . "WHERE matches.batch_id=? AND catalog_id=? ORDER BY samples.id"; if (!($db['mat_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT col, rank_1, rank_2, rank_3, rank_4 FROM catalog_snps " . "WHERE batch_id=? AND tag_id=? ORDER BY col"; if (!($db['snp_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT allele FROM catalog_alleles " . "WHERE batch_id=? AND tag_id=? "; if (!($db['all_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT geno_map FROM markers " . "WHERE batch_id=? AND catalog_id=? "; if (!($db['map_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT pop_id, pop_name FROM populations " . "WHERE batch_id=?"; if (!($db['pop_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT count(batch_id) as cnt FROM sumstats " . "WHERE batch_id=? AND tag_id=?"; if (!($db['stats_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT count(batch_id) as cnt FROM fst " . "WHERE batch_id=? AND tag_id=?"; if (!($db['fst_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT count(batch_id) as cnt FROM hapstats " . "WHERE batch_id=? AND tag_id=?"; if (!($db['hapstats_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT count(batch_id) as cnt FROM phist " . "WHERE batch_id=? AND tag_id=?"; if (!($db['phist_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT marker, catalog_genotypes.sample_id, file, " . "catalog_genotypes.genotype, genotype_corrections.genotype as corrected " . "FROM catalog_genotypes " . "LEFT JOIN genotype_corrections ON " . "(genotype_corrections.catalog_id=catalog_genotypes.catalog_id AND " . "genotype_corrections.sample_id=catalog_genotypes.sample_id AND " . "genotype_corrections.batch_id=catalog_genotypes.batch_id) " . "JOIN samples ON (catalog_genotypes.sample_id=samples.id) " . "JOIN catalog_index ON (catalog_genotypes.catalog_id=catalog_index.tag_id AND " . "catalog_genotypes.batch_id=catalog_index.batch_id) " . "WHERE catalog_genotypes.batch_id=? and catalog_genotypes.catalog_id=? " . "ORDER BY catalog_genotypes.sample_id"; if (!($db['geno_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); // // Check for the existence of SNP summary statistics or Fst data. // $snp_sumstats = 0; $snp_fst_vals = 0; $hap_sumstats = 0; $hap_fst_vals = 0; if ($batch_type == "population") { if (!$db['stats_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['stats_sth'], __FILE__, __LINE__); if (!$db['stats_sth']->execute()) write_db_error($db['stats_sth'], __FILE__, __LINE__); $res = $db['stats_sth']->get_result(); if ($row = $res->fetch_assoc()) { if ($row['cnt'] > 0) $snp_sumstats = 1; } if (!$db['fst_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['fst_sth'], __FILE__, __LINE__); if (!$db['fst_sth']->execute()) write_db_error($db['fst_sth'], __FILE__, __LINE__); $res = $db['fst_sth']->get_result(); if ($row = $res->fetch_assoc()) { if ($row['cnt'] > 0) $snp_fst_vals = 1; } if (!$db['hapstats_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['hapstats_sth'], __FILE__, __LINE__); if (!$db['hapstats_sth']->execute()) write_db_error($db['hapstats_sth'], __FILE__, __LINE__); $res = $db['hapstats_sth']->get_result(); if ($row = $res->fetch_assoc()) { if ($row['cnt'] > 0) $hap_sumstats = 1; } if (!$db['phist_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['phist_sth'], __FILE__, __LINE__); if (!$db['phist_sth']->execute()) write_db_error($db['phist_sth'], __FILE__, __LINE__); $res = $db['phist_sth']->get_result(); if ($row = $res->fetch_assoc()) { if ($row['cnt'] > 0) $hap_fst_vals = 1; } } // // Fetch population names if available. // $pop_names = array(); if ($batch_type == "population") { if (!$db['pop_sth']->bind_param("i", $batch_id)) write_db_error($db['pop_sth'], __FILE__, __LINE__); if (!$db['pop_sth']->execute()) write_db_error($db['pop_sth'], __FILE__, __LINE__); $res = $db['pop_sth']->get_result(); while ($row = $res->fetch_assoc()) $pop_names[$row['pop_id']] = $row['pop_name']; } if (!$db['snp_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['snp_sth'], __FILE__, __LINE__); if (!$db['snp_sth']->execute()) write_db_error($db['snp_sth'], __FILE__, __LINE__); $res = $db['snp_sth']->get_result(); $json_str = "{" . "\"path\": \"$root_path\"," . "\"batch_id\": \"$batch_id\"," . "\"db\": \"$database\"," . "\"id\": \"$tag_id\"," . "\"type\": \"$batch_type\"," . "\"snpstat\": \"$snp_sumstats\"," . "\"snpfst\": \"$snp_fst_vals\"," . "\"hapstat\": \"$snp_sumstats\"," . "\"hapfst\": \"$snp_fst_vals\","; $json_str .= "\"snps\": ["; $rows = 0; while ($row = $res->fetch_assoc()) { $json_str .= "{" . "\"col\": \"$row[col]\"," . "\"rank_1\": \"$row[rank_1]\"," . "\"rank_2\": \"$row[rank_2]\"," . "\"rank_3\": \"$row[rank_3]\"," . "\"rank_4\": \"$row[rank_4]\"" . "},"; $rows++; } if ($rows > 0) $json_str = substr($json_str, 0, -1); $json_str .= "]," . "\"alleles\": ["; if (!$db['map_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['map_sth'], __FILE__, __LINE__); if (!$db['map_sth']->execute()) write_db_error($db['map_sth'], __FILE__, __LINE__); $res = $db['map_sth']->get_result(); if ($res->num_rows > 0) $row = $res->fetch_assoc(); $rows = 0; if ($res->num_rows > 0 && strlen($row['geno_map']) > 0) { $map = array(); $genos = explode(";", $row['geno_map']); $i = 0; foreach ($genos as $g) { if (strlen($g) == 0) continue; $m = explode(":", $g); $map[$m[0]] = $m[1]; $alleles[$m[0]] = $colors[$i % $color_size]; $i++; } asort($map); foreach ($map as $hapl => $geno) { $json_str .= "{" . "\"gtype\": \"$geno\"," . "\"htype\": \"$hapl\"" . "},"; } $rows++; } else { if (!$db['all_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['all_sth'], __FILE__, __LINE__); if (!$db['all_sth']->execute()) write_db_error($db['all_sth'], __FILE__, __LINE__); $res = $db['all_sth']->get_result(); $gtypes = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ"; $i = 0; while ($row = $res->fetch_assoc()) { $json_str .= "{" . "\"gtype\": \"" . $gtypes[$i % 52] . "\"," . "\"htype\": \"$row[allele]\"" . "},"; $i++; $rows++; } } if ($rows > 0) $json_str = substr($json_str, 0, -1); $json_str .= "]," . "\"popkey\": {"; $htypes = array(); $gtypes = array(); // // Fetch and record Observed Haplotypes // if (!$db['mat_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['mat_sth'], __FILE__, __LINE__); if (!$db['mat_sth']->execute()) write_db_error($db['mat_sth'], __FILE__, __LINE__); $res = $db['mat_sth']->get_result(); while ($row = $res->fetch_assoc()) { $a = array('id' => $row['id'], 'file' => $row['file'], 'allele' => $row['allele'], 'tag_id' => $row['tag_id'], 'depth' => $row['depth'], 'lnl' => $row['lnl'], 'pop_id' => $row['pop_id']); if (!isset($htypes[$row['pop_id']])) $htypes[$row['pop_id']] = array(); if (!isset($htypes[$row['pop_id']][$row['file']])) $htypes[$row['pop_id']][$row['file']] = array(); array_push($htypes[$row['pop_id']][$row['file']], $a); } // // Fetch and record Genotypes // if (!$db['geno_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['geno_sth'], __FILE__, __LINE__); if (!$db['geno_sth']->execute()) write_db_error($db['geno_sth'], __FILE__, __LINE__); $res = $db['geno_sth']->get_result(); while ($row = $res->fetch_assoc()) { $gtypes[$row['file']] = array('id' => $row['sample_id'], 'file' => $row['file'], 'genotype' => $row['genotype'], 'corrected' => $row['corrected'], 'marker' => $row['marker']); } ksort($htypes); // // Print the population key. // foreach ($htypes as $pop_id => $population) { if (isset($pop_names[$pop_id])) $json_str .= "\"$pop_id\": \"$pop_names[$pop_id]\","; else $json_str .= "\"$pop_id\": \"\","; } $json_str = substr($json_str, 0, -1); $json_str .= "}," . "\"populations\": {"; // // Print the observed haplotypes grouped by population. // foreach ($htypes as $pop_id => $population) { $json_str .= "\"$pop_id\": ["; foreach ($population as $sample => $match) { $genotype = ""; $corrected = 0; $marker = ""; if (count($gtypes) > 0 && isset($gtypes[$sample])) { $marker = $gtypes[$sample]['marker']; if (strlen($gtypes[$sample]['corrected']) > 0) { $genotype = $gtypes[$sample]['corrected']; $corrected = 1; } else { $genotype = $gtypes[$sample]['genotype']; } } $json_str .= "{" . "\"sample\": \"$sample\"," . "\"sample_id\": \"" . $match[0]['id'] . "\"," . "\"lnl\": \"" . $match[0]['lnl'] . "\","; if ($batch_type == "map") { $json_str .= "\"marker\": \"$marker\"," . "\"genotype\": \"$genotype\"," . "\"corrected\": \"$corrected\","; } $json_str .= "\"obshap\": ["; foreach ($match as $m) { $json_str .= "{" . "\"tag_id\": \"$m[tag_id]\"," . "\"allele\": \"$m[allele]\"," . "\"depth\": \"$m[depth]\"" . "},"; } $json_str = substr($json_str, 0, -1); $json_str .= "]},"; } $json_str = substr($json_str, 0, -1); $json_str .= "],"; } $json_str = substr($json_str, 0, -1); $json_str .= "}}"; echo $json_str; ?> stacks-1.44/php/population_view.js000644 000765 000024 00000120174 12663632701 020066 0ustar00catchenstaff000000 000000 var colors = { 'a' : "#0074D9", // Blue 'b' : "#FF4136", // Red 'c' : "#008000", // Dark Green 'd' : "#FF851B", // Orange 'e' : "#001f3f", // Navy 'f' : "#85144b", // Maroon 'g' : "#F012BE", // Fuchsia 'h' : "#39CCCC", // Teal 'i' : "#3D9970", // Olive 'j' : "#01FF70", // Lime 'k' : "#FFDC00", // Yellow 'l' : "#B10DC9", // Purple 'm' : "#111111", // Black 'n' : "#7FDBFF", // Aqua 'o' : "#AAAAAA", // Gray 'p' : "#DDDDDD", // Silver 'q' : "#2ECC40", // Green 'r' : "#c00000", 's' : "#ffc600", 't' : "#29356c", 'u' : "#860000", 'v' : "#dc6200", 'w' : "#4b398e", 'x' : "#008f56", 'y' : "#bf1e25", 'z' : "#4cb8ff", 'A' : "#26081C", 'B' : "#C44900", 'C' : "#036016", 'D' : "#936426", 'E' : "#E87EA1", 'F' : "#692E38", 'G' : "#19423F", 'H' : "#625F64", 'I' : "#432534", 'J' : "#323845", 'K' : "#758E4F", 'L' : "#B287A3", 'M' : "#C0F8D1", 'N' : "#03440C", 'O' : "#8EDCE6", 'P' : "#D5DCF9", 'Q' : "#A7B0CA", 'R' : "#0CBABA", 'S' : "#725E54", 'T' : "#EE2677", 'U' : "#84A98C", 'V' : "#FCCA46", 'W' : "#A1C181", 'X' : "#443627", 'Y' : "#FE7F2D", 'Z' : "#2EBE62" }; marker_types = { 'ab/--' : ['aa', 'bb', 'ab', '--'], '--/ab' : ['aa', 'bb', 'ab', '--'], 'ab/aa' : ['aa', 'bb', 'ab', '--'], 'aa/ab' : ['aa', 'bb', 'ab', '--'], 'ab/ab' : ['aa', 'ab', 'bb', '--'], 'ab/ac' : ['aa', 'ab', 'ac', 'bc', '--'], 'ab/cd' : ['aa', 'bb', 'cc', 'dd', 'ac', 'ad', 'bc', 'bd', '--'], 'aa/bb' : ['aa', 'bb', 'ab', '--'], 'ab/cc' : ['aa', 'bb', 'ab', 'ac', 'bc', 'cc', '--'], 'cc/ab' : ['aa', 'bb', 'ab', 'ac', 'bc', 'cc', '--'] }; function ajax_locus_population_view(id, root_path, url) { var tr_obj = document.getElementById(id); var caret = document.getElementById(id + "_img"); if (tr_obj.style.display == "none") { tr_obj.style.display = ""; caret.src = root_path + "/caret-d.png"; var div_obj = document.getElementById(id + "_popview_div"); div_obj.innerHTML = ""; // // Make the AJAX query for JSON encoded data for this locus. // $.ajax({ dataType: "json", url: url, success: build_population_view }); } else { tr_obj.style.display = "none"; caret.src = root_path + "/caret-u.png"; } } function ajax_locus_sumstats_view(id, url) { var div_obj = document.getElementById(id + "_sumstat_div"); if (div_obj.style.display == "none") { // // Make the AJAX query for JSON encoded data for this locus. // $.ajax({ dataType: "json", url: url, success: build_sumstats_view }); } } function ajax_locus_fst_view(id, url) { var div_obj = document.getElementById(id + "_fst_div"); if (div_obj.style.display == "none") { // // Make the AJAX query for JSON encoded data for this locus. // $.ajax({ dataType: "json", url: url, success: build_fst_view }); } } function ajax_locus_hapstats_view(id, url) { var div_obj = document.getElementById(id + "_hapstat_div"); if (div_obj.style.display == "none") { // // Make the AJAX query for JSON encoded data for this locus. // $.ajax({ dataType: "json", url: url, success: build_hapstats_view }); } } function ajax_locus_phist_view(id, url) { var div_obj = document.getElementById(id + "_phist_div"); if (div_obj.style.display == "none") { // // Make the AJAX query for JSON encoded data for this locus. // $.ajax({ dataType: "json", url: url, success: build_phist_view }); } } function ajax_locus_stack_view(id, url) { var div_obj = document.getElementById(id + "_stacks_div"); $("html, body").css("cursor", "wait"); if (div_obj.style.display == "none") { // // Make the AJAX query for JSON encoded data for this locus. // $.ajax({ dataType: "json", url: url, success: build_stack_view }); } } function build_population_view(json, status, jqXHR) { var url = json.path + "/stack_view.php" + "?db=" + json.db + "&batch_id=" + json.batch_id + "&tag_id=" + json.id; var html = "
\n" + "
\n" + "
View Stacks
\n"; var snps = write_snps_table(json); html += snps; // // Generate map of genotypes to haplotypes (for use in color picking). // var color_map = {}; for(var i = 0; i < json.alleles.length; i++) color_map[json.alleles[i].htype] = json.alleles[i].gtype; var alleles = write_haplotypes_table(json); html += alleles + "
\n" + "
\n"; var tables = write_population_table(json, color_map); html += tables + "
\n"; $("#" + json.id + "_popview_div").html(html); } function write_snps_table(json) { var html = ""; if (json.snps.length == 0) return html; html += "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n"; for(var i = 0; i < json.snps.length; i++) { var snp = "\n" + "" + "" + "" + ""; html += snp; } html += "
SNPs
ColumnAlleles
" + (i + 1) + "." + json.snps[i].col + "" + json.snps[i].rank_1 + " / " + json.snps[i].rank_2; if (json.snps[i].rank_3.length > 0) snp += " / " + json.snps[i].rank_3; if (json.snps[i].rank_4.length > 0) snp += " / " + json.snps[i].rank_4; snp += "
x
\n"; if (json.snpstat == 1) { url = json.path + "/sumstat_view.php?db=" + json.db + "&batch_id=" + json.batch_id + "&type=" + json.type + "&tag_id=" + json.id; html += "
Summary Stats
\n" + "
\n"; } if (json.snpfst == 1) { url = json.path + "/fst_view.php?db=" + json.db + "&batch_id=" + json.batch_id + "&type=" + json.type + "&tag_id=" + json.id; html += "
FST Stats
\n" + "
\n"; } return html; } function write_haplotypes_table(json) { var html = ""; if (json.alleles.length == 0) return html; html += "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n"; // // Count up the number of occurances of each haplotype. // haplotypes = {}; for (var pop_id in json.populations) { for (var i = 0; i < json.populations[pop_id].length; i++) for (var j = 0; j < json.populations[pop_id][i].obshap.length; j++) if (!(json.populations[pop_id][i].obshap[j].allele in haplotypes)) { haplotypes[json.populations[pop_id][i].obshap[j].allele] = 1; } else { haplotypes[json.populations[pop_id][i].obshap[j].allele]++; } } // // Write out the haplotype table. // for(var i = 0; i < json.alleles.length; i++) { html += "\n" + "\n" + "" + "" + "" + ""; } html += "
Haplotypes
GenotypeHaplotypeCnt
" + (i+1) + "." + json.alleles[i].gtype + "
" + json.alleles[i].htype + "
"; if (json.alleles[i].htype in haplotypes) html += haplotypes[json.alleles[i].htype]; else html += "0"; html += "
x
\n"; if (json.hapstat == 1) { url = json.path + "/hapstat_view.php?db=" + json.db + "&batch_id=" + json.batch_id + "&type=" + json.type + "&tag_id=" + json.id; html += "
Haplotype Stats
\n" + "
\n"; } if (json.hapfst == 1) { url = json.path + "/phist_view.php?db=" + json.db + "&batch_id=" + json.batch_id + "&type=" + json.type + "&tag_id=" + json.id; html += "
ΦST Stats
\n" + "
\n"; } return html; } function write_population_table(json, color_map) { var html = ""; html += "" + "\n" + "\n" + " \n" + "\n"; // // Determine the maximum number of columns in the table if less than 10. // var num_cols = 0; for (var pop_id in json.populations) num_cols = json.populations[pop_id].length > num_cols ? json.populations[pop_id].length : num_cols; num_cols = num_cols < 10 ? num_cols : 10; // // Iterate over each population. // for (var pop_id in json.populations) { html += "\n"; if (Object.keys(json.populations).length > 1) { html += "\n" + "\n" + "\n"; } var col_index = 0; var hap_index = 0; for (var i = 0; i < json.populations[pop_id].length; i++) { col_index++; var sample = json.populations[pop_id][i]; var uniq_id = sample.sample_id + "|" + sample.obshap[0].tag_id; html += " \n"; if (col_index % num_cols == 0) html += "\n" + "\n"; } while (col_index % num_cols != 0) { html += " \n"; col_index++; } } html += "\n"; return html; } function enqueue_sample(cat_id, id) { var cb = document.getElementById(id); var td = document.getElementById(id + "_td"); var sel = document.getElementById(cat_id + "_selected"); var checks = Number(sel.value); if (cb.checked == false) { cb.checked = true; td.style.backgroundColor = "#ffffa8"; checks++; } else { cb.checked = false; td.style.backgroundColor = "#ffffff"; checks--; } if (checks > 0) $("#" + cat_id + "_viewstat_stacks").css("display", ""); else $("#" + cat_id + "_viewstat_stacks").css("display", "none"); sel.value = checks; } function build_stack_list(id, url) { var samples = new Array(); var tags = new Array(); $('input[type=checkbox]').each(function () { if (this.name == "stack_view" && this.checked) { var parts = this.id.split("|"); samples.push(parts[0]); tags.push(parts[1]); } }); url += "&samples=" + samples.join(",") + "&tags=" + tags.join(","); ajax_locus_stack_view(id, url); } function print_population_name(pop_id, json) { var pop_str = json.popkey[pop_id].length > 0 ? json.popkey[pop_id] : "Population " + pop_id; var html = "
" + pop_str + "
\n" + "
\n" + "
\n" + " \n" + " \n" + " save|cancel\n" + " \n" + "
\n"; return html; } function generate_element_select(name, marker, selected_ele, change_js, blur_js) { var script_code = ""; if (change_js.length > 0 && blur_js.length > 0) { script_code = " onchange=\"" + change_js + "\" onblur=\"" + blur_js + "\""; } else if (change_js.length > 0) { script_code = " onchange=\"" + change_js + "\""; } else if (blur_js.length > 0) { script_code = " onblur=\"" + blur_js + "\""; } var ctl = " \n"; return ctl; } function highlight_snp(id) { var span_obj = document.getElementById(id); span_obj.className = "rank_1_hi"; } function unhighlight_snp(id) { var span_obj = document.getElementById(id); span_obj.className = "rank_1"; } function unhighlight_haplotypes(cat_id) { $("#alleles_table_" + cat_id + " tr").css("border", ""); $("#alleles_table_" + cat_id + " tr > td").css("background-color", ""); $("#alleles_table_" + cat_id + " tr td:first-child").css("color", ""); $("#alleles_table_" + cat_id + " tr td:last-child").css("color", ""); var table_obj = document.getElementById("locus_gtypes_" + cat_id); var spans = table_obj.getElementsByTagName("span"); for(var i = 0; i < spans.length; i++) { spans[i].className = ""; } $("div.close_x").css("display", "none"); } function highlight_haplotypes(cat_id, haplotype, tr_id) { unhighlight_haplotypes(cat_id) $("#" + tr_id).css("border", "4px solid #a93535"); $("#" + tr_id + " > td").css("background-color", "#aaa"); $("#" + tr_id + " td:first-child").css("color", "#fff"); $("#" + tr_id + " td:last-child").css("color", "#fff"); var table_obj = document.getElementById("locus_gtypes_" + cat_id); var spans = table_obj.getElementsByTagName("span"); for(var i = 0; i < spans.length; i++) { var parts = spans[i].id.split("_"); if (parts[0] == "haphi" && parts[1] == haplotype) { spans[i].className = "haphi"; } } $("#" + tr_id + "_close").css("display", ""); $("#" + tr_id + "_close").bind("click", {cat_id: cat_id}, function(event) { unhighlight_haplotypes(event.data.cat_id); event.stopPropagation(); }); } function build_sumstats_view(json, status, jqXHR) { var html = "
x
\n" + "
\n" + " View:\n" + " \n" + " Haplotypes\n" + " \n" + " Allele Depths\n" + " \n" + " LnLs\n"; if (json.type == "map") { html += " " + "Genotypes"; } html += "
"; html += print_population_name(pop_id, json); html += "
" + "
" + "" + sample['sample'].replace("_", " ") + "
\n"; var hap_strs = []; var dep_strs = []; for (var j = 0; j < sample.obshap.length; j++) { hap_index++; var c = colors[color_map[sample.obshap[j].allele]]; var url = json.path + "/stack_view.php" + "?db=" + json.db + "&batch_id=" + json.batch_id + "&samples=" + sample.sample_id + "&tags=" + sample.obshap[j].tag_id + "&tag_id=" + json.id; var a = ""; hap_strs.push("" + a + sample.obshap[j].allele + ""); dep_strs.push("" + a + sample.obshap[j].depth + ""); } var hap_str = hap_strs.join(" / "); var dep_str = dep_strs.join(" / "); var gen_str = ""; if (json.type == "map" && sample.genotype.length > 0) { var id = "gtype_" + json.batch_id + "_" + json.id + "_" + sample.sample_id; var url = json.path + "/correct_genotype.php?db=" + json.db + "&batch_id=" + json.batch_id + "&tag_id=" + json.id + "&sample_id=" + sample.sample_id; var jscript = "correct_genotype('" + id + "', '" + url + "')"; var blur_js = "cancel_correction('" + id + "')"; var sel = generate_element_select(id, sample.marker, sample.genotype.toLowerCase(), jscript, blur_js); var genotype = sample.corrected == 1 ? "" + sample.genotype + "" : sample.genotype; gen_str = "
" + "\n" + "
\n" + sel + "
"; } html += "
" + hap_str + "
" + "
" + dep_str + "
" + "
" + sample.lnl + "
" + gen_str + "
\n" + "\n" + "\n" + "\n" + "\n" + " \n" + " \n"; for (var i = 0; i < json.sumstats.length; i++) { html += "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n"; } html += "
SNP Summary Statistics
Pop\n" + " BP\n" + " Column\n" + " Allele 1\n" + " Allele 2\n" + " P\n" + " N\n" + " Obs Het\n" + " Obs Hom\n" + " Exp Het\n" + " Exp Hom\n" + " π\n" + " FIS\n" + "
" + (i + 1) + "." + json.sumstats[i].pop_id + "" + json.sumstats[i].bp + "" + json.sumstats[i].col + "" + json.sumstats[i].p_allele + "" + json.sumstats[i].q_allele + "" + json.sumstats[i].p + "" + json.sumstats[i].n + "" + json.sumstats[i].obshet + "" + json.sumstats[i].obshom + "" + json.sumstats[i].exphet + "" + json.sumstats[i].exphom + "" + json.sumstats[i].pi + "" + json.sumstats[i].fis + "
\n"; $("#" + json.id + "_sumstat_div").html(html); // // We need to position the sumstat overlay. Identify the right-most // border of the containing div. // parent_div = "#" + json.id + "_popview_div"; t = $(parent_div + " table.snps").position(); p = $(parent_div + " div.comp_view").position(); top_coord = t.top + 32; right_coord = p.left + $(parent_div + " div.comp_view").width() + 25; $("#" + json.id + "_sumstat_div").css({top: top_coord, left: right_coord}); // // Set a maximum height for the containing div. // h = $(parent_div).height() - 50; $("#" + json.id + "_sumstat_div").css("max-height", h); // // Make sure the Fst and Hapstat divs are closed. // unhighlight_snp_row(json.id); close_hapstat_view(json.id); close_phist_view(json.id); close_stack_view(json.id); // // Bind the escape key to close this popup. // $(document).keyup(function(event){ if(event.keyCode === 27) close_sumstat_view(json.id); }); // // Display the Sumstats div. // $("#" + json.id + "_sumstat_div").css("display", ""); $("#" + json.id + "_viewstat_sumstat").css({"color": "#a93535", "border-color": "#a93535", "background-color": "#ffffff"}); } function close_sumstat_view(id) { $("#" + id + "_sumstat_div").css("display", "none"); $("#" + id + "_viewstat_sumstat").css({"color": "", "border-color": "", "background-color": ""}); } function build_hapstats_view(json, status, jqXHR) { var html = "
x
\n" + "\n" + "\n" + "\n" + "\n" + "\n" + " \n" + " \n"; for (var i = 0; i < json.hapstats.length; i++) { html += "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n" + "\n"; } html += "
Haplotype Summary Statistics
Pop\n" + " BP\n" + " N\n" + " Haplotype Cnt\n" + " Gene Diversity\n" + " Haplotype Diversity\n" + "
" + (i + 1) + "." + json.hapstats[i].pop_id + "" + json.hapstats[i].bp + "" + json.hapstats[i].n + "" + json.hapstats[i].hapcnt + "" + json.hapstats[i].genediv + "" + json.hapstats[i].hapdiv + "
\n"; $("#" + json.id + "_hapstat_div").html(html); // // We need to position the sumstat overlay. Identify the right-most // border of the containing div. // parent_div = "#" + json.id + "_popview_div"; t = $(parent_div + " table.snps").position(); p = $(parent_div + " div.comp_view").position(); top_coord = t.top + 32; right_coord = p.left + $(parent_div + " div.comp_view").width() + 25; $("#" + json.id + "_hapstat_div").css({top: top_coord, left: right_coord}); // // Set a maximum height for the containing div. // h = $(parent_div).height() - 50; $("#" + json.id + "_hapstat_div").css("max-height", h); // // Make sure the Fst, Phist, and Sumstat divs are closed. // unhighlight_snp_row(json.id); close_sumstat_view(json.id); close_phist_view(json.id); close_stack_view(json.id); // // Bind the escape key to close this popup. // $(document).keyup(function(event){ if(event.keyCode === 27) close_hapstat_view(json.id); }); // // Display the Hapstats div. // $("#" + json.id + "_hapstat_div").css("display", ""); $("#" + json.id + "_viewstat_hapstat").css({"color": "#a93535", "border-color": "#a93535", "background-color": "#ffffff"}); } function close_hapstat_view(id) { $("#" + id + "_hapstat_div").css("display", "none"); $("#" + id + "_viewstat_hapstat").css({"color": "", "border-color": "", "background-color": ""}); } function build_fst_view(json, status, jqXHR) { var html = "
x
\n" + "\n" + "\n" + " \n" + " \n" + "\n" + "
FSTπoverall
\n"; // // Create a population id => name map. // var pop_names = []; for (var pop_id in json.popkey) pop_names[pop_id] = json.popkey[pop_id]; var keys = Object.keys(pop_names); var pop_cnt = keys.length; var cols = Object.keys(json.columns); for (var col in json.columns) { var matrix = []; for (var i = 0; i < keys.length; i++) matrix[keys[i]] = []; for (var i = 0; i < json.columns[col].length; i++) { var s = json.columns[col][i]; matrix[s.pid_1][s.pid_2] = s; matrix[s.pid_2][s.pid_1] = s; } html += "\n" + "\n" + " \n"; for (var i = 0; i < pop_cnt; i++) html += " \n"; html += "\n"; for (var i = 0; i < pop_cnt; i++) { html += "\n" + " \n"; var str = ""; var cssclass = ""; for (var j = 0; j < pop_cnt; j++) { if (i < j) { cssclass = "class=\"pi\""; if (keys[j] in matrix[keys[i]]) str = matrix[keys[i]][keys[j]]['pi_o']; } else { cssclass = "class=\"fst\""; if (keys[j] in matrix[keys[i]]) str = "" + matrix[keys[i]][keys[j]]['fst'] + ""; } if (i == j) html += " \n"; else if (!(keys[j] in matrix[keys[i]])) html += " \n"; else { var id = json.id + "_" + col + "_" + i + "_" + j; var arg = "'" + json.id + "', '" + col + "', '" + i + "', '" + j + "'"; html += " \n"; } } html += "\n"; } html += "
 " + pop_names[keys[i]] + "
" + pop_names[keys[i]] + "  " + str + "
\n"; } html += "\n"; $("#" + json.id + "_fst_div").html(html); // // We need to make the first Fst table visible. // highlight_snp_row(json.id, cols[0]); // // Hook up event handlers to handle clicking on other SNPs to change the Fst table in view. // for (var i = 0; i < cols.length; i++) { var tr_id = "#" + json.id + "_snps_tr_" + cols[i]; $(tr_id).bind("click", {cat_id: json.id, col: cols[i]}, function(event) { change_highlighted_snp_row(event.data.cat_id, event.data.col); }); } // // We need to position the fst overlay. Identify the right-most // border of the containing div. // parent_div = "#" + json.id + "_popview_div"; t = $(parent_div + " table.snps").position(); p = $(parent_div + " div.comp_view").position(); top_coord = t.top + 32; right_coord = p.left + $(parent_div + " div.comp_view").width() + 25; $("#" + json.id + "_fst_div").css({top: top_coord, left: right_coord}); // // Set a maximum height for the containing div. // h = $(parent_div).height() - 32; $("#" + json.id + "_fst_div").css("max-height", h); // // Make sure the Sumstats, Hapstats, and Phist divs are closed. // close_sumstat_view(json.id); close_hapstat_view(json.id); close_phist_view(json.id); close_stack_view(json.id); // // Bind the escape key to close this popup. // $(document).keyup(function(event){ if(event.keyCode === 27) unhighlight_snp_row(json.id); }); // // Display the Fst div. // $("#" + json.id + "_fst_div").css("display", ""); $("#" + json.id + "_viewstat_fststat").css({"color": "#a93535", "border-color": "#a93535", "background-color": "#ffffff"}); } function close_fst_view(id) { $("#" + id + "_fst_div").css("display", "none"); $("#" + id + "_viewstat_fststat").css({"color": "", "border-color": "", "background-color": ""}); // // Remove handlers for clicking on SNP rows. // $("#snps_table_" + id + " tr").unbind("click"); } function build_phist_view(json, status, jqXHR) { var html = "
x
\n" + "\n" + "\n" + " \n" + " \n" + "\n" + "
ΦSTFST’
\n"; // // Create a population id => name map. // var pop_names = []; for (var pop_id in json.popkey) pop_names[pop_id] = json.popkey[pop_id]; var keys = Object.keys(pop_names); var pop_cnt = keys.length; var matrix = []; for (var i = 0; i < keys.length; i++) matrix[keys[i]] = []; for (var i = 0; i < json.phist.length; i++) { var s = json.phist[i]; matrix[s.pid_1][s.pid_2] = s; matrix[s.pid_2][s.pid_1] = s; } html += "\n" + "\n" + " \n"; for (var i = 0; i < pop_cnt; i++) html += " \n"; html += "\n"; for (var i = 0; i < pop_cnt; i++) { html += "\n" + " \n"; var str = ""; var cssclass = ""; for (var j = 0; j < pop_cnt; j++) { if (i < j) { cssclass = "class=\"pi\""; if (keys[j] in matrix[keys[i]]) str = matrix[keys[i]][keys[j]]['fstp']; } else { cssclass = "class=\"fst\""; if (keys[j] in matrix[keys[i]]) str = matrix[keys[i]][keys[j]]['phist']; } if (i == j) html += " \n"; else if (!(keys[j] in matrix[keys[i]])) html += " \n"; else { var id = json.id + "_" + i + "_" + j; var arg = "'" + json.id + "', '', '" + i + "', '" + j + "'"; html += " \n"; } } html += "\n"; } html += "
 " + pop_names[keys[i]] + "
" + pop_names[keys[i]] + "  " + str + "
\n"; $("#" + json.id + "_phist_div").html(html); // // We need to position the fst overlay. Identify the right-most // border of the containing div. // parent_div = "#" + json.id + "_popview_div"; t = $(parent_div + " table.snps").position(); p = $(parent_div + " div.comp_view").position(); top_coord = t.top + 32; right_coord = p.left + $(parent_div + " div.comp_view").width() + 25; $("#" + json.id + "_phist_div").css({top: top_coord, left: right_coord}); // // Set a maximum height for the containing div. // h = $(parent_div).height() - 32; $("#" + json.id + "_phist_div").css("max-height", h); // // Make sure the Sumstats, Hapstats, and Fst divs are closed. // close_sumstat_view(json.id); close_hapstat_view(json.id); unhighlight_snp_row(json.id); close_stack_view(json.id); // // Bind the escape key to close this popup. // $(document).keyup(function(event){ if(event.keyCode === 27) close_phist_view(json.id); }); // // Display the Fst div. // $("#" + json.id + "_phist_div").css("display", ""); $("#" + json.id + "_viewstat_phistat").css({"color": "#a93535", "border-color": "#a93535", "background-color": "#ffffff"}); } function close_phist_view(id) { $("#" + id + "_phist_div").css("display", "none"); $("#" + id + "_viewstat_phistat").css({"color": "", "border-color": "", "background-color": ""}); } function highlight_fst_cells(id, snp, row, col) { var cell_1, cell_2; if (snp.length == 0) { cell_1 = document.getElementById(id + "_" + row + "_" + col); cell_2 = document.getElementById(id + "_" + col + "_" + row); } else { cell_1 = document.getElementById(id + "_" + snp + "_" + row + "_" + col); cell_2 = document.getElementById(id + "_" + snp + "_" + col + "_" + row); } if (row < col) { cell_1.style.backgroundColor = "#f1592a"; cell_1.style.color = "#ffffff"; cell_2.style.backgroundColor = "#24aae2"; cell_2.style.color = "#ffffff"; } else { cell_1.style.backgroundColor = "#24aae2"; cell_1.style.color = "#ffffff"; cell_2.style.backgroundColor = "#f1592a"; cell_2.style.color = "#ffffff"; } } function unhighlight_fst_cells(id, snp, row, col) { var cell_1, cell_2; if (snp.length == 0) { cell_1 = document.getElementById(id + "_" + row + "_" + col); cell_2 = document.getElementById(id + "_" + col + "_" + row); } else { cell_1 = document.getElementById(id + "_" + snp + "_" + row + "_" + col); cell_2 = document.getElementById(id + "_" + snp + "_" + col + "_" + row); } if (row < col) { cell_1.style.backgroundColor = "#fdc4b8"; cell_1.style.color = "#000000"; cell_2.style.backgroundColor = "#aee2f3"; cell_2.style.color = "#000000"; } else { cell_1.style.backgroundColor = "#aee2f3"; cell_1.style.color = "#000000"; cell_2.style.backgroundColor = "#fdc4b8"; cell_2.style.color = "#000000"; } } function unhighlight_snp_row(cat_id) { $("#snps_table_" + cat_id + " tr").css("border", ""); $("#snps_table_" + cat_id + " tr > td").css("background-color", ""); $("#snps_table_" + cat_id + " tr td:first-child").css("color", ""); $("#snps_table_" + cat_id + " tr td:last-child").css("color", ""); // // Make the Fst tables invisible. // close_fst_view(cat_id); $("div.close_x").css("display", "none"); } function highlight_snp_row(cat_id, col) { unhighlight_snp_row(cat_id) var tr_id = cat_id + "_snps_tr_" + col; $("#" + tr_id).css("border", "4px solid #a93535"); $("#" + tr_id + " > td").css("background-color", "#aaa"); $("#" + tr_id + " td:first-child").css("color", "#fff"); $("#" + tr_id + " td:last-child").css("color", "#fff"); // // Make the Fst table visible. // $("#" + cat_id + "_" + col + "_fst").css("display", ""); $("#" + tr_id + "_close").css("display", ""); $("#" + tr_id + "_close").bind("click", {cat_id: cat_id}, function(event) { unhighlight_snp_row(event.data.cat_id); event.stopPropagation(); }); } function change_highlighted_snp_row(cat_id, col) { $("#snps_table_" + cat_id + " tr").css("border", ""); $("#snps_table_" + cat_id + " tr > td").css("background-color", ""); $("#snps_table_" + cat_id + " tr td:first-child").css("color", ""); $("#snps_table_" + cat_id + " tr td:last-child").css("color", ""); $("div.close_x").css("display", "none"); var tr_id = cat_id + "_snps_tr_" + col; $("#" + tr_id).css("border", "4px solid #a93535"); $("#" + tr_id + " > td").css("background-color", "#aaa"); $("#" + tr_id + " td:first-child").css("color", "#fff"); $("#" + tr_id + " td:last-child").css("color", "#fff"); // // Hide the Fst tables; // $("#" + cat_id + "_fst_div table").css("display", "none"); // // Make the Fst table visible. // $("table.fst_key").css("display", ""); $("#" + cat_id + "_" + col + "_fst").css("display", ""); $("#" + tr_id + "_close").css("display", ""); $("#" + tr_id + "_close").bind("click", {cat_id: cat_id}, function(event) { unhighlight_snp_row(event.data.cat_id); event.stopPropagation(); }); } function build_stack_view(json, status, jqXHR) { var html = "
x
\n"; for (var i = 0; i < json.stacks.length; i++) { html += "\n" + "\n" + " \n" + "\n" + "\n" + " \n" + " \n" + " \n" + " \n" + "\n" + "\n" + " \n" + " \n" + " \n" + " \n" + "\n" + "\n" + " \n" + " \n" + " \n" + " \n" + "\n"; // // Print out the primary stack components. // seq_cnt = 1; for (var j = 0; j < json.stacks[i].primary.length; j++) { var bg = (j + 1) % 2 == 1 ? "style=\"background-color: #dddddd;\"" : ""; for (var k = 0; k < json.stacks[i].primary[j].ids.length; k++) { html += "\n" + " \n" + " \n" + " \n" + " \n" + "\n"; seq_cnt++; } } // // Print out the secondary stack components. // for (var j = 0; j < json.stacks[i].secondary.length; j++) { html += "\n" + " \n" + " \n" + " \n" + " \n" + "\n"; seq_cnt++; } html += "
" + json.stacks[i].sample_name + " [#" + json.stacks[i].sample_id + "]; Stack " + json.stacks[i].tag_id + "
   " + json.stacks[i].scale + "
 consensus" + json.stacks[i].consensus + "
 model" + json.stacks[i].model + "
" + seq_cnt + ".primary" + json.stacks[i].primary[j].ids[k] + "" + json.stacks[i].primary[j].seq + "
" + seq_cnt + ".secondary" + json.stacks[i].secondary[j].id + "" + json.stacks[i].secondary[j].seq + "
\n"; } $("#" + json.id + "_stacks_div").html(html); // // Set a maximum height/width for the containing div. // var parent_div = "#" + json.id + "_popview_div"; var p = $(parent_div).position(); var h = $(parent_div).parent().height(); var w = $(window).height(); h = h > w ? w : h; var div_h = Math.round(h * 0.96); $("#" + json.id + "_stacks_div").css("max-height", div_h); $("html, body").css("cursor", "auto"); // // Display the Stacks div. // $("#" + json.id + "_stacks_div").css("display", ""); // // Make sure the Fst, Phist, and Sumstat divs are closed. // unhighlight_snp_row(json.id); close_sumstat_view(json.id); close_hapstat_view(json.id); close_phist_view(json.id); // // Bind the escape key to close this popup. // $(document).keyup(function(event){ if(event.keyCode === 27) close_stack_view(json.id); }); } function close_stack_view(id) { $("#" + id + "_stacks_div").css("display", "none"); } stacks-1.44/php/samples.php000644 000765 000024 00000011026 12663632701 016454 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); $batch_id = isset($_GET['id']) ? $_GET['id'] : 0; $database = isset($_GET['db']) ? $_GET['db'] : ""; // Connect to the database $db = db_connect($database); // // Prepare some SQL queries // $query = "SELECT id, date, description FROM batches WHERE id=?"; if (!($db['batch_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT id, sample_id, type, file FROM samples " . "WHERE batch_id=? ORDER BY id"; if (!($db['samp_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT COUNT(tag_id) AS count FROM tag_index " . "WHERE batch_id=? AND sample_id=?"; if (!($db['count_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT snps FROM tag_index " . "WHERE batch_id=? AND sample_id=? AND snps>0"; if (!($db['snp_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); // // Pull information about this batch // if (!$db['batch_sth']->bind_param("i", $batch_id)) write_db_error($db['batch_sth'], __FILE__, __LINE__); if (!$db['batch_sth']->execute()) write_db_error($db['batch_sth'], __FILE__, __LINE__); $res = $db['batch_sth']->get_result(); $row = $res->fetch_assoc(); $batch = array(); $batch['id'] = $row['id']; $batch['desc'] = $row['description']; $batch['date'] = $row['date']; $page_title = "RAD-Tag Samples"; write_header($page_title); echo <<< EOQ

RAD-Tag Samples for batch #$batch[id] [$batch[date]; $batch[desc]]

EOQ; if (!$db['samp_sth']->bind_param("i", $batch_id)) write_db_error($db['samp_sth'], __FILE__, __LINE__); if (!$db['samp_sth']->execute()) write_db_error($db['samp_sth'], __FILE__, __LINE__); $res = $db['samp_sth']->get_result(); $tag_id = 0; if (!$db['count_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['count_sth'], __FILE__, __LINE__); if (!$db['snp_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['count_sth'], __FILE__, __LINE__); while ($row = $res->fetch_assoc()) { $snps = 0; $poly = 0; $tag_id = $row['id']; // // Query the database to determine how many tags belong to this sample. // if (!$db['count_sth']->execute()) write_db_error($db['count_sth'], __FILE__, __LINE__); $count_res = $db['count_sth']->get_result(); $count_row = $count_res->fetch_assoc(); $count = $count_row['count']; // // Query the database to find how many SNPs were found in this sample. // if (!$db['snp_sth']->execute()) write_db_error($db['snp_sth'], __FILE__, __LINE__); $count_res = $db['snp_sth']->get_result(); while ($count_row = $count_res->fetch_assoc()) { $snps += $count_row['snps']; $poly += $count_row['snps'] > 0 ? 1 : 0; } print "\n" . " \n" . " \n" . " \n" . " \n" . " \n" . " \n" . "\n"; } echo <<< EOQ
Id Type Unique Stacks Polymorphic Loci SNPs Found Source
$row[id]" . ucfirst($row['type']) . "" . $count . "" . $poly . "" . $snps . "" . $row['file'] . "
EOQ; write_footer(); ?> stacks-1.44/php/sequence_blast.php000644 000765 000024 00000011740 12663652222 020010 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); $database = isset($_GET['db']) ? $_GET['db'] : ""; $tag_id = isset($_GET['tag_id']) ? $_GET['tag_id'] : 0; $batch_id = isset($_GET['batch_id']) ? $_GET['batch_id'] : 0; $page = isset($_GET['p']) ? $_GET['p'] : 1; $per_page = isset($_GET['pp']) ? $_GET['pp'] : 10; // Connect to the database $db = db_connect($database); // Save these variables for automatic URL formation later on. $display = array(); $display['db'] = $database; $display['tag_id'] = $tag_id; $display['batch_id'] = $batch_id; $display['p'] = $page; $display['pp'] = $per_page; // // Prepare some SQL queries // $query = "SELECT id, catalog_id, seq_id, type FROM sequence " . "WHERE batch_id=? AND catalog_id=?"; if (!($db['seq_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT id, catalog_id, seq_id, algorithm, query_id, query_len, hit_id, hit_len, score, e_value, percent_ident, hsp_rank, " . "aln_len, query_aln_start, query_aln_end, hit_aln_start, hit_aln_end " . "FROM sequence_blast " . "WHERE batch_id=? AND catalog_id=?"; if (!($db['blast_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $page_title = "Catalog RAD-Tag Sequence/BLAST Hits Viewer"; write_compact_header($page_title); if (!$db['seq_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['seq_sth'], __FILE__, __LINE__); if (!$db['seq_sth']->execute()) write_db_error($db['seq_sth'], __FILE__, __LINE__); $res = $db['seq_sth']->get_result(); $seqs = array(); // Add the marker itself to the sequence array, it won't // be directly listed in the sequence table. $a = array('id' => -1, 'catalog_id' => $tag_id, 'seq_id' => "", 'type' => "se_radtag"); array_push($seqs, $a); while ($row = $res->fetch_assoc()) { $a = array('id' => $row['id'], 'catalog_id' => $row['catalog_id'], 'seq_id' => $row['seq_id'], 'type' => $row['type']); array_push($seqs, $a); } if (!$db['blast_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['blast_sth'], __FILE__, __LINE__); if (!$db['blast_sth']->execute()) write_db_error($db['blast_sth'], __FILE__, __LINE__); $res = $db['blast_sth']->get_result(); $hits = array(); while ($row = $res->fetch_assoc()) { $a = array('sql_id' => $row['id'], 'query_id' => $row['query_id'], 'hit_id' => $row['hit_id'], 'query_len' => $row['query_len'], 'hit_len' => $row['hit_len'], 'score' => $row['score'], 'e_value' => $row['e_value'], 'pctid' => $row['percent_ident'], 'hsp_rank' => $row['hsp_rank'], 'aln_len' => $row['aln_len']); if (!isset($hits[$row['query_id']])) $hits[$row['query_id']] = array(); array_push($hits[$row['query_id']], $a); } foreach ($seqs as $seq) { if (strlen($seq['seq_id']) > 0) $query_id = $seq['catalog_id'] . "|" . $seq['seq_id']; else $query_id = $seq['catalog_id']; $hsps = $hits[$query_id]; $i = 0; if ($seq['type'] == "se_radtag" && !isset($hsps)) continue; else if ($seq['type'] == "se_radtag") print "

$seq[seq_id] [$seq[type]]

\n"; else echo <<< EOQ

$seq[seq_id] [$seq[type]]

EOQ; if (!isset($hsps)) { print "\n"; continue; } echo <<< EOQ EOQ; foreach ($hsps as $hsp) { $i++; echo <<< EOQ EOQ; } print "
  Hit ID Score E-Value Percent Ident Aln Len Query Len Hit Len HSP Rank
$i $hsp[hit_id] $hsp[score] $hsp[e_value] $hsp[pctid] $hsp[aln_len] $hsp[query_len] $hsp[hit_len] $hsp[hsp_rank]
\n"; } write_compact_footer(); ?> stacks-1.44/php/stack_view.php000644 000765 000024 00000014356 12663356440 017163 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); $tag_str = isset($_GET['tags']) ? $_GET['tags'] : ""; $batch_id = isset($_GET['batch_id']) ? $_GET['batch_id'] : 0; $cat_id = isset($_GET['tag_id']) ? $_GET['tag_id'] : ""; $sample_str = isset($_GET['samples']) ? $_GET['samples'] : ""; $database = isset($_GET['db']) ? $_GET['db'] : ""; $tags = explode(",", $tag_str); $samples = explode(",", $sample_str); // Connect to the database $db = db_connect($database); // // Prepare some SQL queries // $query = "SELECT tag_id, sub_id, relationship, seq_id, seq, deleveraged, blacklisted, removed, file, pop_id " . "FROM unique_tags " . "JOIN samples ON (unique_tags.sample_id=samples.id) " . "JOIN batches ON (samples.batch_id=batches.id) " . "WHERE batch_id=? AND unique_tags.sample_id=? AND tag_id=?"; if (!($db['seq_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT col, rank_1, rank_2 FROM snps " . "JOIN samples ON (snps.sample_id=samples.id) " . "JOIN batches ON (samples.batch_id=batches.id) " . "WHERE batch_id=? AND snps.sample_id=? AND tag_id=? AND snps.type='E' ORDER BY col"; if (!($db['snp_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT col FROM catalog_snps " . "WHERE batch_id=? AND tag_id=?"; if (!($db['cat_snp_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $json_str = "{" . "\"id\": \"$cat_id\"," . "\"stacks\": ["; $sample_id = 0; $tag_id = 0; if (!$db['snp_sth']->bind_param("iii", $batch_id, $sample_id, $tag_id)) write_db_error($db['snp_sth'], __FILE__, __LINE__); if (!$db['cat_snp_sth']->bind_param("ii", $batch_id, $cat_id)) write_db_error($db['cat_snp_sth'], __FILE__, __LINE__); if (!$db['seq_sth']->bind_param("iii", $batch_id, $sample_id, $tag_id)) write_db_error($db['seq_sth'], __FILE__, __LINE__); for ($i = 0; $i < count($tags); $i++) { $sample_id = $samples[$i]; $tag_id = $tags[$i]; // // Fetch and store the SNPs. // $snps = array(); if (!$db['snp_sth']->execute()) write_db_error($db['snp_sth'], __FILE__, __LINE__); $res = $db['snp_sth']->get_result(); while ($row = $res->fetch_assoc()) { $snps[$row['col']] = array('col' => $row['col'], 'rank_1' => $row['rank_1'], 'rank_2' => $row['rank_2']); } // // Fetch and store the catalog SNPs. // $cat_snps = array(); if (!$db['snp_sth']->execute()) write_db_error($db['snp_sth'], __FILE__, __LINE__); $res = $db['snp_sth']->get_result(); while ($row = $res->fetch_assoc()) { if (!isset($cat_snps[$row['col']])) $cat_snps[$row['col']] = 0; $cat_snps[$row['col']]++; } // // Fetch the sequences. // $deleveraged = 0; $lumberjackstack = 0; $blacklisted = 0; $consensus = ""; $model = ""; $file = ""; $stacks = array(); $secondary = array(); if (!$db['seq_sth']->execute()) write_db_error($db['seq_sth'], __FILE__, __LINE__); $res = $db['seq_sth']->get_result(); while ($row = $res->fetch_assoc()) { if ($row['relationship'] == "consensus") { $deleveraged = $row['deleveraged']; $lumberjackstack = $row['removed']; $blacklisted = $row['blacklisted']; $consensus = $row['seq']; $file = $row['file']; } else if ($row['relationship'] == "model") { $model = $row['seq']; } else if ($row['relationship'] == "secondary") { array_push($secondary, array('s' => $row['seq'], 'id' => $row['seq_id'])); } else { if (!isset($stacks[$row['sub_id']])) $stacks[$row['sub_id']] = array(); array_push($stacks[$row['sub_id']], array('s' => $row['seq'], 'id' => $row['seq_id'])); } } $con_len = strlen($consensus); $c = print_snps($tags[$i], $consensus, $consensus, $snps, false); $c = addslashes($c); $scale = print_scale($con_len); $scale = addslashes($scale); $json_str .= "{" . "\"sample_id\": \"$sample_id\"," . "\"sample_name\": \"$file\"," . "\"tag_id\": \"$tag_id\"," . "\"consensus\": \"$c\"," . "\"model\": \"$model\"," . "\"scale\": \"$scale\"," . "\"primary\": ["; foreach ($stacks as $sub_id => $stack) { $s = print_snps_errs($consensus, $stack[0]['s'], $snps, $cat_snps); $s = addslashes($s); $json_str .= "{" . "\"seq\": \"$s\"," . "\"ids\": ["; foreach ($stack as $seq) { $json_str .= "\"$seq[id]\","; } if (count($stack) > 0) $json_str = substr($json_str, 0, -1); $json_str .= "]" . "},"; } if (count($stacks) > 0) $json_str = substr($json_str, 0, -1); $json_str .= "]," . "\"secondary\": ["; foreach ($secondary as $seq) { $s = print_snps_errs($consensus, $seq['s'], $snps, $cat_snps); $s = addslashes($s); $json_str .= "{" . "\"id\": \"$seq[id]\"," . "\"seq\": \"$s\"" . "},"; } if (count($secondary) > 0) $json_str = substr($json_str, 0, -1); $json_str .= "]},"; } $json_str = substr($json_str, 0, -1); $json_str .= "]}"; echo $json_str; ?> stacks-1.44/php/stacks.css000644 000765 000024 00000032137 12663632701 016307 0ustar00catchenstaff000000 000000 /* Copyright 2010-2015, Julian Catchen This file is part of Stacks. Stacks is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version. Stacks is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with Stacks. If not, see . */ body { font-family: bitstream vera sans, arial, sans; } a:link { color: #29356c; } a:visited { color: #29356c; } a:hover { color: #a93535; } img { border: 0; } strong { color: #333333; } acronym { border-bottom: 1px dotted; cursor: help; } .s { font-size: small; } div.main { width: 80%; margin-left: auto; margin-right: auto; } #header { background-color: #455372; background-image: url(./images/stacks_bg.png); background-repeat: repeat-x; min-height: 75px; height: 75px; -moz-border-radius: 10px; -webkit-border-radius: 10px; border-radius: 10px; } #header h1 { position: relative; top: 0px; left: 0.75em; font-size: 16pt; } #header h1 a, #header p a { text-decoration: none; color: #ffffff; } #header p a:hover { color: #a93535; } #header p { float: right; position: relative; top: -6em; padding-right: 0.75em; font-size: 10pt; } div.footer { width: 100%; margin-left: auto; margin-right: auto; font-size: small; clear: both; padding-top: 20px; } h3.info_head, h4.info_head, h5.info_head { margin-left: auto; margin-right: auto; width: 80%; padding-bottom: 0px; margin-bottom: 5px; } h3.info_head a:hover, h4.info_head a:hover { cursor: pointer; } table.loc_filter { width: 100%; vertical-align: top; margin-left: auto; margin-right: auto; border: thin solid black; background-color: #eeeeee; padding-right: 10px; height: 160px; } div.filter { width: 80%; margin-left: auto; margin-right: auto; } table.filter { width: 100%; vertical-align: top; margin-left: auto; margin-right: auto; border: thin solid black; background-color: #eeeeee; padding-left: 10px; height: 160px; } table.filter td { border: 0px; padding-top: 6px; padding-bottom: 6px; } table.filter td.active_filter, table.loc_filter td.active_filter { background-color: #a93535; } table.filter td.active_filter a:hover, table.loc_filter td.active_filter a:hover { color: #ffffff; } table.db { margin-left: auto; margin-right: auto; border: 1px solid black; line-height: 1.25; width: 90%; text-align: center; border-spacing: 0px 1px; border-collapse: collapse; } table.db th { border: 1px solid black; line-height: 1; padding: 7px; background-color: #dddddd; } table.db td { border: 1px solid black; padding: 5px; } /* Internal table, should be invisible in terms of borders and the like */ table.int { border: none; border-collapse: collapse; width: 100%; margin-left: auto; margin-right: auto; text-align: center; padding: 0px; margin: 0px; } table.int td { border: none; } td.seq { text-align: center; } td.seq div.seq { margin-left: auto; margin-right: auto; width: 100%; text-align: left; font-family: monospace; } table.radtag { width: 100%; border-collapse: collapse; } table.radtag th { border-bottom: thin solid black; background-color: #cccccc; } table.radtag td { padding: 0; margin: 0; } table.radtag td.num { text-align: center; font-size: smaller; width: 5%; } table.radtag td.con { padding-left: 3em; font-size: smaller; } table.radtag td.primary { padding-left: 3em; font-size: smaller; color: #32773f; width: 15%; } table.radtag td.secondary { padding-left: 3em; font-size: smaller; color: #870214; width: 15%; } table.radtag td.tertiary { padding-left: 3em; font-size: smaller; color: #3d5aff; width: 15%; } table.radtag td.id { font-family: monospace; text-align: right; padding-right: 4px; width: 20%; } table.radtag td.tag { font-family: monospace; text-align: left; border-left: thin solid black; padding-left: 4px; width: 60%; } div.seq_frame_head { margin-left: auto; margin-right: auto; width: 95%; overflow-y: scroll; } div.seq_frame { margin-left: auto; margin-right: auto; width: 95%; overflow-y: scroll; height: 35em; } span.rank_1 { background-color: #ccddd4; color: #a93535; font-weight: bold; } .rank_1_hi { display: inline-block; border: 2px solid #a93535; background-color: #eeeeee; color: #29356c; font-size: 18pt; font-weight: bold; padding: 5px; line-height: 40%; } span.rank_2 { background-color: #ccddd4; color: #29356c; font-weight: bold; } span.err { background-color: #fcdba7; color: #29356c; } span.cat_snp { background-color: #d8f3ff; color: #29356c; } div.footer { clear: both; } tr.catrow { height: 5em; } td.catlink { font-size: small; position: relative; } td.catlink,div.catlink a:hover { cursor: pointer; } span.light_scale { color: #888888; background-color: #e29c71; } span.dark_scale { color: #888888; background-color: #bbd5ff; } table.genotypes td.gtype_toggle { font-size: smaller; text-align: right; border-top: none; border-left: none; border-right: none; } table.genotypes { padding: 1em; margin-top: 0.5em; margin-bottom: 1em; margin-left: auto; margin-right: auto; border-collapse: collapse; } table.genotypes td { padding: 3px; border: thin solid #aaaaaa; text-align: center; max-width: 10%; word-wrap: break-word; vertical-align: top; } table.genotypes td a:hover { cursor: pointer; } table.genotypes div.title { white-space: pre; font-size: smaller; color: #aaaaaa; } table.genotypes div.title:hover { cursor: pointer; color: #a93535; } table.genotypes td div.title input { display:none; } table.genotypes td.pop_id { height: 1.5em; text-align: left; font-size: 12pt; padding-left: 10px; background-color: #eeeeee; } div.pop_annotation { display: inline; } div.pop_annotation:hover { cursor: pointer; } span.corrected { font-weight: bold; color: #870214; } div.ann_parent { position: absolute; float: left; width: 100%; bottom: 0.75em; } a.annotation { font-size: small; } td.export_icon a:hover { cursor: pointer; } div#export_popup { position: absolute; right: 7%; /*top: 2%;*/ border: 2px solid black; background-color: #6f9b6d; width: 350px; height: 250px; font-size: smaller; text-align: center; padding: 1em; -moz-box-shadow: 4px 4px 7px #888; -webkit-box-shadow: 4px 4px 7px #888; box-shadow: 4px 4px 7px #888; } div#export_popup h3 { color: #ffc600; text-decoration: underline; margin-top: 0px; padding-top: 0px; } .r { font-weight: bold; color: #a93535; } #stacks_filter { margin-left: auto; margin-right: auto; width: 85%; padding-bottom: 20px; } div.gloc { margin-top: 0.5em; margin-left: 0.5em; float: left; font-size: smaller; font-family: bitstream vera sans, arial, sans; clear: both; } div.popview { position: relative; } div.comp_view { float: left; margin-left: 1em; } div.matches { overflow: scroll; } div.stack_view { overflow: scroll; position: absolute; width: 98%; background-color: #333; border: 2px solid #ccc; opacity: 0.95; padding: 10px; border-radius: 5px; overflow: scroll; box-shadow: 5px 5px 7px #888888; } div.haplotype_def { margin-left: auto; margin-right: auto; max-width: 150px; word-wrap: break-word; } div.haplotype { max-width: 85px; word-wrap: break-word; } div.lnl { font-size: 10pt; font-family: monospace; } table.snps, table.alleles { border-collapse: collapse; min-width: 225px; } table.snps { margin-top: 2em; } table.alleles { margin-top: 1.5em; margin-bottom: 1.5em; } table.snps th, table.alleles th { height: 1.5em; text-align: left; font-size: 12pt; font-weight: normal; padding-left: 10px; background-color: #eeeeee; border: thin solid #aaaaaa; } table.snps tr.subhead th, table.alleles tr.subhead th { background-color: #ffffff; font-size: x-small; text-decoration: underline; text-align: center; padding-bottom: 0px; border-left: 0; border-right: 0; border-bottom: 0; } table.snps tr:nth-child(3) td, table.alleles tr:nth-child(3) td { border-top: 0; } table.snps tr:nth-child(2n+4) td, table.alleles tr:nth-child(2n+4) td { background-color: #efefef; } table.snps td, table.alleles td { padding: 3px; border: thin solid #aaaaaa; text-align: center; } table.snps tr:first-child th, table.alleles tr:first-child th { border-top: 0; } table.snps tr:last-child td, table.alleles tr:last-child td { border-bottom: 0; } table.snps tr th:first-child, table.alleles tr th:first-child { border-left: 0; } table.snps tr td:first-child, table.alleles tr td:first-child { border-left: 0; font-size: x-small; text-align: left; color: #aaaaaa; vertical-align: bottom; } table.alleles tr td:last-child, table.alleles tr th:last-child { text-align: left; border-right: 0; } table.snps tr td:last-child, table.snps tr th:last-child { border-right: 0; } table.alleles tr:hover td, table.snps tr:hover td { background-color: #aaa; color: #fff; } table.alleles tr:hover td { cursor: pointer; } div.close_x { float: right; font-weight: bold; } div.close_x:hover { cursor: pointer; } .haphi { display: inline-block; border: 3px solid #ef6c00; padding: 7px; line-height: 35%; border-radius: 5px; } div.sumstats { clear: both; padding-top: 2em; } table.sumstats, table.fst { min-width: 250px; border-collapse: collapse; font-size: 10pt; margin-bottom: 2em; } table.sumstats td, table.fst td { height: 2em; border: thin solid #aaa; vertical-align: middle; } table.sumstats td.pop_id, table.fst td.pop_id { width: 5%; background-color: #eeeeee; border-left: 0px; } table.sumstats td.key, table.fst td.key { border-top: 0px; background-color: #eeeeee; padding: 2px; word-wrap: break-word; } table.fst td.fst span:hover { cursor: help; } table.fst td.diagonal { background-color: #888; } table.fst td.fst, table.fst_key td.fst { background-color: #aee2f3; } table.fst td.pi, table.fst_key td.pi { background-color: #fdc4b8; } table.fst_key { font-size: 10pt; width: 100%; margin-top: 1em; margin-bottom: 1em; } table.fst_key td { font-weight: bold; width: 45%; } span.snp { font-weight: normal; } span.snp:hover { color: #a93535; cursor: pointer; } span.snp_sel { font-weight: bold; text-decoration: underline; } div.viewstat { width: 80%; margin-left: auto; margin-right: auto; margin-top: 0.75em; border: 2px solid #29356c; border-radius: 5px; background-color: #eeeeee; color: #29356c; font-size: 11pt; padding: 3px; font-variant: small-caps; } div.viewstat:hover { color: #a93535; border: 2px solid #a93535; background-color: #ffffff; cursor: pointer; } div.top_viewstat { margin-top: 1.5em; margin-bottom: -1em; } div.popup_stat { position: absolute; background-color: #dddddd; border: 2px solid #aaaaaa; opacity: 0.95; padding: 10px; border-radius: 5px; overflow: scroll; box-shadow: 5px 5px 7px #888888; } div.popup_stat_close { float: right; vertical-align: top; color: #a93535; } div.popup_stat_close:hover { color: #29356c; cursor: pointer; } div.stack_view div.popup_stat_close:hover { color: #ffffff; } table.stack { background-color: #ffffff; border-collapse: collapse; margin-left: auto; margin-right: auto; margin-bottom: 2em; } table.stack tr:last-child td { padding-bottom: 5px; } table.stack th { height: 1.5em; text-align: left; font-size: 12pt; font-weight: normal; padding-left: 10px; background-color: #eeeeee; border: thin solid #aaaaaa; } table.stack td { border: 0; padding: 0px; font-size: smaller; } table.stack td.num { padding-left: 10px; text-align: right; font-size: x-small; padding-right: 5px; border-left: thin solid #aaaaaa; } table.stack td.primary { color: #32773f; } table.stack td.secondary { color: #870214; } table.stack td.id { font-family: monospace; text-align: right; padding-right: 4px; font-size: 8pt; } table.stack td.tag { font-family: monospace; text-align: left; border-left: thin solid black; padding-left: 4px; font-size: 10pt; padding-right: 10px; border-right: thin solid #aaaaaa; } i.fa-spinner { padding: 0.5em; }stacks-1.44/php/stacks.js000644 000765 000024 00000010153 12533677757 016145 0ustar00catchenstaff000000 000000 // // Copyright 2010, Julian Catchen // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // function toggle_div(id, root_path, page_state_form) { var tr_obj = document.getElementById(id); var caret = document.getElementById(id + "_img"); if (tr_obj.style.display == "none") { tr_obj.style.display = ""; caret.src = root_path + "/caret-d.png"; update_page_state(page_state_form, id, 1); } else { tr_obj.style.display = "none"; caret.src = root_path + "/caret-u.png"; update_page_state(page_state_form, id, 0); } } function toggle_aln_tr(id, root_path, url) { var tr_obj = document.getElementById(id); var caret = document.getElementById(id + "_img"); if (tr_obj.style.display == "none") { tr_obj.style.display = ""; caret.src = root_path + "/caret-d.png"; // Check if the alignment has been loaded, if not, load // it in the iframe. var iframe_obj = document.getElementById(id + '_iframe'); iframe_obj.src = url; } else { tr_obj.style.display = "none"; caret.src = root_path + "/caret-u.png"; } } function toggle_sel(id) { var div_obj = document.getElementById(id); if (div_obj.style.display == "none") { div_obj.style.display = ""; } else { div_obj.style.display = "none"; } } function set_operation(id, operation) { if (operation == "reset") { var verify = confirm("Are you sure you want to delete all corrections for this marker?"); if (!verify) return; } var form_obj = document.getElementById(id); form_obj.op.value = operation; form_obj.submit(); } function toggle_cb(form_id, value) { var form_obj = document.getElementById(form_id); for(i = 0; i < form_obj.elements.length; i++) if (form_obj.elements[i].value == value) if (form_obj.elements[i].checked == true) form_obj.elements[i].checked = false; else form_obj.elements[i].checked = true; } function toggle_genotypes(catalog_id, table_id, type) { var table_obj = document.getElementById(table_id); divs = table_obj.getElementsByTagName("div"); for(i = 0; i < divs.length; i++) { if (divs[i].id.substr(0, 3) == type) { if (divs[i].style.display == "none") divs[i].style.display = ""; else divs[i].style.display = "none"; } } // // Adjust the height of the parent iframe to possibly accomodate more data // var iframe_obj = parent.document.getElementById(catalog_id + "_iframe"); if (iframe_obj) iframe_obj.style.height = (this.document.body.offsetHeight+25) + 'px'; } function toggle_sumstats(span_obj, col) { var snp_index = document.getElementById("snp_index"); var old_snp = snp_index.value; var old_snp_id = old_snp + "_snp"; var old_span_obj = document.getElementById(old_snp_id); snp_index.value = col; if (col == old_snp) return; span_obj.className = "snp_sel"; old_span_obj.className = "snp"; var div; var div_id; div_id = old_snp + "_sumstats"; div = document.getElementById(div_id); div.style.display = "none"; div_id = old_snp + "_fst"; div = document.getElementById(div_id); div.style.display = "none"; div_id = col + "_sumstats"; div = document.getElementById(div_id); div.style.display = ""; div_id = col + "_fst"; div = document.getElementById(div_id); div.style.display = ""; } stacks-1.44/php/stacks_functions.php000644 000765 000024 00000015735 12663632701 020403 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // function write_header($page_title) { global $version, $site_title; global $root_path, $img_path; echo <<< EOQ $site_title: $page_title
EOQ; } function write_compact_header($page_title) { global $root_path, $css_path, $site_title, $js_path; echo <<< EOQ $site_title: $page_title EOQ; } function write_footer() { echo <<< EOQ
last updated: EOQ; include("last_modified.php"); echo <<< EOQ
EOQ; } function write_compact_footer() { echo <<< EOQ EOQ; } function print_scale($max_len) { if ($max_len == 0) return ""; $class = array("light_scale", "dark_scale"); $str = ""; $c = 0; $str .= ""; for ($i = 0; $i < $max_len; $i++) { $str .= $i % 10; if (($i + 1) % 10 == 0) { $c++; $str .= ""; } } $str .= ""; return $str; } function print_snps($tag_id, $consensus, $seq, $snps, $wrap) { global $display_len; $str = ""; $start = 0; $snp_cnt = count($snps); while (count($snps)) { $snp = array_shift($snps); $con = substr($consensus, $snp['col'], 1); $end = $snp['col']; $s = substr($seq, $start, $end - $start); $str .= $s; $s = substr($seq, $end, 1); if ($con == $s) $str .= "$s"; else if ($s == "N") $str .= "$s"; else $str .= "$s"; $start = $end + 1; } if ($snp_cnt > 0) { $s = substr($seq, $start); $str .= $s; } else { $str = $consensus; } if ($wrap == false || strlen($consensus) <= $display_len) return $str; // // Add line breaks to the sequence // $s = ""; $nuc = 0; $pos = 0; $len = strlen($str); while ($len > $display_len) { for ($pos = 0, $nuc = 0; $pos < $display_len && $nuc < $len; $nuc++, $pos++) { if ($str[$nuc] == '<') { do { $nuc++; } while ($str[$nuc] != '>'); $pos--; } } $s .= substr($str, 0, $nuc) . "
\n"; $str = substr($str, $nuc); $pos = 0; $nuc = 0; $len = strlen($str); } $s .= $str; return $s; } function print_snps_errs($consensus, $sequence, $snps, $cat_snps) { $str = ""; $con = str_split($consensus); $seq = str_split($sequence); for ($i = 0; $i < count($con); $i++) { // Is a SNP defined in this column? if (isset($snps[$i])) { if ($con[$i] == $seq[$i]) $str .= "$seq[$i]"; else if ($seq[$i] == "N") $str .= "$seq[$i]"; else $str .= "$seq[$i]"; } else if (isset($cat_snps[$i])) { $str .= "$seq[$i]"; } else { // Does this nucleotide equal the consensus nucleotide at position $i? if ($con[$i] == $seq[$i] || $seq[$i] == "N") $str .= $seq[$i]; else $str .= "$seq[$i]"; } } return $str; } function generate_key_element_select($name, $elements, $selected_key, $javascript) { $script_code = ""; if (strlen($javascript) > 0) { $script_code = " onchange=\"$javascript\""; } $ctl = " \n"; return $ctl; } function generate_element_select($name, $elements, $selected_ele, $change_js, $blur_js = "") { $script_code = ""; if (strlen($change_js) > 0 && strlen($blur_js) > 0) { $script_code = " onchange=\"$change_js\" onblur=\"$blur_js\""; } else if (strlen($change_js) > 0) { $script_code = " onchange=\"$change_js\""; } else if (strlen($blur_js) > 0) { $script_code = " onblur=\"$blur_js\""; } $ctl = " \n"; return $ctl; } function print_bp($bp) { // Convert the location to be in megabases if ($bp > 1000000) $bp = sprintf("%.02fMb", $bp / 1000000); else if ($bp > 1000) $bp = sprintf("%.1fKb", $bp / 1000); else $bp = sprintf("%dbp", $bp); return $bp; } ?> stacks-1.44/php/sumstat_view.php000644 000765 000024 00000011306 12663632701 017543 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); $database = isset($_GET['db']) ? $_GET['db'] : ""; $tag_id = isset($_GET['tag_id']) ? $_GET['tag_id'] : 0; $batch_id = isset($_GET['batch_id']) ? $_GET['batch_id'] : 0; $batch_type = isset($_GET['type']) ? $_GET['type'] : "map"; // Connect to the database $db = db_connect($database); // Save these variables for automatic URL formation later on. $display = array(); $display['db'] = $database; $display['tag_id'] = $tag_id; $display['batch_id'] = $batch_id; // // Prepare some SQL queries // $query = "SELECT pop_id, pop_name FROM populations " . "WHERE batch_id=?"; if (!($db['pop_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT pop_id, col, bp, p_nuc, q_nuc, n, p, obs_het, obs_hom, exp_het, exp_hom, pi, fis FROM sumstats " . "WHERE batch_id=? AND tag_id=?"; if (!($db['stats_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); // // Fetch population names if available. // $pop_names = array(); if ($batch_type == "population") { if (!$db['pop_sth']->bind_param("i", $batch_id)) write_db_error($db['pop_sth'], __FILE__, __LINE__); if (!$db['pop_sth']->execute()) write_db_error($db['pop_sth'], __FILE__, __LINE__); $res = $db['pop_sth']->get_result(); while ($row = $res->fetch_assoc()) $pop_names[$row['pop_id']] = $row['pop_name']; } if (!$db['stats_sth']->bind_param("ii", $batch_id, $tag_id)) write_db_error($db['stats_sth'], __FILE__, __LINE__); if (!$db['stats_sth']->execute()) write_db_error($db['stats_sth'], __FILE__, __LINE__); $res = $db['stats_sth']->get_result(); $stats = array(); $pops = array(); while ($row = $res->fetch_assoc()) { $a = array('col' => $row['col'], 'bp' => $row['bp'], 'p_nuc' => $row['p_nuc'], 'q_nuc' => $row['q_nuc'], 'n' => $row['n'], 'p' => $row['p'], 'obs_het' => $row['obs_het'], 'obs_hom' => $row['obs_hom'], 'exp_het' => $row['exp_het'], 'exp_hom' => $row['exp_hom'], 'pi' => $row['pi'], 'fis' => $row['fis'], 'pop_id' => $row['pop_id']); if (!isset($stats[$row['col']])) $stats[$row['col']] = array(); $stats[$row['col']][$row['pop_id']] = $a; $pops[$row['pop_id']] = $row['pop_id']; } ksort($stats); ksort($pops); $json_str = "{" . "\"path\": \"$root_path\"," . "\"batch_id\": \"$batch_id\"," . "\"db\": \"$database\"," . "\"id\": \"$tag_id\"," . "\"type\": \"$batch_type\","; $json_str .= "\"sumstats\": ["; foreach ($pops as $pop_id) if (!isset($pop_names[$pop_id])) $pop_names[$pop_id] = $pop_id; $index = count($pop_names) == 0 ? $pops : $pop_names; $rows = 0; foreach ($stats as $col => $stat) { foreach ($index as $pop_id => $pop_name) { if (!isset($stat[$pop_id])) continue; $s = $stat[$pop_id]; $p = $s['p'] < 1 ? sprintf("%.5f", $s['p']) : $s['p']; $ohet = $s['obs_het'] > 0 ? sprintf("%.3f", $s['obs_het']) : $s['obs_het']; $ohom = $s['obs_hom'] < 1 ? sprintf("%.3f", $s['obs_hom']) : $s['obs_hom']; $ehet = $s['exp_het'] > 0 ? sprintf("%.3f", $s['exp_het']) : $s['exp_het']; $ehom = $s['exp_hom'] < 1 ? sprintf("%.3f", $s['exp_hom']) : $s['exp_hom']; $pi = $s['pi'] > 0 ? sprintf("%.3f", $s['pi']) : $s['pi']; $fis = $s['fis'] != 0 ? sprintf("%.3f", $s['fis']) : "0"; $json_str .= "{" . "\"pop_id\": \"$pop_name\"," . "\"col\": \"$s[col]\"," . "\"bp\": \"$s[bp]\"," . "\"p_allele\": \"$s[p_nuc]\"," . "\"q_allele\": \"$s[q_nuc]\"," . "\"p\": \"$p\"," . "\"n\": \"$s[n]\"," . "\"obshet\": \"$ohet\"," . "\"obshom\": \"$ohom\"," . "\"exphet\": \"$ehet\"," . "\"exphom\": \"$ehom\"," . "\"pi\": \"$pi\"," . "\"fis\": \"$fis\"" . "},"; $rows++; } } if ($rows > 0) $json_str = substr($json_str, 0, -1); $json_str .= "]}"; echo $json_str; ?> stacks-1.44/php/tag.php000644 000765 000024 00000026551 12663356440 015577 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); $tag_id = isset($_GET['tag_id']) ? $_GET['tag_id'] : 0; $batch_id = isset($_GET['batch_id']) ? $_GET['batch_id'] : 0; $sample_id = isset($_GET['sample_id']) ? $_GET['sample_id'] : 0; $database = isset($_GET['db']) ? $_GET['db'] : ""; $page = isset($_GET['p']) ? $_GET['p'] : 1; $per_page = isset($_GET['pp']) ? $_GET['pp'] : 10; // Connect to the database $db = db_connect($database); // Save these variables for automatic URL formation later on. $display = array(); $display['tag_id'] = $tag_id; $display['batch_id'] = $batch_id; $display['sample_id'] = $sample_id; $display['db'] = $database; $display['p'] = $page; $display['pp'] = $per_page; // // Prepare some SQL queries // $query = "SELECT batches.id as id, date, description, samples.id as sample_id, samples.sample_id as samp_id, samples.type, file " . "FROM batches " . "JOIN samples ON (batch_id=batches.id) " . "WHERE batches.id=? AND samples.id=?"; if (!($db['batch_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT count(id) as depth FROM unique_tags " . "WHERE relationship!='consensus' AND relationship!='model' AND sample_id=? AND tag_id=?"; if (!($db['depth_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT tag_id, sub_id, relationship, seq_id, seq, deleveraged, blacklisted, removed " . "FROM unique_tags " . "JOIN samples ON (unique_tags.sample_id=samples.id) " . "JOIN batches ON (samples.batch_id=batches.id) " . "WHERE batch_id=? AND unique_tags.sample_id=? AND tag_id=?"; if (!($db['seq_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT col, rank_1, rank_2 FROM snps " . "JOIN samples ON (snps.sample_id=samples.id) " . "JOIN batches ON (samples.batch_id=batches.id) " . "WHERE batch_id=? AND snps.sample_id=? AND tag_id=? AND snps.type='E' ORDER BY col"; if (!($db['snp_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT allele, read_pct FROM alleles " . "WHERE sample_id=? AND tag_id=? "; if (!($db['all_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT catalog_id FROM matches " . "WHERE batch_id=? AND sample_id=? AND tag_id=? " . "GROUP BY catalog_id"; if (!($db['cat_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT col FROM catalog_snps " . "WHERE batch_id=? AND tag_id=?"; if (!($db['cat_snp_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); // // Pull information about this batch // if (!$db['batch_sth']->bind_param("ii", $batch_id, $sample_id)) write_db_error($db['batch_sth'], __FILE__, __LINE__); if (!$db['batch_sth']->execute()) write_db_error($db['batch_sth'], __FILE__, __LINE__); $res = $db['batch_sth']->get_result(); $row = $res->fetch_assoc(); $batch = array(); $batch['id'] = $row['id']; $batch['desc'] = $row['description']; $batch['date'] = $row['date']; $batch['file'] = $row['file']; $batch['sample_id'] = $row['sample_id']; $batch['samp_id'] = $row['samp_id']; $batch['sample_type'] = $row['type']; $page_title = "RAD-Tag Viewer"; write_header($page_title, $batch); echo <<< EOQ

Batch #$batch[id] [$batch[date]; $batch[desc]]

RAD-Tag Sample #$batch[samp_id] [$batch[file]]

Sequence #$display[tag_id]

EOQ; // // Fetch the seqeunces // $deleveraged = 0; $lumberjackstack = 0; $blacklisted = 0; $seqs = array('consensus' => array(), 'primary' => array(), 'secondary' => array(), 'model' => array()); if (!$db['seq_sth']->bind_param("iii", $batch_id, $sample_id, $tag_id)) write_db_error($db['seq_sth'], __FILE__, __LINE__); if (!$db['seq_sth']->execute()) write_db_error($db['seq_sth'], __FILE__, __LINE__); $res = $db['seq_sth']->get_result(); while ($row = $res->fetch_assoc()) { array_push($seqs[$row['relationship']], array('s' => $row['seq'], 'id' => $row['seq_id'], 'sub_id' => $row['sub_id'])); if ($row['relationship'] == "consensus") { $deleveraged = $row['deleveraged']; $lumberjackstack = $row['removed']; $blacklisted = $row['blacklisted']; } } if (!$db['cat_sth']->bind_param("iii", $batch_id, $sample_id, $tag_id)) write_db_error($db['cat_sth'], __FILE__, __LINE__); if (!$db['cat_sth']->execute()) write_db_error($db['cat_sth'], __FILE__, __LINE__); $res = $db['cat_sth']->get_result(); $row = $res->fetch_assoc(); if (isset($row['catalog_id'])) $catalog_id = $row['catalog_id']; else $catalog_id = -1; print "\n"; if (!$db['depth_sth']->bind_param("ii", $sample_id, $tag_id)) write_db_error($db['depth_sth'], __FILE__, __LINE__); if (!$db['depth_sth']->execute()) write_db_error($db['depth_sth'], __FILE__, __LINE__); $res = $db['depth_sth']->get_result(); $row = $res->fetch_assoc(); echo <<< EOQ EOQ; $snps = array(); if (!$db['snp_sth']->bind_param("iii", $batch_id, $sample_id, $tag_id)) write_db_error($db['snp_sth'], __FILE__, __LINE__); if (!$db['snp_sth']->execute()) write_db_error($db['snp_sth'], __FILE__, __LINE__); $res = $db['snp_sth']->get_result(); while ($row = $res->fetch_assoc()) { $snps[$row['col']] = array('col' => $row['col'], 'rank_1' => $row['rank_1'], 'rank_2' => $row['rank_2']); } $cat_snps = array(); if ($catalog_id >= 0) { if (!$db['cat_snp_sth']->bind_param("ii", $batch_id, $catalog_id)) write_db_error($db['cat_snp_sth'], __FILE__, __LINE__); if (!$db['cat_snp_sth']->execute()) write_db_error($db['cat_snp_sth'], __FILE__, __LINE__); $res = $db['cat_snp_sth']->get_result(); while ($row = $res->fetch_assoc()) { if (!isset($cat_snps[$row['col']])) $cat_snps[$row['col']] = 0; $cat_snps[$row['col']]++; } } print " \n"; if (!$db['all_sth']->bind_param("ii", $sample_id, $tag_id)) write_db_error($db['all_sth'], __FILE__, __LINE__); if (!$db['all_sth']->execute()) write_db_error($db['all_sth'], __FILE__, __LINE__); $res = $db['all_sth']->get_result(); $alleles = array(); $i = 0; print " \n"; $deleveraged == 1 ? print " " : print " "; $lumberjackstack == 1 ? print " " : print " "; $blacklisted == 1 ? print " " : print " "; echo <<< EOQ
Catalog ID Depth SNPs Alleles Deleveraged? Lumberjackstack? Blacklisted?
\n"; if ($catalog_id >= 0) print " #$catalog_id\n"; print " $row[depth]x \n". " \n"; foreach ($snps as $snp) { print "\n" . " \n" . "\n"; } print "
Column: $snp[col]$snp[rank_1]/$snp[rank_2]
\n" . " 		\n" . " \n"; while ($row = $res->fetch_assoc()) { $alleles[$row['allele']] = $colors[$i % $color_size]; print " \n" . " \n" . " \n" . " \n"; $i++; } print "
$row[allele]" . sprintf("%.2f%%", $row['read_pct']) . "
\n" . " 		TrueFalseTrueFalseTrueFalse
EOQ; $con_len = isset($seqs['consensus'][0]) ? strlen($seqs['consensus'][0]['s']) : 0; $con_seq = isset($seqs['consensus'][0]) ? $seqs['consensus'][0]['s'] : ""; $s = print_scale($con_len); print "\n" . " \n" . " \n" . " \n" . " \n" . "\n"; $s = print_snps($display['tag_id'], $con_seq, $con_seq, $snps, false); print "\n" . " \n" . " \n" . " \n" . " \n" . "\n"; $s = isset($seqs['model'][0]) ? $seqs['model'][0]['s'] : ""; echo <<< EOQ
  Relationship Seq ID Sequence
   " . $s . "
 consensus" . (isset($seqs['consensus'][0]) ? $seqs['consensus'][0]['id'] : "") . "" . $s . "
EOQ; $i = 1; foreach (array('primary', 'secondary') as $key) { foreach ($seqs[$key] as $seq) { $s = print_snps_errs($con_seq, $seq['s'], $snps, $cat_snps); if ($key == "primary" && $seq['sub_id'] % 2 == 1) $bg = "style=\"background-color: #dddddd;\""; else $bg = ""; print "\n" . " \n" . " \n" . " \n" . " \n" . "\n"; $i++; } } echo <<< EOQ
  model $s
$i$key$seq[id]$s
EOQ; write_footer(); ?> stacks-1.44/php/tags.php000644 000765 000024 00000045310 12663356440 015754 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); $database = isset($_GET['db']) ? $_GET['db'] : ""; $batch_id = isset($_GET['batch_id']) ? $_GET['batch_id'] : 0; $sample_id = isset($_GET['sample_id']) ? $_GET['sample_id'] : 0; $page = isset($_GET['p']) ? $_GET['p'] : 1; $per_page = isset($_GET['pp']) ? $_GET['pp'] : 10; // Connect to the database $db = db_connect($database); // Save these variables for automatic URL formation later on. $display = array(); $display['batch_id'] = $batch_id; $display['sample_id'] = $sample_id; $display['db'] = $database; $display['p'] = $page; $display['pp'] = $per_page; $display['filter_type'] = array(); // // Process the filtering parameters // $param = array(); process_filter($display); prepare_filter_parameters($display, $param); // // Prepare some SQL queries // $query = "SELECT batches.id as id, date, description, samples.id as sample_id, file FROM batches " . "JOIN samples ON (batch_id=batches.id) " . "WHERE batches.id=? AND samples.id=?"; if (!($db['batch_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT COUNT(tag_id) as count FROM tag_index " . "WHERE batch_id=? AND sample_id=?"; $query .= apply_query_filters($display); if (!($db['count_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT depth as count FROM tag_index " . "WHERE batch_id=? AND sample_id=? AND tag_id=?"; if (!($db['depth_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $query = "SELECT col, rank_2 FROM snps " . "JOIN samples ON (snps.sample_id=samples.id) " . "JOIN batches ON (samples.batch_id=batches.id) " . "WHERE snps.type='E' AND batch_id=? AND snps.sample_id=? AND tag_id=? ORDER BY col"; if (!($db['snp_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); // // Pull information about this batch // if (!$db['batch_sth']->bind_param("ii", $batch_id, $sample_id)) write_db_error($db['batch_sth'], __FILE__, __LINE__); if (!$db['batch_sth']->execute()) write_db_error($db['batch_sth'], __FILE__, __LINE__); $res = $db['batch_sth']->get_result(); $row = $res->fetch_assoc(); $batch = array(); $batch['id'] = $row['id']; $batch['desc'] = $row['description']; $batch['date'] = $row['date']; $batch['sample_id'] = $row['sample_id']; $batch['file'] = $row['file']; $page_title = "RAD-Tag Sample Viewer"; write_header($page_title, $batch); echo <<< EOQ

Batch #$batch[id] [$batch[date]; $batch[desc]]

EOQ; write_filter(); echo <<< EOQ

RAD-Tag Sample #$batch[sample_id] [$batch[file]]

EOQ; $query = "SELECT con_tag_id as id, tag_index.tag_id as tag_id, " . "depth, seq, catalog_id, " . "tag_index.deleveraged, tag_index.removed, tag_index.blacklisted " . "FROM tag_index " . "JOIN unique_tags ON (con_tag_id=unique_tags.id) " . "WHERE tag_index.batch_id=? AND tag_index.sample_id=?"; $query .= apply_query_filters($display); $query .= " LIMIT " . ($start_group - 1) . ", " . $display['pp']; if (!($db['tag_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); array_unshift($param, $db['tag_sth']); call_user_func_array("mysqli_stmt_bind_param", $param); array_shift($param); if (!$db['tag_sth']->execute()) write_db_error($db['tag_sth'], __FILE__, __LINE__); $res = $db['tag_sth']->get_result(); $tag_id = 0; if (!$db['snp_sth']->bind_param("iii", $batch_id, $sample_id, $tag_id)) write_db_error($db['snp_sth'], __FILE__, __LINE__); while ($row = $res->fetch_assoc()) { $tag_id = $row['tag_id']; // // Query the database to find how many SNPs were found in this sample. // $snps = array(); if (!$db['snp_sth']->execute()) write_db_error($db['snp_sth'], __FILE__, __LINE__); $snp_res = $db['snp_sth']->get_result(); while ($snp_row = $snp_res->fetch_assoc()) { array_push($snps, array('col' => $snp_row['col'], 'rank' => $snp_row['rank_2'])); } print "\n" . " \n" . " \n"; if (count($snps) == 0) print " \n"; else print " \n"; $s = print_snps($row['tag_id'], $row['seq'], $row['seq'], $snps, true); print " \n"; $row['catalog_id'] > 0 ? print " \n" : print " \n"; print "\n"; } print "\n" . "
EOQ; // // Figure out how many results there are (including filtering) // and write out the proper pagination links // array_unshift($param, $db['count_sth']); call_user_func_array("mysqli_stmt_bind_param", $param); array_shift($param); if (!$db['count_sth']->execute()) write_db_error($db['count_sth'], __FILE__, __LINE__); $res = $db['count_sth']->get_result(); $row = $res->fetch_assoc(); $pagination_count = $row['count']; $start_group = 0; $end_group = 0; write_pagination($pagination_count, $start_group, $end_group, "tags.php"); echo <<< EOQ
Id Depth SNP Consensus Catalog ID
" . "$row[tag_id]$row[depth]NoYes [" . count($snps) . "nuc]
" . $s . "
$row[catalog_id]absent
\n"; write_pagination($pagination_count, $start_group, $end_group, "tags.php"); echo <<< EOQ
EOQ; write_footer(); function generate_hidden_form_vars($var) { global $root_path, $display; $vars = ""; foreach ($display as $key => $d) { if (strstr($key, $var)) continue; if (is_array($d)) { foreach ($d as $e) { $vars .= " \n"; } } else { $vars .= " \n"; } } return $vars; } function generate_per_page_select($name, $per_page) { $pages = array("10", "50", "100", "all"); $ctl = " \n"; return $ctl; } function generate_url($destination) { global $root_path, $display; $url = "href=\"" . $root_path . "/" . $destination . "?"; foreach ($display as $key => $d) { if (is_array($d)) { foreach ($d as $e) $url .= "{$key}[]=$e&"; } else { $url .= "$key=$d&"; } } // Remove the hanging '&' $url = substr($url, 0, -1); $url .= "\""; return $url; } function generate_page_list($page, $num_pages, $destination) { global $display; $page_list = ""; if ($page <= 4) { for ($i = 1; $i < $page; $i++) { $display['p'] = $i; $p = generate_url($destination); $page_list .= "$i\n"; } } else { $display['p'] = 1; $p = generate_url($destination); $page_list .= "1 ...\n"; foreach (array($page - 3, $page - 2, $page - 1) as $i) { $display['p'] = $i; $p = generate_url($destination); $page_list .= "$i\n"; } } $page_list .= " $page\n"; if ($page <= $num_pages - 4) { for ($i = $page + 1; $i <= $page + 3; $i++) { $display['p'] = $i; $p = generate_url($destination); $page_list .= "$i\n"; } $display['p'] = $num_pages; $p = generate_url($destination); $page_list .= "... $num_pages\n"; } else { for ($i = $page + 1; $i <= $num_pages; $i++) { $display['p'] = $i; $p = generate_url($destination); $page_list .= "$i\n"; } } $display['p'] = $page; $page_list = "\n" . $page_list . "\n"; return $page_list; } function write_pagination($num_tags, &$start_gene, &$end_gene, $destination) { global $img_path, $root_path, $display; $cur_page = $display['p']; $page = $display['p']; $per_page = $display['pp']; if ($per_page == "all") $per_page = $num_tags; // // First figure out the total number of pages. If there are // additional genes left over, add an extra page. // $num_pages = floor($num_tags / $per_page); $num_pages += $num_tags % $per_page >= 1 ? 1 : 0; if ($page > $num_pages) { $page = $num_pages; $cur_page = $num_pages; } // Determine the start and end gene numbers $start_gene = 1 + (($page - 1) * $per_page); $end_gene = $start_gene + $per_page > $num_tags ? $num_tags : ($start_gene + $per_page - 1); // Generate the URLs for our links $display['p'] -= 1; $prev_page = generate_url($destination); $display['p'] += 2; $next_page = generate_url($destination); $display['p'] = $cur_page; print "\n" . "\n" . "\n"; if ($num_pages > 1) { $page_list = generate_page_list($page, $num_pages, $destination); print $page_list; } $hidden_vars = generate_hidden_form_vars("pp"); $per_page_ctl = generate_per_page_select("pp", $display['pp']); echo <<< EOQ
\n"; if ($page == 1) { if ($num_pages == 1) { echo <<< EOQ No Previous Page $page No Next Page EOQ; } else { echo <<< EOQ No Previous Page $page View Next Page EOQ; } } else if ($page == $num_pages) { echo <<< EOQ View Previous Page $page No Next Page EOQ; } else { echo <<< EOQ View Previous Page $page View Next Page EOQ; } print " ($num_tags tags)\n" . "
$hidden_vars tags per page   $per_page_ctl
EOQ; } function write_filter() { global $img_path, $root_path, $display; $hidden_vars = generate_hidden_form_vars("filter"); $filters = array("depth" => array(), "snps" => array(), "tagid" => array(), "black" => array(), "delv" => array(), "rem" => array()); $ele_name = isset($display['filter_depth']) ? $display['filter_depth'] : ""; $depth_ctl = generate_element_select("filter_depth", array(1, 5, 10, 20), $ele_name, ""); $ele_name = isset($display['filter_snps']) ? $display['filter_snps'] : ""; $snps_ctl = generate_element_select("filter_snps", array(1, 2, 3, 4, 5), $ele_name, ""); $ele_name = isset($display['filter_delv']) ? $display['filter_delv'] : ""; $delv_ctl = generate_key_element_select("filter_delv", array(1 => "True", 0 => "False"), $ele_name, ""); $ele_name = isset($display['filter_rem']) ? $display['filter_rem'] : ""; $rem_ctl = generate_key_element_select("filter_rem", array(1 => "True", 0 => "False"), $ele_name, ""); $ele_name = isset($display['filter_black']) ? $display['filter_black'] : ""; $black_ctl = generate_key_element_select("filter_black", array(1 => "True", 0 => "False"), $ele_name, ""); if (isset($display['filter_type'])) { foreach ($filters as $key => $f) if (in_array($key, $display['filter_type'])) { $filters[$key]['sel'] = "checked=\"checked\""; $filters[$key]['tr'] = "class=\"active_filter\""; } else { $filters[$key]['sel'] = ""; $filters[$key]['tr'] = ""; } } else { $filters['none']['sel'] = "checked=\"checked\""; } $tagid = isset($display['filter_tagid']) ? $display['filter_tagid'] : ""; echo <<< EOQ

Filter Results

$hidden_vars
Filter by Tag ID:
Filter by depth of coverage: $depth_ctl
Filter by presence of SNPs: $snps_ctl
View deleveraged tags: $delv_ctl
View blacklisted tags: $black_ctl
View lumberjack stacks: $rem_ctl
EOQ; } function process_filter(&$display_params) { if (!isset($_GET['filter_type'])) return; foreach ($_GET['filter_type'] as $filter) { array_push($display_params['filter_type'], $filter); if ($filter == "depth") { $display_params['filter_depth'] = $_GET['filter_depth']; } else if ($filter == "snps") { $display_params['filter_snps'] = $_GET['filter_snps']; } else if ($filter == "tagid") { $display_params['filter_tagid'] = $_GET['filter_tagid']; } else if ($filter == "delv") { $display_params['filter_delv'] = $_GET['filter_delv']; } else if ($filter == "black") { $display_params['filter_black'] = $_GET['filter_black']; } else if ($filter == "rem") { $display_params['filter_rem'] = $_GET['filter_rem']; } } } function prepare_filter_parameters(&$display_params, &$param) { $filters = $display_params['filter_type']; $param[] = &$display_params['batch_id']; $param[] = &$display_params['sample_id']; $typestr = "ii"; if (!isset($filters)) { array_unshift($param, $typestr); return; } foreach ($filters as $filter) { if ($filter == "snps") { $param[] = &$display_params['filter_snps']; $typestr .= "i"; } else if ($filter == "depth") { $param[] = &$display_params['filter_depth']; $typestr .= "i"; } else if ($filter == "tagid") { $param[] = &$display_params['filter_tagid']; $typestr .= "i"; } else if ($filter == "delv") { $param[] = &$display_params['filter_delv']; $typestr .= "i"; } else if ($filter == "black") { $param[] = &$display_params['filter_black']; $typestr .= "i"; } else if ($filter == "rem") { $param[] = &$display_params['filter_rem']; $typestr .= "i"; } } array_unshift($param, $typestr); } function apply_query_filters($display_params) { $sql_filters = array("depth" => "(depth >= ?)", "snps" => "(snps >= ?)", "tagid" => "(tag_index.tag_id = ?)", "delv" => "(tag_index.deleveraged = ?)", "black" => "(tag_index.blacklisted = ?)", "rem" => "(tag_index.removed = ?)"); $filters = $display_params['filter_type']; $query = ""; if (count($filters) > 0) { $query = " AND "; while (count($filters) > 0) { $filter = array_shift($filters); $query .= $sql_filters[$filter]; $query .= count($filters) > 0 ? " AND " : ""; } } return $query; } ?> stacks-1.44/php/version.php000644 000765 000024 00000000000 12335173442 016461 0ustar00catchenstaff000000 000000 stacks-1.44/php/view_sequence.php000644 000765 000024 00000003772 12663652222 017663 0ustar00catchenstaff000000 000000 // // This file is part of Stacks. // // Stacks is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // Stacks is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with Stacks. If not, see . // require_once("header.php"); $database = isset($_GET['db']) ? $_GET['db'] : ""; $seq_id = isset($_GET['id']) ? $_GET['id'] : 0; // Connect to the database $db = db_connect($database); // Save these variables for automatic URL formation later on. $display = array(); $display['db'] = $database; $display['seq_id'] = $seq_id; // // Prepare some SQL queries // $query = "SELECT id, catalog_id, seq_id, type, seq FROM sequence " . "WHERE id=?"; if (!($db['seq_sth'] = $db['dbh']->prepare($query))) write_db_error($db['dbh'], __FILE__, __LINE__); $page_title = "Catalog RAD-Tag Sequence Viewer"; write_compact_header($page_title); if (!$db['seq_sth']->bind_param("i", $seq_id)) write_db_error($db['seq_sth'], __FILE__, __LINE__); if (!$db['seq_sth']->execute()) write_db_error($db['seq_sth'], __FILE__, __LINE__); $res = $db['seq_sth']->get_result(); $row = $res->fetch_assoc(); // // Line wrap the sequence to 80 characters // $full_seq = $row['seq']; $seq = ""; do { $seq .= substr($full_seq, 0, 80) . "\n"; $full_seq = substr($full_seq, 80); } while (strlen($full_seq) > 80); if (strlen($seq) > 0) { $seq .= $full_seq . "\n"; } echo <<< EOQ 
>$row[catalog_id]|$row[seq_id]|$row[type]
$seq
EOQ; write_compact_footer(); ?> stacks-1.44/php/images/caret-d.png000644 000765 000024 00000000241 12335173442 017564 0ustar00catchenstaff000000 000000 ‰PNG  IHDR ‰^ëPLTEŔŔŔ+ס¦tRNS@ćŘfbKGDH pHYs  šśtIMEÔ *Ť.IDATxÚc`€‚c 9  0{¤ŃÜIEND®B`‚stacks-1.44/php/images/caret-u.png000644 000765 000024 00000000245 12335173442 017611 0ustar00catchenstaff000000 000000 ‰PNG  IHDR ‰^ëPLTEŔŔŔ+ס¦tRNS@ćŘfbKGDH pHYs  šśtIMEÔ 7°IĹľIDATxÚc`  6˛ ŠÝq@ĆIEND®B`‚stacks-1.44/php/images/excel_icon.png000644 000765 000024 00000005057 12335173442 020367 0ustar00catchenstaff000000 000000 ‰PNG  IHDR##ŮłYbKGD˙˙˙ ˝§“ pHYs  šśtIME× 7@OŇr ĽIDATXĂĄßo×uÇ?gfvůcµ¤–¤HI¤dÉrÇŽk'vóqP4nP }Hß źŇ×ţy Š~I˘/y)ú‰Ť"vë$hë4±#)ŞYµeŠ”ř›Ë]îěěĚ˝çś<Ě’–[,Đ{wîĚ=ß{î9ßď9+<ŕS„pŁČ«Ç1Ç\@ŔÁqđń w9š÷z1 RÔ•‰FÂüĚěË"ňĄ˙mOćÎĆŽŻov™iMăÔ`ÜkŁő·C¸oÎę[®xâ&¬Ýۦ‘&|ć©G9·Řą!"OÜo/{Ť°ĽŘáâůţżź*DŠ Z§:ĽöúŻxú“ů„»Ż‹ČĘŃ3ÉČŇ&YÚř?ňąŢ›sw°’ĄĹ9.?ň(ŻľvťŐ»÷–Ý}ëŘ^Qćß{{ďíŻ»¸ 抚 7Ö~‚şq{ŻŤ† S#şb1m€žxäI–/aÉîMîe¸*g—ćč ăĹ˝Á—>Ď⑇˛Á¨{ů?o˝Śdŕbh„*B$†HYUŚĘ‚ŞŞ@Ą!T”Şěďmň_ýK–/!Ś#WNđˇ$”––Î1>Á÷_yťŻ>—.»űĎłR IŚg.=ÇĹÎĚ sÇĚ0Ź¨jŠąÖ÷UA„»Ý5ţöď˙ŠT÷í˙ĂĐÔG–\ó&&ĺ(Ę„ď|÷{|ëĎţčé,3 ”śj´čLÍ˙öć¨8ę’qîn)†#Bíˇß‚âdiJ§3Ëë˙ýEŁ˘˘(úäŃYŰđŹ/˝JŤEúÚăżúEČ ĂCȦůܥϱ4łČk7˙‹_Ż]Ă$ˇ7čń…Ç~źÎ\‹Ę+d'uĽřŤ ¤©đé§>•+Ń hL¨0Äśď˙ř—ô7o‘™+#->bkwĂô€‰é„őŤ›Ľłs‹‡fâćć› 'şĽ»·Áúť >qůSt¬Eĺ#+şy  VGĚPu#$ ‚] ’˛|zšÎlF¬O#L„DbRġÓÖ6_yęËLÇiĎg?ţY´XJ ÁEV°s‰š*ÂÝť’‘á ¸ëŰV·T@P#F0«7’ …Z±Ô>ÇţΗyéęż “ÎŇÂ9vĽ»µCřĆłĎóµO}…Ťa2¤ôśąSÂ|«ćÎĽt5rľ3u|Dy?°¸0ÉB«ć«…X)ćĚÁjj'“ü=-Ă!ĂĆ T”üŰŐĺď^ţ’i§“µŮX»ÇŹw^ĺ÷>˙,çfÎĐŰÜĂGMFąŃËűé´ŚąS1=¬H‡ĎLÓa7çÂŇi;Í:͵˘,)ę` î‚»ąQUˇ˘ĐŰű{,Ř"Í<wÎÍ?DĎúlŻmł±±ÁÝ·îRU ß}ç˙ü¤É;=a»VĘa98]ĎQU˘w”ĐŁ˝-%Ü޲˛0‹¨â&§µwÝÉđZQĹRľţ…?ĺkĎţ jŠş‚;ŁŞäÚęUÖWß可8ć SÍIN/LSV¬Ä1sĺ“üîG?Ť&V‘&SܸűSB5`öT“ĄŮ&ĐdŁQPV9WVÚ$€»ű9—ĎťfeaÜ­´Ž-Ô8ŽťĚͧŠężö27WŻC@ÝP‹¸+QKŞÇŢrÔ¸ŇË»ôóśµí’›wŽ*úĂŔÁ P ·°‰ćÜŮ.Ůííđë‰ZFîu ^™ŻµĎę Vw !›hNĄN_ac—ťŤ­Z(Ôů°‚I3ˇŠJQF’ÔŔ °Ë­'Y^8Ëâü$íĐ$;(¨bÁ…łł`Ž›±×_ăĚ™ÎÎLÄ D µ7'Ő1š]\ĽôĚéI>ęýü­Ű÷ž>öĐy†QXŰîsńĚ Ť„Ű÷z4&&8ÝĘhˇtúŇäüÜ$âu…Ľ?\bţt‹ł­Ś BwwH©J‘¸¦8öáegkrö™[ďl:‰09Ů&±ş:Xëe$T•Ý|‚ĐŤt6¨TĆA>˘} Š21ŃdżčÍOröL‹…ą"ŽF;ŽĂ1—×ŔÁFĄ®,N5IÓ&¸Pĺ‘Mp~©Á0í|Äěéŕ,´ÚÜŢęÓH„ÖÜ4Żţâ6Ď=}‰ůN›JV& ć5rńt\ ívFł‘ŕ“Ýś)ŚĺÎ4ý‘ňúz—ĆäűýCś&íÖ­Fdf¶A‘Ź(JEµö ž .;jâú`0>ć\0Ś^ąţö>B$lu#•¶ö†€ÉđÔ<˙quÄŐ«|ü‘”U]ëQćXZgGż3 > Ô©‡Ő/%Kť6Ť´vkUö’qiq†~ĄlíäxtÜ$­yDµ¤._ÝQ˘I€Łc@âśä™ZěÔ Şbdj"ˇŮ„,5$qfŰ“dAŮÜŃŻ mÎAáÎ”Ô ›ŠN‹s Fař¦áObô1AÝv[HÜ›î`ć•!§‘ŠTĺnH}ą™Nnţ­Ď,ŮŘeEIEND®B`‚stacks-1.44/php/images/l-arrow-disabled.png000644 000765 000024 00000011364 12335173442 021405 0ustar00catchenstaff000000 000000 ‰PNG  IHDR"ŔN' pHYs  šśbiCCPPhotoshop ICC profilexÚ­Wy8Ôí×?ß1Ć>ö±3–,ŮĆľed'; ĂXĆ:Ť} zDIŠH‘˘$”˝BĄ¨DYž¬©ČҢy˙ĐÓs]ż÷zß÷ź÷üő9ç:ç|Î}źűşÎ}¸ITj "ŁbiλÍđž^Ţx¦×Ŕś€p‘c¨¦NNvđ?Ęúk@^*“¨ÔNĄ3“Ĺ ¦ľńmáÚ‹uťżŕ,ÍÓËQľml|ŰŘřb©±H(đ†’ČH*(Ń\ť‰H5`C¶q`¶q/`ăCb O™Ŕ´€1"Ĺ`•€LŽ ŚŔž@©DFF“¸:@>J‹ŕšeO/oüvÉä }4ćíż¶0I€š;˘’˙ÚäÚIu;˙µ­:îyL°†: ěfŚtúŞ,ÓI€­:ýGťľu €a  #"0Ž˙űľ¤ŕ˙Ň·Ďü[J•ÁĘxSĚTÇÜĹ2Âú‰m…+Ŕ)ÉĄĚ­ĎcĘkÇçĘďó¤Q…cDbEăÄĹS$2đ™’™RG¤ŹĘßqFöĽ\ą|µÂÍťMŠ=JCĘ“*ëV5)u 7Í­ríÇ:›z ú$ó»ĆŤ$Ś)&wÍ؉űĚo[ňě¦Y Ú¨Úłűâ`çX»‡Ó9Âĺ±›{„G›ÖŰooµĎW_˘ßQ˙ţÎ@GňÉ ç!¬ˇ¦”¸°ÚđÉHŢ(óčęŤýocб„8żřĽ„ű‰óÉÜ)F©Ôei=écSg-^Î\ÉZĎŢú ťĂ~”;—?OěüqŤ|¦V'÷śr?í}Ć÷l`!Ą(Ľ8üĺ<Ą$¬4ô©Ě»Üĺ˘m…É%µË •rU;®HU㯊]şÎ»†«–ý&ăÍ_u›ő«·oĎ6ŚŢl|z·»©ąąľĺĆ˝«÷/¶žm;Ţ~´#łó`Wr7íAÔèG=¤^b®oęń­'éOmúúçž5<Ďp”\|Ńń˛ŕ•ßňĐ÷×O† GHo”ß|{űh4Ě}\l|f˘ćoÚ;íwß&[¦R¦ő§×ggSçLß3Ľď›Ďűŕđ‘ëăËO… ž‹‚‹#KEź=——‡ľ¬8¬r¬ö}Í\3]ŁŻ·m$ojm.«ůţCáÇÜĎŠ-˙_żŢŇ÷Óé(4ÇȇágâdfcafŮbýĆö™}†ă ¶źł“«ž»‚ç4o_8nź€Ł QHOX[DITULE\[Âo!i,Ą+­,Ł´CIv§A^I°S[‘¨dŁě©BQM"䨕Ş7j<×\ŇćÖŃÔuŃKÖżb0lČl¤eĽß¤Ćtž¨hnqÓrÓĘÂ:×fÄNÍ>ŐaĐIdOsŁ+âfď~ŇcĚKÁ›ş÷–φŻ®ß˙vŇV 9>čnđj¨ %,ěZřÇHŮ(źč"ękoŚklAÜăćD»¤ăÉS¦S?řś¶–ţ-ăűÁ_‡3Ůłxł…Žý%“C8Şźk–g}Ěń¸gľĎ ˙ňÉ°Sѧ©ghgŁ ŁŠ¨Ĺ‘ç‚Ďű—x–:]°(Ó.Wľ¨\ˇtIá˛BĄ|•ě™jé«2פŻKÜ­®ş)XÇ_Ďs‹ó6göG#Ç]î&\ł`‹ŕ=‘űřVą6•v­˝Nł.‡nď{î~„´ŇÓ×[ŇůŘô‰Ŕ“ŧű‹źE>' ¬öż¨x™đĘaHjhíußpÉHôÓ·ÜoÇGoŚĄŽ[MpOŚţ]ń.lReryŞiúÄLč¬ńśŔÜâűóŢ>Z|úôqáţⱥ}źU>˙XîůrrĹwUauĺ뽵ěu§ ÁŤ±ÍŠo”ď„ď›?î˙Ěزř…ýőśąÝÔĂ2ú+ăgĚ"Ó"ó"Ëë(Ű+ö'­ŘZÎ2®\î2Żź“7(p]0SČG Â(2,Z)'n!Á)1ŽŻ”Ś”Ň—ĆHż–ąş#MÖMNIžYţ˝BĎÎjĹcJ±Ę{UĚTU˘j,jęłšZuÚĄ:9şńz$};]»đ†ĂeŁ!ă;&§LăÍ(DsK ]KŮÝ‚VŚVëÖlŢŘöÚ5ÚW99ć8%î qvw±s5pSr—đŔyrxˇĽľzŘ;ĺóvß+ß~żn˙6Rk@Kŕ-ňĄ ś`Z_¨E5L8ś)|-b&r0Ş-ş–Z±˙,íŻÄŘĐ8ßx‡ŁDŐ$ÉdľtĘję»ĎŇÚŇŻgś>~(ř°c¦f.ÉŢ8˛ř×tÎŰŁą=y-ÇnŻÎ/9‘[~’zŠ|ÚíŚĺYÍB™"ţbĆâŻçfĎŹ”ô—¶]¸QVZ~ębVEęĄčˤJŹŞ=W¬« ®J]kc×ďß(ŞˇŐ:ŢT©ĂÖ-Ő÷ßşz;·!úŽKŁÖ]ÜÝŤ¦±ćÖ–Ň{i÷I­Ä¶íŚíÓÝťU]™Ý¤†…®?č©ěÍí+yÜůdę)ý™đs˝ďÁÔ%/{_-ľvÉ}ót”wĚĽůoŢwé“_§Îbç®Ě;}äţôz±öóŮ/«Ąkősßń?ýĄŃéŰłŁ Pŕ9 ŕ\ ż Ď€«pâpŐ¤+ëY€ĺ˙3?ĚŔ â  Ćŕ  —ŕĚ"¬:âü…4!Pb(7ÔqÔcf+†\†A´0šŚ®GÓ/1~Ă8a®1ˇ™HL̒̇™çYěYęY…Y3YŘ<ŘşŘŐŮK98â8¦°ŽŘVNEÎ".&®®In'îVEž"^FŢ8ŢY>ľN~9Ž—…ű&!đNĐKđąĐnˇá]ÂŤ"Ú"-˘F˘]bVbâžâł4 :>OR\ň¶”µÔ¤tŠ źLý§+˛…rFrÓňy z s;Ď(š+®*U)űŞŞ ¨ć,Ô@­U=]ĂL“AłK+KŰNG\Ą»¤7¬ßep{WĄáiŁ,ă“0Ó@ł}Ds+ 3KăÝ&V&Ö&6»mťěĽí¨Ž‡ť ÷Ô;÷ąLş!îröž ^ŢĂ>śűô}ýNűw‘VĺČľA…ÁýˇĘî°śđţH®(÷čręgšiLněßńš 9‰cÉĘ)GRçÓĚÓK3~r?Üś%‘ťvämŽńŃ;y>ÇEó‡ ŠOś‘>ű±¨ö\Z‰ý޲ɋŤ— *ÉW¬®Ş\ç»ńŁv©îý­™†‰Ćů¦µ{ě­ŇíťŃÝéË{Zú&ž2=# xľ8ńŞwó†8š;>öNv*ufčýÎ>M/Ů,_]e[‹Ůůnůó>ťľý“VŕĐ;‚pşáÂ!‘H2„â@Yˇ˛QŹX &ŃŠčdtŁ c(c ‹ˇ`zr>1;17°łaůÂęÍÚ˦ĹVÁÎĂ~}ť#„ă-ÖŰÉ©ĹyŤKšë7÷î_<ń<+ĽŢIľ˝|Żůřă,qť†M‚:‚w…4„j…Ő„oŠ¨‰4‰‰v‹ŮŠ ‰ű‰/I¤âYđE’;%;¤Ü¤¤łeÄešw¸ďX—-”3•ĎPSxľ3QŻŘ«Ł,©üJ%SUCu†P¨fĄ¶ĄŢ¨Ą)Ł9˘uBŰ^§ł®;Ą÷DżŃŕĘ®s†9FÉĆ&~¦f{Vć†:–š»5­´­µl m-íśě˝‚“śňöT9·» »n¸‹yyFzy?őaÜGđőđ;âß@š ä#[ĄׇĚQ¤ĂHáeQ˘Ń~ÔĘýK1ę±Iq°‰îI%Éł©Ú˛Ó^dHŚ8ԜɑEĘ®=ň=Çţč­<ëc[ůŐľ§„N?={¸H§xí|ciT™|ů|EÝĺ„*“jŢ«ďŻßŻ9s3ľ>ŕ¶íí»jÍ*÷4[ Ű=;ÉݦŐz„űŘo=Ý|¶4đáĹ×W?‡ŮŢŹjŽ{üť2Y1ýr=Żů1|ˇaiă‹áęÁµ› ßi?_ýé?€,č#„B&TB,"8„Ä UČJĺ†*DM0Č2Ä2Ż~­]Ë[§lmrmN|»ń=ů‡őOźă[Ążşét€í} X‰ŃŃ4ĽŃţ%2"î.`§ÄZ» Ó,ťŔ f˘€‘ Š±půŤu)–Ö۱ 5ÖÉďäPWŔ ´0’ŤÓośá`8ä89ČÜâwěeZśłČ áѶÎŔ€ŚEąýĂ5ďňŹ˙:™dn ˘(Lr(ŃaŰ%v@sŔC DCD (Đ@DŢX A$"€ű!(@† ůq@K  B ”3üwwPţ !ÚŹry+8ľ(:Iß=”pťđđđż˝m˙0AÔźLŰě˙čhY´:Z m†6DˇőŹĆˇ…A­‰ÖE›˘ŤŃh-´Ţł…» ˛˙T€˙“qhňµ*C0€ń1s@H?ʡă`{WŔp”ř´o¦dü绊 JŚ FS“h”ĐXĽ)•„'FGRăbhJxë¨@%Ľ: đ_şLTÄl°Ă­gAMAŘëőŞ cHRMmsŽâů†™xÔN3Ľycá,üIDATxÚĽ–ËNë:†/ۉe‚„A…T €÷đP0A€ •‹PAĄ©/gŔq¶Óö@Ďfk[Š’¦ńú–˙u±™÷Ţă/ö·´ę‡Ţ{8çşßÎ9üŽŻ´*Č9"Âóó3ŽŽŽđńń±ŕÄ*C| F‰777¨Ş ăńmŰ"I0Ćŕ˝cěgŔ ÷óós\^^˘( @۶0Ć€čS¤pď%Č'Äw°étŠłł3<>>bggRJXkaŚµv),9çŔëŢ-CĽ`<ăřřmŰbww·“NOO!Ąő Ć@ď=ˇµ†÷DÔĆ°Ńh„““$I‚˘(zńÜŢŢĆl6ű2K‰MÓŕíí J©.Öb>9Ľ÷¨ëUU!ĎsdYÖIľ%"¤iúerpΡ”Âl61śs0Ć~ťs°ÖâââWWW(ŠI’ô`1t•rňŢw@Ć8çáĆŞŞÂőő5c˙»Ć– cL*ďýgá{ďaŚAžçRâýý}ĺşúnXka­íVLáÁŤŤ ŕőőOOOťî?ó-âT¶ÖBkŤápÉd‚¦iŔ[¨3"çDô嵬‰‹Ť(Ą°··‡ş®q{{‹˛,Á9‡sśsŚFŁŮ獦Ó)’$éfŚAA$IşŚ xxx@]×(ËJ)Ŕd2Áúú:Ň4íÉl !¦)´ÖȲ¬§’^¤” -Ş,KH)qŹ­­-äy"‚ÖZë^k '8ž$ „ťr]ÎÇ)L €R MÓt5VB1?G)%¤”BüZa¬y<)–:X[[ĂÝÝ^^^0ˇµţ,ćČ`Ë9ď]˝ÎÇ`8ěűűűŘÜÜD–ePJAŃĆÇwöţëLş1¦ŰűŚ1ź ř_@€…Ěďń.ŇÝż:DhčˇcÄqŠëńÇGŚXŠ7ÔX˛?zL í/nôóŮĽęřgŤĆ3p©­ĆIEND®B`‚stacks-1.44/php/images/l-arrow.png000644 000765 000024 00000011576 12335173442 017645 0ustar00catchenstaff000000 000000 ‰PNG  IHDR­˘Ůp pHYs  šśbiCCPPhotoshop ICC profilexÚ­Wy8Ôí×?ß1Ć>ö±3–,ŮĆľed'; ĂXĆ:Ť} zDIŠH‘˘$”˝BĄ¨DYž¬©ČҢy˙ĐÓs]ż÷zß÷ź÷üő9ç:ç|Î}źűşÎ}¸ITj "ŁbiλÍđž^Ţx¦×Ŕś€p‘c¨¦NNvđ?Ęúk@^*“¨ÔNĄ3“Ĺ ¦ľńmáÚ‹uťżŕ,ÍÓËQľml|ŰŘřb©±H(đ†’ČH*(Ń\ť‰H5`C¶q`¶q/`ăCb O™Ŕ´€1"Ĺ`•€LŽ ŚŔž@©DFF“¸:@>J‹ŕšeO/oüvÉä }4ćíż¶0I€š;˘’˙ÚäÚIu;˙µ­:îyL°†: ěfŚtúŞ,ÓI€­:ýGťľu €a  #"0Ž˙űľ¤ŕ˙Ň·Ďü[J•ÁĘxSĚTÇÜĹ2Âú‰m…+Ŕ)ÉĄĚ­ĎcĘkÇçĘďó¤Q…cDbEăÄĹS$2đ™’™RG¤ŹĘßqFöĽ\ą|µÂÍťMŠ=JCĘ“*ëV5)u 7Í­ríÇ:›z ú$ó»ĆŤ$Ś)&wÍ؉űĚo[ňě¦Y Ú¨Úłűâ`çX»‡Ó9Âĺ±›{„G›ÖŰooµĎW_˘ßQ˙ţÎ@GňÉ ç!¬ˇ¦”¸°ÚđÉHŢ(óčęŤýocб„8żřĽ„ű‰óÉÜ)F©Ôei=écSg-^Î\ÉZĎŢú ťĂ~”;—?OěüqŤ|¦V'÷śr?í}Ć÷l`!Ą(Ľ8üĺ<Ą$¬4ô©Ě»Üĺ˘m…É%µË •rU;®HU㯊]şÎ»†«–ý&ăÍ_u›ő«·oĎ6ŚŢl|z·»©ąąľĺĆ˝«÷/¶žm;Ţ~´#łó`Wr7íAÔèG=¤^b®oęń­'éOmúúçž5<Ďp”\|Ńń˛ŕ•ßňĐ÷×O† GHo”ß|{űh4Ě}\l|f˘ćoÚ;íwß&[¦R¦ő§×ggSçLß3Ľď›Ďűŕđ‘ëăËO… ž‹‚‹#KEź=——‡ľ¬8¬r¬ö}Í\3]ŁŻ·m$ojm.«ůţCáÇÜĎŠ-˙_żŢŇ÷Óé(4ÇȇágâdfcafŮbýĆö™}†ă ¶źł“«ž»‚ç4o_8nź€Ł QHOX[DITULE\[Âo!i,Ą+­,Ł´CIv§A^I°S[‘¨dŁě©BQM"䨕Ş7j<×\ŇćÖŃÔuŃKÖżb0lČl¤eĽß¤Ćtž¨hnqÓrÓĘÂ:×fÄNÍ>ŐaĐIdOsŁ+âfď~ŇcĚKÁ›ş÷–φŻ®ß˙vŇV 9>čnđj¨ %,ěZřÇHŮ(źč"ękoŚklAÜăćD»¤ăÉS¦S?řś¶–ţ-ăűÁ_‡3Ůłxł…Žý%“C8Şźk–g}Ěń¸gľĎ ˙ňÉ°Sѧ©ghgŁ ŁŠ¨Ĺ‘ç‚Ďű—x–:]°(Ó.Wľ¨\ˇtIá˛BĄ|•ě™jé«2פŻKÜ­®ş)XÇ_Ďs‹ó6göG#Ç]î&\ł`‹ŕ=‘űřVą6•v­˝Nł.‡nď{î~„´ŇÓ×[ŇůŘô‰Ŕ“ŧű‹źE>' ¬öż¨x™đĘaHjhíußpÉHôÓ·ÜoÇGoŚĄŽ[MpOŚţ]ń.lReryŞiúÄLč¬ńśŔÜâűóŢ>Z|úôqáţⱥ}źU>˙XîůrrĹwUauĺ뽵ěu§ ÁŤ±ÍŠo”ď„ď›?î˙Ěزř…ýőśąÝÔĂ2ú+ăgĚ"Ó"ó"Ëë(Ű+ö'­ŘZÎ2®\î2Żź“7(p]0SČG Â(2,Z)'n!Á)1ŽŻ”Ś”Ň—ĆHż–ąş#MÖMNIžYţ˝BĎÎjĹcJ±Ę{UĚTU˘j,jęłšZuÚĄ:9şńz$};]»đ†ĂeŁ!ă;&§LăÍ(DsK ]KŮÝ‚VŚVëÖlŢŘöÚ5ÚW99ć8%î qvw±s5pSr—đŔyrxˇĽľzŘ;ĺóvß+ß~żn˙6Rk@Kŕ-ňĄ ś`Z_¨E5L8ś)|-b&r0Ş-ş–Z±˙,íŻÄŘĐ8ßx‡ŁDŐ$ÉdľtĘję»ĎŇÚŇŻgś>~(ř°c¦f.ÉŢ8˛ř×tÎŰŁą=y-ÇnŻÎ/9‘[~’zŠ|ÚíŚĺYÍB™"ţbĆâŻçfĎŹ”ô—¶]¸QVZ~ębVEęĄčˤJŹŞ=W¬« ®J]kc×ďß(ŞˇŐ:ŢT©ĂÖ-Ő÷ßşz;·!úŽKŁÖ]ÜÝŤ¦±ćÖ–Ň{i÷I­Ä¶íŚíÓÝťU]™Ý¤†…®?č©ěÍí+yÜůdę)ý™đs˝ďÁÔ%/{_-ľvÉ}ót”wĚĽůoŢwé“_§Îbç®Ě;}äţôz±öóŮ/«Ąkősßń?ýĄŃéŰłŁ Pŕ9 ŕ\ ż Ď€«pâpŐ¤+ëY€ĺ˙3?ĚŔ â  Ćŕ  —ŕĚ"¬:âü…4!Pb(7ÔqÔcf+†\†A´0šŚ®GÓ/1~Ă8a®1ˇ™HL̒̇™çYěYęY…Y3YŘ<ŘşŘŐŮK98â8¦°ŽŘVNEÎ".&®®In'îVEž"^FŢ8ŢY>ľN~9Ž—…ű&!đNĐKđąĐnˇá]ÂŤ"Ú"-˘F˘]bVbâžâł4 :>OR\ň¶”µÔ¤tŠ źLý§+˛…rFrÓňy z s;Ď(š+®*U)űŞŞ ¨ć,Ô@­U=]ĂL“AłK+KŰNG\Ą»¤7¬ßep{WĄáiŁ,ă“0Ó@ł}Ds+ 3KăÝ&V&Ö&6»mťěĽí¨Ž‡ť ÷Ô;÷ąLş!îröž ^ŢĂ>śűô}ýNűw‘VĺČľA…ÁýˇĘî°śđţH®(÷čręgšiLněßńš 9‰cÉĘ)GRçÓĚÓK3~r?Üś%‘ťvämŽńŃ;y>ÇEó‡ ŠOś‘>ű±¨ö\Z‰ý޲ɋŤ— *ÉW¬®Ş\ç»ńŁv©îý­™†‰Ćů¦µ{ě­ŇíťŃÝéË{Zú&ž2=# xľ8ńŞwó†8š;>öNv*ufčýÎ>M/Ů,_]e[‹Ůůnůó>ťľý“VŕĐ;‚pşáÂ!‘H2„â@Yˇ˛QŹX &ŃŠčdtŁ c(c ‹ˇ`zr>1;17°łaůÂęÍÚ˦ĹVÁÎĂ~}ť#„ă-ÖŰÉ©ĹyŤKšë7÷î_<ń<+ĽŢIľ˝|Żůřă,qť†M‚:‚w…4„j…Ő„oŠ¨‰4‰‰v‹ŮŠ ‰ű‰/I¤âYđE’;%;¤Ü¤¤łeÄešw¸ďX—-”3•ĎPSxľ3QŻŘ«Ł,©üJ%SUCu†P¨fĄ¶ĄŢ¨Ą)Ł9˘uBŰ^§ł®;Ą÷DżŃŕĘ®s†9FÉĆ&~¦f{Vć†:–š»5­´­µl m-íśě˝‚“śňöT9·» »n¸‹yyFzy?őaÜGđőđ;âß@š ä#[ĄׇĚQ¤ĂHáeQ˘Ń~ÔĘýK1ę±Iq°‰îI%Éł©Ú˛Ó^dHŚ8ԜɑEĘ®=ň=Çţč­<ëc[ůŐľ§„N?={¸H§xí|ciT™|ů|EÝĺ„*“jŢ«ďŻßŻ9s3ľ>ŕ¶íí»jÍ*÷4[ Ű=;ÉݦŐz„űŘo=Ý|¶4đáĹ×W?‡ŮŢŹjŽ{üť2Y1ýr=Żů1|ˇaiă‹áęÁµ› ßi?_ýé?€,č#„B&TB,"8„Ä UČJĺ†*DM0Č2Ä2Ż~­]Ë[§lmrmN|»ń=ů‡őOźă[Ążşét€í} X‰ŃŃ4ĽŃţ%2"î.`§ÄZ» Ó,ťŔ f˘€‘ Š±půŤu)–Ö۱ 5ÖÉďäPWŔ ´0’ŤÓośá`8ä89ČÜâwěeZśłČ áѶÎŔ€ŚEąýĂ5ďňŹ˙:™dn ˘(Lr(ŃaŰ%v@sŔC DCD (Đ@DŢX A$"€ű!(@† ůq@K  B ”3üwwPţ !ÚŹry+8ľ(:Iß=”pťđđđż˝m˙0AÔźLŰě˙čhY´:Z m†6DˇőŹĆˇ…A­‰ÖE›˘ŤŃh-´Ţł…» ˛˙T€˙“qhňµ*C0€ń1s@H?ʡă`{WŔp”ř´o¦dü绊 JŚ FS“h”ĐXĽ)•„'FGRăbhJxë¨@%Ľ: đ_şLTÄl°Ă­gAMAŘëőŞ cHRMmsŽâů†™xÔN3Ľycá,†IDATxÚ¬•=K3KÇw^7»y3qQ—@°-ěŚoů©ěŇę±llL-Ö‚…`Qµ *ŘX(""˘Ĺ8fwggžâä.ą>7ŢÇË=Ĺ°,3ż=çĚ˙i­Ť˙;đźlúé‡ń÷,Ą””!d†”R)ő'č UJEQ†!Ąôęęjiiéĺĺ^ţ+—ö"J)Ă0äś×ëőőőő»»;Ďó˛Ů,Bc ą˙Ş” ĂPJ)ĄÜŘŘŘÚÚšžžÖZ7›MÇqc†aŔjлׯи‰Q5›ÍµµµóóóąąąL&ăűţÇÇG»ÝţB„€ŽÇE ¸AZë(Š‚ 0 ăööveeĹóĽR©¤µfŚضťN§)Ą„¸Š˘hyyąP(†Aˇ1QJB¨Ńh¬®®öőőMMMI)µÖAĚÎζZ­(Š~Wçüäääţţ~``§ÝעµŢÝÝ­ŐjŁŁŁĹb˛ Ą4—Ëý㵦™Ďç[­–‚sNˇP5¤ąąąą˝˝]*•rą\LŚą˝”â(BČ4Íc\«Ővvv*• B( ĂYHk-„đ}ß4MĄ¨bdd$ťN?<<Bţß}ß‚¬AFźźźăăă‹‹‹777———śóďĺý{„a›Çr ‚ŔuÝjµúôôtzzŠ1ŽSUqÎYŹĐZw'AAYŚ±D"Ásgaaˇ^Żďďď—ËĺD"!Ą4MłŃh<>>Ć'Á{}}Ífł”RĐ/…Ś8ç¶mgłŮ ÚívĄR9;;ŰŰŰ›źźw!ôüü<<<śĎçŁ(4Ć1fŰv>źw]·X,B(Ą!Šh2™TJufĆĺr9“ÉMLLŚŤŤ1Ɔ††\×…$ŕ!l–ËĺR©”iš‚ťácL)eŚMNN:Žs||Ün· !ýýý®ëš¦iš&pa›mŰÉd2“ÉX–Ĺ9Çwl·c uĹ …ÂáááőőuµZu]ײ,˲cGRĘ9OüN·OŔŻľď !<Ď{{{ó<xqq133388hŰv7´»25„ôĹ|ńdBĽżż !„RĘd2I)…JA'q÷aí ˝x tT} „X–•JĄŔÇśs˲€H)í>ňE§¨×€)żĄ4.ó{żˇoţb`ä(Š”RÝ„’ż‰_ăÎýŠ.’IEND®B`‚stacks-1.44/php/images/r-arrow-disabled.png000644 000765 000024 00000011405 12335173442 021407 0ustar00catchenstaff000000 000000 ‰PNG  IHDR"ŔN' pHYs  šśbiCCPPhotoshop ICC profilexÚ­Wy8Ôí×?ß1Ć>ö±3–,ŮĆľed'; ĂXĆ:Ť} zDIŠH‘˘$”˝BĄ¨DYž¬©ČҢy˙ĐÓs]ż÷zß÷ź÷üő9ç:ç|Î}źűşÎ}¸ITj "ŁbiλÍđž^Ţx¦×Ŕś€p‘c¨¦NNvđ?Ęúk@^*“¨ÔNĄ3“Ĺ ¦ľńmáÚ‹uťżŕ,ÍÓËQľml|ŰŘřb©±H(đ†’ČH*(Ń\ť‰H5`C¶q`¶q/`ăCb O™Ŕ´€1"Ĺ`•€LŽ ŚŔž@©DFF“¸:@>J‹ŕšeO/oüvÉä }4ćíż¶0I€š;˘’˙ÚäÚIu;˙µ­:îyL°†: ěfŚtúŞ,ÓI€­:ýGťľu €a  #"0Ž˙űľ¤ŕ˙Ň·Ďü[J•ÁĘxSĚTÇÜĹ2Âú‰m…+Ŕ)ÉĄĚ­ĎcĘkÇçĘďó¤Q…cDbEăÄĹS$2đ™’™RG¤ŹĘßqFöĽ\ą|µÂÍťMŠ=JCĘ“*ëV5)u 7Í­ríÇ:›z ú$ó»ĆŤ$Ś)&wÍ؉űĚo[ňě¦Y Ú¨Úłűâ`çX»‡Ó9Âĺ±›{„G›ÖŰooµĎW_˘ßQ˙ţÎ@GňÉ ç!¬ˇ¦”¸°ÚđÉHŢ(óčęŤýocб„8żřĽ„ű‰óÉÜ)F©Ôei=écSg-^Î\ÉZĎŢú ťĂ~”;—?OěüqŤ|¦V'÷śr?í}Ć÷l`!Ą(Ľ8üĺ<Ą$¬4ô©Ě»Üĺ˘m…É%µË •rU;®HU㯊]şÎ»†«–ý&ăÍ_u›ő«·oĎ6ŚŢl|z·»©ąąľĺĆ˝«÷/¶žm;Ţ~´#łó`Wr7íAÔèG=¤^b®oęń­'éOmúúçž5<Ďp”\|Ńń˛ŕ•ßňĐ÷×O† GHo”ß|{űh4Ě}\l|f˘ćoÚ;íwß&[¦R¦ő§×ggSçLß3Ľď›Ďűŕđ‘ëăËO… ž‹‚‹#KEź=——‡ľ¬8¬r¬ö}Í\3]ŁŻ·m$ojm.«ůţCáÇÜĎŠ-˙_żŢŇ÷Óé(4ÇȇágâdfcafŮbýĆö™}†ă ¶źł“«ž»‚ç4o_8nź€Ł QHOX[DITULE\[Âo!i,Ą+­,Ł´CIv§A^I°S[‘¨dŁě©BQM"䨕Ş7j<×\ŇćÖŃÔuŃKÖżb0lČl¤eĽß¤Ćtž¨hnqÓrÓĘÂ:×fÄNÍ>ŐaĐIdOsŁ+âfď~ŇcĚKÁ›ş÷–φŻ®ß˙vŇV 9>čnđj¨ %,ěZřÇHŮ(źč"ękoŚklAÜăćD»¤ăÉS¦S?řś¶–ţ-ăűÁ_‡3Ůłxł…Žý%“C8Şźk–g}Ěń¸gľĎ ˙ňÉ°Sѧ©ghgŁ ŁŠ¨Ĺ‘ç‚Ďű—x–:]°(Ó.Wľ¨\ˇtIá˛BĄ|•ě™jé«2פŻKÜ­®ş)XÇ_Ďs‹ó6göG#Ç]î&\ł`‹ŕ=‘űřVą6•v­˝Nł.‡nď{î~„´ŇÓ×[ŇůŘô‰Ŕ“ŧű‹źE>' ¬öż¨x™đĘaHjhíußpÉHôÓ·ÜoÇGoŚĄŽ[MpOŚţ]ń.lReryŞiúÄLč¬ńśŔÜâűóŢ>Z|úôqáţⱥ}źU>˙XîůrrĹwUauĺ뽵ěu§ ÁŤ±ÍŠo”ď„ď›?î˙Ěزř…ýőśąÝÔĂ2ú+ăgĚ"Ó"ó"Ëë(Ű+ö'­ŘZÎ2®\î2Żź“7(p]0SČG Â(2,Z)'n!Á)1ŽŻ”Ś”Ň—ĆHż–ąş#MÖMNIžYţ˝BĎÎjĹcJ±Ę{UĚTU˘j,jęłšZuÚĄ:9şńz$};]»đ†ĂeŁ!ă;&§LăÍ(DsK ]KŮÝ‚VŚVëÖlŢŘöÚ5ÚW99ć8%î qvw±s5pSr—đŔyrxˇĽľzŘ;ĺóvß+ß~żn˙6Rk@Kŕ-ňĄ ś`Z_¨E5L8ś)|-b&r0Ş-ş–Z±˙,íŻÄŘĐ8ßx‡ŁDŐ$ÉdľtĘję»ĎŇÚŇŻgś>~(ř°c¦f.ÉŢ8˛ř×tÎŰŁą=y-ÇnŻÎ/9‘[~’zŠ|ÚíŚĺYÍB™"ţbĆâŻçfĎŹ”ô—¶]¸QVZ~ębVEęĄčˤJŹŞ=W¬« ®J]kc×ďß(ŞˇŐ:ŢT©ĂÖ-Ő÷ßşz;·!úŽKŁÖ]ÜÝŤ¦±ćÖ–Ň{i÷I­Ä¶íŚíÓÝťU]™Ý¤†…®?č©ěÍí+yÜůdę)ý™đs˝ďÁÔ%/{_-ľvÉ}ót”wĚĽůoŢwé“_§Îbç®Ě;}äţôz±öóŮ/«Ąkősßń?ýĄŃéŰłŁ Pŕ9 ŕ\ ż Ď€«pâpŐ¤+ëY€ĺ˙3?ĚŔ â  Ćŕ  —ŕĚ"¬:âü…4!Pb(7ÔqÔcf+†\†A´0šŚ®GÓ/1~Ă8a®1ˇ™HL̒̇™çYěYęY…Y3YŘ<ŘşŘŐŮK98â8¦°ŽŘVNEÎ".&®®In'îVEž"^FŢ8ŢY>ľN~9Ž—…ű&!đNĐKđąĐnˇá]ÂŤ"Ú"-˘F˘]bVbâžâł4 :>OR\ň¶”µÔ¤tŠ źLý§+˛…rFrÓňy z s;Ď(š+®*U)űŞŞ ¨ć,Ô@­U=]ĂL“AłK+KŰNG\Ą»¤7¬ßep{WĄáiŁ,ă“0Ó@ł}Ds+ 3KăÝ&V&Ö&6»mťěĽí¨Ž‡ť ÷Ô;÷ąLş!îröž ^ŢĂ>śűô}ýNűw‘VĺČľA…ÁýˇĘî°śđţH®(÷čręgšiLněßńš 9‰cÉĘ)GRçÓĚÓK3~r?Üś%‘ťvämŽńŃ;y>ÇEó‡ ŠOś‘>ű±¨ö\Z‰ý޲ɋŤ— *ÉW¬®Ş\ç»ńŁv©îý­™†‰Ćů¦µ{ě­ŇíťŃÝéË{Zú&ž2=# xľ8ńŞwó†8š;>öNv*ufčýÎ>M/Ů,_]e[‹Ůůnůó>ťľý“VŕĐ;‚pşáÂ!‘H2„â@Yˇ˛QŹX &ŃŠčdtŁ c(c ‹ˇ`zr>1;17°łaůÂęÍÚ˦ĹVÁÎĂ~}ť#„ă-ÖŰÉ©ĹyŤKšë7÷î_<ń<+ĽŢIľ˝|Żůřă,qť†M‚:‚w…4„j…Ő„oŠ¨‰4‰‰v‹ŮŠ ‰ű‰/I¤âYđE’;%;¤Ü¤¤łeÄešw¸ďX—-”3•ĎPSxľ3QŻŘ«Ł,©üJ%SUCu†P¨fĄ¶ĄŢ¨Ą)Ł9˘uBŰ^§ł®;Ą÷DżŃŕĘ®s†9FÉĆ&~¦f{Vć†:–š»5­´­µl m-íśě˝‚“śňöT9·» »n¸‹yyFzy?őaÜGđőđ;âß@š ä#[ĄׇĚQ¤ĂHáeQ˘Ń~ÔĘýK1ę±Iq°‰îI%Éł©Ú˛Ó^dHŚ8ԜɑEĘ®=ň=Çţč­<ëc[ůŐľ§„N?={¸H§xí|ciT™|ů|EÝĺ„*“jŢ«ďŻßŻ9s3ľ>ŕ¶íí»jÍ*÷4[ Ű=;ÉݦŐz„űŘo=Ý|¶4đáĹ×W?‡ŮŢŹjŽ{üť2Y1ýr=Żů1|ˇaiă‹áęÁµ› ßi?_ýé?€,č#„B&TB,"8„Ä UČJĺ†*DM0Č2Ä2Ż~­]Ë[§lmrmN|»ń=ů‡őOźă[Ążşét€í} X‰ŃŃ4ĽŃţ%2"î.`§ÄZ» Ó,ťŔ f˘€‘ Š±půŤu)–Ö۱ 5ÖÉďäPWŔ ´0’ŤÓośá`8ä89ČÜâwěeZśłČ áѶÎŔ€ŚEąýĂ5ďňŹ˙:™dn ˘(Lr(ŃaŰ%v@sŔC DCD (Đ@DŢX A$"€ű!(@† ůq@K  B ”3üwwPţ !ÚŹry+8ľ(:Iß=”pťđđđż˝m˙0AÔźLŰě˙čhY´:Z m†6DˇőŹĆˇ…A­‰ÖE›˘ŤŃh-´Ţł…» ˛˙T€˙“qhňµ*C0€ń1s@H?ʡă`{WŔp”ř´o¦dü绊 JŚ FS“h”ĐXĽ)•„'FGRăbhJxë¨@%Ľ: đ_şLTÄl°Ă­gAMAŘëőŞ cHRMmsŽâů†™xÔN3Ľycá, IDATxÚĽ–ËKë@ĹĎ<[ŁZ°µ ŠŹ…ú˙oĹť‚YąĐŤÖGDM›vwáťąÓ˝ ‚Ů$í÷űgÎ ±ÖZüâ"ż ¤M˙`­EŁ1MrntÁ'“ ÎÎÎđôôJ)Ś1KyÓʬµ¨Ş ÷÷÷x}}ĹŃŃŚ1 ô#BHł ĂŕácŚR ÓéI’ MS\]]!Ë2_ĺ˙*Ąuł!„ř'üî*Ą µFEŘŘŘŔíí-.//QU•˙ÝWPb­µĆ@Y–¸¸¸Ąô“8ÂJ§Ó)”RŘÜÜ„1„ ‡CH)qzzŠ$I>ŞˇÔ·ŮŤ18Ťp~~Ž^Ż÷®6KB „đ‰ąnEŞŞprr‚n·[ %Zk«µ†1Ăá×××č÷űĐZ/%¤p1Ćđöö†˘(pxx­­-B@)őIq'wĄfłŮ\ëš.­5Úí6„Ȳ eYbooĎWJuÁµÖPJýŘIŚ1R˘ßďăććY–â‹ ˇú–iăR~IŢßß!„Ŕúú:”Rľc<Ěě;ˇ, ˘”âńńăńűűűHÓJ)/0^'÷ń»Ŕ®3~C˙UbžçĐZcgg­V ZkH)?[ĄRJŚÇcÜÝÝÍ áÖn·ŃívˇµĄZkäyÎ9¶··Ç1(Ą`ŚyÁe»şşŠăăc”e‰ŞŞ0›ÍjŰ\Uʲô[a2™ Ďs$I‚^Ż‡8ŽE¤”ŕśű˝HůdŚAV«Ćâ8†RjΦÜAc(ŠŁŃťNťNQaeeĹCťAĚíCJ)8çľ×śsď—áľtáěěáá/// X[[CEăqCJ !Ä×-eŚAJéç±čüÎÜűççg!°»»ë+ aśóO0tmrpr0§HgÖZ MS0Ć|˛áÜ\ĽĹłŃßi‹ę ΕRsáÜŹ$IăKTx>†nŠ‰â«t-ü ÖřÖ¶xŞ;“pŁXś×Ź‹'IxÖÍ«nýÝŕY4»;;ŐIEND®B`‚stacks-1.44/php/images/r-arrow.png000644 000765 000024 00000011647 12335173442 017652 0ustar00catchenstaff000000 000000 ‰PNG  IHDR­˘Ůp pHYs  šśbiCCPPhotoshop ICC profilexÚ­Wy8Ôí×?ß1Ć>ö±3–,ŮĆľed'; ĂXĆ:Ť} zDIŠH‘˘$”˝BĄ¨DYž¬©ČҢy˙ĐÓs]ż÷zß÷ź÷üő9ç:ç|Î}źűşÎ}¸ITj "ŁbiλÍđž^Ţx¦×Ŕś€p‘c¨¦NNvđ?Ęúk@^*“¨ÔNĄ3“Ĺ ¦ľńmáÚ‹uťżŕ,ÍÓËQľml|ŰŘřb©±H(đ†’ČH*(Ń\ť‰H5`C¶q`¶q/`ăCb O™Ŕ´€1"Ĺ`•€LŽ ŚŔž@©DFF“¸:@>J‹ŕšeO/oüvÉä }4ćíż¶0I€š;˘’˙ÚäÚIu;˙µ­:îyL°†: ěfŚtúŞ,ÓI€­:ýGťľu €a  #"0Ž˙űľ¤ŕ˙Ň·Ďü[J•ÁĘxSĚTÇÜĹ2Âú‰m…+Ŕ)ÉĄĚ­ĎcĘkÇçĘďó¤Q…cDbEăÄĹS$2đ™’™RG¤ŹĘßqFöĽ\ą|µÂÍťMŠ=JCĘ“*ëV5)u 7Í­ríÇ:›z ú$ó»ĆŤ$Ś)&wÍ؉űĚo[ňě¦Y Ú¨Úłűâ`çX»‡Ó9Âĺ±›{„G›ÖŰooµĎW_˘ßQ˙ţÎ@GňÉ ç!¬ˇ¦”¸°ÚđÉHŢ(óčęŤýocб„8żřĽ„ű‰óÉÜ)F©Ôei=écSg-^Î\ÉZĎŢú ťĂ~”;—?OěüqŤ|¦V'÷śr?í}Ć÷l`!Ą(Ľ8üĺ<Ą$¬4ô©Ě»Üĺ˘m…É%µË •rU;®HU㯊]şÎ»†«–ý&ăÍ_u›ő«·oĎ6ŚŢl|z·»©ąąľĺĆ˝«÷/¶žm;Ţ~´#łó`Wr7íAÔèG=¤^b®oęń­'éOmúúçž5<Ďp”\|Ńń˛ŕ•ßňĐ÷×O† GHo”ß|{űh4Ě}\l|f˘ćoÚ;íwß&[¦R¦ő§×ggSçLß3Ľď›Ďűŕđ‘ëăËO… ž‹‚‹#KEź=——‡ľ¬8¬r¬ö}Í\3]ŁŻ·m$ojm.«ůţCáÇÜĎŠ-˙_żŢŇ÷Óé(4ÇȇágâdfcafŮbýĆö™}†ă ¶źł“«ž»‚ç4o_8nź€Ł QHOX[DITULE\[Âo!i,Ą+­,Ł´CIv§A^I°S[‘¨dŁě©BQM"䨕Ş7j<×\ŇćÖŃÔuŃKÖżb0lČl¤eĽß¤Ćtž¨hnqÓrÓĘÂ:×fÄNÍ>ŐaĐIdOsŁ+âfď~ŇcĚKÁ›ş÷–φŻ®ß˙vŇV 9>čnđj¨ %,ěZřÇHŮ(źč"ękoŚklAÜăćD»¤ăÉS¦S?řś¶–ţ-ăűÁ_‡3Ůłxł…Žý%“C8Şźk–g}Ěń¸gľĎ ˙ňÉ°Sѧ©ghgŁ ŁŠ¨Ĺ‘ç‚Ďű—x–:]°(Ó.Wľ¨\ˇtIá˛BĄ|•ě™jé«2פŻKÜ­®ş)XÇ_Ďs‹ó6göG#Ç]î&\ł`‹ŕ=‘űřVą6•v­˝Nł.‡nď{î~„´ŇÓ×[ŇůŘô‰Ŕ“ŧű‹źE>' ¬öż¨x™đĘaHjhíußpÉHôÓ·ÜoÇGoŚĄŽ[MpOŚţ]ń.lReryŞiúÄLč¬ńśŔÜâűóŢ>Z|úôqáţⱥ}źU>˙XîůrrĹwUauĺ뽵ěu§ ÁŤ±ÍŠo”ď„ď›?î˙Ěزř…ýőśąÝÔĂ2ú+ăgĚ"Ó"ó"Ëë(Ű+ö'­ŘZÎ2®\î2Żź“7(p]0SČG Â(2,Z)'n!Á)1ŽŻ”Ś”Ň—ĆHż–ąş#MÖMNIžYţ˝BĎÎjĹcJ±Ę{UĚTU˘j,jęłšZuÚĄ:9şńz$};]»đ†ĂeŁ!ă;&§LăÍ(DsK ]KŮÝ‚VŚVëÖlŢŘöÚ5ÚW99ć8%î qvw±s5pSr—đŔyrxˇĽľzŘ;ĺóvß+ß~żn˙6Rk@Kŕ-ňĄ ś`Z_¨E5L8ś)|-b&r0Ş-ş–Z±˙,íŻÄŘĐ8ßx‡ŁDŐ$ÉdľtĘję»ĎŇÚŇŻgś>~(ř°c¦f.ÉŢ8˛ř×tÎŰŁą=y-ÇnŻÎ/9‘[~’zŠ|ÚíŚĺYÍB™"ţbĆâŻçfĎŹ”ô—¶]¸QVZ~ębVEęĄčˤJŹŞ=W¬« ®J]kc×ďß(ŞˇŐ:ŢT©ĂÖ-Ő÷ßşz;·!úŽKŁÖ]ÜÝŤ¦±ćÖ–Ň{i÷I­Ä¶íŚíÓÝťU]™Ý¤†…®?č©ěÍí+yÜůdę)ý™đs˝ďÁÔ%/{_-ľvÉ}ót”wĚĽůoŢwé“_§Îbç®Ě;}äţôz±öóŮ/«Ąkősßń?ýĄŃéŰłŁ Pŕ9 ŕ\ ż Ď€«pâpŐ¤+ëY€ĺ˙3?ĚŔ â  Ćŕ  —ŕĚ"¬:âü…4!Pb(7ÔqÔcf+†\†A´0šŚ®GÓ/1~Ă8a®1ˇ™HL̒̇™çYěYęY…Y3YŘ<ŘşŘŐŮK98â8¦°ŽŘVNEÎ".&®®In'îVEž"^FŢ8ŢY>ľN~9Ž—…ű&!đNĐKđąĐnˇá]ÂŤ"Ú"-˘F˘]bVbâžâł4 :>OR\ň¶”µÔ¤tŠ źLý§+˛…rFrÓňy z s;Ď(š+®*U)űŞŞ ¨ć,Ô@­U=]ĂL“AłK+KŰNG\Ą»¤7¬ßep{WĄáiŁ,ă“0Ó@ł}Ds+ 3KăÝ&V&Ö&6»mťěĽí¨Ž‡ť ÷Ô;÷ąLş!îröž ^ŢĂ>śűô}ýNűw‘VĺČľA…ÁýˇĘî°śđţH®(÷čręgšiLněßńš 9‰cÉĘ)GRçÓĚÓK3~r?Üś%‘ťvämŽńŃ;y>ÇEó‡ ŠOś‘>ű±¨ö\Z‰ý޲ɋŤ— *ÉW¬®Ş\ç»ńŁv©îý­™†‰Ćů¦µ{ě­ŇíťŃÝéË{Zú&ž2=# xľ8ńŞwó†8š;>öNv*ufčýÎ>M/Ů,_]e[‹Ůůnůó>ťľý“VŕĐ;‚pşáÂ!‘H2„â@Yˇ˛QŹX &ŃŠčdtŁ c(c ‹ˇ`zr>1;17°łaůÂęÍÚ˦ĹVÁÎĂ~}ť#„ă-ÖŰÉ©ĹyŤKšë7÷î_<ń<+ĽŢIľ˝|Żůřă,qť†M‚:‚w…4„j…Ő„oŠ¨‰4‰‰v‹ŮŠ ‰ű‰/I¤âYđE’;%;¤Ü¤¤łeÄešw¸ďX—-”3•ĎPSxľ3QŻŘ«Ł,©üJ%SUCu†P¨fĄ¶ĄŢ¨Ą)Ł9˘uBŰ^§ł®;Ą÷DżŃŕĘ®s†9FÉĆ&~¦f{Vć†:–š»5­´­µl m-íśě˝‚“śňöT9·» »n¸‹yyFzy?őaÜGđőđ;âß@š ä#[ĄׇĚQ¤ĂHáeQ˘Ń~ÔĘýK1ę±Iq°‰îI%Éł©Ú˛Ó^dHŚ8ԜɑEĘ®=ň=Çţč­<ëc[ůŐľ§„N?={¸H§xí|ciT™|ů|EÝĺ„*“jŢ«ďŻßŻ9s3ľ>ŕ¶íí»jÍ*÷4[ Ű=;ÉݦŐz„űŘo=Ý|¶4đáĹ×W?‡ŮŢŹjŽ{üť2Y1ýr=Żů1|ˇaiă‹áęÁµ› ßi?_ýé?€,č#„B&TB,"8„Ä UČJĺ†*DM0Č2Ä2Ż~­]Ë[§lmrmN|»ń=ů‡őOźă[Ążşét€í} X‰ŃŃ4ĽŃţ%2"î.`§ÄZ» Ó,ťŔ f˘€‘ Š±půŤu)–Ö۱ 5ÖÉďäPWŔ ´0’ŤÓośá`8ä89ČÜâwěeZśłČ áѶÎŔ€ŚEąýĂ5ďňŹ˙:™dn ˘(Lr(ŃaŰ%v@sŔC DCD (Đ@DŢX A$"€ű!(@† ůq@K  B ”3üwwPţ !ÚŹry+8ľ(:Iß=”pťđđđż˝m˙0AÔźLŰě˙čhY´:Z m†6DˇőŹĆˇ…A­‰ÖE›˘ŤŃh-´Ţł…» ˛˙T€˙“qhňµ*C0€ń1s@H?ʡă`{WŔp”ř´o¦dü绊 JŚ FS“h”ĐXĽ)•„'FGRăbhJxë¨@%Ľ: đ_şLTÄl°Ă­gAMAŘëőŞ cHRMmsŽâů†™xÔN3Ľycá,ŻIDATxÚ¬U=K+Mž™=ű‘ÍĆŤë*¨(ń‚ BéÔB+K±°±Jg!bcˇť? ŕ°€lR)4XŠżŁ’ťÍÎěîÜbĽ{˝zŻŻÂ{Še`fźsćśçy !Đ˙ä[§żXů PEQ!„0ĆaĆqü9:ů ˘âááˇT*Őj5ŕśGQÇńgż}QµŰmJi˝^™ťť=88‚ ŐjA ó} ň.AÇqËBâ8Ă1F)}~~Îĺrů|~{{{gg' CÎ9çüŻő’·]Ă“_!·ă8ćśAŕy^ŽăŚŹŹďíímmm===EQ$[ń®ĹXÖ…şĽĽÜÜÜYc†aE///”ŇééiÎ9ƸZ­Ú¶˝ľľ>44„Ň4MQŚń+h†a !ŞŐęĘĘĘää$c쏴#„E±, „c8==m6›kkkĹbQˇiČĂ /Čó<Ďqśžžž >©\pÎGGGĎĎĎ766JĄŇÜÜcL„'(Ą­V+ÔWΰ,«\.7ŤĄĄ%ü+ěšĹw%Ë9·m{jjjww·\.BäĐBňć_·ţ<Eą¸¸Čd2ĂĂĂ”R Ą8çß‚Ă«ŞZ«Őnnn–——óůĽďűŞŞľ‚ľ…ŞŞí)ĆXÁKČ€ŞT*ľď/,,ôööĘ­×é#„!`Űöőőőţţľ´Ś·ž„1Žă¸żżżX,2Ć Ýnš¦9??ďş®ŞŞ†a¨Ş*“!DQBČŕŕŕęęjŁŃh6›žç˝• !äńńńęę c¬iÚýý}ĄRÉĺr®ëfłY۶MÓ” „€lŤa–eőőő†áş.Ą4±8) MÓîîît]Ż×ëŐjµP( Ék‰›N§]!DÓ´T*eŰ6Ć8•Jůľ/H*8Ćç\Q”ăăăłłł™™™?~8ŽÓÝÝÝŐŐ•Íf-Ëz}HĄRRÂétZÚŹÔç\ćŕśźśśX–µ¸¸č8N&“q]×uÝŽŽŽt:­ëz‚‰EI¶JűO@)Ą···GGGcccş®ëşnŰvgg§lĄ®ëRťż©’8«Jľ‰f¤I{žE‘çy`š¦iš™LF'ń‘?@?ľkr!_$)6ß÷Ąu*Š"‹5 ăťă%]îíŕ7‘% EUU iâ{Ąü§‰$ĚëűC¬î_ˇź wŰźx­¶IEND®B`‚stacks-1.44/php/images/stacks_bg.png000644 000765 000024 00000002035 12335173442 020210 0ustar00catchenstaff000000 000000 ‰PNG  IHDR#Kú~>nîiCCPICC Profilex…TĎkAţ6n©Đ"Zk˛x"IY«hEÔ6ýbk ۶Ed3IÖn6ëî&µĄäâŃ*ŢEíˇ˙€zđd/J…ZE(Ţ«(bˇ-ńÍnL¶ĄęŔÎ~óŢ7ď}ovß rŇ4ő€ä ÇR˘il|BjüŽ˘ A4%UŰěN$Asů{çŘz[VĂ{űw˛w­šŇ¶š„ý@ŕGšŮ*°ďq Y<ߡ)ÇtßăŘňěŹ9NyxÁµ+=ÄY"|@5-ÎM¸SÍ%Ó@H8”őqR>ś×‹”×infĆČ˝O¦»Ěî«bˇśNö˝ťô~NłŢ>Â! ­?F¸ŤžőŚŐ?âaá¤ćĆ=5ôř`·©ř5Â_M'˘TqŮ. ń®ýVňJ‚p8Ęda€sZHO×Lnřş‡}&ׯâwVQáyŹgŢÔÝďEÚŻ0  š HPEa°P@†<14˛r?#«“{2u$j»tbD±A{6Ü=·Q¤Ý<ţ("q”Cµ’üAţ*ŻÉOĺyůË\°ŘV÷”­›šşňŕ;Ĺ噹×ÓČăsM^|•Ôv“WG–¬yzĽšě?ěW—1ć‚5ŹÄs°űń-_•Ě—)ŚĹăUóęK„uZ17ßźl;=â.Ď.µÖs­‰‹7V›—gýjHű“ćUůO^őńügÍÄcâ)1&vŠç!‰—Ĺ.ń’ŘK« â`mÇť•†)Ňm‘ú$Ő``šĽő/]?[x˝F őQ”ĚŇT‰÷Â*d4ťąoúŹŰÇüä÷Šçź(/lŕČ™şmSqďˇeĄns®żŃ}đ¶nkŁ~8üX<«­R5ź Ľv‡zč)Ó––Í9R‡,ź“şéĘbRĚPŰCRR×%×eKł™UbévŘ™Ónˇ9B÷ħJe“úŻń°ý°Růů¬RŮ~NÖ—úoŔĽýEŔx‹‰ pHYs  šśŐIDAThí±@ …G-9w'îż 08ÖĄü2ł„G€fď^z±˝?ߣ Ď RÇż ‹InHF2‰@ŇÍŚd¤›É$I73’I’nf$“$ÝĚH&Hş™‘L"t3#™D éfćdşĚź×Be¦kăiK…ĘLsLŞ:mâ”Cł DfSš÷ë]ĄM#öłĎHf&suŕ}¬r…H+Ěß„Ů4ąDĄ‘ť&Tź!íM˝sFl;đ>şŰAs]!Ň«űumJĚa—+Đpú¶«He5ńîćIEND®B`‚stacks-1.44/php/images/stacks_logo_rev_small.png000644 000765 000024 00000014166 12335173442 022634 0ustar00catchenstaff000000 000000 ‰PNG  IHDR|AúŔÍîiCCPICC Profilex…TĎkAţ6n©Đ"Zk˛x"IY«hEÔ6ýbk ۶Ed3IÖn6ëî&µĄäâŃ*ŢEíˇ˙€zđd/J…ZE(Ţ«(bˇ-ńÍnL¶ĄęŔÎ~óŢ7ď}ovß rŇ4ő€ä ÇR˘il|BjüŽ˘ A4%UŰěN$Asů{çŘz[VĂ{űw˛w­šŇ¶š„ý@ŕGšŮ*°ďq Y<ߡ)ÇtßăŘňěŹ9NyxÁµ+=ÄY"|@5-ÎM¸SÍ%Ó@H8”őqR>ś×‹”×infĆČ˝O¦»Ěî«bˇśNö˝ťô~NłŢ>Â! ­?F¸ŤžőŚŐ?âaá¤ćĆ=5ôř`·©ř5Â_M'˘TqŮ. ń®ýVňJ‚p8Ęda€sZHO×Lnřş‡}&ׯâwVQáyŹgŢÔÝďEÚŻ0  š HPEa°P@†<14˛r?#«“{2u$j»tbD±A{6Ü=·Q¤Ý<ţ("q”Cµ’üAţ*ŻÉOĺyůË\°ŘV÷”­›šşňŕ;Ĺ噹×ÓČăsM^|•Ôv“WG–¬yzĽšě?ěW—1ć‚5ŹÄs°űń-_•Ě—)ŚĹăUóęK„uZ17ßźl;=â.Ď.µÖs­‰‹7V›—gýjHű“ćUůO^őńügÍÄcâ)1&vŠç!‰—Ĺ.ń’ŘK« â`mÇť•†)Ňm‘ú$Ő``šĽő/]?[x˝F őQ”ĚŇT‰÷Â*d4ťąoúŹŰÇüä÷Šçź(/lŕČ™şmSqďˇeĄns®żŃ}đ¶nkŁ~8üX<«­R5ź Ľv‡zč)Ó––Í9R‡,ź“şéĘbRĚPŰCRR×%×eKł™UbévŘ™Ónˇ9B÷ħJe“úŻń°ý°Růů¬RŮ~NÖ—úoŔĽýEŔx‹‰ pHYsgźŇR.IDATxí› ťey€ĎžÝ„% „pÉXÂݨT" T*E…)hŠ„ˇ`§Ni±H™ˇGEÄ©Š "Bi(·JB…ŕŔ€ĐŠX.5á „p IövNźç?˙»ůöĎžł›M¶2çežýîß÷~ďűÝÎ9ˇ­Z­–ĆPÚé[F#}4ŞŚ¦a«M} ´Ť±ĂëŹÜ*iŠĘMµ5hÓ,ĐrxÓLßś[oŽÝ›6jËáM3}sn9Ľ9voÚ¨c<˛ JFóŮĎŹdŁi7ĆSzgw?Ö·˙qŁ4Ń:ÚőŹ˛m«Y Śőçđ6Ć•Ń»»µĂGcąmĆz‡·śÖŔřÍ(j=Úšaő&ŽŮrxŤßŚˇÇúHźŔ¤dcîâ¸óߤ]O3Ś˛%Ź9ÖďĆxľ´7Ćáao-kÉf¶ŔXżŇ7łş­î6Őc˝Ă·BÁNÍ_C»wň.÷jJßHÚ éżďŹµĂýŇek¨ľUz«4‘˙†Ťw¸×Á;ŮáÎĺm÷ĹŃHŽôX©áÔgNęm±rSĄŢFq¬Ű€ ÷up‡kGmÖi´Ă=ŽÄU:ÜJŤĆ4T˝ˇĘH&Ű4ŁŚDą:ub®ę~\;‚vĽţüĘŘtSŽ÷z×:/UĘ<ĎdCó˝c{A GGY8k{Ę$íÇúŤ$Śö•ŢjTńm^önô›—č¨í.„eđ¶r¸N ş:? *ľ/LbQŻŔř/¸ž†hÇ–źĄ]Š‹Ŕ‰Ú ˇŢbx§ĎőŻb“6çĽ b#Dvř ‹;<śmx śďk Íµ˛ęď™ÔMÄ/Ĺŕ„tş Ŕx¤unq˛Ž•ćĹOó¨ňN•6yHÓç”:<ś˝-Ú}ľk©’á„ c,Ç-µµµ•Ú;;łrëP×׿'D•>ę’R¬+Ř6(–mLÚţ]ˇĹ»×ľ=˝Ş6·8–ă†8VŁqÔĎú#Ň1uřî4rK9;á&--Eg»+‡’0ŠGýci…µŹÝ=­ÚľőĚrµş¦żgUwçnűoŰľÝÔőUú{Ću/{b§ęŞŐŇV“&–Ş•ľRĄZSľs××­ĹśT¶X](Ł‘řŐn8§Q˝bYq¸1ě_{ §G±˙ziǵ?NÂn’zőµ—:Z'[4Ńtör6Tůp¸Ç¶ŇhB–E›ré†*ĄăŽËXyýˇóŞĺI‡—Ęť«+o¬čź|Üu{M:čx†®¶±˝K}+_™˛ňÖsŽč]z×[Űď5®Ňű|©m⸶ʊţꛥGf\Z}ÖsŔŚˇ®ö«B\ ® ÉĐŢŰŰŰ?nÜ8u~ Ł®mťôp˘ś{jÄíHO‰ öëµĺ'Ťíź&uĄ¨c1­nŽźęťůžÚ<™Őý xÜ„ˇŁ©b§!áT;TxpőďT›˙ #¦Vşá_ťŃŹłŁ}iŇá˙˛´cĆž»–ĘmĄţŢžŽ)ÓwĚšy–#ĺIS:'}ńĄŢžžR{G{©R©0ťšN•r­nma†l6†?„90 üWĆŮDG8ŮßŔĎ`!h„Ř=D‡ ˘XW|ŚľK3űĐÇç2řo¸mś“¨ĂĆŠ»PÇřđ= üîC{ÚߍđŻ`żÎÁĐńÔç8ćTKU>Žť¤yđň*µ©ă-đ_fďTłP? !}y䇄Ö)-Y˛d«r‚×Ú~úa ¬‚nH%ĘV“ůZO9˛±ópGÂŻÂŻ!ô#ÚPÖRú} öŐž‡iżĆSť÷'}+ôŔpň&ěD˙Ń× yăJ.&Ü9©WÓáńęř<Ż“đz(Ę%d„ľy|?Â;Šë¤W:d}„’†{WŞ•—óFá {fT&TąP(Q Ęű{»çSż(a€‹ĺµt˙eD˘ďéÄG:Áč/uŘÍdn“÷F‹ľ]đ±č˙řó2Ň…µŰ}F˙#qř¸ĽŤí®€7˛v´ß÷‘~ †çě&*ĘCdL»˝´ăŔ#đdđ(ńHńůx\\Ţ ŠGžG‹uŠb~v´•;ĆßKüsŕ‘e[Ź,®‰¬­u|züzś9–ÇŽa[©\ľ›0d‘Ă#‘‡Ď>OŔkyžýŽÇ\ŢóGĂép1xf÷ˇ:ăľ®ďĂnęň4,‚*ĄJą\*Ď$~ xť„|śăzÄ;˙áÄ:qŚ[WŰžfŃ&×ŔI° ´·z«Ó·aĐž¦×µp?řÝ…ývÁÁpLß!ŢďhW[=qÔyTĽ Š»<]áw“>˘n¶ę,X¦ŁżÝí]°'LóAé­ŐG ç‚;xŘö‚.HW˙q¤C â©cź§ ńÓĐݧ•W…ňě–Gߡ»'×B@*^ąd§Lě°´Źđ38ÓťöYň¬:©'Ůbâ;uŇëń+¤Cň±+ÎqW°®'FčyńTśŰźAŞ[ÄcÜ?‚­aŔᡤ•O‚"˘Łî„ ë€Đ6*”öa^“Ć“‰+ѧKű‰zĆDťü5 q˛Ł‰0ęÔ Ď¦N*Žů°~č:źW çĽęőůł©óSp“¸pÍot¤Ç]¶:Ťú!ął«Ú¤ ˘Ż°éo€z^KŃ6 㣧»÷w–ň#đÎĽď ď†nş÷.QĽWŻ„»ŕ3 Ä˝_KŐî3ďďhĄxç{—E™őěŰ´¨O*ÖUÔC˝­cýČ'š‰ý(ËáŮ,¶ľŻüŁ^–ëÝ­Ľü§8Oĺ:đł¬cy§:Žú¨č¨Ţ­oÂŁŕ¸ćő!ˇ źI#ëJp<ű÷~ý%ü)<öëXözxCĹű‰xŹ;¶uE]Ă>ö%™X)•Ôa÷Sp|~•WÚ Ĺmś:ń Ň7Ăeŕ)íä …Sĺ-Ke¸2ÇU_'â„ŤÇ$‰fb~CCúhKe`âdF˝ýO†X`Żż”¨SK ţk}ťď&őoTźâ쇨}©v5qRÚQg?C9›ěAňTž ]Ź%ýůĎ€újǢ¨Côń(ó¸„ôů`{ë{Bś .š[áßŕĐţáŰgó­çpʳɦ Ś{WŘá-p ̇i cť„»ŢĹđď°TÖ2ŰnŠ„łícř, {ßh¬zâ÷ęPO<Ε0Ş;1aVaţŘŹ»Ń+O‰Ĺć›bOĐń‘Gt±Ě9η_Áćó¶ű¶ÝřóWpü®›ŔŻ‚•¬˝† ÇkxŰx\^ź‚§Ac™o=ă'’iR‹Žúoôëâ<—Ýzá|8ÜŮŤśMń€óŚE]Ý!©t§‰ÍwśĹŕNtAą;Ý<~)ň=ÚÔ SOôö·ž'č©đ$ v? öýqř1üŽĹöĺF;<«•ü±bď¬3†/Ăp{ÄX6¦ÂrŘŃŮ™˘„—Ă©CÍ.¸gŕ𔱍Fqn]Ŕ´ůEq1Ż(Ű›D›FačďÂü1č A{©ďl¸ŽßL:ÝüˇÄ Ą} áźËůˇ}+Ž©źĚ» ÜőWBeĐg4>»mL:>ÇÚć?@ń‹Ĺ/ěß –Ąőţ‚´ź‹ýŽw*XĎĎŤéřégĐôK”ö'Â,(~ţŹ~n˘LYW ˛_˘,¨ӗ׹Ť0ĘŇ϶‘×(Luô9ś>ź…řÄ1”řnü›}§6‹Ľ4T/ÇŠĹ°Ńń”UĘ˙Ô×|w /Ú3óşž îŔ{áH¸^W|HědĂGYƸ+ŇLâÓ`R!o¸dôŐ¨ž»vŰĽÂŮ„KŔ]ď®Vľ'f±ő§ZžÜ S)ćhßw;ůcp3(^Wkóвö0 ‹ĆŚ<˘ĂŠQŠÖ*Q”¨ů¦‹y–…>Ć=–v7‚h űťŻ‚‹ĘĹĄÎb»XtÖ5O)ęg^”-5‘Č.Ä»ňtÔIŠ7ĆřCŤQ¬ ó·ü5¸€'@8ý[Ä÷:ÜŘÚÍz†Ö =#îőp=PX {Ě%ą›U4»kíDC7 ŮN™] všż^y§ĄhÓ‘}ŮĆx¤»ĚHd)q'Ą¨«F4ŚÉG»uäĹ,ňČ7ú øĂÎk5˛íµ“ă‹cDźD‡uT¬çÝ:ߢÓŐw&ř±k;ĐöbutS¦ă8®ý†îjăŢŮkyó¸đO—ÉNČx4¸şw‘ť8 8uĹxş{d…bB˙CťĺPIE%âřŚÇ’ˇŹÇRBáZަoŚy†ALď]Q0DFűeĎĺĺq*OÚqÝ>®ě×úó7n}ő˙4xő¨—ůő$Ęâqĺă톼˛óÖ>‡Ŕ×ň<ű»č'Ë]JęZÎŕE·šĚřHfąşU¨ď"±î€ż„ť@q@q ¬A×ćĎ€o°Š\‘Ö çÜH\q‚©ŢČr&QťMzŹß…@C…Ńn Ď“ł0ϵ®ňa87‹Őî@űµLŚ;çŐ?_˶©h;ővŽçŔă Žć+gÁiY¬6žQwîQ°/¨Gč`»żˇw¶˘Ť\ô!^ŻĹ+ďČőż¸e<Ĺú[ř řkPÔ‹Đ—µżöřRVüE(^Â˙IÜ_ȬŻçxuHŢ*PâĄ~ńř×(¶ńőů¦ßŃ&^č?#/~Í ť"Üź˛…˘nńŻ@ľJ<ę©[ĽÂŐk (ń˛őŐ>âSD´34ďTxBľH$­S|Ą/¦|çĽNŃ.‡‘sô7n Zyť`Dźţ"gťgátQV µ™˙Ö@‰_=B|`ĺŻLN"w×sVţ–‚?*¸˛¦‚«Ç:Š+Mńz\±öáîs(q”-&ţĚWł+ňXđ … Ľ>_‡‹Á#÷řŘź:řBż®űt§¨×!p"ě !¶‰“,ň í'äA"ßóŔťä.ń(?>÷Â2Pě{ÄüÝńóž,J:oÓ1wăé¦Ă>Îýŕ"ëňSó‰´«o‹‰ŕňĎpÜ ŹŔK`_~ëř8 ş` ŘFŻ4¨­Cv§…¸ş{"Q'LwµUž†ěłaúyÔ1b'? B\ͱű"/űżG$ô;€ř‹yşĄź__"˝ b%çŐŞż%b–…ü‘¬ĎŐ©»Íß˝ŻŤŠ„ë ‘ śČĂDşŔľŁżă‰+öŁ<Ĺný8ýŚ» ÝÝ= m”+Á˛˝ NˇtlO˝×`měBťH”Ąa6pž1™p6ěNxĽKA J,$ď*ŰyŻéTťăZ«Aqë;†ý›ď˘>n ęÚŔůď ¶×kÁ>žËC‚Ll§mŠs×nľOt˛Îómă<ŠőČĘň -sNSÁ~]ŕn¦b»mÉۦsÔńúĆşż2`óÔáÚ(&ÝČ0Ń‘ˇŽ¶®©„3†šxôˇ.EKă5ŰÉP°]Ńp楢n¶nŰÔ›űpcŘv(±ťRĎ.ęĄ~#ŃÍ~ÔO; رčp+)lÇ©ˇ„Ć„uFţĆ}¸¸˘/ű7.ö+CILÚzN$Ú™o<=e¬“Ž±1ú†ăé"ë7ƱOe¸EžŽőÓÓ#ë¤Îuv>!¶pńÔęeć;®yCεžĂ©ß’-Ńé*ÚçךSÁ-‡ ˛Ą'[ßŇ=\_ËáléÉ–Ă·tć×rxÁ [zň˙Béuď#đIEND®B`‚stacks-1.44/htslib/._bgzf.c000644 000765 000024 00000000265 12750743246 016315 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/bgzf.c000644 000765 000024 00000121646 12750743246 016107 0ustar00catchenstaff000000 000000 /* The MIT License Copyright (c) 2008 Broad Institute / Massachusetts Institute of Technology 2011, 2012 Attractive Chaos Copyright (C) 2009, 2013-2016 Genome Research Ltd Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include #include #include #include #include #include #include #include "htslib/hts.h" #include "htslib/bgzf.h" #include "htslib/hfile.h" #define BGZF_CACHE #define BGZF_MT #define BLOCK_HEADER_LENGTH 18 #define BLOCK_FOOTER_LENGTH 8 /* BGZF/GZIP header (speciallized from RFC 1952; little endian): +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ | 31|139| 8| 4| 0| 0|255| 6| 66| 67| 2|BLK_LEN| +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ BGZF extension: ^ ^ ^ ^ | | | | FLG.EXTRA XLEN B C BGZF format is compatible with GZIP. It limits the size of each compressed block to 2^16 bytes and adds and an extra "BC" field in the gzip header which records the size. */ static const uint8_t g_magic[19] = "\037\213\010\4\0\0\0\0\0\377\6\0\102\103\2\0\0\0"; #ifdef BGZF_CACHE typedef struct { int size; uint8_t *block; int64_t end_offset; } cache_t; #include "htslib/khash.h" KHASH_MAP_INIT_INT64(cache, cache_t) #endif typedef struct { uint64_t uaddr; // offset w.r.t. uncompressed data uint64_t caddr; // offset w.r.t. compressed data } bgzidx1_t; struct __bgzidx_t { int noffs, moffs; // the size of the index, n:used, m:allocated bgzidx1_t *offs; // offsets uint64_t ublock_addr; // offset of the current block (uncompressed data) }; void bgzf_index_destroy(BGZF *fp); int bgzf_index_add_block(BGZF *fp); static inline void packInt16(uint8_t *buffer, uint16_t value) { buffer[0] = value; buffer[1] = value >> 8; } static inline int unpackInt16(const uint8_t *buffer) { return buffer[0] | buffer[1] << 8; } static inline void packInt32(uint8_t *buffer, uint32_t value) { buffer[0] = value; buffer[1] = value >> 8; buffer[2] = value >> 16; buffer[3] = value >> 24; } static const char *bgzf_zerr(int errnum, z_stream *zs) { static char buffer[32]; /* Return zs->msg if available. zlib doesn't set this very reliably. Looking at the source suggests that it may get set to a useful message for deflateInit2, inflateInit2 and inflate when it returns Z_DATA_ERROR. For inflate with other return codes, deflate, deflateEnd and inflateEnd it doesn't appear to be useful. For the likely non-useful cases, the caller should pass NULL into zs. */ if (zs && zs->msg) return zs->msg; // gzerror OF((gzFile file, int *errnum) switch (errnum) { case Z_ERRNO: return strerror(errno); case Z_STREAM_ERROR: return "invalid parameter/compression level, or inconsistent stream state"; case Z_DATA_ERROR: return "invalid or incomplete IO"; case Z_MEM_ERROR: return "out of memory"; case Z_BUF_ERROR: return "progress temporarily not possible, or in() / out() returned an error"; case Z_VERSION_ERROR: return "zlib version mismatch"; case Z_OK: // 0: maybe gzgets error Z_NULL default: snprintf(buffer, sizeof(buffer), "[%d] unknown", errnum); return buffer; // FIXME: Not thread-safe. } } static BGZF *bgzf_read_init(hFILE *hfpr) { BGZF *fp; uint8_t magic[18]; ssize_t n = hpeek(hfpr, magic, 18); if (n < 0) return NULL; fp = (BGZF*)calloc(1, sizeof(BGZF)); if (fp == NULL) return NULL; fp->is_write = 0; fp->is_compressed = (n==2 && magic[0]==0x1f && magic[1]==0x8b); fp->uncompressed_block = malloc(BGZF_MAX_BLOCK_SIZE); fp->compressed_block = malloc(BGZF_MAX_BLOCK_SIZE); fp->is_compressed = (n==18 && magic[0]==0x1f && magic[1]==0x8b) ? 1 : 0; fp->is_gzip = ( !fp->is_compressed || ((magic[3]&4) && memcmp(&magic[12], "BC\2\0",4)==0) ) ? 0 : 1; #ifdef BGZF_CACHE fp->cache = kh_init(cache); #endif return fp; } // get the compress level from the mode string: compress_level==-1 for the default level, -2 plain uncompressed static int mode2level(const char *__restrict mode) { int i, compress_level = -1; for (i = 0; mode[i]; ++i) if (mode[i] >= '0' && mode[i] <= '9') break; if (mode[i]) compress_level = (int)mode[i] - '0'; if (strchr(mode, 'u')) compress_level = -2; return compress_level; } static BGZF *bgzf_write_init(const char *mode) { BGZF *fp; fp = (BGZF*)calloc(1, sizeof(BGZF)); if (fp == NULL) goto mem_fail; fp->is_write = 1; int compress_level = mode2level(mode); if ( compress_level==-2 ) { fp->is_compressed = 0; return fp; } fp->is_compressed = 1; fp->uncompressed_block = malloc(BGZF_MAX_BLOCK_SIZE); if (fp->uncompressed_block == NULL) goto mem_fail; fp->compressed_block = malloc(BGZF_MAX_BLOCK_SIZE); if (fp->compressed_block == NULL) goto mem_fail; fp->compress_level = compress_level < 0? Z_DEFAULT_COMPRESSION : compress_level; // Z_DEFAULT_COMPRESSION==-1 if (fp->compress_level > 9) fp->compress_level = Z_DEFAULT_COMPRESSION; if ( strchr(mode,'g') ) { // gzip output fp->is_gzip = 1; fp->gz_stream = (z_stream*)calloc(1,sizeof(z_stream)); if (fp->gz_stream == NULL) goto mem_fail; fp->gz_stream->zalloc = NULL; fp->gz_stream->zfree = NULL; fp->gz_stream->msg = NULL; int ret = deflateInit2(fp->gz_stream, fp->compress_level, Z_DEFLATED, 15|16, 8, Z_DEFAULT_STRATEGY); if (ret!=Z_OK) { if (hts_verbose >= 1) { fprintf(stderr, "[E::%s] deflateInit2 failed: %s\n", __func__, bgzf_zerr(ret, fp->gz_stream)); } goto fail; } } return fp; mem_fail: if (hts_verbose >= 1) { fprintf(stderr, "[E::%s] %s\n", __func__, strerror(errno)); } fail: if (fp != NULL) { free(fp->uncompressed_block); free(fp->compressed_block); free(fp->gz_stream); free(fp); } return NULL; } BGZF *bgzf_open(const char *path, const char *mode) { BGZF *fp = 0; assert(compressBound(BGZF_BLOCK_SIZE) < BGZF_MAX_BLOCK_SIZE); if (strchr(mode, 'r')) { hFILE *fpr; if ((fpr = hopen(path, mode)) == 0) return 0; fp = bgzf_read_init(fpr); if (fp == 0) { hclose_abruptly(fpr); return NULL; } fp->fp = fpr; } else if (strchr(mode, 'w') || strchr(mode, 'a')) { hFILE *fpw; if ((fpw = hopen(path, mode)) == 0) return 0; fp = bgzf_write_init(mode); if (fp == NULL) return NULL; fp->fp = fpw; } else { errno = EINVAL; return 0; } fp->is_be = ed_is_big(); return fp; } BGZF *bgzf_dopen(int fd, const char *mode) { BGZF *fp = 0; assert(compressBound(BGZF_BLOCK_SIZE) < BGZF_MAX_BLOCK_SIZE); if (strchr(mode, 'r')) { hFILE *fpr; if ((fpr = hdopen(fd, mode)) == 0) return 0; fp = bgzf_read_init(fpr); if (fp == 0) { hclose_abruptly(fpr); return NULL; } // FIXME this closes fd fp->fp = fpr; } else if (strchr(mode, 'w') || strchr(mode, 'a')) { hFILE *fpw; if ((fpw = hdopen(fd, mode)) == 0) return 0; fp = bgzf_write_init(mode); if (fp == NULL) return NULL; fp->fp = fpw; } else { errno = EINVAL; return 0; } fp->is_be = ed_is_big(); return fp; } BGZF *bgzf_hopen(hFILE *hfp, const char *mode) { BGZF *fp = NULL; assert(compressBound(BGZF_BLOCK_SIZE) < BGZF_MAX_BLOCK_SIZE); if (strchr(mode, 'r')) { fp = bgzf_read_init(hfp); if (fp == NULL) return NULL; } else if (strchr(mode, 'w') || strchr(mode, 'a')) { fp = bgzf_write_init(mode); if (fp == NULL) return NULL; } else { errno = EINVAL; return 0; } fp->fp = hfp; fp->is_be = ed_is_big(); return fp; } int bgzf_compress(void *_dst, size_t *dlen, const void *src, size_t slen, int level) { uint32_t crc; z_stream zs; uint8_t *dst = (uint8_t*)_dst; // compress the body zs.zalloc = NULL; zs.zfree = NULL; zs.msg = NULL; zs.next_in = (Bytef*)src; zs.avail_in = slen; zs.next_out = dst + BLOCK_HEADER_LENGTH; zs.avail_out = *dlen - BLOCK_HEADER_LENGTH - BLOCK_FOOTER_LENGTH; int ret = deflateInit2(&zs, level, Z_DEFLATED, -15, 8, Z_DEFAULT_STRATEGY); // -15 to disable zlib header/footer if (ret!=Z_OK) { if (hts_verbose >= 1) { fprintf(stderr, "[E::%s] deflateInit2 failed: %s\n", __func__, bgzf_zerr(ret, &zs)); } return -1; } if ((ret = deflate(&zs, Z_FINISH)) != Z_STREAM_END) { if (hts_verbose >= 1) { fprintf(stderr, "[E::%s] deflate failed: %s\n", __func__, bgzf_zerr(ret, ret == Z_DATA_ERROR ? &zs : NULL)); } return -1; } if ((ret = deflateEnd(&zs)) != Z_OK) { if (hts_verbose >= 1) { fprintf(stderr, "[E::%s] deflateEnd failed: %s\n", __func__, bgzf_zerr(ret, NULL)); } return -1; } *dlen = zs.total_out + BLOCK_HEADER_LENGTH + BLOCK_FOOTER_LENGTH; // write the header memcpy(dst, g_magic, BLOCK_HEADER_LENGTH); // the last two bytes are a place holder for the length of the block packInt16(&dst[16], *dlen - 1); // write the compressed length; -1 to fit 2 bytes // write the footer crc = crc32(crc32(0L, NULL, 0L), (Bytef*)src, slen); packInt32((uint8_t*)&dst[*dlen - 8], crc); packInt32((uint8_t*)&dst[*dlen - 4], slen); return 0; } static int bgzf_gzip_compress(BGZF *fp, void *_dst, size_t *dlen, const void *src, size_t slen, int level) { uint8_t *dst = (uint8_t*)_dst; z_stream *zs = fp->gz_stream; int flush = slen ? Z_NO_FLUSH : Z_FINISH; zs->next_in = (Bytef*)src; zs->avail_in = slen; zs->next_out = dst; zs->avail_out = *dlen; int ret = deflate(zs, flush); if (ret == Z_STREAM_ERROR) { if (hts_verbose >= 1) { fprintf(stderr, "[E::%s] deflate failed: %s\n", __func__, bgzf_zerr(ret, NULL)); } return -1; } *dlen = *dlen - zs->avail_out; return 0; } // Deflate the block in fp->uncompressed_block into fp->compressed_block. Also adds an extra field that stores the compressed block length. static int deflate_block(BGZF *fp, int block_length) { size_t comp_size = BGZF_MAX_BLOCK_SIZE; int ret; if ( !fp->is_gzip ) ret = bgzf_compress(fp->compressed_block, &comp_size, fp->uncompressed_block, block_length, fp->compress_level); else ret = bgzf_gzip_compress(fp, fp->compressed_block, &comp_size, fp->uncompressed_block, block_length, fp->compress_level); if ( ret != 0 ) { if (hts_verbose >= 3) { fprintf(stderr, "[E::%s] compression error %d\n", __func__, ret); } fp->errcode |= BGZF_ERR_ZLIB; return -1; } fp->block_offset = 0; return comp_size; } // Inflate the block in fp->compressed_block into fp->uncompressed_block static int inflate_block(BGZF* fp, int block_length) { z_stream zs; zs.zalloc = NULL; zs.zfree = NULL; zs.msg = NULL; zs.next_in = (Bytef*)fp->compressed_block + 18; zs.avail_in = block_length - 16; zs.next_out = (Bytef*)fp->uncompressed_block; zs.avail_out = BGZF_MAX_BLOCK_SIZE; int ret = inflateInit2(&zs, -15); if (ret != Z_OK) { if (hts_verbose >= 1) { fprintf(stderr, "[E::%s] inflateInit2 failed: %s\n", __func__, bgzf_zerr(ret, &zs)); } fp->errcode |= BGZF_ERR_ZLIB; return -1; } if ((ret = inflate(&zs, Z_FINISH)) != Z_STREAM_END) { if (hts_verbose >= 1) { fprintf(stderr, "[E::%s] inflate failed: %s\n", __func__, bgzf_zerr(ret, ret == Z_DATA_ERROR ? &zs : NULL)); } if ((ret = inflateEnd(&zs)) != Z_OK) { if (hts_verbose >= 2) { fprintf(stderr, "[E::%s] inflateEnd failed: %s\n", __func__, bgzf_zerr(ret, NULL)); } } fp->errcode |= BGZF_ERR_ZLIB; return -1; } if ((ret = inflateEnd(&zs)) != Z_OK) { if (hts_verbose >= 1) { fprintf(stderr, "[E::%s] inflateEnd failed: %s\n", __func__, bgzf_zerr(ret, NULL)); } fp->errcode |= BGZF_ERR_ZLIB; return -1; } return zs.total_out; } static int inflate_gzip_block(BGZF *fp, int cached) { int ret = Z_OK; do { if ( !cached && fp->gz_stream->avail_out!=0 ) { fp->gz_stream->avail_in = hread(fp->fp, fp->compressed_block, BGZF_BLOCK_SIZE); if ( fp->gz_stream->avail_in<=0 ) return fp->gz_stream->avail_in; if ( fp->gz_stream->avail_in==0 ) break; fp->gz_stream->next_in = fp->compressed_block; } else cached = 0; do { fp->gz_stream->next_out = (Bytef*)fp->uncompressed_block + fp->block_offset; fp->gz_stream->avail_out = BGZF_MAX_BLOCK_SIZE - fp->block_offset; fp->gz_stream->msg = NULL; ret = inflate(fp->gz_stream, Z_NO_FLUSH); if ( ret==Z_BUF_ERROR ) continue; // non-critical error if ( ret<0 ) { if (hts_verbose >= 1) { fprintf(stderr, "[E::%s] inflate failed: %s\n", __func__, bgzf_zerr(ret, ret == Z_DATA_ERROR ? fp->gz_stream : NULL)); } fp->errcode |= BGZF_ERR_ZLIB; return -1; } unsigned int have = BGZF_MAX_BLOCK_SIZE - fp->gz_stream->avail_out; if ( have ) return have; } while ( fp->gz_stream->avail_out == 0 ); } while (ret != Z_STREAM_END); return BGZF_MAX_BLOCK_SIZE - fp->gz_stream->avail_out; } // Returns: 0 on success (BGZF header); -1 on non-BGZF GZIP header; -2 on error static int check_header(const uint8_t *header) { if ( header[0] != 31 || header[1] != 139 || header[2] != 8 ) return -2; return ((header[3] & 4) != 0 && unpackInt16((uint8_t*)&header[10]) == 6 && header[12] == 'B' && header[13] == 'C' && unpackInt16((uint8_t*)&header[14]) == 2) ? 0 : -1; } #ifdef BGZF_CACHE static void free_cache(BGZF *fp) { khint_t k; khash_t(cache) *h = (khash_t(cache)*)fp->cache; if (fp->is_write) return; for (k = kh_begin(h); k < kh_end(h); ++k) if (kh_exist(h, k)) free(kh_val(h, k).block); kh_destroy(cache, h); } static int load_block_from_cache(BGZF *fp, int64_t block_address) { khint_t k; cache_t *p; khash_t(cache) *h = (khash_t(cache)*)fp->cache; k = kh_get(cache, h, block_address); if (k == kh_end(h)) return 0; p = &kh_val(h, k); if (fp->block_length != 0) fp->block_offset = 0; fp->block_address = block_address; fp->block_length = p->size; memcpy(fp->uncompressed_block, p->block, BGZF_MAX_BLOCK_SIZE); if ( hseek(fp->fp, p->end_offset, SEEK_SET) < 0 ) { // todo: move the error up fprintf(stderr,"Could not hseek to %"PRId64"\n", p->end_offset); exit(1); } return p->size; } static void cache_block(BGZF *fp, int size) { int ret; khint_t k; cache_t *p; khash_t(cache) *h = (khash_t(cache)*)fp->cache; if (BGZF_MAX_BLOCK_SIZE >= fp->cache_size) return; if ((kh_size(h) + 1) * BGZF_MAX_BLOCK_SIZE > (uint32_t)fp->cache_size) { /* A better way would be to remove the oldest block in the * cache, but here we remove a random one for simplicity. This * should not have a big impact on performance. */ for (k = kh_begin(h); k < kh_end(h); ++k) if (kh_exist(h, k)) break; if (k < kh_end(h)) { free(kh_val(h, k).block); kh_del(cache, h, k); } } k = kh_put(cache, h, fp->block_address, &ret); if (ret == 0) return; // if this happens, a bug! p = &kh_val(h, k); p->size = fp->block_length; p->end_offset = fp->block_address + size; p->block = (uint8_t*)malloc(BGZF_MAX_BLOCK_SIZE); memcpy(kh_val(h, k).block, fp->uncompressed_block, BGZF_MAX_BLOCK_SIZE); } #else static void free_cache(BGZF *fp) {} static int load_block_from_cache(BGZF *fp, int64_t block_address) {return 0;} static void cache_block(BGZF *fp, int size) {} #endif int bgzf_read_block(BGZF *fp) { uint8_t header[BLOCK_HEADER_LENGTH], *compressed_block; int count, size = 0, block_length, remaining; // Reading an uncompressed file if ( !fp->is_compressed ) { count = hread(fp->fp, fp->uncompressed_block, BGZF_MAX_BLOCK_SIZE); if (count < 0) // Error { fp->errcode |= BGZF_ERR_IO; return -1; } else if (count == 0) // EOF { fp->block_length = 0; return 0; } if (fp->block_length != 0) fp->block_offset = 0; fp->block_address += count; fp->block_length = count; return 0; } // Reading compressed file int64_t block_address; block_address = htell(fp->fp); if ( fp->is_gzip && fp->gz_stream ) // is this is a initialized gzip stream? { count = inflate_gzip_block(fp, 0); if ( count<0 ) { fp->errcode |= BGZF_ERR_ZLIB; return -1; } fp->block_length = count; fp->block_address = block_address; return 0; } if (fp->cache_size && load_block_from_cache(fp, block_address)) return 0; count = hread(fp->fp, header, sizeof(header)); if (count == 0) { // no data read fp->block_length = 0; return 0; } int ret; if ( count != sizeof(header) || (ret=check_header(header))==-2 ) { fp->errcode |= BGZF_ERR_HEADER; return -1; } if ( ret==-1 ) { // GZIP, not BGZF uint8_t *cblock = (uint8_t*)fp->compressed_block; memcpy(cblock, header, sizeof(header)); count = hread(fp->fp, cblock+sizeof(header), BGZF_BLOCK_SIZE - sizeof(header)) + sizeof(header); int nskip = 10; // Check optional fields to skip: FLG.FNAME,FLG.FCOMMENT,FLG.FHCRC,FLG.FEXTRA // Note: Some of these fields are untested, I did not have appropriate data available if ( header[3] & 0x4 ) // FLG.FEXTRA { nskip += unpackInt16(&cblock[nskip]) + 2; } if ( header[3] & 0x8 ) // FLG.FNAME { while ( nskip= count ) { fp->errcode |= BGZF_ERR_HEADER; return -1; } fp->is_gzip = 1; fp->gz_stream = (z_stream*) calloc(1,sizeof(z_stream)); int ret = inflateInit2(fp->gz_stream, -15); if (ret != Z_OK) { if (hts_verbose >= 1) { fprintf(stderr, "[E::%s] inflateInit2 failed: %s", __func__, bgzf_zerr(ret, fp->gz_stream)); } fp->errcode |= BGZF_ERR_ZLIB; return -1; } fp->gz_stream->avail_in = count - nskip; fp->gz_stream->next_in = cblock + nskip; count = inflate_gzip_block(fp, 1); if ( count<0 ) { fp->errcode |= BGZF_ERR_ZLIB; return -1; } fp->block_length = count; fp->block_address = block_address; if ( fp->idx_build_otf ) return -1; // cannot build index for gzip return 0; } size = count; block_length = unpackInt16((uint8_t*)&header[16]) + 1; // +1 because when writing this number, we used "-1" compressed_block = (uint8_t*)fp->compressed_block; memcpy(compressed_block, header, BLOCK_HEADER_LENGTH); remaining = block_length - BLOCK_HEADER_LENGTH; count = hread(fp->fp, &compressed_block[BLOCK_HEADER_LENGTH], remaining); if (count != remaining) { fp->errcode |= BGZF_ERR_IO; return -1; } size += count; if ((count = inflate_block(fp, block_length)) < 0) { if (hts_verbose >= 2) fprintf(stderr, "[E::%s] inflate_block error %d\n", __func__, count); fp->errcode |= BGZF_ERR_ZLIB; return -1; } if (fp->block_length != 0) fp->block_offset = 0; // Do not reset offset if this read follows a seek. fp->block_address = block_address; fp->block_length = count; if ( fp->idx_build_otf ) { bgzf_index_add_block(fp); fp->idx->ublock_addr += count; } cache_block(fp, size); return 0; } ssize_t bgzf_read(BGZF *fp, void *data, size_t length) { ssize_t bytes_read = 0; uint8_t *output = (uint8_t*)data; if (length <= 0) return 0; assert(fp->is_write == 0); while (bytes_read < length) { int copy_length, available = fp->block_length - fp->block_offset; uint8_t *buffer; if (available <= 0) { int ret = bgzf_read_block(fp); if (ret != 0) { if (hts_verbose >= 2) { fprintf(stderr, "[E::%s] bgzf_read_block error %d after %zd of %zu bytes\n", __func__, ret, bytes_read, length); } fp->errcode |= BGZF_ERR_ZLIB; return -1; } available = fp->block_length - fp->block_offset; if (available <= 0) break; } copy_length = length - bytes_read < available? length - bytes_read : available; buffer = (uint8_t*)fp->uncompressed_block; memcpy(output, buffer + fp->block_offset, copy_length); fp->block_offset += copy_length; output += copy_length; bytes_read += copy_length; } if (fp->block_offset == fp->block_length) { fp->block_address = htell(fp->fp); fp->block_offset = fp->block_length = 0; } fp->uncompressed_address += bytes_read; return bytes_read; } ssize_t bgzf_raw_read(BGZF *fp, void *data, size_t length) { return hread(fp->fp, data, length); } #ifdef BGZF_MT typedef struct { struct bgzf_mtaux_t *mt; void *buf; int i, errcode, toproc, compress_level; } worker_t; typedef struct bgzf_mtaux_t { int n_threads, n_blks, curr, done; volatile int proc_cnt; void **blk; int *len; worker_t *w; pthread_t *tid; pthread_mutex_t lock; pthread_cond_t cv; } mtaux_t; static int worker_aux(worker_t *w) { int i, stop = 0; // wait for condition: to process or all done pthread_mutex_lock(&w->mt->lock); while (!w->toproc && !w->mt->done) pthread_cond_wait(&w->mt->cv, &w->mt->lock); if (w->mt->done) stop = 1; w->toproc = 0; pthread_mutex_unlock(&w->mt->lock); if (stop) return 1; // to quit the thread w->errcode = 0; for (i = w->i; i < w->mt->curr; i += w->mt->n_threads) { size_t clen = BGZF_MAX_BLOCK_SIZE; int ret = bgzf_compress(w->buf, &clen, w->mt->blk[i], w->mt->len[i], w->compress_level); if (ret != 0) { if (hts_verbose >= 2) fprintf(stderr, "[E::%s] bgzf_compress error %d\n", __func__, ret); w->errcode |= BGZF_ERR_ZLIB; // Report error // We're not going to do any more, so set remaining lengths to 0 for (; i < w->mt->curr; i += w->mt->n_threads) w->mt->len[i] = 0; break; // Give up } else { memcpy(w->mt->blk[i], w->buf, clen); w->mt->len[i] = clen; } } __sync_fetch_and_add(&w->mt->proc_cnt, 1); return 0; } static void *mt_worker(void *data) { while (worker_aux((worker_t*)data) == 0); return 0; } int bgzf_mt(BGZF *fp, int n_threads, int n_sub_blks) { int i; mtaux_t *mt; pthread_attr_t attr; if (!fp->is_write || fp->mt || n_threads <= 1) return -1; mt = (mtaux_t*)calloc(1, sizeof(mtaux_t)); mt->n_threads = n_threads; mt->n_blks = n_threads * n_sub_blks; mt->len = (int*)calloc(mt->n_blks, sizeof(int)); mt->blk = (void**)calloc(mt->n_blks, sizeof(void*)); for (i = 0; i < mt->n_blks; ++i) mt->blk[i] = malloc(BGZF_MAX_BLOCK_SIZE); mt->tid = (pthread_t*)calloc(mt->n_threads, sizeof(pthread_t)); // tid[0] is not used, as the worker 0 is launched by the master mt->w = (worker_t*)calloc(mt->n_threads, sizeof(worker_t)); for (i = 0; i < mt->n_threads; ++i) { mt->w[i].i = i; mt->w[i].mt = mt; mt->w[i].compress_level = fp->compress_level; mt->w[i].buf = malloc(BGZF_MAX_BLOCK_SIZE); } pthread_attr_init(&attr); pthread_attr_setdetachstate(&attr, PTHREAD_CREATE_JOINABLE); pthread_mutex_init(&mt->lock, 0); pthread_cond_init(&mt->cv, 0); for (i = 1; i < mt->n_threads; ++i) // worker 0 is effectively launched by the master thread pthread_create(&mt->tid[i], &attr, mt_worker, &mt->w[i]); fp->mt = mt; return 0; } static void mt_destroy(mtaux_t *mt) { int i; // signal all workers to quit pthread_mutex_lock(&mt->lock); mt->done = 1; mt->proc_cnt = 0; pthread_cond_broadcast(&mt->cv); pthread_mutex_unlock(&mt->lock); for (i = 1; i < mt->n_threads; ++i) pthread_join(mt->tid[i], 0); // worker 0 is effectively launched by the master thread // free other data allocated on heap for (i = 0; i < mt->n_blks; ++i) free(mt->blk[i]); for (i = 0; i < mt->n_threads; ++i) free(mt->w[i].buf); free(mt->blk); free(mt->len); free(mt->w); free(mt->tid); pthread_cond_destroy(&mt->cv); pthread_mutex_destroy(&mt->lock); free(mt); } static void mt_queue(BGZF *fp) { mtaux_t *mt = fp->mt; assert(mt->curr < mt->n_blks); // guaranteed by the caller memcpy(mt->blk[mt->curr], fp->uncompressed_block, fp->block_offset); mt->len[mt->curr] = fp->block_offset; fp->block_offset = 0; ++mt->curr; } static int mt_flush_queue(BGZF *fp) { int i; mtaux_t *mt = fp->mt; // signal all the workers to compress pthread_mutex_lock(&mt->lock); for (i = 0; i < mt->n_threads; ++i) mt->w[i].toproc = 1; mt->proc_cnt = 0; pthread_cond_broadcast(&mt->cv); pthread_mutex_unlock(&mt->lock); // worker 0 is doing things here worker_aux(&mt->w[0]); // wait for all the threads to complete while (mt->proc_cnt < mt->n_threads); // dump data to disk for (i = 0; i < mt->n_threads; ++i) fp->errcode |= mt->w[i].errcode; if (fp->errcode == 0) { /* Only try to write if all the threads worked, as otherwise we could get a file with holes in it */ for (i = 0; i < mt->curr; ++i) { if (hwrite(fp->fp, mt->blk[i], mt->len[i]) != mt->len[i]) { fp->errcode |= BGZF_ERR_IO; break; } } } mt->curr = 0; return (fp->errcode == 0)? 0 : -1; } static int lazy_flush(BGZF *fp) { if (fp->mt) { if (fp->block_offset) mt_queue(fp); return (fp->mt->curr < fp->mt->n_blks)? 0 : mt_flush_queue(fp); } else return bgzf_flush(fp); } #else // ~ #ifdef BGZF_MT int bgzf_mt(BGZF *fp, int n_threads, int n_sub_blks) { return 0; } static inline int lazy_flush(BGZF *fp) { return bgzf_flush(fp); } #endif // ~ #ifdef BGZF_MT int bgzf_flush(BGZF *fp) { if (!fp->is_write) return 0; #ifdef BGZF_MT if (fp->mt) { if (fp->block_offset) mt_queue(fp); // guaranteed that assertion does not fail return mt_flush_queue(fp); } #endif while (fp->block_offset > 0) { int block_length; if ( fp->idx_build_otf ) { bgzf_index_add_block(fp); fp->idx->ublock_addr += fp->block_offset; } block_length = deflate_block(fp, fp->block_offset); if (block_length < 0) { if (hts_verbose >= 3) fprintf(stderr, "[E::%s] deflate_block error %d\n", __func__, block_length); return -1; } if (hwrite(fp->fp, fp->compressed_block, block_length) != block_length) { if (hts_verbose >= 1) fprintf(stderr, "[E::%s] hwrite error (wrong size)\n", __func__); fp->errcode |= BGZF_ERR_IO; // possibly truncated file return -1; } fp->block_address += block_length; } return 0; } int bgzf_flush_try(BGZF *fp, ssize_t size) { if (fp->block_offset + size > BGZF_BLOCK_SIZE) return lazy_flush(fp); return 0; } ssize_t bgzf_write(BGZF *fp, const void *data, size_t length) { if ( !fp->is_compressed ) return hwrite(fp->fp, data, length); const uint8_t *input = (const uint8_t*)data; ssize_t remaining = length; assert(fp->is_write); while (remaining > 0) { uint8_t* buffer = (uint8_t*)fp->uncompressed_block; int copy_length = BGZF_BLOCK_SIZE - fp->block_offset; if (copy_length > remaining) copy_length = remaining; memcpy(buffer + fp->block_offset, input, copy_length); fp->block_offset += copy_length; input += copy_length; remaining -= copy_length; if (fp->block_offset == BGZF_BLOCK_SIZE) { if (lazy_flush(fp) != 0) return -1; } } return length - remaining; } ssize_t bgzf_raw_write(BGZF *fp, const void *data, size_t length) { return hwrite(fp->fp, data, length); } int bgzf_close(BGZF* fp) { int ret, block_length; if (fp == 0) return -1; if (fp->is_write && fp->is_compressed) { if (bgzf_flush(fp) != 0) return -1; fp->compress_level = -1; block_length = deflate_block(fp, 0); // write an empty block if (block_length < 0) { if (hts_verbose >= 3) fprintf(stderr, "[E::%s] deflate_block error %d\n", __func__, block_length); return -1; } if (hwrite(fp->fp, fp->compressed_block, block_length) < 0 || hflush(fp->fp) != 0) { if (hts_verbose >= 1) fprintf(stderr, "[E::%s] file write error\n", __func__); fp->errcode |= BGZF_ERR_IO; return -1; } #ifdef BGZF_MT if (fp->mt) mt_destroy(fp->mt); #endif } if ( fp->is_gzip ) { if (!fp->is_write) ret = inflateEnd(fp->gz_stream); else ret = deflateEnd(fp->gz_stream); if (ret != Z_OK && hts_verbose >= 1) fprintf(stderr, "[E::%s] inflateEnd/deflateEnd failed: %s\n", __func__, bgzf_zerr(ret, NULL)); free(fp->gz_stream); } ret = hclose(fp->fp); if (ret != 0) return -1; bgzf_index_destroy(fp); free(fp->uncompressed_block); free(fp->compressed_block); free_cache(fp); free(fp); return 0; } void bgzf_set_cache_size(BGZF *fp, int cache_size) { if (fp) fp->cache_size = cache_size; } int bgzf_check_EOF(BGZF *fp) { uint8_t buf[28]; off_t offset = htell(fp->fp); if (hseek(fp->fp, -28, SEEK_END) < 0) { if (errno == ESPIPE) { hclearerr(fp->fp); return 2; } else return -1; } if ( hread(fp->fp, buf, 28) != 28 ) return -1; if ( hseek(fp->fp, offset, SEEK_SET) < 0 ) return -1; return (memcmp("\037\213\010\4\0\0\0\0\0\377\6\0\102\103\2\0\033\0\3\0\0\0\0\0\0\0\0\0", buf, 28) == 0)? 1 : 0; } int64_t bgzf_seek(BGZF* fp, int64_t pos, int where) { int block_offset; int64_t block_address; if (fp->is_write || where != SEEK_SET) { fp->errcode |= BGZF_ERR_MISUSE; return -1; } block_offset = pos & 0xFFFF; block_address = pos >> 16; if (hseek(fp->fp, block_address, SEEK_SET) < 0) { fp->errcode |= BGZF_ERR_IO; return -1; } fp->block_length = 0; // indicates current block has not been loaded fp->block_address = block_address; fp->block_offset = block_offset; return 0; } int bgzf_is_bgzf(const char *fn) { uint8_t buf[16]; int n; hFILE *fp; if ((fp = hopen(fn, "r")) == 0) return 0; n = hread(fp, buf, 16); if ( hclose(fp) < 0 ) return -1; if (n != 16) return 0; return memcmp(g_magic, buf, 16) == 0? 1 : 0; } int bgzf_getc(BGZF *fp) { int c; if (fp->block_offset >= fp->block_length) { if (bgzf_read_block(fp) != 0) return -2; /* error */ if (fp->block_length == 0) return -1; /* end-of-file */ } c = ((unsigned char*)fp->uncompressed_block)[fp->block_offset++]; if (fp->block_offset == fp->block_length) { fp->block_address = htell(fp->fp); fp->block_offset = 0; fp->block_length = 0; } fp->uncompressed_address++; return c; } #ifndef kroundup32 #define kroundup32(x) (--(x), (x)|=(x)>>1, (x)|=(x)>>2, (x)|=(x)>>4, (x)|=(x)>>8, (x)|=(x)>>16, ++(x)) #endif int bgzf_getline(BGZF *fp, int delim, kstring_t *str) { int l, state = 0; unsigned char *buf = (unsigned char*)fp->uncompressed_block; str->l = 0; do { if (fp->block_offset >= fp->block_length) { if (bgzf_read_block(fp) != 0) { state = -2; break; } if (fp->block_length == 0) { state = -1; break; } } for (l = fp->block_offset; l < fp->block_length && buf[l] != delim; ++l); if (l < fp->block_length) state = 1; l -= fp->block_offset; if (str->l + l + 1 >= str->m) { str->m = str->l + l + 2; kroundup32(str->m); str->s = (char*)realloc(str->s, str->m); } memcpy(str->s + str->l, buf + fp->block_offset, l); str->l += l; fp->block_offset += l + 1; if (fp->block_offset >= fp->block_length) { fp->block_address = htell(fp->fp); fp->block_offset = 0; fp->block_length = 0; } } while (state == 0); if (str->l == 0 && state < 0) return state; fp->uncompressed_address += str->l; if ( delim=='\n' && str->l>0 && str->s[str->l-1]=='\r' ) str->l--; str->s[str->l] = 0; return str->l; } void bgzf_index_destroy(BGZF *fp) { if ( !fp->idx ) return; free(fp->idx->offs); free(fp->idx); fp->idx = NULL; fp->idx_build_otf = 0; } int bgzf_index_build_init(BGZF *fp) { bgzf_index_destroy(fp); fp->idx = (bgzidx_t*) calloc(1,sizeof(bgzidx_t)); if ( !fp->idx ) return -1; fp->idx_build_otf = 1; // build index on the fly return 0; } int bgzf_index_add_block(BGZF *fp) { fp->idx->noffs++; if ( fp->idx->noffs > fp->idx->moffs ) { fp->idx->moffs = fp->idx->noffs; kroundup32(fp->idx->moffs); fp->idx->offs = (bgzidx1_t*) realloc(fp->idx->offs, fp->idx->moffs*sizeof(bgzidx1_t)); if ( !fp->idx->offs ) return -1; } fp->idx->offs[ fp->idx->noffs-1 ].uaddr = fp->idx->ublock_addr; fp->idx->offs[ fp->idx->noffs-1 ].caddr = fp->block_address; return 0; } static inline int fwrite_uint64(uint64_t x, FILE *f) { if (ed_is_big()) x = ed_swap_8(x); if (fwrite(&x, sizeof x, 1, f) != 1) return -1; return 0; } int bgzf_index_dump(BGZF *fp, const char *bname, const char *suffix) { if (bgzf_flush(fp) != 0) return -1; assert(fp->idx); char *tmp = NULL; if ( suffix ) { int blen = strlen(bname); int slen = strlen(suffix); tmp = (char*) malloc(blen + slen + 1); if ( !tmp ) return -1; memcpy(tmp,bname,blen); memcpy(tmp+blen,suffix,slen+1); } FILE *idx = fopen(tmp?tmp:bname,"wb"); if ( tmp ) free(tmp); if ( !idx ) { if (hts_verbose > 1) { fprintf(stderr, "[E::%s] Error opening %s%s : %s\n", __func__, bname, suffix ? suffix : "", strerror(errno)); } return -1; } // Note that the index contains one extra record when indexing files opened // for reading. The terminating record is not present when opened for writing. // This is not a bug. int i; if (fwrite_uint64(fp->idx->noffs - 1, idx) < 0) goto fail; for (i=1; iidx->noffs; i++) { if (fwrite_uint64(fp->idx->offs[i].caddr, idx) < 0) goto fail; if (fwrite_uint64(fp->idx->offs[i].uaddr, idx) < 0) goto fail; } if (fclose(idx) < 0) { if (hts_verbose > 1) { fprintf(stderr, "[E::%s] Error on closing %s%s : %s\n", __func__, bname, suffix ? suffix : "", strerror(errno)); } return -1; } return 0; fail: if (hts_verbose > 1) { fprintf(stderr, "[E::%s] Error writing to %s%s : %s\n", __func__, bname, suffix ? suffix : "", strerror(errno)); } fclose(idx); return -1; } static inline int fread_uint64(uint64_t *xptr, FILE *f) { if (fread(xptr, sizeof *xptr, 1, f) != 1) return -1; if (ed_is_big()) ed_swap_8p(xptr); return 0; } int bgzf_index_load(BGZF *fp, const char *bname, const char *suffix) { char *tmp = NULL; if ( suffix ) { int blen = strlen(bname); int slen = strlen(suffix); tmp = (char*) malloc(blen + slen + 1); if ( !tmp ) return -1; memcpy(tmp,bname,blen); memcpy(tmp+blen,suffix,slen+1); } FILE *idx = fopen(tmp?tmp:bname,"rb"); if ( tmp ) free(tmp); if ( !idx ) { if (hts_verbose > 1) { fprintf(stderr, "[E::%s] Error opening %s%s : %s\n", __func__, bname, suffix ? suffix : "", strerror(errno)); } return -1; } fp->idx = (bgzidx_t*) calloc(1,sizeof(bgzidx_t)); if (fp->idx == NULL) goto fail; uint64_t x; if (fread_uint64(&x, idx) < 0) goto fail; fp->idx->noffs = fp->idx->moffs = x + 1; fp->idx->offs = (bgzidx1_t*) malloc(fp->idx->moffs*sizeof(bgzidx1_t)); if (fp->idx->offs == NULL) goto fail; fp->idx->offs[0].caddr = fp->idx->offs[0].uaddr = 0; int i; for (i=1; iidx->noffs; i++) { if (fread_uint64(&fp->idx->offs[i].caddr, idx) < 0) goto fail; if (fread_uint64(&fp->idx->offs[i].uaddr, idx) < 0) goto fail; } if (fclose(idx) != 0) goto fail; return 0; fail: if (hts_verbose > 1) { fprintf(stderr, "[E::%s] Error reading %s%s : %s\n", __func__, bname, suffix ? suffix : "", strerror(errno)); } fclose(idx); if (fp->idx) { free(fp->idx->offs); free(fp->idx); fp->idx = NULL; } return -1; } int bgzf_useek(BGZF *fp, long uoffset, int where) { if ( !fp->is_compressed ) { if (hseek(fp->fp, uoffset, SEEK_SET) < 0) { fp->errcode |= BGZF_ERR_IO; return -1; } fp->block_length = 0; // indicates current block has not been loaded fp->block_address = uoffset; fp->block_offset = 0; if (bgzf_read_block(fp) < 0) { fp->errcode |= BGZF_ERR_IO; return -1; } fp->uncompressed_address = uoffset; return 0; } if ( !fp->idx ) { fp->errcode |= BGZF_ERR_IO; return -1; } // binary search int ilo = 0, ihi = fp->idx->noffs - 1; while ( ilo<=ihi ) { int i = (ilo+ihi)*0.5; if ( uoffset < fp->idx->offs[i].uaddr ) ihi = i - 1; else if ( uoffset >= fp->idx->offs[i].uaddr ) ilo = i + 1; else break; } int i = ilo-1; if (hseek(fp->fp, fp->idx->offs[i].caddr, SEEK_SET) < 0) { fp->errcode |= BGZF_ERR_IO; return -1; } fp->block_length = 0; // indicates current block has not been loaded fp->block_address = fp->idx->offs[i].caddr; fp->block_offset = 0; if ( bgzf_read_block(fp) < 0 ) { fp->errcode |= BGZF_ERR_IO; return -1; } if ( uoffset - fp->idx->offs[i].uaddr > 0 ) { fp->block_offset = uoffset - fp->idx->offs[i].uaddr; assert( fp->block_offset <= fp->block_length ); // todo: skipped, unindexed, blocks } fp->uncompressed_address = uoffset; return 0; } long bgzf_utell(BGZF *fp) { return fp->uncompressed_address; // currently maintained only when reading } stacks-1.44/htslib/._bgzip.c000644 000765 000024 00000000265 12750743246 016500 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/bgzip.c000644 000765 000024 00000026422 12750743246 016266 0ustar00catchenstaff000000 000000 /* bgzip.c -- Block compression/decompression utility. Copyright (C) 2008, 2009 Broad Institute / Massachusetts Institute of Technology Copyright (C) 2010, 2013-2016 Genome Research Ltd. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notices and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include #include #include #include #include #include #include #include "htslib/bgzf.h" #include "htslib/hts.h" static const int WINDOW_SIZE = 64 * 1024; static void error(const char *format, ...) { va_list ap; va_start(ap, format); vfprintf(stderr, format, ap); va_end(ap); exit(EXIT_FAILURE); } static int confirm_overwrite(const char *fn) { int save_errno = errno; int ret = 0; if (isatty(STDIN_FILENO)) { char c; fprintf(stderr, "[bgzip] %s already exists; do you wish to overwrite (y or n)? ", fn); if (scanf("%c", &c) == 1 && (c == 'Y' || c == 'y')) ret = 1; } errno = save_errno; return ret; } static int bgzip_main_usage(void) { fprintf(stderr, "\n"); fprintf(stderr, "Version: %s\n", hts_version()); fprintf(stderr, "Usage: bgzip [OPTIONS] [FILE] ...\n"); fprintf(stderr, "Options:\n"); fprintf(stderr, " -b, --offset INT decompress at virtual file pointer (0-based uncompressed offset)\n"); fprintf(stderr, " -c, --stdout write on standard output, keep original files unchanged\n"); fprintf(stderr, " -d, --decompress decompress\n"); fprintf(stderr, " -f, --force overwrite files without asking\n"); fprintf(stderr, " -h, --help give this help\n"); fprintf(stderr, " -i, --index compress and create BGZF index\n"); fprintf(stderr, " -I, --index-name FILE name of BGZF index file [file.gz.gzi]\n"); fprintf(stderr, " -r, --reindex (re)index compressed file\n"); fprintf(stderr, " -s, --size INT decompress INT bytes (uncompressed size)\n"); fprintf(stderr, " -@, --threads INT number of compression threads to use [1]\n"); fprintf(stderr, "\n"); return 1; } int main(int argc, char **argv) { int c, compress, pstdout, is_forced, index = 0, reindex = 0; BGZF *fp; void *buffer; long start, end, size; char *index_fname = NULL; int threads = 1; static const struct option loptions[] = { {"help", no_argument, NULL, 'h'}, {"offset", required_argument, NULL, 'b'}, {"stdout", no_argument, NULL, 'c'}, {"decompress", no_argument, NULL, 'd'}, {"force", no_argument, NULL, 'f'}, {"index", no_argument, NULL, 'i'}, {"index-name", required_argument, NULL, 'I'}, {"reindex", no_argument, NULL, 'r'}, {"size", required_argument, NULL, 's'}, {"threads", required_argument, NULL, '@'}, {"version", no_argument, NULL, 1}, {NULL, 0, NULL, 0} }; compress = 1; pstdout = 0; start = 0; size = -1; end = -1; is_forced = 0; while((c = getopt_long(argc, argv, "cdh?fb:@:s:iI:r",loptions,NULL)) >= 0){ switch(c){ case 'd': compress = 0; break; case 'c': pstdout = 1; break; case 'b': start = atol(optarg); compress = 0; pstdout = 1; break; case 's': size = atol(optarg); pstdout = 1; break; case 'f': is_forced = 1; break; case 'i': index = 1; break; case 'I': index_fname = optarg; break; case 'r': reindex = 1; compress = 0; break; case '@': threads = atoi(optarg); break; case 1: printf( "bgzip (htslib) %s\n" "Copyright (C) 2016 Genome Research Ltd.\n", hts_version()); return EXIT_SUCCESS; case 'h': case '?': return bgzip_main_usage(); } } if (size >= 0) end = start + size; if (end >= 0 && end < start) { fprintf(stderr, "[bgzip] Illegal region: [%ld, %ld]\n", start, end); return 1; } if (compress == 1) { struct stat sbuf; int f_src = fileno(stdin); if ( argc>optind ) { if ( stat(argv[optind],&sbuf)<0 ) { fprintf(stderr, "[bgzip] %s: %s\n", strerror(errno), argv[optind]); return 1; } if ((f_src = open(argv[optind], O_RDONLY)) < 0) { fprintf(stderr, "[bgzip] %s: %s\n", strerror(errno), argv[optind]); return 1; } if (pstdout) fp = bgzf_open("-", "w"); else { char *name = malloc(strlen(argv[optind]) + 5); strcpy(name, argv[optind]); strcat(name, ".gz"); fp = bgzf_open(name, is_forced? "w" : "wx"); if (fp == NULL && errno == EEXIST && confirm_overwrite(name)) fp = bgzf_open(name, "w"); if (fp == NULL) { fprintf(stderr, "[bgzip] can't create %s: %s\n", name, strerror(errno)); free(name); return 1; } free(name); } } else if (!pstdout && isatty(fileno((FILE *)stdout)) ) return bgzip_main_usage(); else if ( index && !index_fname ) { fprintf(stderr, "[bgzip] Index file name expected when writing to stdout\n"); return 1; } else fp = bgzf_open("-", "w"); if (threads > 1) bgzf_mt(fp, threads, 256); if ( index ) bgzf_index_build_init(fp); buffer = malloc(WINDOW_SIZE); while ((c = read(f_src, buffer, WINDOW_SIZE)) > 0) if (bgzf_write(fp, buffer, c) < 0) error("Could not write %d bytes: Error %d\n", c, fp->errcode); if ( index ) { if (index_fname) { if (bgzf_index_dump(fp, index_fname, NULL) < 0) error("Could not write index to '%s'\n", index_fname); } else { if (bgzf_index_dump(fp, argv[optind], ".gz.gzi") < 0) error("Could not write index to '%s.gz.gzi'", argv[optind]); } } if (bgzf_close(fp) < 0) error("Close failed: Error %d", fp->errcode); if (argc > optind && !pstdout) unlink(argv[optind]); free(buffer); close(f_src); return 0; } else if ( reindex ) { if ( argc>optind ) { fp = bgzf_open(argv[optind], "r"); if ( !fp ) error("[bgzip] Could not open file: %s\n", argv[optind]); } else { if ( !index_fname ) error("[bgzip] Index file name expected when reading from stdin\n"); fp = bgzf_open("-", "r"); if ( !fp ) error("[bgzip] Could not read from stdin: %s\n", strerror(errno)); } buffer = malloc(BGZF_BLOCK_SIZE); bgzf_index_build_init(fp); int ret; while ( (ret=bgzf_read(fp, buffer, BGZF_BLOCK_SIZE))>0 ) ; free(buffer); if ( ret<0 ) error("Is the file gzipped or bgzipped? The latter is required for indexing.\n"); if ( index_fname ) { if (bgzf_index_dump(fp, index_fname, NULL) < 0) error("Could not write index to '%s'\n", index_fname); } else { if (bgzf_index_dump(fp, argv[optind], ".gzi") < 0) error("Could not write index to '%s.gzi'\n", argv[optind]); } if ( bgzf_close(fp)<0 ) error("Close failed: Error %d\n",fp->errcode); return 0; } else { struct stat sbuf; int f_dst; if ( argc>optind ) { if ( stat(argv[optind],&sbuf)<0 ) { fprintf(stderr, "[bgzip] %s: %s\n", strerror(errno), argv[optind]); return 1; } char *name; int len = strlen(argv[optind]); if ( strcmp(argv[optind]+len-3,".gz") ) { fprintf(stderr, "[bgzip] %s: unknown suffix -- ignored\n", argv[optind]); return 1; } fp = bgzf_open(argv[optind], "r"); if (fp == NULL) { fprintf(stderr, "[bgzip] Could not open file: %s\n", argv[optind]); return 1; } if (pstdout) { f_dst = fileno(stdout); } else { const int wrflags = O_WRONLY | O_CREAT | O_TRUNC; name = strdup(argv[optind]); name[strlen(name) - 3] = '\0'; f_dst = open(name, is_forced? wrflags : wrflags|O_EXCL, 0666); if (f_dst < 0 && errno == EEXIST && confirm_overwrite(name)) f_dst = open(name, wrflags, 0666); if (f_dst < 0) { fprintf(stderr, "[bgzip] can't create %s: %s\n", name, strerror(errno)); free(name); return 1; } free(name); } } else if (!pstdout && isatty(fileno((FILE *)stdin)) ) return bgzip_main_usage(); else { f_dst = fileno(stdout); fp = bgzf_open("-", "r"); if (fp == NULL) { fprintf(stderr, "[bgzip] Could not read from stdin: %s\n", strerror(errno)); return 1; } } buffer = malloc(WINDOW_SIZE); if ( start>0 ) { if ( bgzf_index_load(fp, argv[optind], ".gzi") < 0 ) error("Could not load index: %s.gzi\n", argv[optind]); if ( bgzf_useek(fp, start, SEEK_SET) < 0 ) error("Could not seek to %d-th (uncompressd) byte\n", start); } while (1) { if (end < 0) c = bgzf_read(fp, buffer, WINDOW_SIZE); else c = bgzf_read(fp, buffer, (end - start > WINDOW_SIZE)? WINDOW_SIZE:(end - start)); if (c == 0) break; if (c < 0) error("Could not read %d bytes: Error %d\n", (end - start > WINDOW_SIZE)? WINDOW_SIZE:(end - start), fp->errcode); start += c; if ( write(f_dst, buffer, c) != c ) error("Could not write %d bytes\n", c); if (end >= 0 && start >= end) break; } free(buffer); if (bgzf_close(fp) < 0) error("Close failed: Error %d\n",fp->errcode); if (!pstdout) unlink(argv[optind]); return 0; } return 0; } stacks-1.44/htslib/._cram000755 000765 000024 00000000265 13001012452 016043 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/000755 000765 000024 00000000000 13001012452 015675 5ustar00catchenstaff000000 000000 stacks-1.44/htslib/._faidx.c000644 000765 000024 00000000265 12750743246 016460 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/faidx.c000644 000765 000024 00000035426 12750743246 016252 0ustar00catchenstaff000000 000000 /* faidx.c -- FASTA random access. Copyright (C) 2008, 2009, 2013-2016 Genome Research Ltd. Portions copyright (C) 2011 Broad Institute. Author: Heng Li Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include #include #include #include #include "htslib/bgzf.h" #include "htslib/faidx.h" #include "htslib/hfile.h" #include "htslib/khash.h" #include "htslib/kstring.h" typedef struct { int32_t line_len, line_blen; int64_t len; uint64_t offset; } faidx1_t; KHASH_MAP_INIT_STR(s, faidx1_t) struct __faidx_t { BGZF *bgzf; int n, m; char **name; khash_t(s) *hash; }; #ifndef kroundup32 #define kroundup32(x) (--(x), (x)|=(x)>>1, (x)|=(x)>>2, (x)|=(x)>>4, (x)|=(x)>>8, (x)|=(x)>>16, ++(x)) #endif static inline int fai_insert_index(faidx_t *idx, const char *name, int len, int line_len, int line_blen, uint64_t offset) { if (!name) { fprintf(stderr, "[fai_build_core] malformed line\n"); return -1; } char *name_key = strdup(name); int absent; khint_t k = kh_put(s, idx->hash, name_key, &absent); faidx1_t *v = &kh_value(idx->hash, k); if (! absent) { fprintf(stderr, "[fai_build_core] ignoring duplicate sequence \"%s\" at byte offset %"PRIu64"\n", name, offset); free(name_key); return 0; } if (idx->n == idx->m) { char **tmp; idx->m = idx->m? idx->m<<1 : 16; if (!(tmp = (char**)realloc(idx->name, sizeof(char*) * idx->m))) { fprintf(stderr, "[fai_build_core] out of memory\n"); return -1; } idx->name = tmp; } idx->name[idx->n++] = name_key; v->len = len; v->line_len = line_len; v->line_blen = line_blen; v->offset = offset; return 0; } faidx_t *fai_build_core(BGZF *bgzf) { kstring_t name = { 0, 0, NULL }; int c; int line_len, line_blen, state; int l1, l2; faidx_t *idx; uint64_t offset; int64_t len; idx = (faidx_t*)calloc(1, sizeof(faidx_t)); idx->hash = kh_init(s); len = line_len = line_blen = -1; state = 0; l1 = l2 = -1; offset = 0; while ( (c=bgzf_getc(bgzf))>=0 ) { if (c == '\n') { // an empty line if (state == 1) { offset = bgzf_utell(bgzf); continue; } else if ((state == 0 && len < 0) || state == 2) continue; else if (state == 0) { state = 2; continue; } } if (c == '>') { // fasta header if (len >= 0) { if (fai_insert_index(idx, name.s, len, line_len, line_blen, offset) != 0) goto fail; } name.l = 0; while ((c = bgzf_getc(bgzf)) >= 0) if (! isspace(c)) kputc_(c, &name); else if (name.l > 0 || c == '\n') break; kputsn("", 0, &name); if ( c<0 ) { fprintf(stderr, "[fai_build_core] the last entry has no sequence\n"); goto fail; } if (c != '\n') while ( (c=bgzf_getc(bgzf))>=0 && c != '\n'); state = 1; len = 0; offset = bgzf_utell(bgzf); } else { if (state == 3) { fprintf(stderr, "[fai_build_core] inlined empty line is not allowed in sequence '%s'.\n", name.s); goto fail; } if (state == 2) state = 3; l1 = l2 = 0; do { ++l1; if (isgraph(c)) ++l2; } while ( (c=bgzf_getc(bgzf))>=0 && c != '\n'); if (state == 3 && l2) { fprintf(stderr, "[fai_build_core] different line length in sequence '%s'.\n", name.s); goto fail; } ++l1; len += l2; if (state == 1) line_len = l1, line_blen = l2, state = 0; else if (state == 0) { if (l1 != line_len || l2 != line_blen) state = 2; } } } if (len >= 0) { if (fai_insert_index(idx, name.s, len, line_len, line_blen, offset) != 0) goto fail; } else { goto fail; } free(name.s); return idx; fail: free(name.s); fai_destroy(idx); return NULL; } void fai_save(const faidx_t *fai, FILE *fp) { khint_t k; int i; for (i = 0; i < fai->n; ++i) { faidx1_t x; k = kh_get(s, fai->hash, fai->name[i]); x = kh_value(fai->hash, k); #ifdef _WIN32 fprintf(fp, "%s\t%d\t%ld\t%d\t%d\n", fai->name[i], (int)x.len, (long)x.offset, (int)x.line_blen, (int)x.line_len); #else fprintf(fp, "%s\t%d\t%lld\t%d\t%d\n", fai->name[i], (int)x.len, (long long)x.offset, (int)x.line_blen, (int)x.line_len); #endif } } static faidx_t *fai_read(FILE *fp, const char *fname) { faidx_t *fai; char *buf, *p; int len, line_len, line_blen; #ifdef _WIN32 long offset; #else long long offset; #endif fai = (faidx_t*)calloc(1, sizeof(faidx_t)); fai->hash = kh_init(s); buf = (char*)calloc(0x10000, 1); while (fgets(buf, 0x10000, fp)) { for (p = buf; *p && isgraph(*p); ++p); *p = 0; ++p; #ifdef _WIN32 sscanf(p, "%d%ld%d%d", &len, &offset, &line_blen, &line_len); #else sscanf(p, "%d%lld%d%d", &len, &offset, &line_blen, &line_len); #endif if (fai_insert_index(fai, buf, len, line_len, line_blen, offset) != 0) { free(buf); return NULL; } } free(buf); if (ferror(fp)) { fprintf(stderr, "[fai_load] error while reading \"%s\": %s\n", fname, strerror(errno)); fai_destroy(fai); return NULL; } return fai; } void fai_destroy(faidx_t *fai) { int i; for (i = 0; i < fai->n; ++i) free(fai->name[i]); free(fai->name); kh_destroy(s, fai->hash); if (fai->bgzf) bgzf_close(fai->bgzf); free(fai); } int fai_build(const char *fn) { char *str; BGZF *bgzf; FILE *fp; faidx_t *fai; str = (char*)calloc(strlen(fn) + 5, 1); sprintf(str, "%s.fai", fn); bgzf = bgzf_open(fn, "r"); if ( !bgzf ) { fprintf(stderr, "[fai_build] fail to open the FASTA file %s\n",fn); free(str); return -1; } if ( bgzf->is_compressed ) bgzf_index_build_init(bgzf); fai = fai_build_core(bgzf); if ( !fai ) { if ( bgzf->is_compressed && bgzf->is_gzip ) fprintf(stderr,"Cannot index files compressed with gzip, please use bgzip\n"); bgzf_close(bgzf); free(str); return -1; } if ( bgzf->is_compressed ) { if (bgzf_index_dump(bgzf, fn, ".gzi") < 0) { fprintf(stderr, "[fai_build] fail to make bgzf index %s.gzi\n", fn); fai_destroy(fai); free(str); return -1; } } if (bgzf_close(bgzf) < 0) { fprintf(stderr, "[fai_build] Error on closing %s\n", fn); fai_destroy(fai); free(str); return -1; } fp = fopen(str, "wb"); if ( !fp ) { fprintf(stderr, "[fai_build] fail to write FASTA index %s\n",str); fai_destroy(fai); free(str); return -1; } fai_save(fai, fp); fclose(fp); free(str); fai_destroy(fai); return 0; } static FILE *download_and_open(const char *fn) { const int buf_size = 1 * 1024 * 1024; uint8_t *buf; FILE *fp; hFILE *fp_remote; const char *url = fn; const char *p; int l = strlen(fn); for (p = fn + l - 1; p >= fn; --p) if (*p == '/') break; fn = p + 1; // First try to open a local copy fp = fopen(fn, "r"); if (fp) return fp; // If failed, download from remote and open fp_remote = hopen(url, "rb"); if (fp_remote == 0) { fprintf(stderr, "[download_from_remote] fail to open remote file %s\n",url); return NULL; } if ((fp = fopen(fn, "wb")) == 0) { fprintf(stderr, "[download_from_remote] fail to create file in the working directory %s\n",fn); hclose_abruptly(fp_remote); return NULL; } buf = (uint8_t*)calloc(buf_size, 1); while ((l = hread(fp_remote, buf, buf_size)) > 0) fwrite(buf, 1, l, fp); free(buf); fclose(fp); if (hclose(fp_remote) != 0) fprintf(stderr, "[download_from_remote] fail to close remote file %s\n", url); return fopen(fn, "r"); } faidx_t *fai_load(const char *fn) { char *str; FILE *fp; faidx_t *fai; str = (char*)calloc(strlen(fn) + 5, 1); sprintf(str, "%s.fai", fn); if (hisremote(str)) { fp = download_and_open(str); if ( !fp ) { fprintf(stderr, "[fai_load] failed to open remote FASTA index %s\n", str); free(str); return 0; } } else fp = fopen(str, "rb"); if (fp == 0) { fprintf(stderr, "[fai_load] build FASTA index.\n"); if (fai_build(fn) < 0) { free(str); return 0; } fp = fopen(str, "rb"); if (fp == 0) { fprintf(stderr, "[fai_load] failed to open FASTA index: %s\n", strerror(errno)); free(str); return 0; } } fai = fai_read(fp, str); fclose(fp); free(str); if (fai == NULL) { return NULL; } fai->bgzf = bgzf_open(fn, "rb"); if (fai->bgzf == 0) { fprintf(stderr, "[fai_load] fail to open FASTA file.\n"); return 0; } if ( fai->bgzf->is_compressed==1 ) { if ( bgzf_index_load(fai->bgzf, fn, ".gzi") < 0 ) { fprintf(stderr, "[fai_load] failed to load .gzi index: %s[.gzi]\n", fn); fai_destroy(fai); return NULL; } } return fai; } char *fai_fetch(const faidx_t *fai, const char *str, int *len) { char *s; int c, i, l, k, name_end; khiter_t iter; faidx1_t val; khash_t(s) *h; int beg, end; beg = end = -1; h = fai->hash; name_end = l = strlen(str); s = (char*)malloc(l+1); // remove space for (i = k = 0; i < l; ++i) if (!isspace(str[i])) s[k++] = str[i]; s[k] = 0; l = k; // determine the sequence name for (i = l - 1; i >= 0; --i) if (s[i] == ':') break; // look for colon from the end if (i >= 0) name_end = i; if (name_end < l) { // check if this is really the end int n_hyphen = 0; for (i = name_end + 1; i < l; ++i) { if (s[i] == '-') ++n_hyphen; else if (!isdigit(s[i]) && s[i] != ',') break; } if (i < l || n_hyphen > 1) name_end = l; // malformated region string; then take str as the name s[name_end] = 0; iter = kh_get(s, h, s); if (iter == kh_end(h)) { // cannot find the sequence name iter = kh_get(s, h, str); // try str as the name if (iter == kh_end(h)) { *len = 0; free(s); return 0; } else s[name_end] = ':', name_end = l; } } else iter = kh_get(s, h, str); if(iter == kh_end(h)) { fprintf(stderr, "[fai_fetch] Warning - Reference %s not found in FASTA file, returning empty sequence\n", str); free(s); *len = -2; return 0; }; val = kh_value(h, iter); // parse the interval if (name_end < l) { for (i = k = name_end + 1; i < l; ++i) if (s[i] != ',') s[k++] = s[i]; s[k] = 0; beg = atoi(s + name_end + 1); for (i = name_end + 1; i != k; ++i) if (s[i] == '-') break; end = i < k? atoi(s + i + 1) : val.len; if (beg > 0) --beg; } else beg = 0, end = val.len; if (beg >= val.len) beg = val.len; if (end >= val.len) end = val.len; if (beg > end) beg = end; free(s); // now retrieve the sequence int ret = bgzf_useek(fai->bgzf, val.offset + beg / val.line_blen * val.line_len + beg % val.line_blen, SEEK_SET); if ( ret<0 ) { *len = -1; fprintf(stderr, "[fai_fetch] Error: fai_fetch failed. (Seeking in a compressed, .gzi unindexed, file?)\n"); return NULL; } l = 0; s = (char*)malloc(end - beg + 2); while ( (c=bgzf_getc(fai->bgzf))>=0 && l < end - beg ) if (isgraph(c)) s[l++] = c; s[l] = '\0'; *len = l; return s; } int faidx_fetch_nseq(const faidx_t *fai) { return fai->n; } int faidx_nseq(const faidx_t *fai) { return fai->n; } const char *faidx_iseq(const faidx_t *fai, int i) { return fai->name[i]; } int faidx_seq_len(const faidx_t *fai, const char *seq) { khint_t k = kh_get(s, fai->hash, seq); if ( k == kh_end(fai->hash) ) return -1; return kh_val(fai->hash, k).len; } char *faidx_fetch_seq(const faidx_t *fai, const char *c_name, int p_beg_i, int p_end_i, int *len) { int l, c; khiter_t iter; faidx1_t val; char *seq=NULL; // Adjust position iter = kh_get(s, fai->hash, c_name); if (iter == kh_end(fai->hash)) { *len = -2; fprintf(stderr, "[fai_fetch_seq] The sequence \"%s\" not found\n", c_name); return NULL; } val = kh_value(fai->hash, iter); if(p_end_i < p_beg_i) p_beg_i = p_end_i; if(p_beg_i < 0) p_beg_i = 0; else if(val.len <= p_beg_i) p_beg_i = val.len - 1; if(p_end_i < 0) p_end_i = 0; else if(val.len <= p_end_i) p_end_i = val.len - 1; // Now retrieve the sequence int ret = bgzf_useek(fai->bgzf, val.offset + p_beg_i / val.line_blen * val.line_len + p_beg_i % val.line_blen, SEEK_SET); if ( ret<0 ) { *len = -1; fprintf(stderr, "[fai_fetch_seq] Error: fai_fetch failed. (Seeking in a compressed, .gzi unindexed, file?)\n"); return NULL; } l = 0; seq = (char*)malloc(p_end_i - p_beg_i + 2); while ( (c=bgzf_getc(fai->bgzf))>=0 && l < p_end_i - p_beg_i + 1) if (isgraph(c)) seq[l++] = c; seq[l] = '\0'; *len = l; return seq; } int faidx_has_seq(const faidx_t *fai, const char *seq) { khiter_t iter = kh_get(s, fai->hash, seq); if (iter == kh_end(fai->hash)) return 0; return 1; } stacks-1.44/htslib/._hfile.c000644 000765 000024 00000000265 12750743246 016454 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/hfile.c000644 000765 000024 00000051071 12750743246 016240 0ustar00catchenstaff000000 000000 /* hfile.c -- buffered low-level input/output streams. Copyright (C) 2013-2016 Genome Research Ltd. Author: John Marshall Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include #include #include #include "htslib/hfile.h" #include "hfile_internal.h" #ifndef ENOTSUP #define ENOTSUP EINVAL #endif #ifndef EOVERFLOW #define EOVERFLOW ERANGE #endif #ifndef EPROTONOSUPPORT #define EPROTONOSUPPORT ENOSYS #endif /* hFILE fields are used as follows: char *buffer; // Pointer to the start of the I/O buffer char *begin; // First not-yet-read character / unused position char *end; // First unfilled/unfillable position char *limit; // Pointer to the first position past the buffer const hFILE_backend *backend; // Methods to refill/flush I/O buffer off_t offset; // Offset within the stream of buffer position 0 unsigned at_eof:1;// For reading, whether EOF has been seen int has_errno; // Error number from the last failure on this stream For reading, begin is the first unread character in the buffer and end is the first unfilled position: -----------ABCDEFGHIJKLMNO--------------- ^buffer ^begin ^end ^limit For writing, begin is the first unused position and end is unused so remains equal to buffer: ABCDEFGHIJKLMNOPQRSTUVWXYZ--------------- ^buffer ^begin ^limit ^end Thus if begin > end then there is a non-empty write buffer, if begin < end then there is a non-empty read buffer, and if begin == end then both buffers are empty. In all cases, the stream's file position indicator corresponds to the position pointed to by begin. */ hFILE *hfile_init(size_t struct_size, const char *mode, size_t capacity) { hFILE *fp = (hFILE *) malloc(struct_size); if (fp == NULL) goto error; if (capacity == 0) capacity = 32768; // FIXME For now, clamp input buffer sizes so mpileup doesn't eat memory if (strchr(mode, 'r') && capacity > 32768) capacity = 32768; fp->buffer = (char *) malloc(capacity); if (fp->buffer == NULL) goto error; fp->begin = fp->end = fp->buffer; fp->limit = &fp->buffer[capacity]; fp->offset = 0; fp->at_eof = 0; fp->has_errno = 0; return fp; error: hfile_destroy(fp); return NULL; } void hfile_destroy(hFILE *fp) { int save = errno; if (fp) free(fp->buffer); free(fp); errno = save; } static inline int writebuffer_is_nonempty(hFILE *fp) { return fp->begin > fp->end; } /* Refills the read buffer from the backend (once, so may only partially fill the buffer), returning the number of additional characters read (which might be 0), or negative when an error occurred. */ static ssize_t refill_buffer(hFILE *fp) { ssize_t n; // Move any unread characters to the start of the buffer if (fp->begin > fp->buffer) { fp->offset += fp->begin - fp->buffer; memmove(fp->buffer, fp->begin, fp->end - fp->begin); fp->end = &fp->buffer[fp->end - fp->begin]; fp->begin = fp->buffer; } // Read into the available buffer space at fp->[end,limit) if (fp->at_eof || fp->end == fp->limit) n = 0; else { n = fp->backend->read(fp, fp->end, fp->limit - fp->end); if (n < 0) { fp->has_errno = errno; return n; } else if (n == 0) fp->at_eof = 1; } fp->end += n; return n; } /* Called only from hgetc(), when our buffer is empty. */ int hgetc2(hFILE *fp) { return (refill_buffer(fp) > 0)? (unsigned char) *(fp->begin++) : EOF; } ssize_t hpeek(hFILE *fp, void *buffer, size_t nbytes) { size_t n = fp->end - fp->begin; while (n < nbytes) { ssize_t ret = refill_buffer(fp); if (ret < 0) return ret; else if (ret == 0) break; else n += ret; } if (n > nbytes) n = nbytes; memcpy(buffer, fp->begin, n); return n; } /* Called only from hread(); when called, our buffer is empty and nread bytes have already been placed in the destination buffer. */ ssize_t hread2(hFILE *fp, void *destv, size_t nbytes, size_t nread) { const size_t capacity = fp->limit - fp->buffer; char *dest = (char *) destv; dest += nread, nbytes -= nread; // Read large requests directly into the destination buffer while (nbytes * 2 >= capacity && !fp->at_eof) { ssize_t n = fp->backend->read(fp, dest, nbytes); if (n < 0) { fp->has_errno = errno; return n; } else if (n == 0) fp->at_eof = 1; fp->offset += n; dest += n, nbytes -= n; nread += n; } while (nbytes > 0 && !fp->at_eof) { size_t n; ssize_t ret = refill_buffer(fp); if (ret < 0) return ret; n = fp->end - fp->begin; if (n > nbytes) n = nbytes; memcpy(dest, fp->begin, n); fp->begin += n; dest += n, nbytes -= n; nread += n; } return nread; } /* Flushes the write buffer, fp->[buffer,begin), out through the backend returning 0 on success or negative if an error occurred. */ static ssize_t flush_buffer(hFILE *fp) { const char *buffer = fp->buffer; while (buffer < fp->begin) { ssize_t n = fp->backend->write(fp, buffer, fp->begin - buffer); if (n < 0) { fp->has_errno = errno; return n; } buffer += n; fp->offset += n; } fp->begin = fp->buffer; // Leave the buffer empty return 0; } int hflush(hFILE *fp) { if (flush_buffer(fp) < 0) return EOF; if (fp->backend->flush) { if (fp->backend->flush(fp) < 0) { fp->has_errno = errno; return EOF; } } return 0; } /* Called only from hputc(), when our buffer is already full. */ int hputc2(int c, hFILE *fp) { if (flush_buffer(fp) < 0) return EOF; *(fp->begin++) = c; return c; } /* Called only from hwrite() and hputs2(); when called, our buffer is full and ncopied bytes from the source have already been copied to our buffer. */ ssize_t hwrite2(hFILE *fp, const void *srcv, size_t totalbytes, size_t ncopied) { const char *src = (const char *) srcv; ssize_t ret; const size_t capacity = fp->limit - fp->buffer; size_t remaining = totalbytes - ncopied; src += ncopied; ret = flush_buffer(fp); if (ret < 0) return ret; // Write large blocks out directly from the source buffer while (remaining * 2 >= capacity) { ssize_t n = fp->backend->write(fp, src, remaining); if (n < 0) { fp->has_errno = errno; return n; } fp->offset += n; src += n, remaining -= n; } // Just buffer any remaining characters memcpy(fp->begin, src, remaining); fp->begin += remaining; return totalbytes; } /* Called only from hputs(), when our buffer is already full. */ int hputs2(const char *text, size_t totalbytes, size_t ncopied, hFILE *fp) { return (hwrite2(fp, text, totalbytes, ncopied) >= 0)? 0 : EOF; } off_t hseek(hFILE *fp, off_t offset, int whence) { off_t curpos, pos; if (writebuffer_is_nonempty(fp)) { int ret = flush_buffer(fp); if (ret < 0) return ret; } curpos = htell(fp); // Relative offsets are given relative to the hFILE's stream position, // which may differ from the backend's physical position due to buffering // read-ahead. Correct for this by converting to an absolute position. if (whence == SEEK_CUR) { if (curpos + offset < 0) { // Either a negative offset resulted in a position before the // start of the file, or we overflowed when given a positive offset fp->has_errno = errno = (offset < 0)? EINVAL : EOVERFLOW; return -1; } whence = SEEK_SET; offset = curpos + offset; } // TODO Avoid seeking if the desired position is within our read buffer pos = fp->backend->seek(fp, offset, whence); if (pos < 0) { fp->has_errno = errno; return pos; } // Seeking succeeded, so discard any non-empty read buffer fp->begin = fp->end = fp->buffer; fp->at_eof = 0; fp->offset = pos; return pos; } int hclose(hFILE *fp) { int err = fp->has_errno; if (writebuffer_is_nonempty(fp) && hflush(fp) < 0) err = fp->has_errno; if (fp->backend->close(fp) < 0) err = errno; hfile_destroy(fp); if (err) { errno = err; return EOF; } else return 0; } void hclose_abruptly(hFILE *fp) { int save = errno; if (fp->backend->close(fp) < 0) { /* Ignore subsequent errors */ } hfile_destroy(fp); errno = save; } /*************************** * File descriptor backend * ***************************/ #ifndef _WIN32 #include #include #define HAVE_STRUCT_STAT_ST_BLKSIZE #else #include #define HAVE_CLOSESOCKET #define HAVE_SETMODE #endif #include #include /* For Unix, it doesn't matter whether a file descriptor is a socket. However Windows insists on send()/recv() and its own closesocket() being used when fd happens to be a socket. */ typedef struct { hFILE base; int fd; unsigned is_socket:1; } hFILE_fd; static ssize_t fd_read(hFILE *fpv, void *buffer, size_t nbytes) { hFILE_fd *fp = (hFILE_fd *) fpv; ssize_t n; do { n = fp->is_socket? recv(fp->fd, buffer, nbytes, 0) : read(fp->fd, buffer, nbytes); } while (n < 0 && errno == EINTR); return n; } static ssize_t fd_write(hFILE *fpv, const void *buffer, size_t nbytes) { hFILE_fd *fp = (hFILE_fd *) fpv; ssize_t n; do { n = fp->is_socket? send(fp->fd, buffer, nbytes, 0) : write(fp->fd, buffer, nbytes); } while (n < 0 && errno == EINTR); return n; } static off_t fd_seek(hFILE *fpv, off_t offset, int whence) { hFILE_fd *fp = (hFILE_fd *) fpv; return lseek(fp->fd, offset, whence); } static int fd_flush(hFILE *fpv) { hFILE_fd *fp = (hFILE_fd *) fpv; int ret; do { #ifdef HAVE_FDATASYNC ret = fdatasync(fp->fd); #else ret = fsync(fp->fd); #endif // Ignore invalid-for-fsync(2) errors due to being, e.g., a pipe, // and operation-not-supported errors (Mac OS X) if (ret < 0 && (errno == EINVAL || errno == ENOTSUP)) ret = 0; } while (ret < 0 && errno == EINTR); return ret; } static int fd_close(hFILE *fpv) { hFILE_fd *fp = (hFILE_fd *) fpv; int ret; do { #ifdef HAVE_CLOSESOCKET ret = fp->is_socket? closesocket(fp->fd) : close(fp->fd); #else ret = close(fp->fd); #endif } while (ret < 0 && errno == EINTR); return ret; } static const struct hFILE_backend fd_backend = { fd_read, fd_write, fd_seek, fd_flush, fd_close }; static size_t blksize(int fd) { #ifdef HAVE_STRUCT_STAT_ST_BLKSIZE struct stat sbuf; if (fstat(fd, &sbuf) != 0) return 0; return sbuf.st_blksize; #else return 0; #endif } static hFILE *hopen_fd(const char *filename, const char *mode) { hFILE_fd *fp = NULL; int fd = open(filename, hfile_oflags(mode), 0666); if (fd < 0) goto error; fp = (hFILE_fd *) hfile_init(sizeof (hFILE_fd), mode, blksize(fd)); if (fp == NULL) goto error; fp->fd = fd; fp->is_socket = 0; fp->base.backend = &fd_backend; return &fp->base; error: if (fd >= 0) { int save = errno; (void) close(fd); errno = save; } hfile_destroy((hFILE *) fp); return NULL; } hFILE *hdopen(int fd, const char *mode) { hFILE_fd *fp = (hFILE_fd*) hfile_init(sizeof (hFILE_fd), mode, blksize(fd)); if (fp == NULL) return NULL; fp->fd = fd; fp->is_socket = (strchr(mode, 's') != NULL); fp->base.backend = &fd_backend; return &fp->base; } static hFILE *hopen_fd_fileuri(const char *url, const char *mode) { if (strncmp(url, "file://localhost/", 17) == 0) url += 16; else if (strncmp(url, "file:///", 8) == 0) url += 7; else { errno = EPROTONOSUPPORT; return NULL; } return hopen_fd(url, mode); } static hFILE *hopen_fd_stdinout(const char *mode) { int fd = (strchr(mode, 'r') != NULL)? STDIN_FILENO : STDOUT_FILENO; #if defined HAVE_SETMODE && defined O_BINARY if (setmode(fd, O_BINARY) < 0) return NULL; #endif return hdopen(fd, mode); } int hfile_oflags(const char *mode) { int rdwr = 0, flags = 0; const char *s; for (s = mode; *s; s++) switch (*s) { case 'r': rdwr = O_RDONLY; break; case 'w': rdwr = O_WRONLY; flags |= O_CREAT | O_TRUNC; break; case 'a': rdwr = O_WRONLY; flags |= O_CREAT | O_APPEND; break; case '+': rdwr = O_RDWR; break; #ifdef O_CLOEXEC case 'e': flags |= O_CLOEXEC; break; #endif #ifdef O_EXCL case 'x': flags |= O_EXCL; break; #endif default: break; } #ifdef O_BINARY flags |= O_BINARY; #endif return rdwr | flags; } /********************* * In-memory backend * *********************/ typedef struct { hFILE base; const char *buffer; size_t length, pos; } hFILE_mem; static ssize_t mem_read(hFILE *fpv, void *buffer, size_t nbytes) { hFILE_mem *fp = (hFILE_mem *) fpv; size_t avail = fp->length - fp->pos; if (nbytes > avail) nbytes = avail; memcpy(buffer, fp->buffer + fp->pos, nbytes); fp->pos += nbytes; return nbytes; } static off_t mem_seek(hFILE *fpv, off_t offset, int whence) { hFILE_mem *fp = (hFILE_mem *) fpv; size_t absoffset = (offset >= 0)? offset : -offset; size_t origin; switch (whence) { case SEEK_SET: origin = 0; break; case SEEK_CUR: origin = fp->pos; break; case SEEK_END: origin = fp->length; break; default: errno = EINVAL; return -1; } if ((offset < 0 && absoffset > origin) || (offset >= 0 && absoffset > fp->length - origin)) { errno = EINVAL; return -1; } fp->pos = origin + offset; return fp->pos; } static int mem_close(hFILE *fpv) { return 0; } static const struct hFILE_backend mem_backend = { mem_read, NULL, mem_seek, NULL, mem_close }; static hFILE *hopen_mem(const char *data, const char *mode) { if (strncmp(data, "data:", 5) == 0) data += 5; // TODO Implement write modes, which will require memory allocation if (strchr(mode, 'r') == NULL) { errno = EINVAL; return NULL; } hFILE_mem *fp = (hFILE_mem *) hfile_init(sizeof (hFILE_mem), mode, 0); if (fp == NULL) return NULL; fp->buffer = data; fp->length = strlen(data); fp->pos = 0; fp->base.backend = &mem_backend; return &fp->base; } /***************************************** * Plugin and hopen() backend dispatcher * *****************************************/ #include #include "hts_internal.h" #include "htslib/khash.h" KHASH_MAP_INIT_STR(scheme_string, const struct hFILE_scheme_handler *); static khash_t(scheme_string) *schemes = NULL; struct hFILE_plugin_list { struct hFILE_plugin plugin; struct hFILE_plugin_list *next; }; static struct hFILE_plugin_list *plugins = NULL; static pthread_mutex_t plugins_lock = PTHREAD_MUTEX_INITIALIZER; static void hfile_exit() { pthread_mutex_lock(&plugins_lock); kh_destroy(scheme_string, schemes); while (plugins != NULL) { struct hFILE_plugin_list *p = plugins; if (p->plugin.destroy) p->plugin.destroy(); #ifdef ENABLE_PLUGINS if (p->plugin.obj) close_plugin(p->plugin.obj); #endif plugins = p->next; free(p); } pthread_mutex_unlock(&plugins_lock); pthread_mutex_destroy(&plugins_lock); } void hfile_add_scheme_handler(const char *scheme, const struct hFILE_scheme_handler *handler) { int absent; khint_t k = kh_put(scheme_string, schemes, scheme, &absent); if (absent || handler->priority > kh_value(schemes, k)->priority) { kh_value(schemes, k) = handler; } } static int init_add_plugin(void *obj, int (*init)(struct hFILE_plugin *), const char *pluginname) { struct hFILE_plugin_list *p = malloc (sizeof (struct hFILE_plugin_list)); if (p == NULL) abort(); p->plugin.api_version = 1; p->plugin.obj = obj; p->plugin.name = NULL; p->plugin.destroy = NULL; int ret = (*init)(&p->plugin); if (ret != 0) { if (hts_verbose >= 4) fprintf(stderr, "[W::load_hfile_plugins] " "initialisation failed for plugin \"%s\": %d\n", pluginname, ret); free(p); return ret; } if (hts_verbose >= 5) fprintf(stderr, "[M::load_hfile_plugins] loaded \"%s\"\n", pluginname); p->next = plugins, plugins = p; return 0; } static void load_hfile_plugins() { static const struct hFILE_scheme_handler data = { hopen_mem, hfile_always_local, "built-in", 80 }, file = { hopen_fd_fileuri, hfile_always_local, "built-in", 80 }; schemes = kh_init(scheme_string); if (schemes == NULL) abort(); hfile_add_scheme_handler("data", &data); hfile_add_scheme_handler("file", &file); init_add_plugin(NULL, hfile_plugin_init_net, "knetfile"); #ifdef ENABLE_PLUGINS struct hts_path_itr path; const char *pluginname; hts_path_itr_setup(&path, NULL, NULL, "hfile_", 6, NULL, 0); while ((pluginname = hts_path_itr_next(&path)) != NULL) { void *obj; int (*init)(struct hFILE_plugin *) = (int (*)(struct hFILE_plugin *)) load_plugin(&obj, pluginname, "hfile_plugin_init"); if (init) { if (init_add_plugin(obj, init, pluginname) != 0) close_plugin(obj); } } #else #ifdef HAVE_IRODS init_add_plugin(NULL, hfile_plugin_init_irods, "iRODS"); #endif #ifdef HAVE_LIBCURL init_add_plugin(NULL, hfile_plugin_init_libcurl, "libcurl"); #endif #endif // In the unlikely event atexit() fails, it's better to succeed here and // carry on; then eventually when the program exits, we'll merely close // down the plugins uncleanly, as if we had aborted. (void) atexit(hfile_exit); } /* A filename like "foo:bar" in which we don't recognise the scheme is either an ordinary file or an indication of a missing or broken plugin. Try to open it as an ordinary file; but if there's no such file, set errno distinctively to make the plugin issue apparent. */ static hFILE *hopen_unknown_scheme(const char *fname, const char *mode) { hFILE *fp = hopen_fd(fname, mode); if (fp == NULL && errno == ENOENT) errno = EPROTONOSUPPORT; return fp; } /* Returns the appropriate handler, or NULL if the string isn't an URL. */ static const struct hFILE_scheme_handler *find_scheme_handler(const char *s) { static const struct hFILE_scheme_handler unknown_scheme = { hopen_unknown_scheme, hfile_always_local, "built-in", 0 }; char scheme[12]; int i; for (i = 0; i < sizeof scheme; i++) if (isalnum(s[i]) || s[i] == '+' || s[i] == '-' || s[i] == '.') scheme[i] = tolower(s[i]); else if (s[i] == ':') break; else return NULL; if (i == 0 || i >= sizeof scheme) return NULL; scheme[i] = '\0'; pthread_mutex_lock(&plugins_lock); if (! schemes) load_hfile_plugins(); pthread_mutex_unlock(&plugins_lock); khint_t k = kh_get(scheme_string, schemes, scheme); return (k != kh_end(schemes))? kh_value(schemes, k) : &unknown_scheme; } hFILE *hopen(const char *fname, const char *mode) { const struct hFILE_scheme_handler *handler = find_scheme_handler(fname); if (handler) return handler->open(fname, mode); else if (strcmp(fname, "-") == 0) return hopen_fd_stdinout(mode); else return hopen_fd(fname, mode); } int hfile_always_local (const char *fname) { return 0; } int hfile_always_remote(const char *fname) { return 1; } int hisremote(const char *fname) { const struct hFILE_scheme_handler *handler = find_scheme_handler(fname); return handler? handler->isremote(fname) : 0; } stacks-1.44/htslib/._hfile_internal.h000644 000765 000024 00000000265 12750743246 020355 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/hfile_internal.h000644 000765 000024 00000012775 12750743246 020151 0ustar00catchenstaff000000 000000 /* hfile_internal.h -- internal parts of low-level input/output streams. Copyright (C) 2013-2015 Genome Research Ltd. Author: John Marshall Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #ifndef HFILE_INTERNAL_H #define HFILE_INTERNAL_H #include "htslib/hfile.h" #ifdef __cplusplus extern "C" { #endif struct hFILE_backend { /* As per read(2), returning the number of bytes read (possibly 0) or negative (and setting errno) on errors. Front-end code will call this repeatedly if necessary to attempt to get the desired byte count. */ ssize_t (*read)(hFILE *fp, void *buffer, size_t nbytes) HTS_RESULT_USED; /* As per write(2), returning the number of bytes written or negative (and setting errno) on errors. Front-end code will call this repeatedly if necessary until the desired block is written or an error occurs. */ ssize_t (*write)(hFILE *fp, const void *buffer, size_t nbytes) HTS_RESULT_USED; /* As per lseek(2), returning the resulting offset within the stream or negative (and setting errno) on errors. */ off_t (*seek)(hFILE *fp, off_t offset, int whence) HTS_RESULT_USED; /* Performs low-level flushing, if any, e.g., fsync(2); for writing streams only. Returns 0 for success or negative (and sets errno) on errors. */ int (*flush)(hFILE *fp) HTS_RESULT_USED; /* Closes the underlying stream (for output streams, the buffer will already have been flushed), returning 0 for success or negative (and setting errno) on errors, as per close(2). */ int (*close)(hFILE *fp) HTS_RESULT_USED; }; /* May be called by hopen_*() functions to decode a fopen()-style mode into open(2)-style flags. */ int hfile_oflags(const char *mode); /* Must be called by hopen_*() functions to allocate the hFILE struct and set up its base. Capacity is a suggested buffer size (e.g., via fstat(2)) or 0 for a default-sized buffer. */ hFILE *hfile_init(size_t struct_size, const char *mode, size_t capacity); /* May be called by hopen_*() functions to undo the effects of hfile_init() in the event opening the stream subsequently fails. (This is safe to use even if fp is NULL. This takes care to preserve errno.) */ void hfile_destroy(hFILE *fp); struct hFILE_scheme_handler { /* Opens a stream when dispatched by hopen(); should call hfile_init() to malloc a struct "derived" from hFILE and initialise it appropriately, including setting base.backend to its own backend vector. */ hFILE *(*open)(const char *filename, const char *mode) HTS_RESULT_USED; /* Returns whether the URL denotes remote storage when dispatched by hisremote(). For simple cases, use one of hfile_always_*() below. */ int (*isremote)(const char *filename) HTS_RESULT_USED; /* The name of the plugin or other code providing this handler. */ const char *provider; /* If multiple handlers are registered for the same scheme, the one with the highest priority is used; range is 0 (lowest) to 100 (highest). */ int priority; }; /* May be used as an isremote() function in simple cases. */ extern int hfile_always_local (const char *fname); extern int hfile_always_remote(const char *fname); /* Should be called by plugins for each URL scheme they wish to handle. */ void hfile_add_scheme_handler(const char *scheme, const struct hFILE_scheme_handler *handler); struct hFILE_plugin { /* On entry, HTSlib's plugin API version (currently 1). */ int api_version; /* On entry, the plugin's handle as returned by dlopen() etc. */ void *obj; /* The plugin should fill this in with its (human-readable) name. */ const char *name; /* The plugin may wish to fill in a function to be called on closing. */ void (*destroy)(void); }; #ifdef ENABLE_PLUGINS #define PLUGIN_GLOBAL(identifier,suffix) identifier /* Plugins must define an entry point with this signature. */ extern int hfile_plugin_init(struct hFILE_plugin *self); #else #define PLUGIN_GLOBAL(identifier,suffix) identifier##suffix /* Only plugins distributed within the HTSlib source that might be built even with --disable-plugins need to use PLUGIN_GLOBAL and be listed here; others can simply define hfile_plugin_init(). */ extern int hfile_plugin_init_irods(struct hFILE_plugin *self); extern int hfile_plugin_init_libcurl(struct hFILE_plugin *self); #endif /* This one is never built as a separate plugin. */ extern int hfile_plugin_init_net(struct hFILE_plugin *self); #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/._hfile_irods.c000644 000765 000024 00000000265 12750743246 017654 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/hfile_irods.c000644 000765 000024 00000016303 12750743246 017437 0ustar00catchenstaff000000 000000 /* hfile_irods.c -- iRODS backend for low-level file streams. Copyright (C) 2013, 2015 Genome Research Ltd. Author: John Marshall Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include #include #include "hfile_internal.h" #include "htslib/hts.h" // for hts_version() and hts_verbose #include "htslib/kstring.h" #include #include #include #include #include #include #include typedef struct { hFILE base; int descriptor; } hFILE_irods; static int status_errno(int status) { switch (status) { case SYS_NO_API_PRIV: return EACCES; case SYS_MALLOC_ERR: return ENOMEM; case SYS_OUT_OF_FILE_DESC: return ENFILE; case SYS_BAD_FILE_DESCRIPTOR: return EBADF; case CAT_NO_ACCESS_PERMISSION: return EACCES; case CAT_INVALID_AUTHENTICATION: return EACCES; case CAT_NO_ROWS_FOUND: return ENOENT; case CATALOG_ALREADY_HAS_ITEM_BY_THAT_NAME: return EEXIST; default: return EIO; } } static void set_errno(int status) { int err = abs(status) % 1000; errno = err? err : status_errno(status); } static struct { rcComm_t *conn; rodsEnv env; } irods = { NULL }; static void irods_exit() { if (irods.conn) { (void) rcDisconnect(irods.conn); } irods.conn = NULL; } static int irods_init() { kstring_t useragent = { 0, 0, NULL }; struct sigaction pipehandler; rErrMsg_t err; int ret, pipehandler_ret; if (hts_verbose >= 5) rodsLogLevel(hts_verbose); ret = getRodsEnv(&irods.env); if (ret < 0) goto error; // Set iRODS User-Agent, if our caller hasn't already done so. kputs("htslib/", &useragent); kputs(hts_version(), &useragent); (void) setenv(SP_OPTION, useragent.s, 0); free(useragent.s); // Prior to iRODS 4.1, rcConnect() (even if it fails) installs its own // SIGPIPE handler, which just prints a message and otherwise ignores the // signal. Most actual SIGPIPEs encountered will pertain to e.g. stdout // rather than iRODS's connection, so we save and restore the existing // state (by default, termination; or as already set by our caller). pipehandler_ret = sigaction(SIGPIPE, NULL, &pipehandler); irods.conn = rcConnect(irods.env.rodsHost, irods.env.rodsPort, irods.env.rodsUserName, irods.env.rodsZone, NO_RECONN, &err); if (pipehandler_ret == 0) sigaction(SIGPIPE, &pipehandler, NULL); if (irods.conn == NULL) { ret = err.status; goto error; } if (strcmp(irods.env.rodsUserName, PUBLIC_USER_NAME) != 0) { #if defined IRODS_VERSION_INTEGER && IRODS_VERSION_INTEGER >= 4000000 ret = clientLogin(irods.conn, NULL, NULL); #else ret = clientLogin(irods.conn); #endif if (ret != 0) goto error; } return 0; error: if (irods.conn) { (void) rcDisconnect(irods.conn); } irods.conn = NULL; set_errno(ret); return -1; } static ssize_t irods_read(hFILE *fpv, void *buffer, size_t nbytes) { hFILE_irods *fp = (hFILE_irods *) fpv; openedDataObjInp_t args; bytesBuf_t buf; int ret; memset(&args, 0, sizeof args); args.l1descInx = fp->descriptor; args.len = nbytes; buf.buf = buffer; buf.len = nbytes; ret = rcDataObjRead(irods.conn, &args, &buf); if (ret < 0) set_errno(ret); return ret; } static ssize_t irods_write(hFILE *fpv, const void *buffer, size_t nbytes) { hFILE_irods *fp = (hFILE_irods *) fpv; openedDataObjInp_t args; bytesBuf_t buf; int ret; memset(&args, 0, sizeof args); args.l1descInx = fp->descriptor; args.len = nbytes; buf.buf = (void *) buffer; // ...the iRODS API is not const-correct here buf.len = nbytes; ret = rcDataObjWrite(irods.conn, &args, &buf); if (ret < 0) set_errno(ret); return ret; } static off_t irods_seek(hFILE *fpv, off_t offset, int whence) { hFILE_irods *fp = (hFILE_irods *) fpv; openedDataObjInp_t args; fileLseekOut_t *out = NULL; int ret; memset(&args, 0, sizeof args); args.l1descInx = fp->descriptor; args.offset = offset; args.whence = whence; ret = rcDataObjLseek(irods.conn, &args, &out); if (out) { offset = out->offset; free(out); } else offset = -1; if (ret < 0) { set_errno(ret); return -1; } return offset; } static int irods_close(hFILE *fpv) { hFILE_irods *fp = (hFILE_irods *) fpv; openedDataObjInp_t args; int ret; memset(&args, 0, sizeof args); args.l1descInx = fp->descriptor; ret = rcDataObjClose(irods.conn, &args); if (ret < 0) set_errno(ret); return ret; } static const struct hFILE_backend irods_backend = { irods_read, irods_write, irods_seek, NULL, irods_close }; hFILE *hopen_irods(const char *filename, const char *mode) { hFILE_irods *fp; rodsPath_t path; dataObjInp_t args; int ret; // Initialise the iRODS connection if this is the first use. if (irods.conn == NULL) { if (irods_init() < 0) return NULL; } if (strncmp(filename, "irods:", 6) == 0) filename += 6; else { errno = EINVAL; return NULL; } fp = (hFILE_irods *) hfile_init(sizeof (hFILE_irods), mode, 0); if (fp == NULL) return NULL; strncpy(path.inPath, filename, MAX_NAME_LEN-1); path.inPath[MAX_NAME_LEN-1] = '\0'; ret = parseRodsPath(&path, &irods.env); if (ret < 0) goto error; memset(&args, 0, sizeof args); strcpy(args.objPath, path.outPath); args.openFlags = hfile_oflags(mode); if (args.openFlags & O_CREAT) { args.createMode = 0666; addKeyVal(&args.condInput, DEST_RESC_NAME_KW,irods.env.rodsDefResource); } ret = rcDataObjOpen(irods.conn, &args); if (ret < 0) goto error; fp->descriptor = ret; fp->base.backend = &irods_backend; return &fp->base; error: hfile_destroy((hFILE *) fp); set_errno(ret); return NULL; } int PLUGIN_GLOBAL(hfile_plugin_init,_irods)(struct hFILE_plugin *self) { static const struct hFILE_scheme_handler handler = { hopen_irods, hfile_always_remote, "iRODS", 50 }; self->name = "iRODS"; hfile_add_scheme_handler("irods", &handler); self->destroy = irods_exit; return 0; } stacks-1.44/htslib/._hfile_libcurl.c000644 000765 000024 00000000265 12750743246 020170 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/hfile_libcurl.c000644 000765 000024 00000062312 12750743246 017754 0ustar00catchenstaff000000 000000 /* hfile_libcurl.c -- libcurl backend for low-level file streams. Copyright (C) 2015 Genome Research Ltd. Author: John Marshall Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include #include #include #include #include #include #include "hfile_internal.h" #include "htslib/hts.h" // for hts_version() and hts_verbose #include "htslib/kstring.h" #include typedef struct { hFILE base; CURL *easy; struct curl_slist *headers; off_t file_size; struct { union { char *rd; const char *wr; } ptr; size_t len; } buffer; CURLcode final_result; // easy result code for finished transfers // Flags for communicating with libcurl callbacks: unsigned paused : 1; // callback tells us that it has paused transfer unsigned closing : 1; // informs callback that hclose() has been invoked unsigned finished : 1; // wait_perform() tells us transfer is complete } hFILE_libcurl; static int http_status_errno(int status) { if (status >= 500) switch (status) { case 501: return ENOSYS; case 503: return EBUSY; case 504: return ETIMEDOUT; default: return EIO; } else if (status >= 400) switch (status) { case 401: return EPERM; case 403: return EACCES; case 404: return ENOENT; case 405: return EROFS; case 407: return EPERM; case 408: return ETIMEDOUT; case 410: return ENOENT; default: return EINVAL; } else return 0; } static int easy_errno(CURL *easy, CURLcode err) { long lval; switch (err) { case CURLE_OK: return 0; case CURLE_UNSUPPORTED_PROTOCOL: case CURLE_URL_MALFORMAT: return EINVAL; #if LIBCURL_VERSION_NUM >= 0x071505 case CURLE_NOT_BUILT_IN: return ENOSYS; #endif case CURLE_COULDNT_RESOLVE_PROXY: case CURLE_COULDNT_RESOLVE_HOST: case CURLE_FTP_CANT_GET_HOST: return EDESTADDRREQ; // Lookup failure case CURLE_COULDNT_CONNECT: case CURLE_SEND_ERROR: case CURLE_RECV_ERROR: if (curl_easy_getinfo(easy, CURLINFO_OS_ERRNO, &lval) == CURLE_OK) return lval; else return ECONNABORTED; case CURLE_REMOTE_ACCESS_DENIED: case CURLE_LOGIN_DENIED: case CURLE_TFTP_PERM: return EACCES; case CURLE_PARTIAL_FILE: return EPIPE; case CURLE_HTTP_RETURNED_ERROR: if (curl_easy_getinfo(easy, CURLINFO_RESPONSE_CODE, &lval) == CURLE_OK) return http_status_errno(lval); else return EIO; case CURLE_OUT_OF_MEMORY: return ENOMEM; case CURLE_OPERATION_TIMEDOUT: return ETIMEDOUT; case CURLE_RANGE_ERROR: return ESPIPE; case CURLE_SSL_CONNECT_ERROR: // TODO return SSL error buffer messages return ECONNABORTED; case CURLE_FILE_COULDNT_READ_FILE: case CURLE_TFTP_NOTFOUND: return ENOENT; case CURLE_TOO_MANY_REDIRECTS: return ELOOP; case CURLE_FILESIZE_EXCEEDED: return EFBIG; case CURLE_REMOTE_DISK_FULL: return ENOSPC; case CURLE_REMOTE_FILE_EXISTS: return EEXIST; default: return EIO; } } static int multi_errno(CURLMcode errm) { switch (errm) { case CURLM_CALL_MULTI_PERFORM: case CURLM_OK: return 0; case CURLM_BAD_HANDLE: case CURLM_BAD_EASY_HANDLE: case CURLM_BAD_SOCKET: return EBADF; case CURLM_OUT_OF_MEMORY: return ENOMEM; default: return EIO; } } static struct { CURLM *multi; kstring_t useragent; int nrunning; unsigned perform_again : 1; } curl = { NULL, { 0, 0, NULL }, 0, 0 }; static void libcurl_exit() { (void) curl_multi_cleanup(curl.multi); curl.multi = NULL; free(curl.useragent.s); curl.useragent.l = curl.useragent.m = 0; curl.useragent.s = NULL; curl_global_cleanup(); } static void process_messages() { CURLMsg *msg; int remaining; while ((msg = curl_multi_info_read(curl.multi, &remaining)) != NULL) { hFILE_libcurl *fp = NULL; curl_easy_getinfo(msg->easy_handle, CURLINFO_PRIVATE, (char **) &fp); switch (msg->msg) { case CURLMSG_DONE: fp->finished = 1; fp->final_result = msg->data.result; break; default: break; } } } static int wait_perform() { fd_set rd, wr, ex; int maxfd, nrunning; long timeout; CURLMcode errm; FD_ZERO(&rd); FD_ZERO(&wr); FD_ZERO(&ex); if (curl_multi_fdset(curl.multi, &rd, &wr, &ex, &maxfd) != CURLM_OK) maxfd = -1, timeout = 1000; else if (maxfd < 0) timeout = 100; // as recommended by curl_multi_fdset(3) else { if (curl_multi_timeout(curl.multi, &timeout) != CURLM_OK) timeout = 1000; else if (timeout < 0) timeout = 10000; // as recommended by curl_multi_timeout(3) } if (timeout > 0 && ! curl.perform_again) { struct timeval tval; tval.tv_sec = (timeout / 1000); tval.tv_usec = (timeout % 1000) * 1000; if (select(maxfd + 1, &rd, &wr, &ex, &tval) < 0) return -1; } errm = curl_multi_perform(curl.multi, &nrunning); curl.perform_again = 0; if (errm == CURLM_CALL_MULTI_PERFORM) curl.perform_again = 1; else if (errm != CURLM_OK) { errno = multi_errno(errm); return -1; } if (nrunning < curl.nrunning) process_messages(); return 0; } static size_t recv_callback(char *ptr, size_t size, size_t nmemb, void *fpv) { hFILE_libcurl *fp = (hFILE_libcurl *) fpv; size_t n = size * nmemb; if (n > fp->buffer.len) { fp->paused = 1; return CURL_WRITEFUNC_PAUSE; } else if (n == 0) return 0; memcpy(fp->buffer.ptr.rd, ptr, n); fp->buffer.ptr.rd += n; fp->buffer.len -= n; return n; } static ssize_t libcurl_read(hFILE *fpv, void *bufferv, size_t nbytes) { hFILE_libcurl *fp = (hFILE_libcurl *) fpv; char *buffer = (char *) bufferv; CURLcode err; fp->buffer.ptr.rd = buffer; fp->buffer.len = nbytes; fp->paused = 0; err = curl_easy_pause(fp->easy, CURLPAUSE_CONT); if (err != CURLE_OK) { errno = easy_errno(fp->easy, err); return -1; } while (! fp->paused && ! fp->finished) if (wait_perform() < 0) return -1; nbytes = fp->buffer.ptr.rd - buffer; fp->buffer.ptr.rd = NULL; fp->buffer.len = 0; if (fp->finished && fp->final_result != CURLE_OK) { errno = easy_errno(fp->easy, fp->final_result); return -1; } return nbytes; } static size_t send_callback(char *ptr, size_t size, size_t nmemb, void *fpv) { hFILE_libcurl *fp = (hFILE_libcurl *) fpv; size_t n = size * nmemb; if (fp->buffer.len == 0) { // Send buffer is empty; normally pause, or signal EOF if we're closing if (fp->closing) return 0; else { fp->paused = 1; return CURL_READFUNC_PAUSE; } } if (n > fp->buffer.len) n = fp->buffer.len; memcpy(ptr, fp->buffer.ptr.wr, n); fp->buffer.ptr.wr += n; fp->buffer.len -= n; return n; } static ssize_t libcurl_write(hFILE *fpv, const void *bufferv, size_t nbytes) { hFILE_libcurl *fp = (hFILE_libcurl *) fpv; const char *buffer = (const char *) bufferv; CURLcode err; fp->buffer.ptr.wr = buffer; fp->buffer.len = nbytes; fp->paused = 0; err = curl_easy_pause(fp->easy, CURLPAUSE_CONT); if (err != CURLE_OK) { errno = easy_errno(fp->easy, err); return -1; } while (! fp->paused && ! fp->finished) if (wait_perform() < 0) return -1; nbytes = fp->buffer.ptr.wr - buffer; fp->buffer.ptr.wr = NULL; fp->buffer.len = 0; if (fp->finished && fp->final_result != CURLE_OK) { errno = easy_errno(fp->easy, fp->final_result); return -1; } return nbytes; } static off_t libcurl_seek(hFILE *fpv, off_t offset, int whence) { hFILE_libcurl *fp = (hFILE_libcurl *) fpv; CURLcode err; CURLMcode errm; off_t origin, pos; switch (whence) { case SEEK_SET: origin = 0; break; case SEEK_CUR: errno = ENOSYS; return -1; case SEEK_END: if (fp->file_size < 0) { errno = ESPIPE; return -1; } origin = fp->file_size; break; default: errno = EINVAL; return -1; } // Check 0 <= origin+offset < fp->file_size carefully, avoiding overflow if ((offset < 0)? origin + offset < 0 : (fp->file_size >= 0 && offset > fp->file_size - origin)) { errno = EINVAL; return -1; } pos = origin + offset; errm = curl_multi_remove_handle(curl.multi, fp->easy); if (errm != CURLM_OK) { errno = multi_errno(errm); return -1; } curl.nrunning--; // TODO If we seem to be doing random access, use CURLOPT_RANGE to do // limited reads (e.g. about a BAM block!) so seeking can reuse the // existing connection more often. if (pos <= 2147483647) err = curl_easy_setopt(fp->easy, CURLOPT_RESUME_FROM, (long) pos); else err = curl_easy_setopt(fp->easy, CURLOPT_RESUME_FROM_LARGE, (curl_off_t) pos); if (err != CURLE_OK) { errno = easy_errno(fp->easy, err); return -1; } fp->buffer.len = 0; fp->paused = fp->finished = 0; errm = curl_multi_add_handle(curl.multi, fp->easy); if (errm != CURLM_OK) { errno = multi_errno(errm); return -1; } curl.nrunning++; err = curl_easy_pause(fp->easy, CURLPAUSE_CONT); if (err != CURLE_OK) { errno = easy_errno(fp->easy, err); return -1; } while (! fp->paused && ! fp->finished) if (wait_perform() < 0) return -1; if (fp->finished && fp->final_result != CURLE_OK) { errno = easy_errno(fp->easy, fp->final_result); return -1; } return pos; } static int libcurl_close(hFILE *fpv) { hFILE_libcurl *fp = (hFILE_libcurl *) fpv; CURLcode err; CURLMcode errm; int save_errno = 0; // Before closing the file, unpause it and perform on it so that uploads // have the opportunity to signal EOF to the server -- see send_callback(). fp->buffer.len = 0; fp->closing = 1; fp->paused = 0; err = curl_easy_pause(fp->easy, CURLPAUSE_CONT); if (err != CURLE_OK) save_errno = easy_errno(fp->easy, err); while (save_errno == 0 && ! fp->paused && ! fp->finished) if (wait_perform() < 0) save_errno = errno; if (fp->finished && fp->final_result != CURLE_OK) save_errno = easy_errno(fp->easy, fp->final_result); errm = curl_multi_remove_handle(curl.multi, fp->easy); if (errm != CURLM_OK && save_errno == 0) save_errno = multi_errno(errm); curl.nrunning--; curl_easy_cleanup(fp->easy); if (save_errno) { errno = save_errno; return -1; } else return 0; } static const struct hFILE_backend libcurl_backend = { libcurl_read, libcurl_write, libcurl_seek, NULL, libcurl_close }; static int add_header(hFILE_libcurl *fp, const char *header) { struct curl_slist *list = curl_slist_append(fp->headers, header); if (list == NULL) { errno = ENOMEM; return -1; } fp->headers = list; return 0; } static int add_s3_settings(hFILE_libcurl *fp, const char *url, kstring_t *message); hFILE *hopen_libcurl(const char *url, const char *modes) { hFILE_libcurl *fp; char mode; const char *s; CURLcode err; CURLMcode errm; int save; if ((s = strpbrk(modes, "rwa+")) != NULL) { mode = *s; if (strpbrk(&s[1], "rwa+")) mode = 'e'; } else mode = '\0'; if (mode != 'r' && mode != 'w') { errno = EINVAL; return NULL; } fp = (hFILE_libcurl *) hfile_init(sizeof (hFILE_libcurl), modes, 0); if (fp == NULL) return NULL; fp->easy = curl_easy_init(); if (fp->easy == NULL) { errno = ENOMEM; goto error; } fp->headers = NULL; fp->file_size = -1; fp->buffer.ptr.rd = NULL; fp->buffer.len = 0; fp->final_result = (CURLcode) -1; fp->paused = fp->closing = fp->finished = 0; // Make a route to the hFILE_libcurl* given just a CURL* easy handle err = curl_easy_setopt(fp->easy, CURLOPT_PRIVATE, fp); if (mode == 'r') { err |= curl_easy_setopt(fp->easy, CURLOPT_WRITEFUNCTION, recv_callback); err |= curl_easy_setopt(fp->easy, CURLOPT_WRITEDATA, fp); } else { err |= curl_easy_setopt(fp->easy, CURLOPT_READFUNCTION, send_callback); err |= curl_easy_setopt(fp->easy, CURLOPT_READDATA, fp); err |= curl_easy_setopt(fp->easy, CURLOPT_UPLOAD, 1L); if (add_header(fp, "Transfer-Encoding: chunked") < 0) goto error; } if (tolower(url[0]) == 's' && url[1] == '3') { // Construct the HTTP-Method/Content-MD5/Content-Type part of the // message to be signed. This will be destroyed by add_s3_settings(). kstring_t message = { 0, 0, NULL }; kputs((mode == 'r')? "GET\n" : "PUT\n", &message); kputc('\n', &message); kputc('\n', &message); if (add_s3_settings(fp, url, &message) < 0) goto error; } else err |= curl_easy_setopt(fp->easy, CURLOPT_URL, url); err |= curl_easy_setopt(fp->easy, CURLOPT_USERAGENT, curl.useragent.s); if (fp->headers) err |= curl_easy_setopt(fp->easy, CURLOPT_HTTPHEADER, fp->headers); err |= curl_easy_setopt(fp->easy, CURLOPT_FOLLOWLOCATION, 1L); err |= curl_easy_setopt(fp->easy, CURLOPT_FAILONERROR, 1L); if (hts_verbose >= 8) err |= curl_easy_setopt(fp->easy, CURLOPT_VERBOSE, 1L); if (err != 0) { errno = ENOSYS; goto error; } errm = curl_multi_add_handle(curl.multi, fp->easy); if (errm != CURLM_OK) { errno = multi_errno(errm); goto error; } curl.nrunning++; while (! fp->paused && ! fp->finished) if (wait_perform() < 0) goto error_remove; if (fp->finished && fp->final_result != CURLE_OK) { errno = easy_errno(fp->easy, fp->final_result); goto error_remove; } if (mode == 'r') { double dval; if (curl_easy_getinfo(fp->easy, CURLINFO_CONTENT_LENGTH_DOWNLOAD, &dval) == CURLE_OK && dval >= 0.0) fp->file_size = (off_t) (dval + 0.1); } fp->base.backend = &libcurl_backend; return &fp->base; error_remove: save = errno; (void) curl_multi_remove_handle(curl.multi, fp->easy); curl.nrunning--; errno = save; error: save = errno; curl_easy_cleanup(fp->easy); if (fp->headers) curl_slist_free_all(fp->headers); hfile_destroy((hFILE *) fp); errno = save; return NULL; } int PLUGIN_GLOBAL(hfile_plugin_init,_libcurl)(struct hFILE_plugin *self) { static const struct hFILE_scheme_handler handler = { hopen_libcurl, hfile_always_remote, "libcurl", 50 }; const curl_version_info_data *info; const char * const *protocol; CURLcode err; err = curl_global_init(CURL_GLOBAL_ALL); if (err != CURLE_OK) { errno = easy_errno(NULL, err); return -1; } curl.multi = curl_multi_init(); if (curl.multi == NULL) { curl_global_cleanup(); errno = EIO; return -1; } info = curl_version_info(CURLVERSION_NOW); ksprintf(&curl.useragent, "htslib/%s libcurl/%s", hts_version(), info->version); curl.nrunning = 0; curl.perform_again = 0; self->name = "libcurl"; self->destroy = libcurl_exit; for (protocol = info->protocols; *protocol; protocol++) hfile_add_scheme_handler(*protocol, &handler); hfile_add_scheme_handler("s3", &handler); hfile_add_scheme_handler("s3+http", &handler); if (info->features & CURL_VERSION_SSL) hfile_add_scheme_handler("s3+https", &handler); return 0; } /******************* * Rewrite S3 URLs * *******************/ #if defined HAVE_COMMONCRYPTO #include #define DIGEST_BUFSIZ CC_SHA1_DIGEST_LENGTH static size_t s3_sign(unsigned char *digest, kstring_t *key, kstring_t *message) { CCHmac(kCCHmacAlgSHA1, key->s, key->l, message->s, message->l, digest); return CC_SHA1_DIGEST_LENGTH; } #elif defined HAVE_HMAC #include #define DIGEST_BUFSIZ EVP_MAX_MD_SIZE static size_t s3_sign(unsigned char *digest, kstring_t *key, kstring_t *message) { unsigned int len; HMAC(EVP_sha1(), key->s, key->l, (unsigned char *) message->s, message->l, digest, &len); return len; } #else #error No HMAC() routine found by configure #endif static void urldecode_kput(const char *s, int len, hFILE_libcurl *fp, kstring_t *str) { if (memchr(s, '%', len) != NULL) { int len2; char *s2 = curl_easy_unescape(fp->easy, s, len, &len2); if (s2 == NULL) abort(); kputsn(s2, len2, str); curl_free(s2); } else kputsn(s, len, str); } static void base64_kput(const unsigned char *data, size_t len, kstring_t *str) { static const char base64[] = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"; size_t i = 0; unsigned x = 0; int bits = 0, pad = 0; while (bits || i < len) { if (bits < 6) { x <<= 8, bits += 8; if (i < len) x |= data[i++]; else pad++; } bits -= 6; kputc(base64[(x >> bits) & 63], str); } str->l -= pad; kputsn("==", pad, str); } static int is_dns_compliant(const char *s0, const char *slim) { int has_nondigit = 0, len = 0; const char *s; for (s = s0; s < slim; len++, s++) if (islower(*s)) has_nondigit = 1; else if (*s == '-') { has_nondigit = 1; if (s == s0 || s+1 == slim) return 0; } else if (isdigit(*s)) ; else if (*s == '.') { if (s == s0 || ! isalnum(s[-1])) return 0; if (s+1 == slim || ! isalnum(s[1])) return 0; } else return 0; return has_nondigit && len >= 3 && len <= 63; } static FILE *expand_tilde_open(const char *fname, const char *mode) { FILE *fp; if (strncmp(fname, "~/", 2) == 0) { kstring_t full_fname = { 0, 0, NULL }; const char *home = getenv("HOME"); if (! home) return NULL; kputs(home, &full_fname); kputs(&fname[1], &full_fname); fp = fopen(full_fname.s, mode); free(full_fname.s); } else fp = fopen(fname, mode); return fp; } static void parse_ini(const char *fname, const char *section, ...) { kstring_t line = { 0, 0, NULL }; int active = 1; // Start active, so global properties are accepted char *s; FILE *fp = expand_tilde_open(fname, "r"); if (fp == NULL) return; while (line.l = 0, kgetline(&line, (kgets_func *) fgets, fp) >= 0) if (line.s[0] == '[' && (s = strchr(line.s, ']')) != NULL) { *s = '\0'; active = (strcmp(&line.s[1], section) == 0); } else if (active && (s = strpbrk(line.s, ":=")) != NULL) { const char *key = line.s, *value = &s[1], *akey; va_list args; while (isspace(*key)) key++; while (s > key && isspace(s[-1])) s--; *s = '\0'; while (isspace(*value)) value++; while (line.l > 0 && isspace(line.s[line.l-1])) line.s[--line.l] = '\0'; va_start(args, section); while ((akey = va_arg(args, const char *)) != NULL) { kstring_t *avar = va_arg(args, kstring_t *); if (strcmp(key, akey) == 0) { kputs(value, avar); break; } } va_end(args); } fclose(fp); free(line.s); } static void parse_simple(const char *fname, kstring_t *id, kstring_t *secret) { kstring_t text = { 0, 0, NULL }; char *s; size_t len; FILE *fp = expand_tilde_open(fname, "r"); if (fp == NULL) return; while (kgetline(&text, (kgets_func *) fgets, fp) >= 0) kputc(' ', &text); fclose(fp); s = text.s; while (isspace(*s)) s++; kputsn(s, len = strcspn(s, " \t"), id); s += len; while (isspace(*s)) s++; kputsn(s, strcspn(s, " \t"), secret); free(text.s); } static int add_s3_settings(hFILE_libcurl *fp, const char *s3url, kstring_t *message) { int ret, save; const char *bucket, *path; char date_hdr[40]; CURLcode err; kstring_t url = { 0, 0, NULL }; kstring_t profile = { 0, 0, NULL }; kstring_t id = { 0, 0, NULL }; kstring_t secret = { 0, 0, NULL }; kstring_t token = { 0, 0, NULL }; kstring_t token_hdr = { 0, 0, NULL }; kstring_t auth_hdr = { 0, 0, NULL }; time_t now = time(NULL); #ifdef HAVE_GMTIME_R struct tm tm_buffer; struct tm *tm = gmtime_r(&now, &tm_buffer); #else struct tm *tm = gmtime(&now); #endif strftime(date_hdr, sizeof date_hdr, "Date: %a, %d %b %Y %H:%M:%S GMT", tm); if (add_header(fp, date_hdr) < 0) goto error; kputs(&date_hdr[6], message); kputc('\n', message); // Our S3 URL format is s3[+SCHEME]://[ID[:SECRET[:TOKEN]]@]BUCKET/PATH if (s3url[2] == '+') { bucket = strchr(s3url, ':') + 1; kputsn(&s3url[3], bucket - &s3url[3], &url); } else { kputs("https:", &url); bucket = &s3url[3]; } while (*bucket == '/') kputc(*bucket++, &url); path = bucket + strcspn(bucket, "/?#@"); if (*path == '@') { const char *colon = strpbrk(bucket, ":@"); if (*colon != ':') { urldecode_kput(bucket, colon - bucket, fp, &profile); } else { const char *colon2 = strpbrk(&colon[1], ":@"); urldecode_kput(bucket, colon - bucket, fp, &id); urldecode_kput(&colon[1], colon2 - &colon[1], fp, &secret); if (*colon2 == ':') urldecode_kput(&colon2[1], path - &colon2[1], fp, &token); } bucket = &path[1]; path = bucket + strcspn(bucket, "/?#"); } else { // If the URL has no ID[:SECRET]@, consider environment variables. const char *v; if ((v = getenv("AWS_ACCESS_KEY_ID")) != NULL) kputs(v, &id); if ((v = getenv("AWS_SECRET_ACCESS_KEY")) != NULL) kputs(v, &secret); if ((v = getenv("AWS_SESSION_TOKEN")) != NULL) kputs(v, &token); if ((v = getenv("AWS_DEFAULT_PROFILE")) != NULL) kputs(v, &profile); else if ((v = getenv("AWS_PROFILE")) != NULL) kputs(v, &profile); else kputs("default", &profile); } // Use virtual hosted-style access if possible, otherwise path-style. if (is_dns_compliant(bucket, path)) { kputsn(bucket, path - bucket, &url); kputs(".s3.amazonaws.com", &url); } else { kputs("s3.amazonaws.com/", &url); kputsn(bucket, path - bucket, &url); } kputs(path, &url); if (id.l == 0) { const char *v = getenv("AWS_SHARED_CREDENTIALS_FILE"); parse_ini(v? v : "~/.aws/credentials", profile.s, "aws_access_key_id", &id, "aws_secret_access_key", &secret, "aws_session_token", &token, NULL); } if (id.l == 0) parse_ini("~/.s3cfg", profile.s, "access_key", &id, "secret_key", &secret, "access_token", &token, NULL); if (id.l == 0) parse_simple("~/.awssecret", &id, &secret); if (token.l > 0) { kputs("x-amz-security-token:", message); kputs(token.s, message); kputc('\n', message); kputs("X-Amz-Security-Token: ", &token_hdr); kputs(token.s, &token_hdr); if (add_header(fp, token_hdr.s) < 0) goto error; } kputc('/', message); kputs(bucket, message); // CanonicalizedResource is '/' + bucket + path err = curl_easy_setopt(fp->easy, CURLOPT_URL, url.s); if (err != CURLE_OK) { errno = easy_errno(fp->easy, err); goto error; } // If we have no id/secret, we can't sign the request but will // still be able to access public data sets. if (id.l > 0 && secret.l > 0) { unsigned char digest[DIGEST_BUFSIZ]; size_t digest_len = s3_sign(digest, &secret, message); kputs("Authorization: AWS ", &auth_hdr); kputs(id.s, &auth_hdr); kputc(':', &auth_hdr); base64_kput(digest, digest_len, &auth_hdr); if (add_header(fp, auth_hdr.s) < 0) goto error; } ret = 0; goto free_and_return; error: ret = -1; free_and_return: save = errno; free(url.s); free(profile.s); free(id.s); free(secret.s); free(token.s); free(token_hdr.s); free(auth_hdr.s); free(message->s); errno = save; return ret; } stacks-1.44/htslib/._hfile_net.c000644 000765 000024 00000000265 12750743246 017322 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/hfile_net.c000644 000765 000024 00000006170 12750743246 017106 0ustar00catchenstaff000000 000000 /* hfile_net.c -- network backend for low-level input/output streams. Copyright (C) 2013-2015 Genome Research Ltd. Author: John Marshall Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include "hfile_internal.h" #include "htslib/knetfile.h" typedef struct { hFILE base; knetFile *netfp; } hFILE_net; static int net_inited = 0; #ifdef _WIN32 static void net_exit(void) { knet_win32_destroy(); } #endif static int net_init(void) { #ifdef _WIN32 if (knet_win32_init() != 0) return -1; // In the unlikely event atexit() fails, it's better to succeed here and // carry on and do the I/O; then eventually when the program exits, we'll // merely have failed to clean up properly, as if we had aborted. (void) atexit(net_exit); #endif net_inited = 1; return 0; } static ssize_t net_read(hFILE *fpv, void *buffer, size_t nbytes) { hFILE_net *fp = (hFILE_net *) fpv; return knet_read(fp->netfp, buffer, nbytes); } static off_t net_seek(hFILE *fpv, off_t offset, int whence) { hFILE_net *fp = (hFILE_net *) fpv; return knet_seek(fp->netfp, offset, whence); } static int net_close(hFILE *fpv) { hFILE_net *fp = (hFILE_net *) fpv; return knet_close(fp->netfp); } static const struct hFILE_backend net_backend = { net_read, NULL, net_seek, NULL, net_close }; hFILE *hopen_net(const char *filename, const char *mode) { hFILE_net *fp; // Do any networking initialisation if this is the first use. if (! net_inited) { if (net_init() < 0) return NULL; } fp = (hFILE_net *) hfile_init(sizeof (hFILE_net), mode, 0); if (fp == NULL) return NULL; fp->netfp = knet_open(filename, mode); if (fp->netfp == NULL) { hfile_destroy((hFILE *) fp); return NULL; } fp->base.backend = &net_backend; return &fp->base; } int hfile_plugin_init_net(struct hFILE_plugin *self) { static const struct hFILE_scheme_handler handler = { hopen_net, hfile_always_remote, "knetfile", 0 }; self->name = "knetfile"; hfile_add_scheme_handler("http", &handler); hfile_add_scheme_handler("ftp", &handler); return 0; } stacks-1.44/htslib/._hts.c000644 000765 000024 00000000265 12750743246 016163 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/hts.c000644 000765 000024 00000173100 12750743246 015745 0ustar00catchenstaff000000 000000 /* hts.c -- format-neutral I/O, indexing, and iterator API functions. Copyright (C) 2008, 2009, 2012-2016 Genome Research Ltd. Copyright (C) 2012, 2013 Broad Institute. Author: Heng Li Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include #include #include #include #include #include #include #include "htslib/bgzf.h" #include "htslib/hts.h" #include "cram/cram.h" #include "htslib/hfile.h" #include "version.h" #include "hts_internal.h" #include "htslib/kseq.h" #define KS_BGZF 1 #if KS_BGZF // bgzf now supports gzip-compressed files, the gzFile branch can be removed KSTREAM_INIT2(, BGZF*, bgzf_read, 65536) #else KSTREAM_INIT2(, gzFile, gzread, 16384) #endif #include "htslib/khash.h" KHASH_INIT2(s2i,, kh_cstr_t, int64_t, 1, kh_str_hash_func, kh_str_hash_equal) int hts_verbose = 3; const char *hts_version() { return HTS_VERSION; } const unsigned char seq_nt16_table[256] = { 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 1, 2, 4, 8, 15,15,15,15, 15,15,15,15, 15, 0 /*=*/,15,15, 15, 1,14, 2, 13,15,15, 4, 11,15,15,12, 15, 3,15,15, 15,15, 5, 6, 8,15, 7, 9, 15,10,15,15, 15,15,15,15, 15, 1,14, 2, 13,15,15, 4, 11,15,15,12, 15, 3,15,15, 15,15, 5, 6, 8,15, 7, 9, 15,10,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15 }; const char seq_nt16_str[] = "=ACMGRSVTWYHKDBN"; const int seq_nt16_int[] = { 4, 0, 1, 4, 2, 4, 4, 4, 3, 4, 4, 4, 4, 4, 4, 4 }; /********************** *** Basic file I/O *** **********************/ static enum htsFormatCategory format_category(enum htsExactFormat fmt) { switch (fmt) { case bam: case sam: case cram: return sequence_data; case vcf: case bcf: return variant_data; case bai: case crai: case csi: case gzi: case tbi: return index_file; case bed: return region_list; case unknown_format: case binary_format: case text_format: case format_maximum: break; } return unknown_category; } // Decompress up to ten or so bytes by peeking at the file, which must be // positioned at the start of a GZIP block. static size_t decompress_peek(hFILE *fp, unsigned char *dest, size_t destsize) { // Typically at most a couple of hundred bytes of input are required // to get a few bytes of output from inflate(), so hopefully this buffer // size suffices in general. unsigned char buffer[512]; z_stream zs; ssize_t npeek = hpeek(fp, buffer, sizeof buffer); if (npeek < 0) return 0; zs.zalloc = NULL; zs.zfree = NULL; zs.next_in = buffer; zs.avail_in = npeek; zs.next_out = dest; zs.avail_out = destsize; if (inflateInit2(&zs, 31) != Z_OK) return 0; while (zs.total_out < destsize) if (inflate(&zs, Z_SYNC_FLUSH) != Z_OK) break; destsize = zs.total_out; inflateEnd(&zs); return destsize; } // Parse "x.y" text, taking care because the string is not NUL-terminated // and filling in major/minor only when the digits are followed by a delimiter, // so we don't misread "1.10" as "1.1" due to reaching the end of the buffer. static void parse_version(htsFormat *fmt, const unsigned char *u, const unsigned char *ulim) { const char *s = (const char *) u; const char *slim = (const char *) ulim; short v; fmt->version.major = fmt->version.minor = -1; for (v = 0; s < slim && isdigit_c(*s); s++) v = 10 * v + *s - '0'; if (s < slim) { fmt->version.major = v; if (*s == '.') { s++; for (v = 0; s < slim && isdigit_c(*s); s++) v = 10 * v + *s - '0'; if (s < slim) fmt->version.minor = v; } else fmt->version.minor = 0; } } int hts_detect_format(hFILE *hfile, htsFormat *fmt) { unsigned char s[21]; ssize_t len = hpeek(hfile, s, 18); if (len < 0) return -1; if (len >= 2 && s[0] == 0x1f && s[1] == 0x8b) { // The stream is either gzip-compressed or BGZF-compressed. // Determine which, and decompress the first few bytes. fmt->compression = (len >= 18 && (s[3] & 4) && memcmp(&s[12], "BC\2\0", 4) == 0)? bgzf : gzip; len = decompress_peek(hfile, s, sizeof s); } else { fmt->compression = no_compression; len = hpeek(hfile, s, sizeof s); } if (len < 0) return -1; fmt->compression_level = -1; fmt->specific = NULL; if (len >= 6 && memcmp(s,"CRAM",4) == 0 && s[4]>=1 && s[4]<=3 && s[5]<=1) { fmt->category = sequence_data; fmt->format = cram; fmt->version.major = s[4], fmt->version.minor = s[5]; fmt->compression = custom; return 0; } else if (len >= 4 && s[3] <= '\4') { if (memcmp(s, "BAM\1", 4) == 0) { fmt->category = sequence_data; fmt->format = bam; // TODO Decompress enough to pick version from @HD-VN header fmt->version.major = 1, fmt->version.minor = -1; return 0; } else if (memcmp(s, "BAI\1", 4) == 0) { fmt->category = index_file; fmt->format = bai; fmt->version.major = -1, fmt->version.minor = -1; return 0; } else if (memcmp(s, "BCF\4", 4) == 0) { fmt->category = variant_data; fmt->format = bcf; fmt->version.major = 1, fmt->version.minor = -1; return 0; } else if (memcmp(s, "BCF\2", 4) == 0) { fmt->category = variant_data; fmt->format = bcf; fmt->version.major = s[3]; fmt->version.minor = (len >= 5 && s[4] <= 2)? s[4] : 0; return 0; } else if (memcmp(s, "CSI\1", 4) == 0) { fmt->category = index_file; fmt->format = csi; fmt->version.major = 1, fmt->version.minor = -1; return 0; } else if (memcmp(s, "TBI\1", 4) == 0) { fmt->category = index_file; fmt->format = tbi; fmt->version.major = -1, fmt->version.minor = -1; return 0; } } else if (len >= 16 && memcmp(s, "##fileformat=VCF", 16) == 0) { fmt->category = variant_data; fmt->format = vcf; if (len >= 21 && s[16] == 'v') parse_version(fmt, &s[17], &s[len]); else fmt->version.major = fmt->version.minor = -1; return 0; } else if (len >= 4 && s[0] == '@' && (memcmp(s, "@HD\t", 4) == 0 || memcmp(s, "@SQ\t", 4) == 0 || memcmp(s, "@RG\t", 4) == 0 || memcmp(s, "@PG\t", 4) == 0)) { fmt->category = sequence_data; fmt->format = sam; // @HD-VN is not guaranteed to be the first tag, but then @HD is // not guaranteed to be present at all... if (len >= 9 && memcmp(s, "@HD\tVN:", 7) == 0) parse_version(fmt, &s[7], &s[len]); else fmt->version.major = 1, fmt->version.minor = -1; return 0; } else { // Various possibilities for tab-delimited text: // .crai (gzipped tab-delimited six columns: seqid 5*number) // .bed ([3..12] tab-delimited columns) // .bedpe (>= 10 tab-delimited columns) // .sam (tab-delimited >= 11 columns: seqid number seqid...) // FIXME For now, assume it's SAM fmt->category = sequence_data; fmt->format = sam; fmt->version.major = 1, fmt->version.minor = -1; return 0; } fmt->category = unknown_category; fmt->format = unknown_format; fmt->version.major = fmt->version.minor = -1; fmt->compression = no_compression; return 0; } char *hts_format_description(const htsFormat *format) { kstring_t str = { 0, 0, NULL }; switch (format->format) { case sam: kputs("SAM", &str); break; case bam: kputs("BAM", &str); break; case cram: kputs("CRAM", &str); break; case vcf: kputs("VCF", &str); break; case bcf: if (format->version.major == 1) kputs("Legacy BCF", &str); else kputs("BCF", &str); break; case bai: kputs("BAI", &str); break; case crai: kputs("CRAI", &str); break; case csi: kputs("CSI", &str); break; case tbi: kputs("Tabix", &str); break; default: kputs("unknown", &str); break; } if (format->version.major >= 0) { kputs(" version ", &str); kputw(format->version.major, &str); if (format->version.minor >= 0) { kputc('.', &str); kputw(format->version.minor, &str); } } switch (format->compression) { case custom: kputs(" compressed", &str); break; case gzip: kputs(" gzip-compressed", &str); break; case bgzf: switch (format->format) { case bam: case bcf: case csi: case tbi: // These are by definition BGZF, so just use the generic term kputs(" compressed", &str); break; default: kputs(" BGZF-compressed", &str); break; } break; default: break; } switch (format->category) { case sequence_data: kputs(" sequence", &str); break; case variant_data: kputs(" variant calling", &str); break; case index_file: kputs(" index", &str); break; case region_list: kputs(" genomic region", &str); break; default: break; } if (format->compression == no_compression) switch (format->format) { case sam: case crai: case vcf: case bed: kputs(" text", &str); break; default: kputs(" data", &str); break; } else kputs(" data", &str); return ks_release(&str); } htsFile *hts_open_format(const char *fn, const char *mode, const htsFormat *fmt) { char smode[102], *cp, *cp2, *mode_c; htsFile *fp = NULL; hFILE *hfile; char fmt_code = '\0'; strncpy(smode, mode, 100); smode[100]=0; if ((cp = strchr(smode, ','))) *cp = '\0'; // Migrate format code (b or c) to the end of the smode buffer. for (cp2 = cp = smode; *cp; cp++) { if (*cp == 'b') fmt_code = 'b'; else if (*cp == 'c') fmt_code = 'c'; else *cp2++ = *cp; } mode_c = cp2; *cp2++ = fmt_code; *cp2++ = 0; *cp2++ = 0; // Set or reset the format code if opts->format is used if (fmt && fmt->format != unknown_format) *mode_c = "\0g\0\0b\0c\0\0b\0g\0\0"[fmt->format]; hfile = hopen(fn, smode); if (hfile == NULL) goto error; fp = hts_hopen(hfile, fn, smode); if (fp == NULL) goto error; if (fmt && fmt->specific) if (hts_opt_apply(fp, fmt->specific) != 0) goto error; return fp; error: if (hts_verbose >= 2) fprintf(stderr, "[E::%s] fail to open file '%s'\n", __func__, fn); if (hfile) hclose_abruptly(hfile); return NULL; } htsFile *hts_open(const char *fn, const char *mode) { return hts_open_format(fn, mode, NULL); } /* * Splits str into a prefix, delimiter ('\0' or delim), and suffix, writing * the prefix in lowercase into buf and returning a pointer to the suffix. * On return, buf is always NUL-terminated; thus assumes that the "keyword" * prefix should be one of several known values of maximum length buflen-2. * (If delim is not found, returns a pointer to the '\0'.) */ static const char * scan_keyword(const char *str, char delim, char *buf, size_t buflen) { size_t i = 0; while (*str && *str != delim) { if (i < buflen-1) buf[i++] = tolower_c(*str); str++; } buf[i] = '\0'; return *str? str+1 : str; } /* * Parses arg and appends it to the option list. * * Returns 0 on success; * -1 on failure. */ int hts_opt_add(hts_opt **opts, const char *c_arg) { hts_opt *o, *t; char *val; if (!c_arg) return -1; if (!(o = malloc(sizeof(*o)))) return -1; if (!(o->arg = strdup(c_arg))) { free(o); return -1; } if (!(val = strchr(o->arg, '='))) val = "1"; // assume boolean else *val++ = '\0'; if (strcmp(o->arg, "decode_md") == 0 || strcmp(o->arg, "DECODE_MD") == 0) o->opt = CRAM_OPT_DECODE_MD, o->val.i = atoi(val); else if (strcmp(o->arg, "verbosity") == 0 || strcmp(o->arg, "VERBOSITY") == 0) o->opt = CRAM_OPT_VERBOSITY, o->val.i = atoi(val); else if (strcmp(o->arg, "seqs_per_slice") == 0 || strcmp(o->arg, "SEQS_PER_SLICE") == 0) o->opt = CRAM_OPT_SEQS_PER_SLICE, o->val.i = atoi(val); else if (strcmp(o->arg, "slices_per_container") == 0 || strcmp(o->arg, "SLICES_PER_CONTAINER") == 0) o->opt = CRAM_OPT_SLICES_PER_CONTAINER, o->val.i = atoi(val); else if (strcmp(o->arg, "embed_ref") == 0 || strcmp(o->arg, "EMBED_REF") == 0) o->opt = CRAM_OPT_EMBED_REF, o->val.i = atoi(val); else if (strcmp(o->arg, "no_ref") == 0 || strcmp(o->arg, "NO_REF") == 0) o->opt = CRAM_OPT_NO_REF, o->val.i = atoi(val); else if (strcmp(o->arg, "ignore_md5") == 0 || strcmp(o->arg, "IGNORE_MD5") == 0) o->opt = CRAM_OPT_IGNORE_MD5, o->val.i = atoi(val); else if (strcmp(o->arg, "use_bzip2") == 0 || strcmp(o->arg, "USE_BZIP2") == 0) o->opt = CRAM_OPT_USE_BZIP2, o->val.i = atoi(val); else if (strcmp(o->arg, "use_rans") == 0 || strcmp(o->arg, "USE_RANS") == 0) o->opt = CRAM_OPT_USE_RANS, o->val.i = atoi(val); else if (strcmp(o->arg, "use_lzma") == 0 || strcmp(o->arg, "USE_LZMA") == 0) o->opt = CRAM_OPT_USE_LZMA, o->val.i = atoi(val); else if (strcmp(o->arg, "reference") == 0 || strcmp(o->arg, "REFERENCE") == 0) o->opt = CRAM_OPT_REFERENCE, o->val.s = val; else if (strcmp(o->arg, "version") == 0 || strcmp(o->arg, "VERSION") == 0) o->opt = CRAM_OPT_VERSION, o->val.s =val; else if (strcmp(o->arg, "multi_seq_per_slice") == 0 || strcmp(o->arg, "MULTI_SEQ_PER_SLICE") == 0) o->opt = CRAM_OPT_MULTI_SEQ_PER_SLICE, o->val.i = atoi(val); else if (strcmp(o->arg, "nthreads") == 0 || strcmp(o->arg, "NTHREADS") == 0) o->opt = HTS_OPT_NTHREADS, o->val.i = atoi(val); else if (strcmp(o->arg, "required_fields") == 0 || strcmp(o->arg, "REQUIRED_FIELDS") == 0) o->opt = CRAM_OPT_REQUIRED_FIELDS, o->val.i = strtol(val, NULL, 0); else { fprintf(stderr, "Unknown option '%s'\n", o->arg); free(o->arg); free(o); return -1; } o->next = NULL; // Append; assumes small list. if (*opts) { t = *opts; while (t->next) t = t->next; t->next = o; } else { *opts = o; } return 0; } /* * Applies an hts_opt option list to a given htsFile. * * Returns 0 on success * -1 on failure */ int hts_opt_apply(htsFile *fp, hts_opt *opts) { hts_opt *last = NULL; for (; opts; opts = (last=opts)->next) if (hts_set_opt(fp, opts->opt, opts->val) != 0) return -1; return 0; } /* * Frees an hts_opt list. */ void hts_opt_free(hts_opt *opts) { hts_opt *last = NULL; while (opts) { opts = (last=opts)->next; free(last->arg); free(last); } } /* * Tokenise options as (key(=value)?,)*(key(=value)?)? * NB: No provision for ',' appearing in the value! * Add backslashing rules? * * This could be used as part of a general command line option parser or * as a string concatenated onto the file open mode. * * Returns 0 on success * -1 on failure. */ int hts_parse_opt_list(htsFormat *fmt, const char *str) { while (str && *str) { const char *str_start; int len; char arg[8001]; while (*str && *str == ',') str++; for (str_start = str; *str && *str != ','; str++); len = str - str_start; // Produce a nul terminated copy of the option strncpy(arg, str_start, len < 8000 ? len : 8000); arg[len < 8000 ? len : 8000] = '\0'; if (hts_opt_add((hts_opt **)&fmt->specific, arg) != 0) return -1; if (*str) str++; } return 0; } /* * Accepts a string file format (sam, bam, cram, vcf, bam) optionally * followed by a comma separated list of key=value options and splits * these up into the fields of htsFormat struct. * * format is assumed to be already initialised, either to blank * "unknown" values or via previous hts_opt_add calls. * * Returns 0 on success * -1 on failure. */ int hts_parse_format(htsFormat *format, const char *str) { char fmt[8]; const char *cp = scan_keyword(str, ',', fmt, sizeof fmt); format->version.minor = 0; // unknown format->version.major = 0; // unknown if (strcmp(fmt, "sam") == 0) { format->category = sequence_data; format->format = sam; format->compression = no_compression;; format->compression_level = 0; } else if (strcmp(fmt, "bam") == 0) { format->category = sequence_data; format->format = bam; format->compression = bgzf; format->compression_level = -1; } else if (strcmp(fmt, "cram") == 0) { format->category = sequence_data; format->format = cram; format->compression = custom; format->compression_level = -1; } else if (strcmp(fmt, "vcf") == 0) { format->category = variant_data; format->format = vcf; format->compression = no_compression;; format->compression_level = 0; } else if (strcmp(fmt, "bcf") == 0) { format->category = variant_data; format->format = bcf; format->compression = bgzf; format->compression_level = -1; } else { return -1; } return hts_parse_opt_list(format, cp); } /* * Tokenise options as (key(=value)?,)*(key(=value)?)? * NB: No provision for ',' appearing in the value! * Add backslashing rules? * * This could be used as part of a general command line option parser or * as a string concatenated onto the file open mode. * * Returns 0 on success * -1 on failure. */ static int hts_process_opts(htsFile *fp, const char *opts) { htsFormat fmt; fmt.specific = NULL; if (hts_parse_opt_list(&fmt, opts) != 0) return -1; if (hts_opt_apply(fp, fmt.specific) != 0) { hts_opt_free(fmt.specific); return -1; } hts_opt_free(fmt.specific); return 0; } htsFile *hts_hopen(struct hFILE *hfile, const char *fn, const char *mode) { htsFile *fp = (htsFile*)calloc(1, sizeof(htsFile)); char simple_mode[101], *cp, *opts; simple_mode[100] = '\0'; if (fp == NULL) goto error; fp->fn = strdup(fn); fp->is_be = ed_is_big(); // Split mode into simple_mode,opts strings if ((cp = strchr(mode, ','))) { strncpy(simple_mode, mode, cp-mode <= 100 ? cp-mode : 100); simple_mode[cp-mode] = '\0'; opts = cp+1; } else { strncpy(simple_mode, mode, 100); opts = NULL; } if (strchr(simple_mode, 'r')) { if (hts_detect_format(hfile, &fp->format) < 0) goto error; } else if (strchr(simple_mode, 'w') || strchr(simple_mode, 'a')) { htsFormat *fmt = &fp->format; fp->is_write = 1; if (strchr(simple_mode, 'b')) fmt->format = binary_format; else if (strchr(simple_mode, 'c')) fmt->format = cram; else fmt->format = text_format; if (strchr(simple_mode, 'z')) fmt->compression = bgzf; else if (strchr(simple_mode, 'g')) fmt->compression = gzip; else if (strchr(simple_mode, 'u')) fmt->compression = no_compression; else { // No compression mode specified, set to the default for the format switch (fmt->format) { case binary_format: fmt->compression = bgzf; break; case cram: fmt->compression = custom; break; case text_format: fmt->compression = no_compression; break; default: abort(); } } // Fill in category (if determinable; e.g. 'b' could be BAM or BCF) fmt->category = format_category(fmt->format); fmt->version.major = fmt->version.minor = -1; fmt->compression_level = -1; fmt->specific = NULL; } else goto error; switch (fp->format.format) { case binary_format: case bam: case bcf: fp->fp.bgzf = bgzf_hopen(hfile, simple_mode); if (fp->fp.bgzf == NULL) goto error; fp->is_bin = 1; break; case cram: fp->fp.cram = cram_dopen(hfile, fn, simple_mode); if (fp->fp.cram == NULL) goto error; if (!fp->is_write) cram_set_option(fp->fp.cram, CRAM_OPT_DECODE_MD, 1); fp->is_cram = 1; break; case text_format: case sam: case vcf: if (!fp->is_write) { #if KS_BGZF BGZF *gzfp = bgzf_hopen(hfile, simple_mode); #else // TODO Implement gzip hFILE adaptor hclose(hfile); // This won't work, especially for stdin gzFile gzfp = strcmp(fn, "-")? gzopen(fn, "rb") : gzdopen(fileno(stdin), "rb"); #endif if (gzfp) fp->fp.voidp = ks_init(gzfp); else goto error; } else if (fp->format.compression != no_compression) { fp->fp.bgzf = bgzf_hopen(hfile, simple_mode); if (fp->fp.bgzf == NULL) goto error; } else fp->fp.hfile = hfile; break; default: goto error; } if (opts) hts_process_opts(fp, opts); return fp; error: if (hts_verbose >= 2) fprintf(stderr, "[E::%s] fail to open file '%s'\n", __func__, fn); if (fp) { free(fp->fn); free(fp->fn_aux); free(fp); } return NULL; } int hts_close(htsFile *fp) { int ret, save; switch (fp->format.format) { case binary_format: case bam: case bcf: ret = bgzf_close(fp->fp.bgzf); break; case cram: if (!fp->is_write) { switch (cram_eof(fp->fp.cram)) { case 2: fprintf(stderr, "[W::%s] EOF marker is absent. The input is probably truncated.\n", __func__); break; case 0: /* not at EOF, but may not have wanted all seqs */ default: /* case 1, expected EOF */ break; } } ret = cram_close(fp->fp.cram); break; case text_format: case sam: case vcf: if (!fp->is_write) { #if KS_BGZF BGZF *gzfp = ((kstream_t*)fp->fp.voidp)->f; ret = bgzf_close(gzfp); #else gzFile gzfp = ((kstream_t*)fp->fp.voidp)->f; ret = gzclose(gzfp); #endif ks_destroy((kstream_t*)fp->fp.voidp); } else if (fp->format.compression != no_compression) ret = bgzf_close(fp->fp.bgzf); else ret = hclose(fp->fp.hfile); break; default: ret = -1; break; } save = errno; free(fp->fn); free(fp->fn_aux); free(fp->line.s); free(fp); errno = save; return ret; } const htsFormat *hts_get_format(htsFile *fp) { return fp? &fp->format : NULL; } const char *hts_format_file_extension(const htsFormat *format) { if (!format) return "?"; switch (format->format) { case sam: return "sam"; case bam: return "bam"; case bai: return "bai"; case cram: return "cram"; case crai: return "crai"; case vcf: return "vcf"; case bcf: return "bcf"; case csi: return "csi"; case gzi: return "gzi"; case tbi: return "tbi"; case bed: return "bed"; default: return "?"; } } int hts_set_opt(htsFile *fp, enum hts_fmt_option opt, ...) { int r; va_list args; if (opt == HTS_OPT_NTHREADS) { va_start(args, opt); int nthreads = va_arg(args, int); va_end(args); return hts_set_threads(fp, nthreads); } if (fp->format.format != cram) return 0; va_start(args, opt); r = cram_set_voption(fp->fp.cram, opt, args); va_end(args); return r; } int hts_set_threads(htsFile *fp, int n) { if (fp->format.compression == bgzf) { return bgzf_mt(fp->fp.bgzf, n, 256); } else if (fp->format.format == cram) { return hts_set_opt(fp, CRAM_OPT_NTHREADS, n); } else return 0; } int hts_set_fai_filename(htsFile *fp, const char *fn_aux) { free(fp->fn_aux); if (fn_aux) { fp->fn_aux = strdup(fn_aux); if (fp->fn_aux == NULL) return -1; } else fp->fn_aux = NULL; if (fp->format.format == cram) if (cram_set_option(fp->fp.cram, CRAM_OPT_REFERENCE, fp->fn_aux)) return -1; return 0; } // For VCF/BCF backward sweeper. Not exposing these functions because their // future is uncertain. Things will probably have to change with hFILE... BGZF *hts_get_bgzfp(htsFile *fp) { if ( fp->is_bin ) return fp->fp.bgzf; else return ((kstream_t*)fp->fp.voidp)->f; } int hts_useek(htsFile *fp, long uoffset, int where) { if ( fp->is_bin ) return bgzf_useek(fp->fp.bgzf, uoffset, where); else { ks_rewind((kstream_t*)fp->fp.voidp); ((kstream_t*)fp->fp.voidp)->seek_pos = uoffset; return bgzf_useek(((kstream_t*)fp->fp.voidp)->f, uoffset, where); } } long hts_utell(htsFile *fp) { if ( fp->is_bin ) return bgzf_utell(fp->fp.bgzf); else return ((kstream_t*)fp->fp.voidp)->seek_pos; } int hts_getline(htsFile *fp, int delimiter, kstring_t *str) { int ret, dret; ret = ks_getuntil((kstream_t*)fp->fp.voidp, delimiter, str, &dret); ++fp->lineno; return ret; } char **hts_readlist(const char *string, int is_file, int *_n) { int m = 0, n = 0, dret; char **s = 0; if ( is_file ) { #if KS_BGZF BGZF *fp = bgzf_open(string, "r"); #else gzFile fp = gzopen(string, "r"); #endif if ( !fp ) return NULL; kstream_t *ks; kstring_t str; str.s = 0; str.l = str.m = 0; ks = ks_init(fp); while (ks_getuntil(ks, KS_SEP_LINE, &str, &dret) >= 0) { if (str.l == 0) continue; n++; hts_expand(char*,n,m,s); s[n-1] = strdup(str.s); } ks_destroy(ks); #if KS_BGZF bgzf_close(fp); #else gzclose(fp); #endif free(str.s); } else { const char *q = string, *p = string; while ( 1 ) { if (*p == ',' || *p == 0) { n++; hts_expand(char*,n,m,s); s[n-1] = (char*)calloc(p - q + 1, 1); strncpy(s[n-1], q, p - q); q = p + 1; } if ( !*p ) break; p++; } } s = (char**)realloc(s, n * sizeof(char*)); *_n = n; return s; } char **hts_readlines(const char *fn, int *_n) { int m = 0, n = 0, dret; char **s = 0; #if KS_BGZF BGZF *fp = bgzf_open(fn, "r"); #else gzFile fp = gzopen(fn, "r"); #endif if ( fp ) { // read from file kstream_t *ks; kstring_t str; str.s = 0; str.l = str.m = 0; ks = ks_init(fp); while (ks_getuntil(ks, KS_SEP_LINE, &str, &dret) >= 0) { if (str.l == 0) continue; if (m == n) { m = m? m<<1 : 16; s = (char**)realloc(s, m * sizeof(char*)); } s[n++] = strdup(str.s); } ks_destroy(ks); #if KS_BGZF bgzf_close(fp); #else gzclose(fp); #endif s = (char**)realloc(s, n * sizeof(char*)); free(str.s); } else if (*fn == ':') { // read from string const char *q, *p; for (q = p = fn + 1;; ++p) if (*p == ',' || *p == 0) { if (m == n) { m = m? m<<1 : 16; s = (char**)realloc(s, m * sizeof(char*)); } s[n] = (char*)calloc(p - q + 1, 1); strncpy(s[n++], q, p - q); q = p + 1; if (*p == 0) break; } } else return 0; s = (char**)realloc(s, n * sizeof(char*)); *_n = n; return s; } // DEPRECATED: To be removed in a future HTSlib release int hts_file_type(const char *fname) { int len = strlen(fname); if ( !strcasecmp(".vcf.gz",fname+len-7) ) return FT_VCF_GZ; if ( !strcasecmp(".vcf",fname+len-4) ) return FT_VCF; if ( !strcasecmp(".bcf",fname+len-4) ) return FT_BCF_GZ; if ( !strcmp("-",fname) ) return FT_STDIN; hFILE *f = hopen(fname, "r"); if (f == NULL) return 0; htsFormat fmt; if (hts_detect_format(f, &fmt) < 0) { hclose_abruptly(f); return 0; } if (hclose(f) < 0) return 0; switch (fmt.format) { case vcf: return (fmt.compression == no_compression)? FT_VCF : FT_VCF_GZ; case bcf: return (fmt.compression == no_compression)? FT_BCF : FT_BCF_GZ; default: return 0; } } /**************** *** Indexing *** ****************/ #define HTS_MIN_MARKER_DIST 0x10000 // Finds the special meta bin // ((1<<(3 * n_lvls + 3)) - 1) / 7 + 1 #define META_BIN(idx) ((idx)->n_bins + 1) #define pair64_lt(a,b) ((a).u < (b).u) #include "htslib/ksort.h" KSORT_INIT(_off, hts_pair64_t, pair64_lt) typedef struct { int32_t m, n; uint64_t loff; hts_pair64_t *list; } bins_t; #include "htslib/khash.h" KHASH_MAP_INIT_INT(bin, bins_t) typedef khash_t(bin) bidx_t; typedef struct { int32_t n, m; uint64_t *offset; } lidx_t; struct __hts_idx_t { int fmt, min_shift, n_lvls, n_bins; uint32_t l_meta; int32_t n, m; uint64_t n_no_coor; bidx_t **bidx; lidx_t *lidx; uint8_t *meta; struct { uint32_t last_bin, save_bin; int last_coor, last_tid, save_tid, finished; uint64_t last_off, save_off; uint64_t off_beg, off_end; uint64_t n_mapped, n_unmapped; } z; // keep internal states }; static inline void insert_to_b(bidx_t *b, int bin, uint64_t beg, uint64_t end) { khint_t k; bins_t *l; int absent; k = kh_put(bin, b, bin, &absent); l = &kh_value(b, k); if (absent) { l->m = 1; l->n = 0; l->list = (hts_pair64_t*)calloc(l->m, sizeof(hts_pair64_t)); } if (l->n == l->m) { l->m <<= 1; l->list = (hts_pair64_t*)realloc(l->list, l->m * sizeof(hts_pair64_t)); } l->list[l->n].u = beg; l->list[l->n++].v = end; } static inline void insert_to_l(lidx_t *l, int64_t _beg, int64_t _end, uint64_t offset, int min_shift) { int i, beg, end; beg = _beg >> min_shift; end = (_end - 1) >> min_shift; if (l->m < end + 1) { int old_m = l->m; l->m = end + 1; kroundup32(l->m); l->offset = (uint64_t*)realloc(l->offset, l->m * sizeof(uint64_t)); memset(l->offset + old_m, 0xff, 8 * (l->m - old_m)); // fill l->offset with (uint64_t)-1 } if (beg == end) { // to save a loop in this case if (l->offset[beg] == (uint64_t)-1) l->offset[beg] = offset; } else { for (i = beg; i <= end; ++i) if (l->offset[i] == (uint64_t)-1) l->offset[i] = offset; } if (l->n < end + 1) l->n = end + 1; } hts_idx_t *hts_idx_init(int n, int fmt, uint64_t offset0, int min_shift, int n_lvls) { hts_idx_t *idx; idx = (hts_idx_t*)calloc(1, sizeof(hts_idx_t)); if (idx == NULL) return NULL; idx->fmt = fmt; idx->min_shift = min_shift; idx->n_lvls = n_lvls; idx->n_bins = ((1<<(3 * n_lvls + 3)) - 1) / 7; idx->z.save_bin = idx->z.save_tid = idx->z.last_tid = idx->z.last_bin = 0xffffffffu; idx->z.save_off = idx->z.last_off = idx->z.off_beg = idx->z.off_end = offset0; idx->z.last_coor = 0xffffffffu; if (n) { idx->n = idx->m = n; idx->bidx = (bidx_t**)calloc(n, sizeof(bidx_t*)); if (idx->bidx == NULL) { free(idx); return NULL; } idx->lidx = (lidx_t*) calloc(n, sizeof(lidx_t)); if (idx->lidx == NULL) { free(idx->bidx); free(idx); return NULL; } } return idx; } static void update_loff(hts_idx_t *idx, int i, int free_lidx) { bidx_t *bidx = idx->bidx[i]; lidx_t *lidx = &idx->lidx[i]; khint_t k; int l; uint64_t offset0 = 0; if (bidx) { k = kh_get(bin, bidx, META_BIN(idx)); if (k != kh_end(bidx)) offset0 = kh_val(bidx, k).list[0].u; for (l = 0; l < lidx->n && lidx->offset[l] == (uint64_t)-1; ++l) lidx->offset[l] = offset0; } else l = 1; for (; l < lidx->n; ++l) // fill missing values if (lidx->offset[l] == (uint64_t)-1) lidx->offset[l] = lidx->offset[l-1]; if (bidx == 0) return; for (k = kh_begin(bidx); k != kh_end(bidx); ++k) // set loff if (kh_exist(bidx, k)) { if ( kh_key(bidx, k) < idx->n_bins ) { int bot_bin = hts_bin_bot(kh_key(bidx, k), idx->n_lvls); // disable linear index if bot_bin out of bounds kh_val(bidx, k).loff = bot_bin < lidx->n ? lidx->offset[bot_bin] : 0; } else kh_val(bidx, k).loff = 0; } if (free_lidx) { free(lidx->offset); lidx->m = lidx->n = 0; lidx->offset = 0; } } static void compress_binning(hts_idx_t *idx, int i) { bidx_t *bidx = idx->bidx[i]; khint_t k; int l, m; if (bidx == 0) return; // merge a bin to its parent if the bin is too small for (l = idx->n_lvls; l > 0; --l) { unsigned start = hts_bin_first(l); for (k = kh_begin(bidx); k != kh_end(bidx); ++k) { bins_t *p, *q; if (!kh_exist(bidx, k) || kh_key(bidx, k) >= idx->n_bins || kh_key(bidx, k) < start) continue; p = &kh_value(bidx, k); if (l < idx->n_lvls && p->n > 1) ks_introsort(_off, p->n, p->list); if ((p->list[p->n - 1].v>>16) - (p->list[0].u>>16) < HTS_MIN_MARKER_DIST) { khint_t kp; kp = kh_get(bin, bidx, hts_bin_parent(kh_key(bidx, k))); if (kp == kh_end(bidx)) continue; q = &kh_val(bidx, kp); if (q->n + p->n > q->m) { q->m = q->n + p->n; kroundup32(q->m); q->list = (hts_pair64_t*)realloc(q->list, q->m * sizeof(hts_pair64_t)); } memcpy(q->list + q->n, p->list, p->n * sizeof(hts_pair64_t)); q->n += p->n; free(p->list); kh_del(bin, bidx, k); } } } k = kh_get(bin, bidx, 0); if (k != kh_end(bidx)) ks_introsort(_off, kh_val(bidx, k).n, kh_val(bidx, k).list); // merge adjacent chunks that start from the same BGZF block for (k = kh_begin(bidx); k != kh_end(bidx); ++k) { bins_t *p; if (!kh_exist(bidx, k) || kh_key(bidx, k) >= idx->n_bins) continue; p = &kh_value(bidx, k); for (l = 1, m = 0; l < p->n; ++l) { if (p->list[m].v>>16 >= p->list[l].u>>16) { if (p->list[m].v < p->list[l].v) p->list[m].v = p->list[l].v; } else p->list[++m] = p->list[l]; } p->n = m + 1; } } void hts_idx_finish(hts_idx_t *idx, uint64_t final_offset) { int i; if (idx == NULL || idx->z.finished) return; // do not run this function on an empty index or multiple times if (idx->z.save_tid >= 0) { insert_to_b(idx->bidx[idx->z.save_tid], idx->z.save_bin, idx->z.save_off, final_offset); insert_to_b(idx->bidx[idx->z.save_tid], META_BIN(idx), idx->z.off_beg, final_offset); insert_to_b(idx->bidx[idx->z.save_tid], META_BIN(idx), idx->z.n_mapped, idx->z.n_unmapped); } for (i = 0; i < idx->n; ++i) { update_loff(idx, i, (idx->fmt == HTS_FMT_CSI)); compress_binning(idx, i); } idx->z.finished = 1; } int hts_idx_push(hts_idx_t *idx, int tid, int beg, int end, uint64_t offset, int is_mapped) { int bin; if (tid<0) beg = -1, end = 0; if (tid >= idx->m) { // enlarge the index int32_t oldm = idx->m; idx->m = idx->m? idx->m<<1 : 2; idx->bidx = (bidx_t**)realloc(idx->bidx, idx->m * sizeof(bidx_t*)); idx->lidx = (lidx_t*) realloc(idx->lidx, idx->m * sizeof(lidx_t)); memset(&idx->bidx[oldm], 0, (idx->m - oldm) * sizeof(bidx_t*)); memset(&idx->lidx[oldm], 0, (idx->m - oldm) * sizeof(lidx_t)); } if (idx->n < tid + 1) idx->n = tid + 1; if (idx->z.finished) return 0; if (idx->z.last_tid != tid || (idx->z.last_tid >= 0 && tid < 0)) { // change of chromosome if ( tid>=0 && idx->n_no_coor ) { if (hts_verbose >= 1) fprintf(stderr,"[E::%s] NO_COOR reads not in a single block at the end %d %d\n", __func__, tid,idx->z.last_tid); return -1; } if (tid>=0 && idx->bidx[tid] != 0) { if (hts_verbose >= 1) fprintf(stderr, "[E::%s] chromosome blocks not continuous\n", __func__); return -1; } idx->z.last_tid = tid; idx->z.last_bin = 0xffffffffu; } else if (tid >= 0 && idx->z.last_coor > beg) { // test if positions are out of order if (hts_verbose >= 1) fprintf(stderr, "[E::%s] unsorted positions\n", __func__); return -1; } if ( tid>=0 ) { if (idx->bidx[tid] == 0) idx->bidx[tid] = kh_init(bin); if ( is_mapped) insert_to_l(&idx->lidx[tid], beg, end, idx->z.last_off, idx->min_shift); // last_off points to the start of the current record } else idx->n_no_coor++; bin = hts_reg2bin(beg, end, idx->min_shift, idx->n_lvls); if ((int)idx->z.last_bin != bin) { // then possibly write the binning index if (idx->z.save_bin != 0xffffffffu) // save_bin==0xffffffffu only happens to the first record insert_to_b(idx->bidx[idx->z.save_tid], idx->z.save_bin, idx->z.save_off, idx->z.last_off); if (idx->z.last_bin == 0xffffffffu && idx->z.save_bin != 0xffffffffu) { // change of chr; keep meta information idx->z.off_end = idx->z.last_off; insert_to_b(idx->bidx[idx->z.save_tid], META_BIN(idx), idx->z.off_beg, idx->z.off_end); insert_to_b(idx->bidx[idx->z.save_tid], META_BIN(idx), idx->z.n_mapped, idx->z.n_unmapped); idx->z.n_mapped = idx->z.n_unmapped = 0; idx->z.off_beg = idx->z.off_end; } idx->z.save_off = idx->z.last_off; idx->z.save_bin = idx->z.last_bin = bin; idx->z.save_tid = tid; } if (is_mapped) ++idx->z.n_mapped; else ++idx->z.n_unmapped; idx->z.last_off = offset; idx->z.last_coor = beg; return 0; } void hts_idx_destroy(hts_idx_t *idx) { khint_t k; int i; if (idx == 0) return; // For HTS_FMT_CRAI, idx actually points to a different type -- see sam.c if (idx->fmt == HTS_FMT_CRAI) { hts_cram_idx_t *cidx = (hts_cram_idx_t *) idx; cram_index_free(cidx->cram); free(cidx); return; } for (i = 0; i < idx->m; ++i) { bidx_t *bidx = idx->bidx[i]; free(idx->lidx[i].offset); if (bidx == 0) continue; for (k = kh_begin(bidx); k != kh_end(bidx); ++k) if (kh_exist(bidx, k)) free(kh_value(bidx, k).list); kh_destroy(bin, bidx); } free(idx->bidx); free(idx->lidx); free(idx->meta); free(idx); } // The optimizer eliminates these ed_is_big() calls; still it would be good to // TODO Determine endianness at configure- or compile-time static inline ssize_t HTS_RESULT_USED idx_write_int32(BGZF *fp, int32_t x) { if (ed_is_big()) x = ed_swap_4(x); return bgzf_write(fp, &x, sizeof x); } static inline ssize_t HTS_RESULT_USED idx_write_uint32(BGZF *fp, uint32_t x) { if (ed_is_big()) x = ed_swap_4(x); return bgzf_write(fp, &x, sizeof x); } static inline ssize_t HTS_RESULT_USED idx_write_uint64(BGZF *fp, uint64_t x) { if (ed_is_big()) x = ed_swap_8(x); return bgzf_write(fp, &x, sizeof x); } static inline void swap_bins(bins_t *p) { int i; for (i = 0; i < p->n; ++i) { ed_swap_8p(&p->list[i].u); ed_swap_8p(&p->list[i].v); } } static int hts_idx_save_core(const hts_idx_t *idx, BGZF *fp, int fmt) { int32_t i, j; #define check(ret) if ((ret) < 0) return -1 check(idx_write_int32(fp, idx->n)); if (fmt == HTS_FMT_TBI && idx->l_meta) check(bgzf_write(fp, idx->meta, idx->l_meta)); for (i = 0; i < idx->n; ++i) { khint_t k; bidx_t *bidx = idx->bidx[i]; lidx_t *lidx = &idx->lidx[i]; // write binning index check(idx_write_int32(fp, bidx? kh_size(bidx) : 0)); if (bidx) for (k = kh_begin(bidx); k != kh_end(bidx); ++k) if (kh_exist(bidx, k)) { bins_t *p = &kh_value(bidx, k); check(idx_write_uint32(fp, kh_key(bidx, k))); if (fmt == HTS_FMT_CSI) check(idx_write_uint64(fp, p->loff)); //int j;for(j=0;jn;++j)fprintf(stderr,"%d,%llx,%d,%llx:%llx\n",kh_key(bidx,k),kh_val(bidx, k).loff,j,p->list[j].u,p->list[j].v); check(idx_write_int32(fp, p->n)); for (j = 0; j < p->n; ++j) { check(idx_write_uint64(fp, p->list[j].u)); check(idx_write_uint64(fp, p->list[j].v)); } } // write linear index if (fmt != HTS_FMT_CSI) { check(idx_write_int32(fp, lidx->n)); for (j = 0; j < lidx->n; ++j) check(idx_write_uint64(fp, lidx->offset[j])); } } check(idx_write_uint64(fp, idx->n_no_coor)); return 0; #undef check } int hts_idx_save(const hts_idx_t *idx, const char *fn, int fmt) { int ret, save; char *fnidx = (char*)calloc(1, strlen(fn) + 5); if (fnidx == NULL) return -1; strcpy(fnidx, fn); switch (fmt) { case HTS_FMT_BAI: strcat(fnidx, ".bai"); break; case HTS_FMT_CSI: strcat(fnidx, ".csi"); break; case HTS_FMT_TBI: strcat(fnidx, ".tbi"); break; default: abort(); } ret = hts_idx_save_as(idx, fn, fnidx, fmt); save = errno; free(fnidx); errno = save; return ret; } int hts_idx_save_as(const hts_idx_t *idx, const char *fn, const char *fnidx, int fmt) { BGZF *fp; #define check(ret) if ((ret) < 0) goto fail if (fnidx == NULL) return hts_idx_save(idx, fn, fmt); fp = bgzf_open(fnidx, (fmt == HTS_FMT_BAI)? "wu" : "w"); if (fp == NULL) return -1; if (fmt == HTS_FMT_CSI) { check(bgzf_write(fp, "CSI\1", 4)); check(idx_write_int32(fp, idx->min_shift)); check(idx_write_int32(fp, idx->n_lvls)); check(idx_write_uint32(fp, idx->l_meta)); if (idx->l_meta) check(bgzf_write(fp, idx->meta, idx->l_meta)); } else if (fmt == HTS_FMT_TBI) { check(bgzf_write(fp, "TBI\1", 4)); } else if (fmt == HTS_FMT_BAI) { check(bgzf_write(fp, "BAI\1", 4)); } else abort(); check(hts_idx_save_core(idx, fp, fmt)); return bgzf_close(fp); #undef check fail: bgzf_close(fp); return -1; } static int hts_idx_load_core(hts_idx_t *idx, BGZF *fp, int fmt) { int32_t i, n, is_be; is_be = ed_is_big(); if (idx == NULL) return -4; for (i = 0; i < idx->n; ++i) { bidx_t *h; lidx_t *l = &idx->lidx[i]; uint32_t key; int j, absent; bins_t *p; h = idx->bidx[i] = kh_init(bin); if (bgzf_read(fp, &n, 4) != 4) return -1; if (is_be) ed_swap_4p(&n); for (j = 0; j < n; ++j) { khint_t k; if (bgzf_read(fp, &key, 4) != 4) return -1; if (is_be) ed_swap_4p(&key); k = kh_put(bin, h, key, &absent); if (absent <= 0) return -3; // Duplicate bin number p = &kh_val(h, k); if (fmt == HTS_FMT_CSI) { if (bgzf_read(fp, &p->loff, 8) != 8) return -1; if (is_be) ed_swap_8p(&p->loff); } else p->loff = 0; if (bgzf_read(fp, &p->n, 4) != 4) return -1; if (is_be) ed_swap_4p(&p->n); p->m = p->n; p->list = (hts_pair64_t*)malloc(p->m * sizeof(hts_pair64_t)); if (p->list == NULL) return -2; if (bgzf_read(fp, p->list, p->n<<4) != p->n<<4) return -1; if (is_be) swap_bins(p); } if (fmt != HTS_FMT_CSI) { // load linear index int j; if (bgzf_read(fp, &l->n, 4) != 4) return -1; if (is_be) ed_swap_4p(&l->n); l->m = l->n; l->offset = (uint64_t*)malloc(l->n * sizeof(uint64_t)); if (l->offset == NULL) return -2; if (bgzf_read(fp, l->offset, l->n << 3) != l->n << 3) return -1; if (is_be) for (j = 0; j < l->n; ++j) ed_swap_8p(&l->offset[j]); for (j = 1; j < l->n; ++j) // fill missing values; may happen given older samtools and tabix if (l->offset[j] == 0) l->offset[j] = l->offset[j-1]; update_loff(idx, i, 1); } } if (bgzf_read(fp, &idx->n_no_coor, 8) != 8) idx->n_no_coor = 0; if (is_be) ed_swap_8p(&idx->n_no_coor); return 0; } static hts_idx_t *hts_idx_load_local(const char *fn) { uint8_t magic[4]; int i, is_be; hts_idx_t *idx = NULL; uint8_t *meta = NULL; BGZF *fp = bgzf_open(fn, "r"); if (fp == NULL) return NULL; is_be = ed_is_big(); if (bgzf_read(fp, magic, 4) != 4) goto fail; if (memcmp(magic, "CSI\1", 4) == 0) { uint32_t x[3], n; if (bgzf_read(fp, x, 12) != 12) goto fail; if (is_be) for (i = 0; i < 3; ++i) ed_swap_4p(&x[i]); if (x[2]) { if ((meta = (uint8_t*)malloc(x[2])) == NULL) goto fail; if (bgzf_read(fp, meta, x[2]) != x[2]) goto fail; } if (bgzf_read(fp, &n, 4) != 4) goto fail; if (is_be) ed_swap_4p(&n); if ((idx = hts_idx_init(n, HTS_FMT_CSI, 0, x[0], x[1])) == NULL) goto fail; idx->l_meta = x[2]; idx->meta = meta; meta = NULL; if (hts_idx_load_core(idx, fp, HTS_FMT_CSI) < 0) goto fail; } else if (memcmp(magic, "TBI\1", 4) == 0) { uint32_t x[8]; if (bgzf_read(fp, x, 32) != 32) goto fail; if (is_be) for (i = 0; i < 8; ++i) ed_swap_4p(&x[i]); if ((idx = hts_idx_init(x[0], HTS_FMT_TBI, 0, 14, 5)) == NULL) goto fail; idx->l_meta = 28 + x[7]; if ((idx->meta = (uint8_t*)malloc(idx->l_meta)) == NULL) goto fail; memcpy(idx->meta, &x[1], 28); if (bgzf_read(fp, idx->meta + 28, x[7]) != x[7]) goto fail; if (hts_idx_load_core(idx, fp, HTS_FMT_TBI) < 0) goto fail; } else if (memcmp(magic, "BAI\1", 4) == 0) { uint32_t n; if (bgzf_read(fp, &n, 4) != 4) goto fail; if (is_be) ed_swap_4p(&n); idx = hts_idx_init(n, HTS_FMT_BAI, 0, 14, 5); if (hts_idx_load_core(idx, fp, HTS_FMT_BAI) < 0) goto fail; } else { errno = EINVAL; goto fail; } bgzf_close(fp); return idx; fail: bgzf_close(fp); hts_idx_destroy(idx); free(meta); return NULL; } void hts_idx_set_meta(hts_idx_t *idx, int l_meta, uint8_t *meta, int is_copy) { if (idx->meta) free(idx->meta); idx->l_meta = l_meta; if (is_copy) { idx->meta = (uint8_t*)malloc(l_meta); memcpy(idx->meta, meta, l_meta); } else idx->meta = meta; } uint8_t *hts_idx_get_meta(hts_idx_t *idx, int *l_meta) { *l_meta = idx->l_meta; return idx->meta; } const char **hts_idx_seqnames(const hts_idx_t *idx, int *n, hts_id2name_f getid, void *hdr) { if ( !idx->n ) { *n = 0; return NULL; } int tid = 0, i; const char **names = (const char**) calloc(idx->n,sizeof(const char*)); for (i=0; in; i++) { bidx_t *bidx = idx->bidx[i]; if ( !bidx ) continue; names[tid++] = getid(hdr,i); } *n = tid; return names; } int hts_idx_get_stat(const hts_idx_t* idx, int tid, uint64_t* mapped, uint64_t* unmapped) { if ( idx->fmt == HTS_FMT_CRAI ) { *mapped = 0; *unmapped = 0; return -1; } bidx_t *h = idx->bidx[tid]; khint_t k = kh_get(bin, h, META_BIN(idx)); if (k != kh_end(h)) { *mapped = kh_val(h, k).list[1].u; *unmapped = kh_val(h, k).list[1].v; return 0; } else { *mapped = 0; *unmapped = 0; return -1; } } uint64_t hts_idx_get_n_no_coor(const hts_idx_t* idx) { return idx->n_no_coor; } /**************** *** Iterator *** ****************/ static inline int reg2bins(int64_t beg, int64_t end, hts_itr_t *itr, int min_shift, int n_lvls) { int l, t, s = min_shift + (n_lvls<<1) + n_lvls; if (beg >= end) return 0; if (end >= 1LL<>s); e = t + (end>>s); n = e - b + 1; if (itr->bins.n + n > itr->bins.m) { itr->bins.m = itr->bins.n + n; kroundup32(itr->bins.m); itr->bins.a = (int*)realloc(itr->bins.a, sizeof(int) * itr->bins.m); } for (i = b; i <= e; ++i) itr->bins.a[itr->bins.n++] = i; } return itr->bins.n; } hts_itr_t *hts_itr_query(const hts_idx_t *idx, int tid, int beg, int end, hts_readrec_func *readrec) { int i, n_off, l, bin; hts_pair64_t *off; khint_t k; bidx_t *bidx; uint64_t min_off; hts_itr_t *iter = 0; if (tid < 0) { int finished0 = 0; uint64_t off0 = (uint64_t)-1; khint_t k; switch (tid) { case HTS_IDX_START: // Find the smallest offset, note that sequence ids may not be ordered sequentially for (i=0; in; i++) { bidx = idx->bidx[i]; k = kh_get(bin, bidx, META_BIN(idx)); if (k == kh_end(bidx)) continue; if ( off0 > kh_val(bidx, k).list[0].u ) off0 = kh_val(bidx, k).list[0].u; } if ( off0==(uint64_t)-1 && idx->n_no_coor ) off0 = 0; // only no-coor reads in this bam break; case HTS_IDX_NOCOOR: if ( idx->n>0 ) { bidx = idx->bidx[idx->n - 1]; k = kh_get(bin, bidx, META_BIN(idx)); if (k != kh_end(bidx)) off0 = kh_val(bidx, k).list[0].v; } if ( off0==(uint64_t)-1 && idx->n_no_coor ) off0 = 0; // only no-coor reads in this bam break; case HTS_IDX_REST: off0 = 0; break; case HTS_IDX_NONE: finished0 = 1; off0 = 0; break; default: return 0; } if (off0 != (uint64_t)-1) { iter = (hts_itr_t*)calloc(1, sizeof(hts_itr_t)); iter->read_rest = 1; iter->finished = finished0; iter->curr_off = off0; iter->readrec = readrec; return iter; } else return 0; } if (beg < 0) beg = 0; if (end < beg) return 0; if (tid >= idx->n || (bidx = idx->bidx[tid]) == NULL) return 0; iter = (hts_itr_t*)calloc(1, sizeof(hts_itr_t)); iter->tid = tid, iter->beg = beg, iter->end = end; iter->i = -1; iter->readrec = readrec; // compute min_off bin = hts_bin_first(idx->n_lvls) + (beg>>idx->min_shift); do { int first; k = kh_get(bin, bidx, bin); if (k != kh_end(bidx)) break; first = (hts_bin_parent(bin)<<3) + 1; if (bin > first) --bin; else bin = hts_bin_parent(bin); } while (bin); if (bin == 0) k = kh_get(bin, bidx, bin); min_off = k != kh_end(bidx)? kh_val(bidx, k).loff : 0; // retrieve bins reg2bins(beg, end, iter, idx->min_shift, idx->n_lvls); for (i = n_off = 0; i < iter->bins.n; ++i) if ((k = kh_get(bin, bidx, iter->bins.a[i])) != kh_end(bidx)) n_off += kh_value(bidx, k).n; if (n_off == 0) return iter; off = (hts_pair64_t*)calloc(n_off, sizeof(hts_pair64_t)); for (i = n_off = 0; i < iter->bins.n; ++i) { if ((k = kh_get(bin, bidx, iter->bins.a[i])) != kh_end(bidx)) { int j; bins_t *p = &kh_value(bidx, k); for (j = 0; j < p->n; ++j) if (p->list[j].v > min_off) off[n_off++] = p->list[j]; } } if (n_off == 0) { free(off); return iter; } ks_introsort(_off, n_off, off); // resolve completely contained adjacent blocks for (i = 1, l = 0; i < n_off; ++i) if (off[l].v < off[i].v) off[++l] = off[i]; n_off = l + 1; // resolve overlaps between adjacent blocks; this may happen due to the merge in indexing for (i = 1; i < n_off; ++i) if (off[i-1].v >= off[i].u) off[i-1].v = off[i].u; // merge adjacent blocks for (i = 1, l = 0; i < n_off; ++i) { if (off[l].v>>16 == off[i].u>>16) off[l].v = off[i].v; else off[++l] = off[i]; } n_off = l + 1; iter->n_off = n_off; iter->off = off; return iter; } void hts_itr_destroy(hts_itr_t *iter) { if (iter) { free(iter->off); free(iter->bins.a); free(iter); } } static inline long long push_digit(long long i, char c) { // ensure subtraction occurs first, avoiding overflow for >= MAX-48 or so int digit = c - '0'; return 10 * i + digit; } long long hts_parse_decimal(const char *str, char **strend, int flags) { long long n = 0; int decimals = 0, e = 0, lost = 0; char sign = '+', esign = '+'; const char *s; while (isspace(*str)) str++; s = str; if (*s == '+' || *s == '-') sign = *s++; while (*s) if (isdigit(*s)) n = push_digit(n, *s++); else if (*s == ',' && (flags & HTS_PARSE_THOUSANDS_SEP)) s++; else break; if (*s == '.') { s++; while (isdigit(*s)) decimals++, n = push_digit(n, *s++); } if (*s == 'E' || *s == 'e') { s++; if (*s == '+' || *s == '-') esign = *s++; while (isdigit(*s)) e = push_digit(e, *s++); if (esign == '-') e = -e; } e -= decimals; while (e > 0) n *= 10, e--; while (e < 0) lost += n % 10, n /= 10, e++; if (lost > 0 && hts_verbose >= 3) fprintf(stderr, "[W::%s] discarding fractional part of %.*s\n", __func__, (int)(s - str), str); if (strend) *strend = (char *) s; else if (*s && hts_verbose >= 2) fprintf(stderr, "[W::%s] ignoring unknown characters after %.*s[%s]\n", __func__, (int)(s - str), str, s); return (sign == '+')? n : -n; } const char *hts_parse_reg(const char *s, int *beg, int *end) { char *hyphen; const char *colon = strrchr(s, ':'); if (colon == NULL) { *beg = 0; *end = INT_MAX; return s + strlen(s); } *beg = hts_parse_decimal(colon+1, &hyphen, HTS_PARSE_THOUSANDS_SEP) - 1; if (*beg < 0) *beg = 0; if (*hyphen == '\0') *end = INT_MAX; else if (*hyphen == '-') *end = hts_parse_decimal(hyphen+1, NULL, HTS_PARSE_THOUSANDS_SEP); else return NULL; if (*beg >= *end) return NULL; return colon; } hts_itr_t *hts_itr_querys(const hts_idx_t *idx, const char *reg, hts_name2id_f getid, void *hdr, hts_itr_query_func *itr_query, hts_readrec_func *readrec) { int tid, beg, end; const char *q; if (strcmp(reg, ".") == 0) return itr_query(idx, HTS_IDX_START, 0, 0, readrec); else if (strcmp(reg, "*") == 0) return itr_query(idx, HTS_IDX_NOCOOR, 0, 0, readrec); q = hts_parse_reg(reg, &beg, &end); if (q) { char *tmp = (char*)alloca(q - reg + 1); strncpy(tmp, reg, q - reg); tmp[q - reg] = 0; tid = getid(hdr, tmp); } else { // not parsable as a region, but possibly a sequence named "foo:a" tid = getid(hdr, reg); beg = 0; end = INT_MAX; } if (tid < 0) return NULL; return itr_query(idx, tid, beg, end, readrec); } int hts_itr_next(BGZF *fp, hts_itr_t *iter, void *r, void *data) { int ret, tid, beg, end; if (iter == NULL || iter->finished) return -1; if (iter->read_rest) { if (iter->curr_off) { // seek to the start if (bgzf_seek(fp, iter->curr_off, SEEK_SET) < 0) return -1; iter->curr_off = 0; // only seek once } ret = iter->readrec(fp, data, r, &tid, &beg, &end); if (ret < 0) iter->finished = 1; iter->curr_tid = tid; iter->curr_beg = beg; iter->curr_end = end; return ret; } if (iter->off == 0) return -1; for (;;) { if (iter->curr_off == 0 || iter->curr_off >= iter->off[iter->i].v) { // then jump to the next chunk if (iter->i == iter->n_off - 1) { ret = -1; break; } // no more chunks if (iter->i < 0 || iter->off[iter->i].v != iter->off[iter->i+1].u) { // not adjacent chunks; then seek if (bgzf_seek(fp, iter->off[iter->i+1].u, SEEK_SET) < 0) return -1; iter->curr_off = bgzf_tell(fp); } ++iter->i; } if ((ret = iter->readrec(fp, data, r, &tid, &beg, &end)) >= 0) { iter->curr_off = bgzf_tell(fp); if (tid != iter->tid || beg >= iter->end) { // no need to proceed ret = -1; break; } else if (end > iter->beg && iter->end > beg) { iter->curr_tid = tid; iter->curr_beg = beg; iter->curr_end = end; return ret; } } else break; // end of file or error } iter->finished = 1; return ret; } /********************** *** Retrieve index *** **********************/ static char *test_and_fetch(const char *fn) { FILE *fp; if (hisremote(fn)) { const int buf_size = 1 * 1024 * 1024; hFILE *fp_remote; uint8_t *buf; int l; const char *p; for (p = fn + strlen(fn) - 1; p >= fn; --p) if (*p == '/') break; ++p; // p now points to the local file name // Attempt to open local file first if ((fp = fopen((char*)p, "rb")) != 0) { fclose(fp); return (char*)p; } // Attempt to open remote file. Stay quiet on failure, it is OK to fail when trying first .csi then .tbi index. if ((fp_remote = hopen(fn, "r")) == 0) return 0; if ((fp = fopen(p, "w")) == 0) { if (hts_verbose >= 1) fprintf(stderr, "[E::%s] fail to create file '%s' in the working directory\n", __func__, p); hclose_abruptly(fp_remote); return 0; } if (hts_verbose >= 3) fprintf(stderr, "[M::%s] downloading file '%s' to local directory\n", __func__, fn); buf = (uint8_t*)calloc(buf_size, 1); while ((l = hread(fp_remote, buf, buf_size)) > 0) fwrite(buf, 1, l, fp); free(buf); fclose(fp); if (hclose(fp_remote) != 0) fprintf(stderr, "[E::%s] fail to close remote file '%s'\n", __func__, fn); return (char*)p; } else { if ((fp = fopen(fn, "rb")) == 0) return 0; fclose(fp); return (char*)fn; } } char *hts_idx_getfn(const char *fn, const char *ext) { int i, l_fn, l_ext; char *fnidx, *ret; l_fn = strlen(fn); l_ext = strlen(ext); fnidx = (char*)calloc(l_fn + l_ext + 1, 1); strcpy(fnidx, fn); strcpy(fnidx + l_fn, ext); if ((ret = test_and_fetch(fnidx)) == 0) { for (i = l_fn - 1; i > 0; --i) if (fnidx[i] == '.') break; strcpy(fnidx + i, ext); ret = test_and_fetch(fnidx); } if (ret == 0) { free(fnidx); return 0; } l_fn = strlen(ret); memmove(fnidx, ret, l_fn + 1); return fnidx; } hts_idx_t *hts_idx_load(const char *fn, int fmt) { char *fnidx; hts_idx_t *idx; fnidx = hts_idx_getfn(fn, ".csi"); if (! fnidx) fnidx = hts_idx_getfn(fn, fmt == HTS_FMT_BAI? ".bai" : ".tbi"); if (fnidx == 0) return 0; idx = hts_idx_load2(fn, fnidx); free(fnidx); return idx; } hts_idx_t *hts_idx_load2(const char *fn, const char *fnidx) { // Check that the index file is up to date, the main file might have changed struct stat stat_idx,stat_main; if ( !stat(fn, &stat_main) && !stat(fnidx, &stat_idx) ) { if ( stat_idx.st_mtime < stat_main.st_mtime ) fprintf(stderr, "Warning: The index file is older than the data file: %s\n", fnidx); } return hts_idx_load_local(fnidx); } stacks-1.44/htslib/._hts_internal.h000644 000765 000024 00000000265 12750743246 020064 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/hts_internal.h000644 000765 000024 00000007244 12750743246 017653 0ustar00catchenstaff000000 000000 /* hts_internal.h -- internal functions; not part of the public API. Copyright (C) 2015-2016 Genome Research Ltd. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #ifndef HTSLIB_HTS_INTERNAL_H #define HTSLIB_HTS_INTERNAL_H #include #include #include "htslib/hts.h" #ifdef __cplusplus extern "C" { #endif // The functions operate on ints such as are returned by fgetc(), // i.e., characters represented as unsigned-char-valued ints, or EOF. // To operate on plain chars (and to avoid warnings on some platforms), // technically one must cast to unsigned char everywhere (see CERT STR37-C) // or less painfully use these *_c() functions that operate on plain chars // (but not EOF, which must be considered separately where it is applicable). // TODO We may eventually wish to implement these functions directly without // using their equivalents, and thus make them immune to locales. static inline int isalnum_c(char c) { return isalnum((unsigned char) c); } static inline int isalpha_c(char c) { return isalpha((unsigned char) c); } static inline int isdigit_c(char c) { return isdigit((unsigned char) c); } static inline int isgraph_c(char c) { return isgraph((unsigned char) c); } static inline int islower_c(char c) { return islower((unsigned char) c); } static inline int isprint_c(char c) { return isprint((unsigned char) c); } static inline int isspace_c(char c) { return isspace((unsigned char) c); } static inline int isupper_c(char c) { return isupper((unsigned char) c); } static inline char tolower_c(char c) { return tolower((unsigned char) c); } static inline char toupper_c(char c) { return toupper((unsigned char) c); } struct cram_fd; char *hts_idx_getfn(const char *fn, const char *ext); // The CRAM implementation stores the loaded index within the cram_fd rather // than separately as is done elsewhere in htslib. So if p is a pointer to // an hts_idx_t with p->fmt == HTS_FMT_CRAI, then it actually points to an // hts_cram_idx_t and should be cast accordingly. typedef struct hts_cram_idx_t { int fmt; struct cram_fd *cram; } hts_cram_idx_t; struct hts_path_itr { kstring_t path, entry; void *dirv; // DIR * privately const char *pathdir, *prefix, *suffix; size_t prefix_len, suffix_len, entry_dir_l; }; void hts_path_itr_setup(struct hts_path_itr *itr, const char *path, const char *builtin_path, const char *prefix, size_t prefix_len, const char *suffix, size_t suffix_len); const char *hts_path_itr_next(struct hts_path_itr *itr); void *load_plugin(void **pluginp, const char *filename, const char *symbol); void *plugin_sym(void *plugin, const char *name, const char **errmsg); void close_plugin(void *plugin); #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/._htsfile.c000644 000765 000024 00000000265 12750743246 017023 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htsfile.c000644 000765 000024 00000015231 12750743246 016605 0ustar00catchenstaff000000 000000 /* htsfile.c -- file identifier and minimal viewer. Copyright (C) 2014-2016 Genome Research Ltd. Author: John Marshall Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include #include #include #include "htslib/hfile.h" #include "htslib/hts.h" #include "htslib/sam.h" #include "htslib/vcf.h" enum { identify, view_headers, view_all } mode = identify; int show_headers = 1; int status = EXIT_SUCCESS; /* Exit status from main */ static htsFile *dup_stdout(const char *mode) { int fd = dup(STDOUT_FILENO); if (fd < 0) { perror("htsfile: Couldn't duplicate stdout"); return NULL; } hFILE *hfp = hdopen(fd, mode); return hfp? hts_hopen(hfp, "-", mode) : NULL; } static int view_sam(hFILE *hfp, const char *filename) { samFile *in = hts_hopen(hfp, filename, "r"); bam_hdr_t *hdr = NULL; samFile *out = NULL; if (in == NULL) { status = EXIT_FAILURE; return 0; } hdr = sam_hdr_read(in); if (hdr == NULL) { status = EXIT_FAILURE; goto clean; } out = dup_stdout("w"); if (out == NULL) { status = EXIT_FAILURE; goto clean; } if (show_headers) { if (sam_hdr_write(out, hdr) != 0) { status = EXIT_FAILURE; goto clean; } } if (mode == view_all) { bam1_t *b = bam_init1(); int ret; while ((ret = sam_read1(in, hdr, b)) >= 0) { if (sam_write1(out, hdr, b) < 0) { status = EXIT_FAILURE; goto clean; } } bam_destroy1(b); if (ret != -1) // eof status = EXIT_FAILURE; } clean: if (hdr != NULL) bam_hdr_destroy(hdr); if (out != NULL && hts_close(out) != 0) status = EXIT_FAILURE; if (hts_close(in) != 0) status = EXIT_FAILURE; return 1; } static int view_vcf(hFILE *hfp, const char *filename) { vcfFile *in = hts_hopen(hfp, filename, "r"); bcf_hdr_t *hdr = NULL; vcfFile *out = NULL; if (in == NULL) { status = EXIT_FAILURE; return 0; } hdr = bcf_hdr_read(in); if (hdr == NULL) { status = EXIT_FAILURE; goto clean; } out = dup_stdout("w"); if (out == NULL) { status = EXIT_FAILURE; goto clean; } if (show_headers) { if (bcf_hdr_write(out, hdr) != 0) { status = EXIT_FAILURE; goto clean; } } if (mode == view_all) { bcf1_t *rec = bcf_init(); while (bcf_read(in, hdr, rec) >= 0) { if (bcf_write(out, hdr, rec) < 0) { status = EXIT_FAILURE; goto clean; } } bcf_destroy(rec); } clean: if (hdr != NULL) bcf_hdr_destroy(hdr); if (out != NULL) hts_close(out); hts_close(in); return 1; } static void usage(FILE *fp, int status) { fprintf(fp, "Usage: htsfile [-chHv] FILE...\n" "Options:\n" " -c, --view Write textual form of FILEs to standard output\n" " -h, --header-only Display only headers in view mode, not records\n" " -H, --no-header Suppress header display in view mode\n" " -v, --verbose Increase verbosity of warnings and diagnostics\n"); exit(status); } int main(int argc, char **argv) { static const struct option options[] = { { "header-only", no_argument, NULL, 'h' }, { "no-header", no_argument, NULL, 'H' }, { "view", no_argument, NULL, 'c' }, { "verbose", no_argument, NULL, 'v' }, { "help", no_argument, NULL, '?' }, { "version", no_argument, NULL, 1 }, { NULL, 0, NULL, 0 } }; int c, i; status = EXIT_SUCCESS; while ((c = getopt_long(argc, argv, "chHv?", options, NULL)) >= 0) switch (c) { case 'c': mode = view_all; break; case 'h': mode = view_headers; show_headers = 1; break; case 'H': show_headers = 0; break; case 'v': hts_verbose++; break; case 1: printf( "htsfile (htslib) %s\n" "Copyright (C) 2016 Genome Research Ltd.\n", hts_version()); exit(EXIT_SUCCESS); break; case '?': usage(stdout, EXIT_SUCCESS); break; default: usage(stderr, EXIT_FAILURE); break; } if (optind == argc) usage(stderr, EXIT_FAILURE); for (i = optind; i < argc; i++) { htsFormat fmt; hFILE *fp = hopen(argv[i], "r"); if (fp == NULL) { fprintf(stderr, "htsfile: can't open \"%s\": %s\n", argv[i], strerror(errno)); status = EXIT_FAILURE; continue; } if (hts_detect_format(fp, &fmt) < 0) { fprintf(stderr, "htsfile: detecting \"%s\" format failed: %s\n", argv[i], strerror(errno)); hclose_abruptly(fp); status = EXIT_FAILURE; continue; } if (mode == identify) { char *description = hts_format_description(&fmt); printf("%s:\t%s\n", argv[i], description); free(description); } else switch (fmt.category) { case sequence_data: if (view_sam(fp, argv[i])) fp = NULL; break; case variant_data: if (view_vcf(fp, argv[i])) fp = NULL; break; default: fprintf(stderr, "htsfile: can't view %s: unknown format\n", argv[i]); status = EXIT_FAILURE; break; } if (fp && hclose(fp) < 0) { fprintf(stderr, "htsfile: closing %s failed\n", argv[i]); status = EXIT_FAILURE; } } return status; } stacks-1.44/htslib/._htslib000755 000765 000024 00000000265 13001012452 016406 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/000755 000765 000024 00000000000 13001012452 016240 5ustar00catchenstaff000000 000000 stacks-1.44/htslib/._htslib.pc.in000644 000765 000024 00000000265 12750743246 017437 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib.pc.in000644 000765 000024 00000000411 12750743246 017213 0ustar00catchenstaff000000 000000 includedir=@includedir@ libdir=@libdir@ Name: htslib Description: C library for high-throughput sequencing data formats Version: @PACKAGE_VERSION@ Cflags: -I${includedir} Libs: -L${libdir} -lhts Libs.private: -L${libdir} -lhts -lm -lpthread Requires.private: zlib stacks-1.44/htslib/._htslib_vars.mk000644 000765 000024 00000000265 12750766322 020072 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib_vars.mk000644 000765 000024 00000005147 12750766322 017661 0ustar00catchenstaff000000 000000 # Makefile variables useful for third-party code using htslib's public API. # # Copyright (C) 2013-2016 Genome Research Ltd. # # Author: John Marshall # # Permission is hereby granted, free of charge, to any person obtaining a copy # of this software and associated documentation files (the "Software"), to deal # in the Software without restriction, including without limitation the rights # to use, copy, modify, merge, publish, distribute, sublicense, and/or sell # copies of the Software, and to permit persons to whom the Software is # furnished to do so, subject to the following conditions: # # The above copyright notice and this permission notice shall be included in # all copies or substantial portions of the Software. # # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL # THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER # LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING # FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER # DEALINGS IN THE SOFTWARE. # These variables can be used to express dependencies on htslib headers. # See htslib.mk for details. htslib_bgzf_h = $(HTSPREFIX)htslib/bgzf.h $(htslib_hts_defs_h) htslib_cram_h = $(HTSPREFIX)htslib/cram.h $(htslib_hts_h) htslib_faidx_h = $(HTSPREFIX)htslib/faidx.h $(htslib_hts_defs_h) htslib_hfile_h = $(HTSPREFIX)htslib/hfile.h $(htslib_hts_defs_h) htslib_hts_h = $(HTSPREFIX)htslib/hts.h $(htslib_hts_defs_h) htslib_hts_defs_h = $(HTSPREFIX)htslib/hts_defs.h htslib_kbitset_h = $(HTSPREFIX)htslib/kbitset.h htslib_kfunc_h = $(HTSPREFIX)htslib/kfunc.h htslib_khash_h = $(HTSPREFIX)htslib/khash.h htslib_khash_str2int_h = $(HTSPREFIX)htslib/khash_str2int.h $(htslib_khash_h) htslib_klist_h = $(HTSPREFIX)htslib/klist.h htslib_knetfile_h = $(HTSPREFIX)htslib/knetfile.h htslib_kseq_h = $(HTSPREFIX)htslib/kseq.h htslib_ksort_h = $(HTSPREFIX)htslib/ksort.h htslib_kstring_h = $(HTSPREFIX)htslib/kstring.h htslib_regidx_h = $(HTSPREFIX)htslib/regidx.h htslib_sam_h = $(HTSPREFIX)htslib/sam.h $(htslib_hts_h) htslib_synced_bcf_reader_h = $(HTSPREFIX)htslib/synced_bcf_reader.h $(htslib_hts_h) $(htslib_vcf_h) $(htslib_tbx_h) htslib_tbx_h = $(HTSPREFIX)htslib/tbx.h $(htslib_hts_h) htslib_vcf_h = $(HTSPREFIX)htslib/vcf.h $(htslib_hts_h) $(htslib_kstring_h) $(htslib_hts_defs_h) htslib_vcf_sweep_h = $(HTSPREFIX)htslib/vcf_sweep.h $(htslib_hts_h) $(htslib_vcf_h) htslib_vcfutils_h = $(HTSPREFIX)htslib/vcfutils.h $(htslib_vcf_h) stacks-1.44/htslib/._INSTALL000644 000765 000024 00000000265 12750743246 016252 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/INSTALL000644 000765 000024 00000010676 12750743246 016044 0ustar00catchenstaff000000 000000 Basic Installation ================== To build and install HTSlib, 'cd' to the htslib-1.x directory containing the package's source and type the following commands: ./configure make make install The './configure' command checks your build environment and allows various optional functionality to be enabled (see Configuration below). If you don't want to select any optional functionality, you may wish to omit configure and just type 'make; make install' as for previous versions of HTSlib. However if the build fails you should run './configure' as it can diagnose the common reasons for build failures. The 'make' command builds the HTSlib library and and various useful utilities: bgzip, htsfile, and tabix. If compilation fails you should run './configure' as it can diagnose problems with your build environment that cause build failures. The 'make install' command installs the libraries, library header files, utilities, several manual pages, and a pkgconfig file to /usr/local. The installation location can be changed by configuring with --prefix=DIR or via 'make prefix=DIR install' (see Installation Locations below). Configuration ============= By default, './configure' examines your build environment, checking for requirements such as the zlib development files, and arranges for a plain HTSlib build. The following configure options can be used to enable various features and specify further optional external requirements: --enable-plugins Use plugins to implement exotic file access protocols and other specialised facilities. This enables such facilities to be developed and packaged outwith HTSlib, and somewhat isolates HTSlib-using programs from their library dependencies. By default, any enabled pluggable facilities (such as iRODS and libcurl file access) are built directly within HTSlib. --with-plugin-dir=DIR Specifies the directory into which plugins built while building HTSlib should be installed; by default, LIBEXECDIR/htslib. --with-plugin-path=DIR:DIR:DIR... Specifies the list of directories that HTSlib will search for plugins. By default, only the directory specified via --with-plugin-dir will be searched; you can use --with-plugin-path='DIR:$(plugindir):DIR' and so on to cause additional directories to be searched. --with-irods[=DIR] Specifies the location of the iRODS client library to use to enable access to data objects stored in iRODS () via file paths like 'irods:DATAOBJ'. DIR is the base of an iRODS source tree such that the library is present as DIR/lib/core/obj/libRodsAPI.* and headers are present under DIR/lib/api/include and so on. If '=DIR' is omitted, $IRODS_HOME will be used as a base directory. --enable-libcurl Use libcurl () to implement network access to remote files via FTP, HTTP, HTTPS, S3, etc. By default, HTSlib uses its own simple networking code to provide access via FTP and HTTP only. The configure script also accepts the usual options and environment variables for tuning installation locations and compilers: type './configure --help' for details. For example, ./configure CC=icc --prefix=/opt/icc-compiled would specify that HTSlib is to be built with icc and installed into bin, lib, etc subdirectories under /opt/icc-compiled. Installation Locations ====================== By default, 'make install' installs HTSlib libraries under /usr/local/lib, HTSlib header files under /usr/local/include, utility programs under /usr/local/bin, etc. (To be precise, the header files are installed within a fixed 'htslib' subdirectory under the specified .../include location.) You can specify a different location to install HTSlib by configuring with --prefix=DIR or specify locations for particular parts of HTSlib by configuring with --libdir=DIR and so on. Type './configure --help' for the full list of such install directory options. Alternatively you can specify different locations at install time by typing 'make prefix=DIR install' or 'make libdir=DIR install' and so on. Consult the list of prefix/exec_prefix/etc variables near the top of the Makefile for the full list of such variables that can be overridden. You can also specify a staging area by typing 'make DESTDIR=DIR install', possibly in conjunction with other --prefix or prefix=DIR settings. For example, make DESTDIR=/tmp/staging prefix=/opt would install into bin, lib, etc subdirectories under /tmp/staging/opt. stacks-1.44/htslib/._kfunc.c000644 000765 000024 00000000265 12750743246 016473 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/kfunc.c000644 000765 000024 00000021716 12750743246 016262 0ustar00catchenstaff000000 000000 /* The MIT License Copyright (C) 2010, 2013 Genome Research Ltd. Copyright (C) 2011 Attractive Chaos Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include "htslib/kfunc.h" /* Log gamma function * \log{\Gamma(z)} * AS245, 2nd algorithm, http://lib.stat.cmu.edu/apstat/245 */ double kf_lgamma(double z) { double x = 0; x += 0.1659470187408462e-06 / (z+7); x += 0.9934937113930748e-05 / (z+6); x -= 0.1385710331296526 / (z+5); x += 12.50734324009056 / (z+4); x -= 176.6150291498386 / (z+3); x += 771.3234287757674 / (z+2); x -= 1259.139216722289 / (z+1); x += 676.5203681218835 / z; x += 0.9999999999995183; return log(x) - 5.58106146679532777 - z + (z-0.5) * log(z+6.5); } /* complementary error function * \frac{2}{\sqrt{\pi}} \int_x^{\infty} e^{-t^2} dt * AS66, 2nd algorithm, http://lib.stat.cmu.edu/apstat/66 */ double kf_erfc(double x) { const double p0 = 220.2068679123761; const double p1 = 221.2135961699311; const double p2 = 112.0792914978709; const double p3 = 33.912866078383; const double p4 = 6.37396220353165; const double p5 = .7003830644436881; const double p6 = .03526249659989109; const double q0 = 440.4137358247522; const double q1 = 793.8265125199484; const double q2 = 637.3336333788311; const double q3 = 296.5642487796737; const double q4 = 86.78073220294608; const double q5 = 16.06417757920695; const double q6 = 1.755667163182642; const double q7 = .08838834764831844; double expntl, z, p; z = fabs(x) * M_SQRT2; if (z > 37.) return x > 0.? 0. : 2.; expntl = exp(z * z * - .5); if (z < 10. / M_SQRT2) // for small z p = expntl * ((((((p6 * z + p5) * z + p4) * z + p3) * z + p2) * z + p1) * z + p0) / (((((((q7 * z + q6) * z + q5) * z + q4) * z + q3) * z + q2) * z + q1) * z + q0); else p = expntl / 2.506628274631001 / (z + 1. / (z + 2. / (z + 3. / (z + 4. / (z + .65))))); return x > 0.? 2. * p : 2. * (1. - p); } /* The following computes regularized incomplete gamma functions. * Formulas are taken from Wiki, with additional input from Numerical * Recipes in C (for modified Lentz's algorithm) and AS245 * (http://lib.stat.cmu.edu/apstat/245). * * A good online calculator is available at: * * http://www.danielsoper.com/statcalc/calc23.aspx * * It calculates upper incomplete gamma function, which equals * kf_gammaq(s,z)*tgamma(s). */ #define KF_GAMMA_EPS 1e-14 #define KF_TINY 1e-290 // regularized lower incomplete gamma function, by series expansion static double _kf_gammap(double s, double z) { double sum, x; int k; for (k = 1, sum = x = 1.; k < 100; ++k) { sum += (x *= z / (s + k)); if (x / sum < KF_GAMMA_EPS) break; } return exp(s * log(z) - z - kf_lgamma(s + 1.) + log(sum)); } // regularized upper incomplete gamma function, by continued fraction static double _kf_gammaq(double s, double z) { int j; double C, D, f; f = 1. + z - s; C = f; D = 0.; // Modified Lentz's algorithm for computing continued fraction // See Numerical Recipes in C, 2nd edition, section 5.2 for (j = 1; j < 100; ++j) { double a = j * (s - j), b = (j<<1) + 1 + z - s, d; D = b + a * D; if (D < KF_TINY) D = KF_TINY; C = b + a / C; if (C < KF_TINY) C = KF_TINY; D = 1. / D; d = C * D; f *= d; if (fabs(d - 1.) < KF_GAMMA_EPS) break; } return exp(s * log(z) - z - kf_lgamma(s) - log(f)); } double kf_gammap(double s, double z) { return z <= 1. || z < s? _kf_gammap(s, z) : 1. - _kf_gammaq(s, z); } double kf_gammaq(double s, double z) { return z <= 1. || z < s? 1. - _kf_gammap(s, z) : _kf_gammaq(s, z); } /* Regularized incomplete beta function. The method is taken from * Numerical Recipe in C, 2nd edition, section 6.4. The following web * page calculates the incomplete beta function, which equals * kf_betai(a,b,x) * gamma(a) * gamma(b) / gamma(a+b): * * http://www.danielsoper.com/statcalc/calc36.aspx */ static double kf_betai_aux(double a, double b, double x) { double C, D, f; int j; if (x == 0.) return 0.; if (x == 1.) return 1.; f = 1.; C = f; D = 0.; // Modified Lentz's algorithm for computing continued fraction for (j = 1; j < 200; ++j) { double aa, d; int m = j>>1; aa = (j&1)? -(a + m) * (a + b + m) * x / ((a + 2*m) * (a + 2*m + 1)) : m * (b - m) * x / ((a + 2*m - 1) * (a + 2*m)); D = 1. + aa * D; if (D < KF_TINY) D = KF_TINY; C = 1. + aa / C; if (C < KF_TINY) C = KF_TINY; D = 1. / D; d = C * D; f *= d; if (fabs(d - 1.) < KF_GAMMA_EPS) break; } return exp(kf_lgamma(a+b) - kf_lgamma(a) - kf_lgamma(b) + a * log(x) + b * log(1.-x)) / a / f; } double kf_betai(double a, double b, double x) { return x < (a + 1.) / (a + b + 2.)? kf_betai_aux(a, b, x) : 1. - kf_betai_aux(b, a, 1. - x); } #ifdef KF_MAIN #include int main(int argc, char *argv[]) { double x = 5.5, y = 3; double a, b; printf("erfc(%lg): %lg, %lg\n", x, erfc(x), kf_erfc(x)); printf("upper-gamma(%lg,%lg): %lg\n", x, y, kf_gammaq(y, x)*tgamma(y)); a = 2; b = 2; x = 0.5; printf("incomplete-beta(%lg,%lg,%lg): %lg\n", a, b, x, kf_betai(a, b, x) / exp(kf_lgamma(a+b) - kf_lgamma(a) - kf_lgamma(b))); return 0; } #endif // log\binom{n}{k} static double lbinom(int n, int k) { if (k == 0 || n == k) return 0; return lgamma(n+1) - lgamma(k+1) - lgamma(n-k+1); } // n11 n12 | n1_ // n21 n22 | n2_ //-----------+---- // n_1 n_2 | n // hypergeometric distribution static double hypergeo(int n11, int n1_, int n_1, int n) { return exp(lbinom(n1_, n11) + lbinom(n-n1_, n_1-n11) - lbinom(n, n_1)); } typedef struct { int n11, n1_, n_1, n; double p; } hgacc_t; // incremental version of hypergenometric distribution static double hypergeo_acc(int n11, int n1_, int n_1, int n, hgacc_t *aux) { if (n1_ || n_1 || n) { aux->n11 = n11; aux->n1_ = n1_; aux->n_1 = n_1; aux->n = n; } else { // then only n11 changed; the rest fixed if (n11%11 && n11 + aux->n - aux->n1_ - aux->n_1) { if (n11 == aux->n11 + 1) { // incremental aux->p *= (double)(aux->n1_ - aux->n11) / n11 * (aux->n_1 - aux->n11) / (n11 + aux->n - aux->n1_ - aux->n_1); aux->n11 = n11; return aux->p; } if (n11 == aux->n11 - 1) { // incremental aux->p *= (double)aux->n11 / (aux->n1_ - n11) * (aux->n11 + aux->n - aux->n1_ - aux->n_1) / (aux->n_1 - n11); aux->n11 = n11; return aux->p; } } aux->n11 = n11; } aux->p = hypergeo(aux->n11, aux->n1_, aux->n_1, aux->n); return aux->p; } double kt_fisher_exact(int n11, int n12, int n21, int n22, double *_left, double *_right, double *two) { int i, j, max, min; double p, q, left, right; hgacc_t aux; int n1_, n_1, n; n1_ = n11 + n12; n_1 = n11 + n21; n = n11 + n12 + n21 + n22; // calculate n1_, n_1 and n max = (n_1 < n1_) ? n_1 : n1_; // max n11, for right tail min = n1_ + n_1 - n; // not sure why n11-n22 is used instead of min(n_1,n1_) if (min < 0) min = 0; // min n11, for left tail *two = *_left = *_right = 1.; if (min == max) return 1.; // no need to do test q = hypergeo_acc(n11, n1_, n_1, n, &aux); // the probability of the current table // left tail p = hypergeo_acc(min, 0, 0, 0, &aux); for (left = 0., i = min + 1; p < 0.99999999 * q && i<=max; ++i) // loop until underflow left += p, p = hypergeo_acc(i, 0, 0, 0, &aux); --i; if (p < 1.00000001 * q) left += p; else --i; // right tail p = hypergeo_acc(max, 0, 0, 0, &aux); for (right = 0., j = max - 1; p < 0.99999999 * q && j>=0; --j) // loop until underflow right += p, p = hypergeo_acc(j, 0, 0, 0, &aux); ++j; if (p < 1.00000001 * q) right += p; else ++j; // two-tail *two = left + right; if (*two > 1.) *two = 1.; // adjust left and right if (abs(i - n11) < abs(j - n11)) right = 1. - left + q; else left = 1.0 - right + q; *_left = left; *_right = right; return q; } stacks-1.44/htslib/._knetfile.c000644 000765 000024 00000000265 12750743246 017166 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/knetfile.c000644 000765 000024 00000044633 12750743246 016760 0ustar00catchenstaff000000 000000 /* The MIT License Copyright (c) 2008 by Genome Research Ltd (GRL). 2010 by Attractive Chaos Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ /* Probably I will not do socket programming in the next few years and therefore I decide to heavily annotate this file, for Linux and Windows as well. -ac */ #include #include #include #include #include #include #include #include #include #ifndef _WIN32 #include #include #include #endif #include "htslib/knetfile.h" /* In winsock.h, the type of a socket is SOCKET, which is: "typedef * u_int SOCKET". An invalid SOCKET is: "(SOCKET)(~0)", or signed * integer -1. In knetfile.c, I use "int" for socket type * throughout. This should be improved to avoid confusion. * * In Linux/Mac, recv() and read() do almost the same thing. You can see * in the header file that netread() is simply an alias of read(). In * Windows, however, they are different and using recv() is mandatory. */ /* This function tests if the file handler is ready for reading (or * writing if is_read==0). */ static int socket_wait(int fd, int is_read) { fd_set fds, *fdr = 0, *fdw = 0; struct timeval tv; int ret; tv.tv_sec = 5; tv.tv_usec = 0; // 5 seconds time out FD_ZERO(&fds); FD_SET(fd, &fds); if (is_read) fdr = &fds; else fdw = &fds; ret = select(fd+1, fdr, fdw, 0, &tv); #ifndef _WIN32 if (ret == -1) perror("select"); #else if (ret == 0) fprintf(stderr, "select time-out\n"); else if (ret == SOCKET_ERROR) fprintf(stderr, "select: %d\n", WSAGetLastError()); #endif return ret; } #ifndef _WIN32 /* This function does not work with Windows due to the lack of * getaddrinfo() in winsock. It is addapted from an example in "Beej's * Guide to Network Programming" (http://beej.us/guide/bgnet/). */ static int socket_connect(const char *host, const char *port) { #define __err_connect(func) do { perror(func); freeaddrinfo(res); return -1; } while (0) int ai_err, on = 1, fd; struct linger lng = { 0, 0 }; struct addrinfo hints, *res = 0; memset(&hints, 0, sizeof(struct addrinfo)); hints.ai_family = AF_UNSPEC; hints.ai_socktype = SOCK_STREAM; /* In Unix/Mac, getaddrinfo() is the most convenient way to get * server information. */ if ((ai_err = getaddrinfo(host, port, &hints, &res)) != 0) { fprintf(stderr, "can't resolve %s:%s: %s\n", host, port, gai_strerror(ai_err)); return -1; } if ((fd = socket(res->ai_family, res->ai_socktype, res->ai_protocol)) == -1) __err_connect("socket"); /* The following two setsockopt() are used by ftplib * (http://nbpfaus.net/~pfau/ftplib/). I am not sure if they * necessary. */ if (setsockopt(fd, SOL_SOCKET, SO_REUSEADDR, &on, sizeof(on)) == -1) __err_connect("setsockopt"); if (setsockopt(fd, SOL_SOCKET, SO_LINGER, &lng, sizeof(lng)) == -1) __err_connect("setsockopt"); if (connect(fd, res->ai_addr, res->ai_addrlen) != 0) __err_connect("connect"); freeaddrinfo(res); return fd; } #else /* MinGW's printf has problem with "%lld" */ char *int64tostr(char *buf, int64_t x) { int cnt; int i = 0; do { buf[i++] = '0' + x % 10; x /= 10; } while (x); buf[i] = 0; for (cnt = i, i = 0; i < cnt/2; ++i) { int c = buf[i]; buf[i] = buf[cnt-i-1]; buf[cnt-i-1] = c; } return buf; } int64_t strtoint64(const char *buf) { int64_t x; for (x = 0; *buf != '\0'; ++buf) x = x * 10 + ((int64_t) *buf - 48); return x; } /* In windows, the first thing is to establish the TCP connection. */ int knet_win32_init() { WSADATA wsaData; return WSAStartup(MAKEWORD(2, 2), &wsaData); } void knet_win32_destroy() { WSACleanup(); } /* A slightly modfied version of the following function also works on * Mac (and presummably Linux). However, this function is not stable on * my Mac. It sometimes works fine but sometimes does not. Therefore for * non-Windows OS, I do not use this one. */ static SOCKET socket_connect(const char *host, const char *port) { #define __err_connect(func) \ do { \ fprintf(stderr, "%s: %d\n", func, WSAGetLastError()); \ return -1; \ } while (0) int on = 1; SOCKET fd; struct linger lng = { 0, 0 }; struct sockaddr_in server; struct hostent *hp = 0; // open socket if ((fd = socket(AF_INET, SOCK_STREAM, IPPROTO_TCP)) == INVALID_SOCKET) __err_connect("socket"); if (setsockopt(fd, SOL_SOCKET, SO_REUSEADDR, (char*)&on, sizeof(on)) == -1) __err_connect("setsockopt"); if (setsockopt(fd, SOL_SOCKET, SO_LINGER, (char*)&lng, sizeof(lng)) == -1) __err_connect("setsockopt"); // get host info if (isalpha(host[0])) hp = gethostbyname(host); else { struct in_addr addr; addr.s_addr = inet_addr(host); hp = gethostbyaddr((char*)&addr, 4, AF_INET); } if (hp == 0) __err_connect("gethost"); // connect server.sin_addr.s_addr = *((unsigned long*)hp->h_addr); server.sin_family= AF_INET; server.sin_port = htons(atoi(port)); if (connect(fd, (struct sockaddr*)&server, sizeof(server)) != 0) __err_connect("connect"); // freehostent(hp); // strangely in MSDN, hp is NOT freed (memory leak?!) return fd; } #endif static off_t my_netread(int fd, void *buf, off_t len) { off_t rest = len, curr, l = 0; /* recv() and read() may not read the required length of data with * one call. They have to be called repeatedly. */ while (rest) { if (socket_wait(fd, 1) <= 0) break; // socket is not ready for reading curr = netread(fd, (void*)((char*)buf + l), rest); /* According to the glibc manual, section 13.2, a zero returned * value indicates end-of-file (EOF), which should mean that * read() will not return zero if EOF has not been met but data * are not immediately available. */ if (curr == 0) break; l += curr; rest -= curr; } return l; } /************************* * FTP specific routines * *************************/ static int kftp_get_response(knetFile *ftp) { #ifndef _WIN32 unsigned char c; #else char c; #endif int n = 0; char *p; if (socket_wait(ftp->ctrl_fd, 1) <= 0) return 0; while (netread(ftp->ctrl_fd, &c, 1)) { // FIXME: this is *VERY BAD* for unbuffered I/O //fputc(c, stderr); if (n >= ftp->max_response) { ftp->max_response = ftp->max_response? ftp->max_response<<1 : 256; ftp->response = (char*)realloc(ftp->response, ftp->max_response); } ftp->response[n++] = c; if (c == '\n') { if (n >= 4 && isdigit(ftp->response[0]) && isdigit(ftp->response[1]) && isdigit(ftp->response[2]) && ftp->response[3] != '-') break; n = 0; continue; } } if (n < 2) return -1; ftp->response[n-2] = 0; return strtol(ftp->response, &p, 0); } static int kftp_send_cmd(knetFile *ftp, const char *cmd, int is_get) { if (socket_wait(ftp->ctrl_fd, 0) <= 0) return -1; // socket is not ready for writing int len = strlen(cmd); if ( netwrite(ftp->ctrl_fd, cmd, len) != len ) return -1; return is_get? kftp_get_response(ftp) : 0; } static int kftp_pasv_prep(knetFile *ftp) { char *p; int v[6]; kftp_send_cmd(ftp, "PASV\r\n", 1); for (p = ftp->response; *p && *p != '('; ++p); if (*p != '(') return -1; ++p; sscanf(p, "%d,%d,%d,%d,%d,%d", &v[0], &v[1], &v[2], &v[3], &v[4], &v[5]); memcpy(ftp->pasv_ip, v, 4 * sizeof(int)); ftp->pasv_port = (v[4]<<8&0xff00) + v[5]; return 0; } static int kftp_pasv_connect(knetFile *ftp) { char host[80], port[10]; if (ftp->pasv_port == 0) { fprintf(stderr, "[kftp_pasv_connect] kftp_pasv_prep() is not called before hand.\n"); return -1; } sprintf(host, "%d.%d.%d.%d", ftp->pasv_ip[0], ftp->pasv_ip[1], ftp->pasv_ip[2], ftp->pasv_ip[3]); sprintf(port, "%d", ftp->pasv_port); ftp->fd = socket_connect(host, port); if (ftp->fd == -1) return -1; return 0; } int kftp_connect(knetFile *ftp) { ftp->ctrl_fd = socket_connect(ftp->host, ftp->port); if (ftp->ctrl_fd == -1) return -1; kftp_get_response(ftp); kftp_send_cmd(ftp, "USER anonymous\r\n", 1); kftp_send_cmd(ftp, "PASS kftp@\r\n", 1); kftp_send_cmd(ftp, "TYPE I\r\n", 1); return 0; } int kftp_reconnect(knetFile *ftp) { if (ftp->ctrl_fd != -1) { netclose(ftp->ctrl_fd); ftp->ctrl_fd = -1; } netclose(ftp->fd); ftp->fd = -1; return kftp_connect(ftp); } // initialize ->type, ->host, ->retr and ->size knetFile *kftp_parse_url(const char *fn, const char *mode) { knetFile *fp; char *p; int l; if (strstr(fn, "ftp://") != fn) return 0; for (p = (char*)fn + 6; *p && *p != '/'; ++p); if (*p != '/') return 0; l = p - fn - 6; fp = (knetFile*)calloc(1, sizeof(knetFile)); fp->type = KNF_TYPE_FTP; fp->fd = -1; /* the Linux/Mac version of socket_connect() also recognizes a port * like "ftp", but the Windows version does not. */ fp->port = strdup("21"); fp->host = (char*)calloc(l + 1, 1); if (strchr(mode, 'c')) fp->no_reconnect = 1; strncpy(fp->host, fn + 6, l); fp->retr = (char*)calloc(strlen(p) + 8, 1); sprintf(fp->retr, "RETR %s\r\n", p); fp->size_cmd = (char*)calloc(strlen(p) + 8, 1); sprintf(fp->size_cmd, "SIZE %s\r\n", p); fp->seek_offset = 0; return fp; } // place ->fd at offset off int kftp_connect_file(knetFile *fp) { int ret; long long file_size; if (fp->fd != -1) { netclose(fp->fd); if (fp->no_reconnect) kftp_get_response(fp); } kftp_pasv_prep(fp); kftp_send_cmd(fp, fp->size_cmd, 1); #ifndef _WIN32 // If the file does not exist, the response will be "550 Could not get file // size". Be silent on failure, hts_idx_load can be trying the existence of .csi or .tbi. if ( sscanf(fp->response,"%*d %lld", &file_size) != 1 ) return -1; #else const char *p = fp->response; while (*p != ' ') ++p; while (*p < '0' || *p > '9') ++p; file_size = strtoint64(p); #endif fp->file_size = file_size; if (fp->offset>=0) { char tmp[32]; #ifndef _WIN32 sprintf(tmp, "REST %lld\r\n", (long long)fp->offset); #else strcpy(tmp, "REST "); int64tostr(tmp + 5, fp->offset); strcat(tmp, "\r\n"); #endif kftp_send_cmd(fp, tmp, 1); } kftp_send_cmd(fp, fp->retr, 0); kftp_pasv_connect(fp); ret = kftp_get_response(fp); if (ret != 150) { fprintf(stderr, "[kftp_connect_file] %s\n", fp->response); netclose(fp->fd); fp->fd = -1; return -1; } fp->is_ready = 1; return 0; } /************************** * HTTP specific routines * **************************/ knetFile *khttp_parse_url(const char *fn, const char *mode) { knetFile *fp; char *p, *proxy, *q; int l; if (strstr(fn, "http://") != fn) return 0; // set ->http_host for (p = (char*)fn + 7; *p && *p != '/'; ++p); l = p - fn - 7; fp = (knetFile*)calloc(1, sizeof(knetFile)); fp->http_host = (char*)calloc(l + 1, 1); strncpy(fp->http_host, fn + 7, l); fp->http_host[l] = 0; for (q = fp->http_host; *q && *q != ':'; ++q); if (*q == ':') *q++ = 0; // get http_proxy proxy = getenv("http_proxy"); // set ->host, ->port and ->path if (proxy == 0) { fp->host = strdup(fp->http_host); // when there is no proxy, server name is identical to http_host name. fp->port = strdup(*q? q : "80"); fp->path = strdup(*p? p : "/"); } else { fp->host = (strstr(proxy, "http://") == proxy)? strdup(proxy + 7) : strdup(proxy); for (q = fp->host; *q && *q != ':'; ++q); if (*q == ':') *q++ = 0; fp->port = strdup(*q? q : "80"); fp->path = strdup(fn); } fp->type = KNF_TYPE_HTTP; fp->ctrl_fd = fp->fd = -1; fp->seek_offset = 0; return fp; } int khttp_connect_file(knetFile *fp) { int ret, l = 0; char *buf, *p; if (fp->fd != -1) netclose(fp->fd); fp->fd = socket_connect(fp->host, fp->port); buf = (char*)calloc(0x10000, 1); // FIXME: I am lazy... But in principle, 64KB should be large enough. l += sprintf(buf + l, "GET %s HTTP/1.0\r\nHost: %s\r\n", fp->path, fp->http_host); l += sprintf(buf + l, "Range: bytes=%lld-\r\n", (long long)fp->offset); l += sprintf(buf + l, "\r\n"); if ( netwrite(fp->fd, buf, l) != l ) { free(buf); return -1; } l = 0; while (netread(fp->fd, buf + l, 1)) { // read HTTP header; FIXME: bad efficiency if (buf[l] == '\n' && l >= 3) if (strncmp(buf + l - 3, "\r\n\r\n", 4) == 0) break; ++l; } buf[l] = 0; if (l < 14) { // prematured header free(buf); netclose(fp->fd); fp->fd = -1; return -1; } ret = strtol(buf + 8, &p, 0); // HTTP return code if (ret == 200 && fp->offset>0) { // 200 (complete result); then skip beginning of the file off_t rest = fp->offset; while (rest) { off_t l = rest < 0x10000? rest : 0x10000; rest -= my_netread(fp->fd, buf, l); } } else if (ret != 206 && ret != 200) { // failed to open file free(buf); netclose(fp->fd); switch (ret) { case 401: errno = EPERM; break; case 403: errno = EACCES; break; case 404: errno = ENOENT; break; case 407: errno = EPERM; break; case 408: errno = ETIMEDOUT; break; case 410: errno = ENOENT; break; case 503: errno = EAGAIN; break; case 504: errno = ETIMEDOUT; break; default: errno = (ret >= 400 && ret < 500)? EINVAL : EIO; break; } fp->fd = -1; return -1; } free(buf); fp->is_ready = 1; return 0; } /******************** * Generic routines * ********************/ knetFile *knet_open(const char *fn, const char *mode) { knetFile *fp = 0; if (mode[0] != 'r') { fprintf(stderr, "[kftp_open] only mode \"r\" is supported.\n"); return 0; } if (strstr(fn, "ftp://") == fn) { fp = kftp_parse_url(fn, mode); if (fp == 0) return 0; if (kftp_connect(fp) == -1) { knet_close(fp); return 0; } kftp_connect_file(fp); } else if (strstr(fn, "http://") == fn) { fp = khttp_parse_url(fn, mode); if (fp == 0) return 0; khttp_connect_file(fp); } else { // local file #ifdef _WIN32 /* In windows, O_BINARY is necessary. In Linux/Mac, O_BINARY may * be undefined on some systems, although it is defined on my * Mac and the Linux I have tested on. */ int fd = open(fn, O_RDONLY | O_BINARY); #else int fd = open(fn, O_RDONLY); #endif if (fd == -1) { perror("open"); return 0; } fp = (knetFile*)calloc(1, sizeof(knetFile)); fp->type = KNF_TYPE_LOCAL; fp->fd = fd; fp->ctrl_fd = -1; } if (fp && fp->fd == -1) { knet_close(fp); return 0; } return fp; } knetFile *knet_dopen(int fd, const char *mode) { knetFile *fp = (knetFile*)calloc(1, sizeof(knetFile)); fp->type = KNF_TYPE_LOCAL; fp->fd = fd; return fp; } ssize_t knet_read(knetFile *fp, void *buf, size_t len) { off_t l = 0; if (fp->fd == -1) return 0; if (fp->type == KNF_TYPE_FTP) { if (fp->is_ready == 0) { if (!fp->no_reconnect) kftp_reconnect(fp); kftp_connect_file(fp); } } else if (fp->type == KNF_TYPE_HTTP) { if (fp->is_ready == 0) khttp_connect_file(fp); } if (fp->type == KNF_TYPE_LOCAL) { // on Windows, the following block is necessary; not on UNIX size_t rest = len; ssize_t curr; while (rest) { do { curr = read(fp->fd, (void*)((char*)buf + l), rest); } while (curr < 0 && EINTR == errno); if (curr < 0) return -1; if (curr == 0) break; l += curr; rest -= curr; } } else l = my_netread(fp->fd, buf, len); fp->offset += l; return l; } off_t knet_seek(knetFile *fp, off_t off, int whence) { if (whence == SEEK_SET && off == fp->offset) return 0; if (fp->type == KNF_TYPE_LOCAL) { /* Be aware that lseek() returns the offset after seeking, while fseek() returns zero on success. */ off_t offset = lseek(fp->fd, off, whence); if (offset == -1) return -1; fp->offset = offset; return fp->offset; } else if (fp->type == KNF_TYPE_FTP) { if (whence == SEEK_CUR) fp->offset += off; else if (whence == SEEK_SET) fp->offset = off; else if (whence == SEEK_END) fp->offset = fp->file_size + off; else return -1; fp->is_ready = 0; return fp->offset; } else if (fp->type == KNF_TYPE_HTTP) { if (whence == SEEK_END) { // FIXME: can we allow SEEK_END in future? fprintf(stderr, "[knet_seek] SEEK_END is not supported for HTTP. Offset is unchanged.\n"); errno = ESPIPE; return -1; } if (whence == SEEK_CUR) fp->offset += off; else if (whence == SEEK_SET) fp->offset = off; else return -1; fp->is_ready = 0; return fp->offset; } errno = EINVAL; fprintf(stderr,"[knet_seek] %s\n", strerror(errno)); return -1; } int knet_close(knetFile *fp) { if (fp == 0) return 0; if (fp->ctrl_fd != -1) netclose(fp->ctrl_fd); // FTP specific if (fp->fd != -1) { /* On Linux/Mac, netclose() is an alias of close(), but on * Windows, it is an alias of closesocket(). */ if (fp->type == KNF_TYPE_LOCAL) close(fp->fd); else netclose(fp->fd); } free(fp->host); free(fp->port); free(fp->response); free(fp->retr); // FTP specific free(fp->path); free(fp->http_host); // HTTP specific free(fp); return 0; } #ifdef KNETFILE_MAIN int main(void) { char *buf; knetFile *fp; int type = 4, l; #ifdef _WIN32 knet_win32_init(); #endif buf = calloc(0x100000, 1); if (type == 0) { fp = knet_open("knetfile.c", "r"); knet_seek(fp, 1000, SEEK_SET); } else if (type == 1) { // NCBI FTP, large file fp = knet_open("ftp://ftp.ncbi.nih.gov/1000genomes/ftp/data/NA12878/alignment/NA12878.chrom6.SLX.SRP000032.2009_06.bam", "r"); knet_seek(fp, 2500000000ll, SEEK_SET); l = knet_read(fp, buf, 255); } else if (type == 2) { fp = knet_open("ftp://ftp.sanger.ac.uk/pub4/treefam/tmp/index.shtml", "r"); knet_seek(fp, 1000, SEEK_SET); } else if (type == 3) { fp = knet_open("http://www.sanger.ac.uk/Users/lh3/index.shtml", "r"); knet_seek(fp, 1000, SEEK_SET); } else if (type == 4) { fp = knet_open("http://www.sanger.ac.uk/Users/lh3/ex1.bam", "r"); knet_read(fp, buf, 10000); knet_seek(fp, 20000, SEEK_SET); knet_seek(fp, 10000, SEEK_SET); l = knet_read(fp, buf+10000, 10000000) + 10000; } if (type != 4 && type != 1) { knet_read(fp, buf, 255); buf[255] = 0; printf("%s\n", buf); } else write(fileno(stdout), buf, l); knet_close(fp); free(buf); return 0; } #endif stacks-1.44/htslib/._kstring.c000644 000765 000024 00000000265 12750743246 017046 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/kstring.c000644 000765 000024 00000016156 12750743246 016637 0ustar00catchenstaff000000 000000 /* The MIT License Copyright (C) 2011 by Attractive Chaos Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include #include #include #include "htslib/kstring.h" int kvsprintf(kstring_t *s, const char *fmt, va_list ap) { va_list args; int l; va_copy(args, ap); l = vsnprintf(s->s + s->l, s->m - s->l, fmt, args); // This line does not work with glibc 2.0. See `man snprintf'. va_end(args); if (l + 1 > s->m - s->l) { s->m = s->l + l + 2; kroundup32(s->m); s->s = (char*)realloc(s->s, s->m); va_copy(args, ap); l = vsnprintf(s->s + s->l, s->m - s->l, fmt, args); va_end(args); } s->l += l; return l; } int ksprintf(kstring_t *s, const char *fmt, ...) { va_list ap; int l; va_start(ap, fmt); l = kvsprintf(s, fmt, ap); va_end(ap); return l; } char *kstrtok(const char *str, const char *sep, ks_tokaux_t *aux) { const char *p, *start; if (sep) { // set up the table if (str == 0 && (aux->tab[0]&1)) return 0; // no need to set up if we have finished aux->finished = 0; if (sep[1]) { aux->sep = -1; aux->tab[0] = aux->tab[1] = aux->tab[2] = aux->tab[3] = 0; for (p = sep; *p; ++p) aux->tab[*p>>6] |= 1ull<<(*p&0x3f); } else aux->sep = sep[0]; } if (aux->finished) return 0; else if (str) aux->p = str - 1, aux->finished = 0; if (aux->sep < 0) { for (p = start = aux->p + 1; *p; ++p) if (aux->tab[*p>>6]>>(*p&0x3f)&1) break; } else { for (p = start = aux->p + 1; *p; ++p) if (*p == aux->sep) break; } aux->p = p; // end of token if (*p == 0) aux->finished = 1; // no more tokens return (char*)start; } // s MUST BE a null terminated string; l = strlen(s) int ksplit_core(char *s, int delimiter, int *_max, int **_offsets) { int i, n, max, last_char, last_start, *offsets, l; n = 0; max = *_max; offsets = *_offsets; l = strlen(s); #define __ksplit_aux do { \ if (_offsets) { \ s[i] = 0; \ if (n == max) { \ int *tmp; \ max = max? max<<1 : 2; \ if ((tmp = (int*)realloc(offsets, sizeof(int) * max))) { \ offsets = tmp; \ } else { \ free(offsets); \ *_offsets = NULL; \ return 0; \ } \ } \ offsets[n++] = last_start; \ } else ++n; \ } while (0) for (i = 0, last_char = last_start = 0; i <= l; ++i) { if (delimiter == 0) { if (isspace(s[i]) || s[i] == 0) { if (isgraph(last_char)) __ksplit_aux; // the end of a field } else { if (isspace(last_char) || last_char == 0) last_start = i; } } else { if (s[i] == delimiter || s[i] == 0) { if (last_char != 0 && last_char != delimiter) __ksplit_aux; // the end of a field } else { if (last_char == delimiter || last_char == 0) last_start = i; } } last_char = s[i]; } *_max = max; *_offsets = offsets; return n; } int kgetline(kstring_t *s, kgets_func *fgets_fn, void *fp) { size_t l0 = s->l; while (s->l == l0 || s->s[s->l-1] != '\n') { if (s->m - s->l < 200) ks_resize(s, s->m + 200); if (fgets_fn(s->s + s->l, s->m - s->l, fp) == NULL) break; s->l += strlen(s->s + s->l); } if (s->l == l0) return EOF; if (s->l > l0 && s->s[s->l-1] == '\n') { s->l--; if (s->l > l0 && s->s[s->l-1] == '\r') s->l--; } s->s[s->l] = '\0'; return 0; } /********************** * Boyer-Moore search * **********************/ typedef unsigned char ubyte_t; // reference: http://www-igm.univ-mlv.fr/~lecroq/string/node14.html static int *ksBM_prep(const ubyte_t *pat, int m) { int i, *suff, *prep, *bmGs, *bmBc; prep = (int*)calloc(m + 256, sizeof(int)); bmGs = prep; bmBc = prep + m; { // preBmBc() for (i = 0; i < 256; ++i) bmBc[i] = m; for (i = 0; i < m - 1; ++i) bmBc[pat[i]] = m - i - 1; } suff = (int*)calloc(m, sizeof(int)); { // suffixes() int f = 0, g; suff[m - 1] = m; g = m - 1; for (i = m - 2; i >= 0; --i) { if (i > g && suff[i + m - 1 - f] < i - g) suff[i] = suff[i + m - 1 - f]; else { if (i < g) g = i; f = i; while (g >= 0 && pat[g] == pat[g + m - 1 - f]) --g; suff[i] = f - g; } } } { // preBmGs() int j = 0; for (i = 0; i < m; ++i) bmGs[i] = m; for (i = m - 1; i >= 0; --i) if (suff[i] == i + 1) for (; j < m - 1 - i; ++j) if (bmGs[j] == m) bmGs[j] = m - 1 - i; for (i = 0; i <= m - 2; ++i) bmGs[m - 1 - suff[i]] = m - 1 - i; } free(suff); return prep; } void *kmemmem(const void *_str, int n, const void *_pat, int m, int **_prep) { int i, j, *prep = 0, *bmGs, *bmBc; const ubyte_t *str, *pat; str = (const ubyte_t*)_str; pat = (const ubyte_t*)_pat; prep = (_prep == 0 || *_prep == 0)? ksBM_prep(pat, m) : *_prep; if (_prep && *_prep == 0) *_prep = prep; bmGs = prep; bmBc = prep + m; j = 0; while (j <= n - m) { for (i = m - 1; i >= 0 && pat[i] == str[i+j]; --i); if (i >= 0) { int max = bmBc[str[i+j]] - m + 1 + i; if (max < bmGs[i]) max = bmGs[i]; j += max; } else return (void*)(str + j); } if (_prep == 0) free(prep); return 0; } char *kstrstr(const char *str, const char *pat, int **_prep) { return (char*)kmemmem(str, strlen(str), pat, strlen(pat), _prep); } char *kstrnstr(const char *str, const char *pat, int n, int **_prep) { return (char*)kmemmem(str, n, pat, strlen(pat), _prep); } /*********************** * The main() function * ***********************/ #ifdef KSTRING_MAIN #include int main() { kstring_t *s; int *fields, n, i; ks_tokaux_t aux; char *p; s = (kstring_t*)calloc(1, sizeof(kstring_t)); // test ksprintf() ksprintf(s, " abcdefg: %d ", 100); printf("'%s'\n", s->s); // test ksplit() fields = ksplit(s, 0, &n); for (i = 0; i < n; ++i) printf("field[%d] = '%s'\n", i, s->s + fields[i]); // test kstrtok() s->l = 0; for (p = kstrtok("ab:cde:fg/hij::k", ":/", &aux); p; p = kstrtok(0, 0, &aux)) { kputsn(p, aux.p - p, s); kputc('\n', s); } printf("%s", s->s); // free free(s->s); free(s); free(fields); { static char *str = "abcdefgcdgcagtcakcdcd"; static char *pat = "cd"; char *ret, *s = str; int *prep = 0; while ((ret = kstrstr(s, pat, &prep)) != 0) { printf("match: %s\n", ret); s = ret + prep[0]; } free(prep); } return 0; } #endif stacks-1.44/htslib/._LICENSE000644 000765 000024 00000000265 12750743246 016226 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/LICENSE000644 000765 000024 00000006724 12750743246 016017 0ustar00catchenstaff000000 000000 [Files in this distribution outwith the cram/ subdirectory are distributed according to the terms of the following MIT/Expat license.] The MIT/Expat License Copyright (C) 2012-2014 Genome Research Ltd. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. [Files within the cram/ subdirectory in this distribution are distributed according to the terms of the following Modified 3-Clause BSD license.] The Modified-BSD License Copyright (C) 2012-2014 Genome Research Ltd. Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. [The use of a range of years within a copyright notice in this distribution should be interpreted as being equivalent to a list of years including the first and last year specified and all consecutive years between them. For example, a copyright notice that reads "Copyright (C) 2005, 2007-2009, 2011-2012" should be interpreted as being identical to a notice that reads "Copyright (C) 2005, 2007, 2008, 2009, 2011, 2012" and a copyright notice that reads "Copyright (C) 2005-2012" should be interpreted as being identical to a notice that reads "Copyright (C) 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012".] stacks-1.44/htslib/._Makefile000644 000765 000024 00000000265 12752657173 016666 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/Makefile000644 000765 000024 00000043430 12752657173 016452 0ustar00catchenstaff000000 000000 # Makefile for htslib, a C library for high-throughput sequencing data formats. # # Copyright (C) 2013-2015 Genome Research Ltd. # # Author: John Marshall # # Permission is hereby granted, free of charge, to any person obtaining a copy # of this software and associated documentation files (the "Software"), to deal # in the Software without restriction, including without limitation the rights # to use, copy, modify, merge, publish, distribute, sublicense, and/or sell # copies of the Software, and to permit persons to whom the Software is # furnished to do so, subject to the following conditions: # # The above copyright notice and this permission notice shall be included in # all copies or substantial portions of the Software. # # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL # THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER # LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING # FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER # DEALINGS IN THE SOFTWARE. CC = gcc AR = ar RANLIB = ranlib CPPFLAGS = # TODO: probably update cram code to make it compile cleanly with -Wc++-compat CFLAGS = -g -Wall -O2 EXTRA_CFLAGS_PIC = -fpic LDFLAGS = LIBS = # For now these don't work too well as samtools also needs to know to # add -lbz2 and -llzma if linking against the static libhts.a library. # TODO This needs configury and adding to htslib.pc.in. # # # Bzip2 support; optionally used by CRAM. # HAVE_LIBBZ2 := $(shell echo -e "\#include \012int main(void){return 0;}" > .test.c && $(CC) $(CFLAGS) $(CPPFLAGS) -o .test .test.c -lbz2 2>/dev/null && echo yes) # ifeq "$(HAVE_LIBBZ2)" "yes" # CPPFLAGS += -DHAVE_LIBBZ2 # LIBS += -lbz2 # endif # # # Lzma support; optionally used by CRAM. # HAVE_LIBLZMA := $(shell echo -e "\#include \012int main(void){return 0;}" > .test.c && $(CC) $(CFLAGS) $(CPPFLAGS) -o .test .test.c -llzma 2>/dev/null && echo yes) # ifeq "$(HAVE_LIBLZMA)" "yes" # CPPFLAGS += -DHAVE_LIBLZMA # LIBS += -llzma # endif prefix = /usr/local exec_prefix = $(prefix) bindir = $(exec_prefix)/bin includedir = $(prefix)/include libdir = $(exec_prefix)/lib libexecdir = $(exec_prefix)/libexec datarootdir = $(prefix)/share mandir = $(datarootdir)/man man1dir = $(mandir)/man1 man5dir = $(mandir)/man5 pkgconfigdir= $(libdir)/pkgconfig MKDIR_P = mkdir -p INSTALL = install -p INSTALL_PROGRAM = $(INSTALL) INSTALL_DATA = $(INSTALL) -m 644 INSTALL_DIR = $(MKDIR_P) -m 755 # Set by config.mk if plugins are enabled plugindir = BUILT_PROGRAMS = \ bgzip \ htsfile \ tabix BUILT_TEST_PROGRAMS = \ test/fieldarith \ test/hfile \ test/sam \ test/test-regidx \ test/test_view \ test/test-vcf-api \ test/test-vcf-sweep # all: lib-static lib-shared $(BUILT_PROGRAMS) plugins $(BUILT_TEST_PROGRAMS) all: lib-static HTSPREFIX = include htslib_vars.mk PACKAGE_VERSION = 1.3.1 LIBHTS_SOVERSION = 1 # $(NUMERIC_VERSION) is for items that must have a numeric X.Y.Z string # even if this is a dirty or untagged Git working tree. NUMERIC_VERSION = $(PACKAGE_VERSION) # If building from a Git repository, replace $(PACKAGE_VERSION) with the Git # description of the working tree: either a release tag with the same value # as $(PACKAGE_VERSION) above, or an exact description likely based on a tag. # Much of this is also GNU Make-specific. If you don't have GNU Make and/or # are not building from a Git repository, comment out this conditional. ifneq "$(wildcard .git)" "" original_version := $(PACKAGE_VERSION) PACKAGE_VERSION := $(shell git describe --always --dirty) # Unless the Git description matches /\d*\.\d*(\.\d*)?/, i.e., is exactly a tag # with a numeric name, revert $(NUMERIC_VERSION) to the original version number # written above, but with the patchlevel field bumped to 255. ifneq "$(subst ..,.,$(subst 0,,$(subst 1,,$(subst 2,,$(subst 3,,$(subst 4,,$(subst 5,,$(subst 6,,$(subst 7,,$(subst 8,,$(subst 9,,$(PACKAGE_VERSION))))))))))))" "." empty := NUMERIC_VERSION := $(subst $(empty) ,.,$(wordlist 1,2,$(subst ., ,$(original_version))) 255) endif # Force version.h to be remade if $(PACKAGE_VERSION) has changed. version.h: $(if $(wildcard version.h),$(if $(findstring "$(PACKAGE_VERSION)",$(shell cat version.h)),,force)) endif version.h: echo '#define HTS_VERSION "$(PACKAGE_VERSION)"' > $@ print-version: @echo $(PACKAGE_VERSION) .SUFFIXES: .bundle .c .o .pico .so .c.o: $(CC) $(CFLAGS) -I. $(CPPFLAGS) -c -o $@ $< .c.pico: $(CC) $(CFLAGS) -I. $(CPPFLAGS) $(EXTRA_CFLAGS_PIC) -c -o $@ $< LIBHTS_OBJS = \ kfunc.o \ knetfile.o \ kstring.o \ bgzf.o \ faidx.o \ hfile.o \ hfile_net.o \ hts.o \ md5.o \ regidx.o \ sam.o \ synced_bcf_reader.o \ vcf_sweep.o \ tbx.o \ vcf.o \ vcfutils.o \ cram/cram_codecs.o \ cram/cram_decode.o \ cram/cram_encode.o \ cram/cram_external.o \ cram/cram_index.o \ cram/cram_io.o \ cram/cram_samtools.o \ cram/cram_stats.o \ cram/files.o \ cram/mFILE.o \ cram/open_trace_file.o \ cram/pooled_alloc.o \ cram/rANS_static.o \ cram/sam_header.o \ cram/string_alloc.o \ cram/thread_pool.o \ cram/vlen.o \ cram/zfio.o PLUGIN_OBJS = cram_h = cram/cram.h $(cram_samtools_h) $(cram_sam_header_h) $(cram_structs_h) $(cram_io_h) cram/cram_encode.h cram/cram_decode.h cram/cram_stats.h cram/cram_codecs.h cram/cram_index.h $(htslib_cram_h) cram_io_h = cram/cram_io.h $(cram_misc_h) cram_misc_h = cram/misc.h cram/os.h cram_sam_header_h = cram/sam_header.h cram/string_alloc.h cram/pooled_alloc.h $(htslib_khash_h) $(htslib_kstring_h) cram_samtools_h = cram/cram_samtools.h $(htslib_sam_h) $(cram_sam_header_h) cram_structs_h = cram/cram_structs.h cram/thread_pool.h cram/string_alloc.h $(htslib_khash_h) cram_open_trace_file_h = cram/open_trace_file.h cram/mFILE.h hfile_internal_h = hfile_internal.h $(htslib_hfile_h) hts_internal_h = hts_internal.h $(htslib_hts_h) # To be effective, config.mk needs to appear after most Makefile variables are # set but before most rules appear, so that it can both use previously-set # variables in its own rules' prerequisites and also update variables for use # in later rules' prerequisites. # If your make doesn't accept -include, change this to 'include' if you are # using the configure script or just comment the line out if you are not. -include config.mk # Usually config.h is generated by running configure or config.status, # but if those aren't used create a default config.h here. config.h: echo '/* Empty config.h generated by Makefile */' > $@ lib-static: libhts.a # $(shell), :=, and ifeq/.../endif are GNU Make-specific. If you don't have # GNU Make, comment out the parts of this conditional that don't apply. PLATFORM := $(shell uname -s) ifeq "$(PLATFORM)" "Darwin" SHLIB_FLAVOUR = dylib lib-shared: libhts.dylib BUILT_PLUGINS = $(PLUGIN_OBJS:.o=.bundle) else SHLIB_FLAVOUR = so lib-shared: libhts.so BUILT_PLUGINS = $(PLUGIN_OBJS:.o=.so) endif plugins: $(BUILT_PLUGINS) libhts.a: $(LIBHTS_OBJS) @-rm -f $@ $(AR) -rc $@ $(LIBHTS_OBJS) -$(RANLIB) $@ # The target here is libhts.so, as that is the built file that other rules # depend upon and that is used when -lhts appears in other program's recipes. # As a byproduct invisible to make, libhts.so.NN is also created, as it is the # file used at runtime (when $LD_LIBRARY_PATH includes the build directory). libhts.so: $(LIBHTS_OBJS:.o=.pico) $(CC) -shared -Wl,-soname,libhts.so.$(LIBHTS_SOVERSION) -pthread $(LDFLAGS) -o $@ $(LIBHTS_OBJS:.o=.pico) -lz -lm $(LIBS) ln -sf $@ libhts.so.$(LIBHTS_SOVERSION) # Similarly this also creates libhts.NN.dylib as a byproduct, so that programs # when run can find this uninstalled shared library (when $DYLD_LIBRARY_PATH # includes this project's build directory). libhts.dylib: $(LIBHTS_OBJS) $(CC) -dynamiclib -install_name $(libdir)/libhts.$(LIBHTS_SOVERSION).dylib -current_version $(NUMERIC_VERSION) -compatibility_version $(LIBHTS_SOVERSION) $(LDFLAGS) -o $@ $(LIBHTS_OBJS) -lz $(LIBS) ln -sf $@ libhts.$(LIBHTS_SOVERSION).dylib .pico.so: $(CC) -shared -Wl,-E -pthread $(LDFLAGS) -o $@ $< $(LIBS) .o.bundle: $(CC) -bundle -Wl,-undefined,dynamic_lookup $(LDFLAGS) -o $@ $< $(LIBS) bgzf.o bgzf.pico: bgzf.c config.h $(htslib_hts_h) $(htslib_bgzf_h) $(htslib_hfile_h) $(htslib_khash_h) kstring.o kstring.pico: kstring.c config.h $(htslib_kstring_h) knetfile.o knetfile.pico: knetfile.c config.h $(htslib_knetfile_h) hfile.o hfile.pico: hfile.c config.h $(htslib_hfile_h) $(hfile_internal_h) $(hts_internal_h) $(htslib_khash_h) hfile_irods.o hfile_irods.pico: hfile_irods.c config.h $(hfile_internal_h) $(htslib_hts_h) $(htslib_kstring_h) hfile_libcurl.o hfile_libcurl.pico: hfile_libcurl.c config.h $(hfile_internal_h) $(htslib_hts_h) $(htslib_kstring_h) hfile_net.o hfile_net.pico: hfile_net.c config.h $(hfile_internal_h) $(htslib_knetfile_h) hts.o hts.pico: hts.c config.h version.h $(htslib_hts_h) $(htslib_bgzf_h) $(cram_h) $(htslib_hfile_h) $(htslib_khash_h) $(htslib_kseq_h) $(htslib_ksort_h) $(hts_internal_h) vcf.o vcf.pico: vcf.c config.h $(htslib_vcf_h) $(htslib_bgzf_h) $(htslib_tbx_h) $(htslib_hfile_h) $(hts_internal_h) $(htslib_khash_h) $(htslib_kseq_h) $(htslib_kstring_h) $(htslib_khash_str2int_h) sam.o sam.pico: sam.c config.h $(htslib_sam_h) $(htslib_bgzf_h) $(cram_h) $(hts_internal_h) $(htslib_hfile_h) $(htslib_khash_h) $(htslib_kseq_h) $(htslib_kstring_h) tbx.o tbx.pico: tbx.c config.h $(htslib_tbx_h) $(htslib_bgzf_h) $(htslib_khash_h) faidx.o faidx.pico: faidx.c config.h $(htslib_bgzf_h) $(htslib_faidx_h) $(htslib_hfile_h) $(htslib_khash_h) $(htslib_kstring_h) synced_bcf_reader.o synced_bcf_reader.pico: synced_bcf_reader.c config.h $(htslib_synced_bcf_reader_h) $(htslib_kseq_h) $(htslib_khash_str2int_h) $(htslib_bgzf_h) vcf_sweep.o vcf_sweep.pico: vcf_sweep.c config.h $(htslib_vcf_sweep_h) $(htslib_bgzf_h) vcfutils.o vcfutils.pico: vcfutils.c config.h $(htslib_vcfutils_h) $(htslib_kbitset_h) kfunc.o kfunc.pico: kfunc.c config.h $(htslib_kfunc_h) regidx.o regidx.pico: regidx.c config.h $(htslib_hts_h) $(htslib_kstring_h) $(htslib_kseq_h) $(htslib_khash_str2int_h) $(htslib_regidx_h) md5.o md5.pico: md5.c config.h $(htslib_hts_h) plugin.o plugin.pico: plugin.c config.h $(hts_internal_h) $(htslib_kstring_h) cram/cram_codecs.o cram/cram_codecs.pico: cram/cram_codecs.c config.h $(cram_h) cram/cram_decode.o cram/cram_decode.pico: cram/cram_decode.c config.h $(cram_h) cram/os.h $(htslib_hts_h) cram/cram_encode.o cram/cram_encode.pico: cram/cram_encode.c config.h $(cram_h) cram/os.h $(htslib_hts_h) cram/cram_external.o cram/cram_external.pico: cram/cram_external.c config.h $(htslib_hfile_h) $(cram_h) cram/cram_index.o cram/cram_index.pico: cram/cram_index.c config.h $(htslib_hfile_h) $(cram_h) cram/os.h cram/zfio.h $(hts_internal_h) cram/cram_io.o cram/cram_io.pico: cram/cram_io.c config.h $(cram_h) cram/os.h $(htslib_hts_h) $(cram_open_trace_file_h) cram/rANS_static.h $(htslib_hfile_h) $(htslib_bgzf_h) $(htslib_faidx_h) cram/cram_samtools.o cram/cram_samtools.pico: cram/cram_samtools.c config.h $(cram_h) $(htslib_sam_h) cram/cram_stats.o cram/cram_stats.pico: cram/cram_stats.c config.h $(cram_h) cram/os.h cram/files.o cram/files.pico: cram/files.c config.h $(cram_misc_h) cram/mFILE.o cram/mFILE.pico: cram/mFILE.c config.h cram/os.h cram/mFILE.h cram/vlen.h cram/open_trace_file.o cram/open_trace_file.pico: cram/open_trace_file.c config.h cram/os.h $(cram_open_trace_file_h) $(cram_misc_h) $(htslib_hfile_h) cram/pooled_alloc.o cram/pooled_alloc.pico: cram/pooled_alloc.c config.h cram/pooled_alloc.h cram/rANS_static.o cram/rANS_static.pico: cram/rANS_static.c config.h cram/rANS_static.h cram/rANS_byte.h cram/sam_header.o cram/sam_header.pico: cram/sam_header.c config.h $(cram_sam_header_h) cram/string_alloc.h cram/string_alloc.o cram/string_alloc.pico: cram/string_alloc.c config.h cram/string_alloc.h cram/thread_pool.o cram/thread_pool.pico: cram/thread_pool.c config.h cram/thread_pool.h cram/vlen.o cram/vlen.pico: cram/vlen.c config.h cram/vlen.h cram/os.h cram/zfio.o cram/zfio.pico: cram/zfio.c config.h cram/os.h cram/zfio.h bgzip: bgzip.o libhts.a $(CC) -pthread $(LDFLAGS) -o $@ bgzip.o libhts.a -lz $(LIBS) htsfile: htsfile.o libhts.a $(CC) -pthread $(LDFLAGS) -o $@ htsfile.o libhts.a -lz $(LIBS) tabix: tabix.o libhts.a $(CC) -pthread $(LDFLAGS) -o $@ tabix.o libhts.a -lz $(LIBS) bgzip.o: bgzip.c config.h $(htslib_bgzf_h) $(htslib_hts_h) htsfile.o: htsfile.c config.h $(htslib_hfile_h) $(htslib_hts_h) $(htslib_sam_h) $(htslib_vcf_h) tabix.o: tabix.c config.h $(htslib_tbx_h) $(htslib_sam_h) $(htslib_vcf_h) $(htslib_kseq_h) $(htslib_bgzf_h) $(htslib_hts_h) $(htslib_regidx_h) # For tests that might use it, set $REF_PATH explicitly to use only reference # areas within the test suite (or set it to ':' to use no reference areas). check test: htsfile $(BUILT_TEST_PROGRAMS) test/fieldarith test/fieldarith.sam test/hfile test/sam test/ce.fa test/faidx.fa test/test-regidx cd test && REF_PATH=: ./test.pl test/fieldarith: test/fieldarith.o libhts.a $(CC) -pthread $(LDFLAGS) -o $@ test/fieldarith.o libhts.a -lz $(LIBS) test/hfile: test/hfile.o libhts.a $(CC) -pthread $(LDFLAGS) -o $@ test/hfile.o libhts.a -lz $(LIBS) test/sam: test/sam.o libhts.a $(CC) -pthread $(LDFLAGS) -o $@ test/sam.o libhts.a -lz $(LIBS) test/test-regidx: test/test-regidx.o libhts.a $(CC) -pthread $(LDFLAGS) -o $@ test/test-regidx.o libhts.a -lz $(LIBS) test/test_view: test/test_view.o libhts.a $(CC) -pthread $(LDFLAGS) -o $@ test/test_view.o libhts.a -lz $(LIBS) test/test-vcf-api: test/test-vcf-api.o libhts.a $(CC) -pthread $(LDFLAGS) -o $@ test/test-vcf-api.o libhts.a -lz $(LIBS) test/test-vcf-sweep: test/test-vcf-sweep.o libhts.a $(CC) -pthread $(LDFLAGS) -o $@ test/test-vcf-sweep.o libhts.a -lz $(LIBS) test/fieldarith.o: test/fieldarith.c config.h $(htslib_sam_h) test/hfile.o: test/hfile.c config.h $(htslib_hfile_h) $(htslib_hts_defs_h) test/sam.o: test/sam.c config.h $(htslib_hts_defs_h) $(htslib_sam_h) $(htslib_faidx_h) $(htslib_kstring_h) test/test-regidx.o: test/test-regidx.c config.h $(htslib_regidx_h) test/test_view.o: test/test_view.c config.h $(cram_h) $(htslib_sam_h) test/test-vcf-api.o: test/test-vcf-api.c config.h $(htslib_hts_h) $(htslib_vcf_h) $(htslib_kstring_h) $(htslib_kseq_h) test/test-vcf-sweep.o: test/test-vcf-sweep.c config.h $(htslib_vcf_sweep_h) # install: libhts.a $(BUILT_PROGRAMS) $(BUILT_PLUGINS) installdirs install-$(SHLIB_FLAVOUR) install-pkgconfig # $(INSTALL_PROGRAM) $(BUILT_PROGRAMS) $(DESTDIR)$(bindir) # if test -n "$(BUILT_PLUGINS)"; then $(INSTALL_PROGRAM) $(BUILT_PLUGINS) $(DESTDIR)$(plugindir); fi # $(INSTALL_DATA) htslib/*.h $(DESTDIR)$(includedir)/htslib # $(INSTALL_DATA) libhts.a $(DESTDIR)$(libdir)/libhts.a # $(INSTALL_DATA) htsfile.1 tabix.1 $(DESTDIR)$(man1dir) # $(INSTALL_DATA) faidx.5 sam.5 vcf.5 $(DESTDIR)$(man5dir) installdirs: $(INSTALL_DIR) $(DESTDIR)$(bindir) $(DESTDIR)$(includedir) $(DESTDIR)$(includedir)/htslib $(DESTDIR)$(libdir) $(DESTDIR)$(man1dir) $(DESTDIR)$(man5dir) $(DESTDIR)$(pkgconfigdir) if test -n "$(plugindir)"; then $(INSTALL_DIR) $(DESTDIR)$(plugindir); fi # After installation, the real file in $(libdir) will be libhts.so.X.Y.Z, # with symlinks libhts.so (used via -lhts during linking of client programs) # and libhts.so.NN (used by client executables at runtime). install-so: libhts.so installdirs $(INSTALL_DATA) libhts.so $(DESTDIR)$(libdir)/libhts.so.$(PACKAGE_VERSION) ln -sf libhts.so.$(PACKAGE_VERSION) $(DESTDIR)$(libdir)/libhts.so ln -sf libhts.so.$(PACKAGE_VERSION) $(DESTDIR)$(libdir)/libhts.so.$(LIBHTS_SOVERSION) install-dylib: libhts.dylib installdirs $(INSTALL_PROGRAM) libhts.dylib $(DESTDIR)$(libdir)/libhts.$(PACKAGE_VERSION).dylib ln -sf libhts.$(PACKAGE_VERSION).dylib $(DESTDIR)$(libdir)/libhts.dylib ln -sf libhts.$(PACKAGE_VERSION).dylib $(DESTDIR)$(libdir)/libhts.$(LIBHTS_SOVERSION).dylib # Substitute these pseudo-autoconf variables only at install time # so that "make install prefix=/prefix/path" etc continue to work. install-pkgconfig: installdirs sed -e 's#@includedir@#$(includedir)#g;s#@libdir@#$(libdir)#g;s#@PACKAGE_VERSION@#$(PACKAGE_VERSION)#g' htslib.pc.in > $(DESTDIR)$(pkgconfigdir)/htslib.pc chmod 644 $(DESTDIR)$(pkgconfigdir)/htslib.pc # A pkg-config file (suitable for copying to $PKG_CONFIG_PATH) that provides # flags for building against the uninstalled library in this build directory. htslib-uninstalled.pc: htslib.pc.in sed -e 's#@includedir@#'`pwd`'#g;s#@libdir@#'`pwd`'#g' htslib.pc.in > $@ testclean: -rm -f test/*.tmp test/*.tmp.* mostlyclean: testclean -rm -f *.o *.pico cram/*.o cram/*.pico test/*.o test/*.dSYM version.h clean: mostlyclean clean-$(SHLIB_FLAVOUR) -rm -f libhts.a $(BUILT_PROGRAMS) $(BUILT_PLUGINS) $(BUILT_TEST_PROGRAMS) distclean maintainer-clean: clean -rm -f config.cache config.h config.log config.mk config.status -rm -f TAGS *-uninstalled.pc clean-so: -rm -f libhts.so libhts.so.* clean-dylib: -rm -f libhts.dylib libhts.*.dylib tags TAGS: ctags -f TAGS *.[ch] cram/*.[ch] htslib/*.h # We recommend libhts-using programs be built against a separate htslib # installation. However if you feel that you must bundle htslib source # code with your program, this hook enables Automake-style "make dist" # for this subdirectory. If you do bundle an htslib snapshot, please # add identifying information to $(PACKAGE_VERSION) as appropriate. # (The wildcards attempt to omit non-exported files (.git*, README.md, # etc) and other detritus that might be in the top-level directory.) distdir: tar -c *.[ch15] [ILMNRcht]*[ELSbcekmnt] | (cd $(distdir) && tar -x) +cd $(distdir) && $(MAKE) distclean force: .PHONY: all check clean distclean distdir force .PHONY: install install-pkgconfig installdirs lib-shared lib-static .PHONY: maintainer-clean mostlyclean plugins print-version tags test testclean .PHONY: clean-so install-so .PHONY: clean-dylib install-dylib stacks-1.44/htslib/._md5.c000644 000765 000024 00000000265 12750743246 016052 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/md5.c000644 000765 000024 00000024472 12750743246 015643 0ustar00catchenstaff000000 000000 /* * Trivial amendments by James Bonfield to provide an * HTSlib interface. 2015. * * Externally our API uses an opaque hts_md5_context structure. * * Internally either this gets defined and used with the routines here * or it remains incomplete and is cast to the OpenSSL MD5_CTX structure * and used by routines from OpenSSL. */ /* * This is an OpenSSL-compatible implementation of the RSA Data Security, Inc. * MD5 Message-Digest Algorithm (RFC 1321). * * Homepage: * http://openwall.info/wiki/people/solar/software/public-domain-source-code/md5 * * Author: * Alexander Peslyak, better known as Solar Designer * * This software was written by Alexander Peslyak in 2001. No copyright is * claimed, and the software is hereby placed in the public domain. * In case this attempt to disclaim copyright and place the software in the * public domain is deemed null and void, then the software is * Copyright (c) 2001 Alexander Peslyak and it is hereby released to the * general public under the following terms: * * Redistribution and use in source and binary forms, with or without * modification, are permitted. * * There's ABSOLUTELY NO WARRANTY, express or implied. * * (This is a heavily cut-down "BSD license".) * * This differs from Colin Plumb's older public domain implementation in that * no exactly 32-bit integer data type is required (any 32-bit or wider * unsigned integer data type will do), there's no compile-time endianness * configuration, and the function prototypes match OpenSSL's. No code from * Colin Plumb's implementation has been reused; this comment merely compares * the properties of the two independent implementations. * * The primary goals of this implementation are portability and ease of use. * It is meant to be fast, but not as fast as possible. Some known * optimizations are not included to reduce source code size and avoid * compile-time configuration. */ #include #include #include "htslib/hts.h" #ifndef HAVE_OPENSSL #include /* Any 32-bit or wider unsigned integer data type will do */ typedef unsigned int hts_md5_u32plus; struct hts_md5_context { hts_md5_u32plus lo, hi; hts_md5_u32plus a, b, c, d; unsigned char buffer[64]; hts_md5_u32plus block[16]; }; /* * The basic MD5 functions. * * F and G are optimized compared to their RFC 1321 definitions for * architectures that lack an AND-NOT instruction, just like in Colin Plumb's * implementation. */ #define F(x, y, z) ((z) ^ ((x) & ((y) ^ (z)))) #define G(x, y, z) ((y) ^ ((z) & ((x) ^ (y)))) #define H(x, y, z) (((x) ^ (y)) ^ (z)) #define H2(x, y, z) ((x) ^ ((y) ^ (z))) #define I(x, y, z) ((y) ^ ((x) | ~(z))) /* * The MD5 transformation for all four rounds. */ #define STEP(f, a, b, c, d, x, t, s) \ (a) += f((b), (c), (d)) + (x) + (t); \ (a) = (((a) << (s)) | (((a) & 0xffffffff) >> (32 - (s)))); \ (a) += (b); /* * SET reads 4 input bytes in little-endian byte order and stores them * in a properly aligned word in host byte order. * * The check for little-endian architectures that tolerate unaligned * memory accesses is just an optimization. Nothing will break if it * doesn't work. */ #if defined(__i386__) || defined(__x86_64__) || defined(__vax__) #define SET(n) \ (*(hts_md5_u32plus *)&ptr[(n) * 4]) #define GET(n) \ SET(n) #else #define SET(n) \ (ctx->block[(n)] = \ (hts_md5_u32plus)ptr[(n) * 4] | \ ((hts_md5_u32plus)ptr[(n) * 4 + 1] << 8) | \ ((hts_md5_u32plus)ptr[(n) * 4 + 2] << 16) | \ ((hts_md5_u32plus)ptr[(n) * 4 + 3] << 24)) #define GET(n) \ (ctx->block[(n)]) #endif /* * This processes one or more 64-byte data blocks, but does NOT update * the bit counters. There are no alignment requirements. */ static const void *body(hts_md5_context *ctx, const void *data, unsigned long size) { const unsigned char *ptr; hts_md5_u32plus a, b, c, d; hts_md5_u32plus saved_a, saved_b, saved_c, saved_d; ptr = (const unsigned char *)data; a = ctx->a; b = ctx->b; c = ctx->c; d = ctx->d; do { saved_a = a; saved_b = b; saved_c = c; saved_d = d; /* Round 1 */ STEP(F, a, b, c, d, SET(0), 0xd76aa478, 7) STEP(F, d, a, b, c, SET(1), 0xe8c7b756, 12) STEP(F, c, d, a, b, SET(2), 0x242070db, 17) STEP(F, b, c, d, a, SET(3), 0xc1bdceee, 22) STEP(F, a, b, c, d, SET(4), 0xf57c0faf, 7) STEP(F, d, a, b, c, SET(5), 0x4787c62a, 12) STEP(F, c, d, a, b, SET(6), 0xa8304613, 17) STEP(F, b, c, d, a, SET(7), 0xfd469501, 22) STEP(F, a, b, c, d, SET(8), 0x698098d8, 7) STEP(F, d, a, b, c, SET(9), 0x8b44f7af, 12) STEP(F, c, d, a, b, SET(10), 0xffff5bb1, 17) STEP(F, b, c, d, a, SET(11), 0x895cd7be, 22) STEP(F, a, b, c, d, SET(12), 0x6b901122, 7) STEP(F, d, a, b, c, SET(13), 0xfd987193, 12) STEP(F, c, d, a, b, SET(14), 0xa679438e, 17) STEP(F, b, c, d, a, SET(15), 0x49b40821, 22) /* Round 2 */ STEP(G, a, b, c, d, GET(1), 0xf61e2562, 5) STEP(G, d, a, b, c, GET(6), 0xc040b340, 9) STEP(G, c, d, a, b, GET(11), 0x265e5a51, 14) STEP(G, b, c, d, a, GET(0), 0xe9b6c7aa, 20) STEP(G, a, b, c, d, GET(5), 0xd62f105d, 5) STEP(G, d, a, b, c, GET(10), 0x02441453, 9) STEP(G, c, d, a, b, GET(15), 0xd8a1e681, 14) STEP(G, b, c, d, a, GET(4), 0xe7d3fbc8, 20) STEP(G, a, b, c, d, GET(9), 0x21e1cde6, 5) STEP(G, d, a, b, c, GET(14), 0xc33707d6, 9) STEP(G, c, d, a, b, GET(3), 0xf4d50d87, 14) STEP(G, b, c, d, a, GET(8), 0x455a14ed, 20) STEP(G, a, b, c, d, GET(13), 0xa9e3e905, 5) STEP(G, d, a, b, c, GET(2), 0xfcefa3f8, 9) STEP(G, c, d, a, b, GET(7), 0x676f02d9, 14) STEP(G, b, c, d, a, GET(12), 0x8d2a4c8a, 20) /* Round 3 */ STEP(H, a, b, c, d, GET(5), 0xfffa3942, 4) STEP(H2, d, a, b, c, GET(8), 0x8771f681, 11) STEP(H, c, d, a, b, GET(11), 0x6d9d6122, 16) STEP(H2, b, c, d, a, GET(14), 0xfde5380c, 23) STEP(H, a, b, c, d, GET(1), 0xa4beea44, 4) STEP(H2, d, a, b, c, GET(4), 0x4bdecfa9, 11) STEP(H, c, d, a, b, GET(7), 0xf6bb4b60, 16) STEP(H2, b, c, d, a, GET(10), 0xbebfbc70, 23) STEP(H, a, b, c, d, GET(13), 0x289b7ec6, 4) STEP(H2, d, a, b, c, GET(0), 0xeaa127fa, 11) STEP(H, c, d, a, b, GET(3), 0xd4ef3085, 16) STEP(H2, b, c, d, a, GET(6), 0x04881d05, 23) STEP(H, a, b, c, d, GET(9), 0xd9d4d039, 4) STEP(H2, d, a, b, c, GET(12), 0xe6db99e5, 11) STEP(H, c, d, a, b, GET(15), 0x1fa27cf8, 16) STEP(H2, b, c, d, a, GET(2), 0xc4ac5665, 23) /* Round 4 */ STEP(I, a, b, c, d, GET(0), 0xf4292244, 6) STEP(I, d, a, b, c, GET(7), 0x432aff97, 10) STEP(I, c, d, a, b, GET(14), 0xab9423a7, 15) STEP(I, b, c, d, a, GET(5), 0xfc93a039, 21) STEP(I, a, b, c, d, GET(12), 0x655b59c3, 6) STEP(I, d, a, b, c, GET(3), 0x8f0ccc92, 10) STEP(I, c, d, a, b, GET(10), 0xffeff47d, 15) STEP(I, b, c, d, a, GET(1), 0x85845dd1, 21) STEP(I, a, b, c, d, GET(8), 0x6fa87e4f, 6) STEP(I, d, a, b, c, GET(15), 0xfe2ce6e0, 10) STEP(I, c, d, a, b, GET(6), 0xa3014314, 15) STEP(I, b, c, d, a, GET(13), 0x4e0811a1, 21) STEP(I, a, b, c, d, GET(4), 0xf7537e82, 6) STEP(I, d, a, b, c, GET(11), 0xbd3af235, 10) STEP(I, c, d, a, b, GET(2), 0x2ad7d2bb, 15) STEP(I, b, c, d, a, GET(9), 0xeb86d391, 21) a += saved_a; b += saved_b; c += saved_c; d += saved_d; ptr += 64; } while (size -= 64); ctx->a = a; ctx->b = b; ctx->c = c; ctx->d = d; return ptr; } void hts_md5_reset(hts_md5_context *ctx) { ctx->a = 0x67452301; ctx->b = 0xefcdab89; ctx->c = 0x98badcfe; ctx->d = 0x10325476; ctx->lo = 0; ctx->hi = 0; } void hts_md5_update(hts_md5_context *ctx, const void *data, unsigned long size) { hts_md5_u32plus saved_lo; unsigned long used, available; saved_lo = ctx->lo; if ((ctx->lo = (saved_lo + size) & 0x1fffffff) < saved_lo) ctx->hi++; ctx->hi += size >> 29; used = saved_lo & 0x3f; if (used) { available = 64 - used; if (size < available) { memcpy(&ctx->buffer[used], data, size); return; } memcpy(&ctx->buffer[used], data, available); data = (const unsigned char *)data + available; size -= available; body(ctx, ctx->buffer, 64); } if (size >= 64) { data = body(ctx, data, size & ~(unsigned long)0x3f); size &= 0x3f; } memcpy(ctx->buffer, data, size); } void hts_md5_final(unsigned char *result, hts_md5_context *ctx) { unsigned long used, available; used = ctx->lo & 0x3f; ctx->buffer[used++] = 0x80; available = 64 - used; if (available < 8) { memset(&ctx->buffer[used], 0, available); body(ctx, ctx->buffer, 64); used = 0; available = 64; } memset(&ctx->buffer[used], 0, available - 8); ctx->lo <<= 3; ctx->buffer[56] = ctx->lo; ctx->buffer[57] = ctx->lo >> 8; ctx->buffer[58] = ctx->lo >> 16; ctx->buffer[59] = ctx->lo >> 24; ctx->buffer[60] = ctx->hi; ctx->buffer[61] = ctx->hi >> 8; ctx->buffer[62] = ctx->hi >> 16; ctx->buffer[63] = ctx->hi >> 24; body(ctx, ctx->buffer, 64); result[0] = ctx->a; result[1] = ctx->a >> 8; result[2] = ctx->a >> 16; result[3] = ctx->a >> 24; result[4] = ctx->b; result[5] = ctx->b >> 8; result[6] = ctx->b >> 16; result[7] = ctx->b >> 24; result[8] = ctx->c; result[9] = ctx->c >> 8; result[10] = ctx->c >> 16; result[11] = ctx->c >> 24; result[12] = ctx->d; result[13] = ctx->d >> 8; result[14] = ctx->d >> 16; result[15] = ctx->d >> 24; memset(ctx, 0, sizeof(*ctx)); } hts_md5_context *hts_md5_init(void) { hts_md5_context *ctx = malloc(sizeof(*ctx)); if (!ctx) return NULL; hts_md5_reset(ctx); return ctx; } #else #include #include /* * Wrappers around the OpenSSL libcrypto.so MD5 implementation. * * These are here to ensure they end up in the symbol table of the * library regardless of the static inline in the headers. */ hts_md5_context *hts_md5_init(void) { MD5_CTX *ctx = malloc(sizeof(*ctx)); if (!ctx) return NULL; MD5_Init(ctx); return (hts_md5_context *)ctx; } void hts_md5_reset(hts_md5_context *ctx) { MD5_Init((MD5_CTX *)ctx); } void hts_md5_update(hts_md5_context *ctx, const void *data, unsigned long size) { MD5_Update((MD5_CTX *)ctx, data, size); } void hts_md5_final(unsigned char *result, hts_md5_context *ctx) { MD5_Final(result, (MD5_CTX *)ctx); } #endif void hts_md5_destroy(hts_md5_context *ctx) { if (!ctx) return; free(ctx); } void hts_md5_hex(char *hex, const unsigned char *digest) { int i; for (i = 0; i < 16; i++) { hex[i*2+0] = "0123456789abcdef"[(digest[i]>>4)&0xf]; hex[i*2+1] = "0123456789abcdef"[digest[i]&0xf]; } hex[32] = 0; } stacks-1.44/htslib/._NEWS000644 000765 000024 00000000265 12750743246 015720 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/NEWS000644 000765 000024 00000011264 12750743246 015504 0ustar00catchenstaff000000 000000 Noteworthy changes in release 1.3.1 (22 April 2016) * Improved error checking and reporting, especially of I/O errors when writing output files (#17, #315, PR #271, PR #317). * Build fixes for 32-bit systems; be sure to run configure to enable large file support and access to 2GiB+ files. * Numerous VCF parsing fixes (#321, #322, #323, #324, #325; PR #370). Particular thanks to Kostya Kortchinsky of the Google Security Team for testing and numerous input parsing bug reports. * HTSlib now prints an informational message when initially creating a CRAM reference cache in the default location under your $HOME directory. (No message is printed if you are using $REF_CACHE to specify a location.) * Avoided rare race condition when caching downloaded CRAM reference sequence files, by using distinctive names for temporary files (in addition to O_EXCL, which has always been used). Occasional corruption would previously occur when multiple tools were simultaneously caching the same reference sequences on an NFS filesystem that did not support O_EXCL (PR #320). * Prevented race condition in file access plugin loading (PR #341). * Fixed mpileup memory leak, so no more "[bam_plp_destroy] memory leak [...] Continue anyway" warning messages (#299). * Various minor CRAM fixes. * Fixed documentation problems #348 and #358. Noteworthy changes in release 1.3 (15 December 2015) * Files can now be accessed via HTTPS and Amazon S3 in addition to HTTP and FTP, when HTSlib is configured to use libcurl for network file access rather than the included basic knetfile networking. * HTSlib can be built to use remote access hFILE backends (such as iRODS and libcurl) via a plugin mechanism. This allows other backends to be easily added and facilitates building tools that use HTSlib, as they don't need to be linked with the backends' various required libraries. * When writing CRAM output, sam_open() etc now default to writing CRAM v3.0 rather than v2.1. * fai_build() and samtools faidx now accept initial whitespace in ">" headers (e.g., "> chr1 description" is taken to refer to "chr1"). * tabix --only-header works again (was broken in 1.2.x; #249). * HTSlib's configure script and Makefile now fully support the standard convention of allowing CC/CPPFLAGS/CFLAGS/LDFLAGS/LIBS to be overridden as needed. Previously the Makefile listened to $(LDLIBS) instead; if you were overriding that, you should now override LIBS rather than LDLIBS. * Fixed bugs #168, #172, #176, #197, #206, #225, #245, #265, #295, and #296. Noteworthy changes in release 1.2.1 (3 February 2015) * Reinstated hts_file_type() and FT_* macros, which were available until 1.1 but briefly removed in 1.2. This function is deprecated and will be removed in a future release -- you should use hts_detect_format() etc instead Noteworthy changes in release 1.2 (2 February 2015) * HTSlib now has a configure script which checks your build environment and allows for selection of optional extras. See INSTALL for details * By default, reference sequences are fetched from the EBI CRAM Reference Registry and cached in your $HOME cache directory. This behaviour can be controlled by setting REF_PATH and REF_CACHE enviroment variables (see the samtools(1) man page for details) * Numerous CRAM improvements: - Support for CRAM v3.0, an upcoming revision to CRAM supporting better compression and per-container checksums - EOF checking for v2.1 and v3.0 (similar to checking BAM EOF blocks) - Non-standard values for PNEXT and TLEN fields are now preserved - hts_set_fai_filename() now provides a reference file when encoding - Generated read names are now numbered from 1, rather than being labelled 'slice:record-in-slice' - Multi-threading and speed improvements * New htsfile command for identifying file formats, and corresponding file format detection APIs * New tabix --regions FILE, --targets FILE options for filtering via BED files * Optional iRODS file access, disabled by default. Configure with --with-irods to enable accessing iRODS data objects directly via 'irods:DATAOBJ' * All occurences of 2^29 in the source have been eliminated, so indexing and querying against reference sequences larger than 512Mbp works (when using CSI indices) * Support for plain GZIP compression in various places * VCF header editing speed improvements * Added seq_nt16_int[] (equivalent to the samtools API's bam_nt16_nt4_table) * Reinstated faidx_fetch_nseq(), which was accidentally removed from 1.1. Now faidx_fetch_nseq() and faidx_nseq() are equivalent; eventually faidx_fetch_nseq() will be deprecated and removed [#156] * Fixed bugs #141, #152, #155, #158, #159, and various memory leaks stacks-1.44/htslib/._plugin.c000644 000765 000024 00000000265 12750743246 016663 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/plugin.c000644 000765 000024 00000011653 12750743246 016451 0ustar00catchenstaff000000 000000 /* plugin.c -- low-level path parsing and plugin functions. Copyright (C) 2015 Genome Research Ltd. Author: John Marshall Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include #include #include #include #include "hts_internal.h" #include "htslib/kstring.h" #ifndef PLUGINPATH #define PLUGINPATH "" #endif #ifdef __APPLE__ #define PLUGIN_EXT ".bundle" #define PLUGIN_EXT_LEN 7 #else #define PLUGIN_EXT ".so" #define PLUGIN_EXT_LEN 3 #endif static DIR *open_nextdir(struct hts_path_itr *itr) { DIR *dir; while (1) { const char *colon = strchr(itr->pathdir, ':'); if (colon == NULL) return NULL; itr->entry.l = 0; kputsn(itr->pathdir, colon - itr->pathdir, &itr->entry); itr->pathdir = &colon[1]; if (itr->entry.l == 0) continue; dir = opendir(itr->entry.s); if (dir) break; if (hts_verbose >= 4) fprintf(stderr, "[W::hts_path_itr] can't scan directory \"%s\": %s\n", itr->entry.s, strerror(errno)); } if (itr->entry.s[itr->entry.l-1] != '/') kputc('/', &itr->entry); itr->entry_dir_l = itr->entry.l; return dir; } void hts_path_itr_setup(struct hts_path_itr *itr, const char *path, const char *builtin_path, const char *prefix, size_t prefix_len, const char *suffix, size_t suffix_len) { itr->prefix = prefix; itr->prefix_len = prefix_len; if (suffix) itr->suffix = suffix, itr->suffix_len = suffix_len; else itr->suffix = PLUGIN_EXT, itr->suffix_len = PLUGIN_EXT_LEN; itr->path.l = itr->path.m = 0; itr->path.s = NULL; itr->entry.l = itr->entry.m = 0; itr->entry.s = NULL; if (! builtin_path) builtin_path = PLUGINPATH; if (! path) { path = getenv("HTS_PATH"); if (! path) path = ""; } while (1) { size_t len = strcspn(path, ":"); if (len == 0) kputs(builtin_path, &itr->path); else kputsn(path, len, &itr->path); kputc(':', &itr->path); path += len; if (*path == ':') path++; else break; } // Note that ':' now terminates entries rather than separates them itr->pathdir = itr->path.s; itr->dirv = open_nextdir(itr); } const char *hts_path_itr_next(struct hts_path_itr *itr) { while (itr->dirv) { struct dirent *e; while ((e = readdir((DIR *) itr->dirv)) != NULL) { size_t d_name_len = strlen(e->d_name); if (strncmp(e->d_name, itr->prefix, itr->prefix_len) == 0 && d_name_len >= itr->suffix_len && strncmp(e->d_name + d_name_len - itr->suffix_len, itr->suffix, itr->suffix_len) == 0) { itr->entry.l = itr->entry_dir_l; kputs(e->d_name, &itr->entry); return itr->entry.s; } } closedir((DIR *) itr->dirv); itr->dirv = open_nextdir(itr); } itr->pathdir = NULL; free(itr->path.s); itr->path.s = NULL; free(itr->entry.s); itr->entry.s = NULL; return NULL; } void *load_plugin(void **pluginp, const char *filename, const char *symbol) { void *lib = dlopen(filename, RTLD_NOW | RTLD_LOCAL); if (lib == NULL) goto error; void *sym = dlsym(lib, symbol); if (sym == NULL) goto error; *pluginp = lib; return sym; error: if (hts_verbose >= 4) fprintf(stderr, "[W::%s] can't load plugin \"%s\": %s\n", __func__, filename, dlerror()); if (lib) dlclose(lib); return NULL; } void *plugin_sym(void *plugin, const char *name, const char **errmsg) { void *sym = dlsym(plugin, name); if (sym == NULL) *errmsg = dlerror(); return sym; } void close_plugin(void *plugin) { if (dlclose(plugin) != 0) { if (hts_verbose >= 4) fprintf(stderr, "[W::%s] dlclose() failed: %s\n", __func__, dlerror()); } } stacks-1.44/htslib/._README000644 000765 000024 00000000265 12750743246 016101 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/README000644 000765 000024 00000000421 12750743246 015656 0ustar00catchenstaff000000 000000 HTSlib is an implementation of a unified C library for accessing common file formats, such as SAM, CRAM, VCF, and BCF, used for high-throughput sequencing data. It is the core library used by samtools and bcftools. See INSTALL for building and installation instructions. stacks-1.44/htslib/._regidx.c000644 000765 000024 00000000265 12750743246 016647 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/regidx.c000644 000765 000024 00000023673 12750743246 016442 0ustar00catchenstaff000000 000000 /* Copyright (C) 2014 Genome Research Ltd. Author: Petr Danecek Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include "htslib/hts.h" #include "htslib/kstring.h" #include "htslib/kseq.h" #include "htslib/khash_str2int.h" #include "htslib/regidx.h" #define LIDX_SHIFT 13 // number of insignificant index bits // List of regions for one chromosome typedef struct { int *idx, nidx; int nregs, mregs; // n:used, m:alloced reg_t *regs; void *payload; } reglist_t; // Container of all sequences struct _regidx_t { int nseq, mseq; // n:used, m:alloced reglist_t *seq; // regions for each sequence void *seq2regs; // hash for fast lookup from chr name to regions char **seq_names; regidx_free_f free; // function to free any data allocated by regidx_parse_f regidx_parse_f parse; // parse one input line void *usr; // user data to pass to regidx_parse_f // temporary data for index initialization kstring_t str; int rid_prev, start_prev, end_prev; int payload_size; void *payload; }; int regidx_seq_nregs(regidx_t *idx, const char *seq) { int iseq; if ( khash_str2int_get(idx->seq2regs, seq, &iseq)!=0 ) return 0; // no such sequence return idx->seq[iseq].nregs; } int regidx_nregs(regidx_t *idx) { int i, nregs = 0; for (i=0; inseq; i++) nregs += idx->seq[i].nregs; return nregs; } char **regidx_seq_names(regidx_t *idx, int *n) { *n = idx->nseq; return idx->seq_names; } int _regidx_build_index(regidx_t *idx) { int iseq; for (iseq=0; iseqnseq; iseq++) { reglist_t *list = &idx->seq[iseq]; int j,k, imax = 0; // max index bin for (j=0; jnregs; j++) { int ibeg = list->regs[j].start >> LIDX_SHIFT; int iend = list->regs[j].end >> LIDX_SHIFT; if ( imax < iend + 1 ) { int old_imax = imax; imax = iend + 1; kroundup32(imax); list->idx = (int*) realloc(list->idx, imax*sizeof(int)); for (k=old_imax; kidx[k] = -1; } if ( ibeg==iend ) { if ( list->idx[ibeg]<0 ) list->idx[ibeg] = j; } else { for (k=ibeg; k<=iend; k++) if ( list->idx[k]<0 ) list->idx[k] = j; } list->nidx = iend + 1; } } return 0; } int regidx_insert(regidx_t *idx, char *line) { if ( !line ) return _regidx_build_index(idx); char *chr_from, *chr_to; reg_t reg; int ret = idx->parse(line,&chr_from,&chr_to,®,idx->payload,idx->usr); if ( ret==-2 ) return -1; // error if ( ret==-1 ) return 0; // skip the line int rid; idx->str.l = 0; kputsn(chr_from, chr_to-chr_from+1, &idx->str); if ( khash_str2int_get(idx->seq2regs, idx->str.s, &rid)!=0 ) { idx->nseq++; int m_prev = idx->mseq; hts_expand0(reglist_t,idx->nseq,idx->mseq,idx->seq); hts_expand0(char*,idx->nseq,m_prev,idx->seq_names); idx->seq_names[idx->nseq-1] = strdup(idx->str.s); rid = khash_str2int_inc(idx->seq2regs, idx->seq_names[idx->nseq-1]); } reglist_t *list = &idx->seq[rid]; list->nregs++; int m_prev = list->mregs; hts_expand(reg_t,list->nregs,list->mregs,list->regs); list->regs[list->nregs-1] = reg; if ( idx->payload_size ) { if ( m_prev < list->mregs ) list->payload = realloc(list->payload,idx->payload_size*list->mregs); memcpy(list->payload + idx->payload_size*(list->nregs-1), idx->payload, idx->payload_size); } if ( idx->rid_prev==rid ) { if ( idx->start_prev > reg.start || (idx->start_prev==reg.start && idx->end_prev>reg.end) ) { fprintf(stderr,"The regions are not sorted: %s:%d-%d is before %s:%d-%d\n", idx->str.s,idx->start_prev+1,idx->end_prev+1,idx->str.s,reg.start+1,reg.end+1); return -1; } } idx->rid_prev = rid; idx->start_prev = reg.start; idx->end_prev = reg.end; return 0; } regidx_t *regidx_init(const char *fname, regidx_parse_f parser, regidx_free_f free_f, size_t payload_size, void *usr_dat) { if ( !parser ) { if ( !fname ) parser = regidx_parse_tab; else { int len = strlen(fname); if ( len>=7 && !strcasecmp(".bed.gz",fname+len-7) ) parser = regidx_parse_bed; else if ( len>=8 && !strcasecmp(".bed.bgz",fname+len-8) ) parser = regidx_parse_bed; else if ( len>=4 && !strcasecmp(".bed",fname+len-4) ) parser = regidx_parse_bed; else parser = regidx_parse_tab; } } regidx_t *idx = (regidx_t*) calloc(1,sizeof(regidx_t)); idx->free = free_f; idx->parse = parser; idx->usr = usr_dat; idx->seq2regs = khash_str2int_init(); idx->rid_prev = -1; idx->start_prev = -1; idx->end_prev = -1; idx->payload_size = payload_size; if ( payload_size ) idx->payload = malloc(payload_size); if ( !fname ) return idx; kstring_t str = {0,0,0}; htsFile *fp = hts_open(fname,"r"); if ( !fp ) goto error; while ( hts_getline(fp, KS_SEP_LINE, &str) > 0 ) { if ( regidx_insert(idx, str.s) ) goto error; } regidx_insert(idx, NULL); free(str.s); hts_close(fp); return idx; error: free(str.s); if ( fp ) hts_close(fp); regidx_destroy(idx); return NULL; } void regidx_destroy(regidx_t *idx) { int i, j; for (i=0; inseq; i++) { reglist_t *list = &idx->seq[i]; if ( idx->free ) { for (j=0; jnregs; j++) idx->free(list->payload + idx->payload_size*j); } free(list->payload); free(list->regs); free(list->idx); } free(idx->seq_names); free(idx->seq); free(idx->str.s); free(idx->payload); khash_str2int_destroy_free(idx->seq2regs); free(idx); } int regidx_overlap(regidx_t *idx, const char *chr, uint32_t from, uint32_t to, regitr_t *itr) { if ( itr ) itr->i = itr->n = 0; int iseq; if ( khash_str2int_get(idx->seq2regs, chr, &iseq)!=0 ) return 0; // no such sequence reglist_t *list = &idx->seq[iseq]; if ( !list->nregs ) return 0; int i, ibeg = from>>LIDX_SHIFT; int ireg = ibeg < list->nidx ? list->idx[ibeg] : list->idx[ list->nidx - 1 ]; if ( ireg < 0 ) { // linear search; if slow, replace with binary search if ( ibeg > list->nidx ) ibeg = list->nidx; for (i=ibeg - 1; i>=0; i--) if ( list->idx[i] >=0 ) break; ireg = i>=0 ? list->idx[i] : 0; } for (i=ireg; inregs; i++) { if ( list->regs[i].start > to ) return 0; // no match if ( list->regs[i].end >= from && list->regs[i].start <= to ) break; // found } if ( i>=list->nregs ) return 0; // no match if ( !itr ) return 1; itr->i = 0; itr->n = list->nregs - i; itr->reg = &idx->seq[iseq].regs[i]; if ( idx->payload_size ) itr->payload = idx->seq[iseq].payload + i*idx->payload_size; else itr->payload = NULL; return 1; } int regidx_parse_bed(const char *line, char **chr_beg, char **chr_end, reg_t *reg, void *payload, void *usr) { char *ss = (char*) line; while ( *ss && isspace(*ss) ) ss++; if ( !*ss ) return -1; // skip blank lines if ( *ss=='#' ) return -1; // skip comments char *se = ss; while ( *se && !isspace(*se) ) se++; if ( !*se ) { fprintf(stderr,"Could not parse bed line: %s\n", line); return -2; } *chr_beg = ss; *chr_end = se-1; ss = se+1; reg->start = hts_parse_decimal(ss, &se, 0); if ( ss==se ) { fprintf(stderr,"Could not parse bed line: %s\n", line); return -2; } ss = se+1; reg->end = hts_parse_decimal(ss, &se, 0) - 1; if ( ss==se ) { fprintf(stderr,"Could not parse bed line: %s\n", line); return -2; } return 0; } int regidx_parse_tab(const char *line, char **chr_beg, char **chr_end, reg_t *reg, void *payload, void *usr) { char *ss = (char*) line; while ( *ss && isspace(*ss) ) ss++; if ( !*ss ) return -1; // skip blank lines if ( *ss=='#' ) return -1; // skip comments char *se = ss; while ( *se && !isspace(*se) ) se++; if ( !*se ) { fprintf(stderr,"Could not parse bed line: %s\n", line); return -2; } *chr_beg = ss; *chr_end = se-1; ss = se+1; reg->start = hts_parse_decimal(ss, &se, 0) - 1; if ( ss==se ) { fprintf(stderr,"Could not parse bed line: %s\n", line); return -2; } if ( !se[0] || !se[1] ) reg->end = reg->start; else { ss = se+1; reg->end = hts_parse_decimal(ss, &se, 0); if ( ss==se ) reg->end = reg->start; else reg->end--; } return 0; } stacks-1.44/htslib/._sam.c000644 000765 000024 00000000265 12750743246 016145 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/sam.c000644 000765 000024 00000207604 12750743246 015736 0ustar00catchenstaff000000 000000 /* sam.c -- SAM and BAM file I/O and manipulation. Copyright (C) 2008-2010, 2012-2016 Genome Research Ltd. Copyright (C) 2010, 2012, 2013 Broad Institute. Author: Heng Li Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include #include #include #include #include "htslib/sam.h" #include "htslib/bgzf.h" #include "cram/cram.h" #include "hts_internal.h" #include "htslib/hfile.h" #include "htslib/khash.h" KHASH_DECLARE(s2i, kh_cstr_t, int64_t) typedef khash_t(s2i) sdict_t; /********************** *** BAM header I/O *** **********************/ bam_hdr_t *bam_hdr_init() { return (bam_hdr_t*)calloc(1, sizeof(bam_hdr_t)); } void bam_hdr_destroy(bam_hdr_t *h) { int32_t i; if (h == NULL) return; if (h->target_name) { for (i = 0; i < h->n_targets; ++i) free(h->target_name[i]); free(h->target_name); free(h->target_len); } free(h->text); free(h->cigar_tab); if (h->sdict) kh_destroy(s2i, (sdict_t*)h->sdict); free(h); } bam_hdr_t *bam_hdr_dup(const bam_hdr_t *h0) { if (h0 == NULL) return NULL; bam_hdr_t *h; if ((h = bam_hdr_init()) == NULL) return NULL; // copy the simple data h->n_targets = h0->n_targets; h->ignore_sam_err = h0->ignore_sam_err; h->l_text = h0->l_text; // Then the pointery stuff h->cigar_tab = NULL; h->sdict = NULL; // TODO Check for memory allocation failures h->text = (char*)calloc(h->l_text + 1, 1); memcpy(h->text, h0->text, h->l_text); h->target_len = (uint32_t*)calloc(h->n_targets, sizeof(uint32_t)); h->target_name = (char**)calloc(h->n_targets, sizeof(char*)); int i; for (i = 0; i < h->n_targets; ++i) { h->target_len[i] = h0->target_len[i]; h->target_name[i] = strdup(h0->target_name[i]); } return h; } static bam_hdr_t *hdr_from_dict(sdict_t *d) { bam_hdr_t *h; khint_t k; h = bam_hdr_init(); h->sdict = d; h->n_targets = kh_size(d); // TODO Check for memory allocation failures h->target_len = (uint32_t*)malloc(sizeof(uint32_t) * h->n_targets); h->target_name = (char**)malloc(sizeof(char*) * h->n_targets); for (k = kh_begin(d); k != kh_end(d); ++k) { if (!kh_exist(d, k)) continue; h->target_name[kh_val(d, k)>>32] = (char*)kh_key(d, k); h->target_len[kh_val(d, k)>>32] = kh_val(d, k)<<32>>32; kh_val(d, k) >>= 32; } return h; } bam_hdr_t *bam_hdr_read(BGZF *fp) { bam_hdr_t *h; char buf[4]; int magic_len, has_EOF; int32_t i, name_len, num_names = 0; size_t bufsize; ssize_t bytes; // check EOF has_EOF = bgzf_check_EOF(fp); if (has_EOF < 0) { perror("[W::bam_hdr_read] bgzf_check_EOF"); } else if (has_EOF == 0 && hts_verbose >= 2) fprintf(stderr, "[W::%s] EOF marker is absent. The input is probably truncated.\n", __func__); // read "BAM1" magic_len = bgzf_read(fp, buf, 4); if (magic_len != 4 || strncmp(buf, "BAM\1", 4)) { if (hts_verbose >= 1) fprintf(stderr, "[E::%s] invalid BAM binary header\n", __func__); return 0; } h = bam_hdr_init(); if (!h) goto nomem; // read plain text and the number of reference sequences bytes = bgzf_read(fp, &h->l_text, 4); if (bytes != 4) goto read_err; if (fp->is_be) ed_swap_4p(&h->l_text); bufsize = ((size_t) h->l_text) + 1; if (bufsize < h->l_text) goto nomem; // so large that adding 1 overflowed h->text = (char*)malloc(bufsize); if (!h->text) goto nomem; h->text[h->l_text] = 0; // make sure it is NULL terminated bytes = bgzf_read(fp, h->text, h->l_text); if (bytes != h->l_text) goto read_err; bytes = bgzf_read(fp, &h->n_targets, 4); if (bytes != 4) goto read_err; if (fp->is_be) ed_swap_4p(&h->n_targets); if (h->n_targets < 0) goto invalid; // read reference sequence names and lengths if (h->n_targets > 0) { h->target_name = (char**)calloc(h->n_targets, sizeof(char*)); if (!h->target_name) goto nomem; h->target_len = (uint32_t*)calloc(h->n_targets, sizeof(uint32_t)); if (!h->target_len) goto nomem; } else { h->target_name = NULL; h->target_len = NULL; } for (i = 0; i != h->n_targets; ++i) { bytes = bgzf_read(fp, &name_len, 4); if (bytes != 4) goto read_err; if (fp->is_be) ed_swap_4p(&name_len); if (name_len <= 0) goto invalid; h->target_name[i] = (char*)malloc(name_len); if (!h->target_name[i]) goto nomem; num_names++; bytes = bgzf_read(fp, h->target_name[i], name_len); if (bytes != name_len) goto read_err; if (h->target_name[i][name_len - 1] != '\0') { /* Fix missing NUL-termination. Is this being too nice? We could alternatively bail out with an error. */ char *new_name; if (name_len == INT32_MAX) goto invalid; new_name = realloc(h->target_name[i], name_len + 1); if (new_name == NULL) goto nomem; h->target_name[i] = new_name; h->target_name[i][name_len] = '\0'; } bytes = bgzf_read(fp, &h->target_len[i], 4); if (bytes != 4) goto read_err; if (fp->is_be) ed_swap_4p(&h->target_len[i]); } return h; nomem: if (hts_verbose >= 1) fprintf(stderr, "[E::%s] out of memory\n", __func__); goto clean; read_err: if (hts_verbose >= 1) { if (bytes < 0) { fprintf(stderr, "[E::%s] error reading BGZF stream\n", __func__); } else { fprintf(stderr, "[E::%s] truncated bam header\n", __func__); } } goto clean; invalid: if (hts_verbose >= 1) { fprintf(stderr, "[E::%s] invalid BAM binary header\n", __func__); } clean: if (h != NULL) { h->n_targets = num_names; // ensure we free only allocated target_names bam_hdr_destroy(h); } return NULL; } int bam_hdr_write(BGZF *fp, const bam_hdr_t *h) { char buf[4]; int32_t i, name_len, x; // write "BAM1" strncpy(buf, "BAM\1", 4); if (bgzf_write(fp, buf, 4) < 0) return -1; // write plain text and the number of reference sequences if (fp->is_be) { x = ed_swap_4(h->l_text); if (bgzf_write(fp, &x, 4) < 0) return -1; if (h->l_text) { if (bgzf_write(fp, h->text, h->l_text) < 0) return -1; } x = ed_swap_4(h->n_targets); if (bgzf_write(fp, &x, 4) < 0) return -1; } else { if (bgzf_write(fp, &h->l_text, 4) < 0) return -1; if (h->l_text) { if (bgzf_write(fp, h->text, h->l_text) < 0) return -1; } if (bgzf_write(fp, &h->n_targets, 4) < 0) return -1; } // write sequence names and lengths for (i = 0; i != h->n_targets; ++i) { char *p = h->target_name[i]; name_len = strlen(p) + 1; if (fp->is_be) { x = ed_swap_4(name_len); if (bgzf_write(fp, &x, 4) < 0) return -1; } else { if (bgzf_write(fp, &name_len, 4) < 0) return -1; } if (bgzf_write(fp, p, name_len) < 0) return -1; if (fp->is_be) { x = ed_swap_4(h->target_len[i]); if (bgzf_write(fp, &x, 4) < 0) return -1; } else { if (bgzf_write(fp, &h->target_len[i], 4) < 0) return -1; } } if (bgzf_flush(fp) < 0) return -1; return 0; } int bam_name2id(bam_hdr_t *h, const char *ref) { sdict_t *d = (sdict_t*)h->sdict; khint_t k; if (h->sdict == 0) { int i, absent; d = kh_init(s2i); for (i = 0; i < h->n_targets; ++i) { k = kh_put(s2i, d, h->target_name[i], &absent); kh_val(d, k) = i; } h->sdict = d; } k = kh_get(s2i, d, ref); return k == kh_end(d)? -1 : kh_val(d, k); } /************************* *** BAM alignment I/O *** *************************/ bam1_t *bam_init1() { return (bam1_t*)calloc(1, sizeof(bam1_t)); } void bam_destroy1(bam1_t *b) { if (b == 0) return; free(b->data); free(b); } bam1_t *bam_copy1(bam1_t *bdst, const bam1_t *bsrc) { uint8_t *data = bdst->data; int m_data = bdst->m_data; // backup data and m_data if (m_data < bsrc->l_data) { // double the capacity m_data = bsrc->l_data; kroundup32(m_data); data = (uint8_t*)realloc(data, m_data); } memcpy(data, bsrc->data, bsrc->l_data); // copy var-len data *bdst = *bsrc; // copy the rest // restore the backup bdst->m_data = m_data; bdst->data = data; return bdst; } bam1_t *bam_dup1(const bam1_t *bsrc) { if (bsrc == NULL) return NULL; bam1_t *bdst = bam_init1(); if (bdst == NULL) return NULL; return bam_copy1(bdst, bsrc); } int bam_cigar2qlen(int n_cigar, const uint32_t *cigar) { int k, l; for (k = l = 0; k < n_cigar; ++k) if (bam_cigar_type(bam_cigar_op(cigar[k]))&1) l += bam_cigar_oplen(cigar[k]); return l; } int bam_cigar2rlen(int n_cigar, const uint32_t *cigar) { int k, l; for (k = l = 0; k < n_cigar; ++k) if (bam_cigar_type(bam_cigar_op(cigar[k]))&2) l += bam_cigar_oplen(cigar[k]); return l; } int32_t bam_endpos(const bam1_t *b) { if (!(b->core.flag & BAM_FUNMAP) && b->core.n_cigar > 0) return b->core.pos + bam_cigar2rlen(b->core.n_cigar, bam_get_cigar(b)); else return b->core.pos + 1; } static inline int aux_type2size(uint8_t type) { switch (type) { case 'A': case 'c': case 'C': return 1; case 's': case 'S': return 2; case 'i': case 'I': case 'f': return 4; case 'd': return 8; case 'Z': case 'H': case 'B': return type; default: return 0; } } static void swap_data(const bam1_core_t *c, int l_data, uint8_t *data, int is_host) { uint8_t *s; uint32_t *cigar = (uint32_t*)(data + c->l_qname); uint32_t i, n; s = data + c->n_cigar*4 + c->l_qname + c->l_qseq + (c->l_qseq + 1)/2; for (i = 0; i < c->n_cigar; ++i) ed_swap_4p(&cigar[i]); while (s < data + l_data) { int size; s += 2; // skip key size = aux_type2size(*s); ++s; // skip type switch (size) { case 1: ++s; break; case 2: ed_swap_2p(s); s += 2; break; case 4: ed_swap_4p(s); s += 4; break; case 8: ed_swap_8p(s); s += 8; break; case 'Z': case 'H': while (*s) ++s; ++s; break; case 'B': size = aux_type2size(*s); ++s; if (is_host) memcpy(&n, s, 4), ed_swap_4p(s); else ed_swap_4p(s), memcpy(&n, s, 4); s += 4; switch (size) { case 1: s += n; break; case 2: for (i = 0; i < n; ++i, s += 2) ed_swap_2p(s); break; case 4: for (i = 0; i < n; ++i, s += 4) ed_swap_4p(s); break; case 8: for (i = 0; i < n; ++i, s += 8) ed_swap_8p(s); break; } break; } } } int bam_read1(BGZF *fp, bam1_t *b) { bam1_core_t *c = &b->core; int32_t block_len, ret, i; uint32_t x[8]; if ((ret = bgzf_read(fp, &block_len, 4)) != 4) { if (ret == 0) return -1; // normal end-of-file else return -2; // truncated } if (bgzf_read(fp, x, 32) != 32) return -3; if (fp->is_be) { ed_swap_4p(&block_len); for (i = 0; i < 8; ++i) ed_swap_4p(x + i); } c->tid = x[0]; c->pos = x[1]; c->bin = x[2]>>16; c->qual = x[2]>>8&0xff; c->l_qname = x[2]&0xff; c->flag = x[3]>>16; c->n_cigar = x[3]&0xffff; c->l_qseq = x[4]; c->mtid = x[5]; c->mpos = x[6]; c->isize = x[7]; b->l_data = block_len - 32; if (b->l_data < 0 || c->l_qseq < 0 || c->l_qname < 1) return -4; if ((char *)bam_get_aux(b) - (char *)b->data > b->l_data) return -4; if (b->m_data < b->l_data) { b->m_data = b->l_data; kroundup32(b->m_data); b->data = (uint8_t*)realloc(b->data, b->m_data); if (!b->data) return -4; } if (bgzf_read(fp, b->data, b->l_data) != b->l_data) return -4; //b->l_aux = b->l_data - c->n_cigar * 4 - c->l_qname - c->l_qseq - (c->l_qseq+1)/2; if (fp->is_be) swap_data(c, b->l_data, b->data, 0); return 4 + block_len; } int bam_write1(BGZF *fp, const bam1_t *b) { const bam1_core_t *c = &b->core; uint32_t x[8], block_len = b->l_data + 32, y; int i, ok; x[0] = c->tid; x[1] = c->pos; x[2] = (uint32_t)c->bin<<16 | c->qual<<8 | c->l_qname; x[3] = (uint32_t)c->flag<<16 | c->n_cigar; x[4] = c->l_qseq; x[5] = c->mtid; x[6] = c->mpos; x[7] = c->isize; ok = (bgzf_flush_try(fp, 4 + block_len) >= 0); if (fp->is_be) { for (i = 0; i < 8; ++i) ed_swap_4p(x + i); y = block_len; if (ok) ok = (bgzf_write(fp, ed_swap_4p(&y), 4) >= 0); swap_data(c, b->l_data, b->data, 1); } else { if (ok) ok = (bgzf_write(fp, &block_len, 4) >= 0); } if (ok) ok = (bgzf_write(fp, x, 32) >= 0); if (ok) ok = (bgzf_write(fp, b->data, b->l_data) >= 0); if (fp->is_be) swap_data(c, b->l_data, b->data, 0); return ok? 4 + block_len : -1; } /******************** *** BAM indexing *** ********************/ static hts_idx_t *bam_index(BGZF *fp, int min_shift) { int n_lvls, i, fmt, ret; bam1_t *b; hts_idx_t *idx; bam_hdr_t *h; h = bam_hdr_read(fp); if (h == NULL) return NULL; if (min_shift > 0) { int64_t max_len = 0, s; for (i = 0; i < h->n_targets; ++i) if (max_len < h->target_len[i]) max_len = h->target_len[i]; max_len += 256; for (n_lvls = 0, s = 1< s; ++n_lvls, s <<= 3); fmt = HTS_FMT_CSI; } else min_shift = 14, n_lvls = 5, fmt = HTS_FMT_BAI; idx = hts_idx_init(h->n_targets, fmt, bgzf_tell(fp), min_shift, n_lvls); bam_hdr_destroy(h); b = bam_init1(); while ((ret = bam_read1(fp, b)) >= 0) { ret = hts_idx_push(idx, b->core.tid, b->core.pos, bam_endpos(b), bgzf_tell(fp), !(b->core.flag&BAM_FUNMAP)); if (ret < 0) goto err; // unsorted } if (ret < -1) goto err; // corrupted BAM file hts_idx_finish(idx, bgzf_tell(fp)); bam_destroy1(b); return idx; err: bam_destroy1(b); hts_idx_destroy(idx); return NULL; } int sam_index_build2(const char *fn, const char *fnidx, int min_shift) { hts_idx_t *idx; htsFile *fp; int ret = 0; if ((fp = hts_open(fn, "r")) == 0) return -2; switch (fp->format.format) { case cram: ret = cram_index_build(fp->fp.cram, fn, fnidx); break; case bam: idx = bam_index(fp->fp.bgzf, min_shift); if (idx) { ret = hts_idx_save_as(idx, fn, fnidx, (min_shift > 0)? HTS_FMT_CSI : HTS_FMT_BAI); hts_idx_destroy(idx); } else ret = -1; break; default: ret = -3; break; } hts_close(fp); return ret; } int sam_index_build(const char *fn, int min_shift) { return sam_index_build2(fn, NULL, min_shift); } // Provide bam_index_build() symbol for binary compability with earlier HTSlib #undef bam_index_build int bam_index_build(const char *fn, int min_shift) { return sam_index_build2(fn, NULL, min_shift); } static int bam_readrec(BGZF *fp, void *ignored, void *bv, int *tid, int *beg, int *end) { bam1_t *b = bv; int ret; if ((ret = bam_read1(fp, b)) >= 0) { *tid = b->core.tid; *beg = b->core.pos; *end = bam_endpos(b); } return ret; } // This is used only with read_rest=1 iterators, so need not set tid/beg/end. static int cram_readrec(BGZF *ignored, void *fpv, void *bv, int *tid, int *beg, int *end) { htsFile *fp = fpv; bam1_t *b = bv; return cram_get_bam_seq(fp->fp.cram, &b); } // This is used only with read_rest=1 iterators, so need not set tid/beg/end. static int sam_bam_cram_readrec(BGZF *bgzfp, void *fpv, void *bv, int *tid, int *beg, int *end) { htsFile *fp = fpv; bam1_t *b = bv; switch (fp->format.format) { case bam: return bam_read1(bgzfp, b); case cram: return cram_get_bam_seq(fp->fp.cram, &b); default: // TODO Need headers available to implement this for SAM files fprintf(stderr, "[sam_bam_cram_readrec] Not implemented for SAM files -- Exiting\n"); abort(); } } hts_idx_t *sam_index_load2(htsFile *fp, const char *fn, const char *fnidx) { switch (fp->format.format) { case bam: return fnidx? hts_idx_load2(fn, fnidx) : hts_idx_load(fn, HTS_FMT_BAI); case cram: { if (cram_index_load(fp->fp.cram, fn, fnidx) < 0) return NULL; // Cons up a fake "index" just pointing at the associated cram_fd: hts_cram_idx_t *idx = malloc(sizeof (hts_cram_idx_t)); if (idx == NULL) return NULL; idx->fmt = HTS_FMT_CRAI; idx->cram = fp->fp.cram; return (hts_idx_t *) idx; } default: return NULL; // TODO Would use tbx_index_load if it returned hts_idx_t } } hts_idx_t *sam_index_load(htsFile *fp, const char *fn) { return sam_index_load2(fp, fn, NULL); } static hts_itr_t *cram_itr_query(const hts_idx_t *idx, int tid, int beg, int end, hts_readrec_func *readrec) { const hts_cram_idx_t *cidx = (const hts_cram_idx_t *) idx; hts_itr_t *iter = (hts_itr_t *) calloc(1, sizeof(hts_itr_t)); if (iter == NULL) return NULL; // Cons up a dummy iterator for which hts_itr_next() will simply invoke // the readrec function: iter->read_rest = 1; iter->off = NULL; iter->bins.a = NULL; iter->readrec = readrec; if (tid >= 0 || tid == HTS_IDX_NOCOOR) { cram_range r = { tid == HTS_IDX_NOCOOR ? -1 : tid, beg+1, end }; int ret = cram_set_option(cidx->cram, CRAM_OPT_RANGE, &r); iter->curr_off = 0; // The following fields are not required by hts_itr_next(), but are // filled in in case user code wants to look at them. iter->tid = tid; iter->beg = beg; iter->end = end; switch (ret) { case 0: break; case -2: // No data vs this ref, so mark iterator as completed. // Same as HTS_IDX_NONE. iter->finished = 1; break; default: free(iter); return NULL; } } else switch (tid) { case HTS_IDX_REST: iter->curr_off = 0; break; case HTS_IDX_NONE: iter->curr_off = 0; iter->finished = 1; break; default: fprintf(stderr, "[cram_itr_query] tid=%d not implemented for CRAM files -- Exiting\n", tid); abort(); break; } return iter; } hts_itr_t *sam_itr_queryi(const hts_idx_t *idx, int tid, int beg, int end) { const hts_cram_idx_t *cidx = (const hts_cram_idx_t *) idx; if (idx == NULL) return hts_itr_query(NULL, tid, beg, end, sam_bam_cram_readrec); else if (cidx->fmt == HTS_FMT_CRAI) return cram_itr_query(idx, tid, beg, end, cram_readrec); else return hts_itr_query(idx, tid, beg, end, bam_readrec); } static int cram_name2id(void *fdv, const char *ref) { cram_fd *fd = (cram_fd *) fdv; return sam_hdr_name2ref(fd->header, ref); } hts_itr_t *sam_itr_querys(const hts_idx_t *idx, bam_hdr_t *hdr, const char *region) { const hts_cram_idx_t *cidx = (const hts_cram_idx_t *) idx; if (cidx->fmt == HTS_FMT_CRAI) return hts_itr_querys(idx, region, cram_name2id, cidx->cram, cram_itr_query, cram_readrec); else return hts_itr_querys(idx, region, (hts_name2id_f)(bam_name2id), hdr, hts_itr_query, bam_readrec); } /********************** *** SAM header I/O *** **********************/ #include "htslib/kseq.h" #include "htslib/kstring.h" bam_hdr_t *sam_hdr_parse(int l_text, const char *text) { const char *q, *r, *p; khash_t(s2i) *d; d = kh_init(s2i); for (p = text; *p; ++p) { if (strncmp(p, "@SQ\t", 4) == 0) { char *sn = 0; int ln = -1; for (q = p + 4;; ++q) { if (strncmp(q, "SN:", 3) == 0) { q += 3; for (r = q; *r != '\t' && *r != '\n' && *r != '\0'; ++r); sn = (char*)calloc(r - q + 1, 1); strncpy(sn, q, r - q); q = r; } else if (strncmp(q, "LN:", 3) == 0) ln = strtol(q + 3, (char**)&q, 10); while (*q != '\t' && *q != '\n' && *q != '\0') ++q; if (*q == '\0' || *q == '\n') break; } p = q; if (sn && ln >= 0) { khint_t k; int absent; k = kh_put(s2i, d, sn, &absent); if (!absent) { if (hts_verbose >= 2) fprintf(stderr, "[W::%s] duplicated sequence '%s'\n", __func__, sn); free(sn); } else kh_val(d, k) = (int64_t)(kh_size(d) - 1)<<32 | ln; } } while (*p != '\0' && *p != '\n') ++p; } return hdr_from_dict(d); } bam_hdr_t *sam_hdr_read(htsFile *fp) { switch (fp->format.format) { case bam: return bam_hdr_read(fp->fp.bgzf); case cram: return cram_header_to_bam(fp->fp.cram->header); case sam: { kstring_t str; bam_hdr_t *h; int has_SQ = 0; str.l = str.m = 0; str.s = 0; while (hts_getline(fp, KS_SEP_LINE, &fp->line) >= 0) { if (fp->line.s[0] != '@') break; if (fp->line.l > 3 && strncmp(fp->line.s,"@SQ",3) == 0) has_SQ = 1; kputsn(fp->line.s, fp->line.l, &str); kputc('\n', &str); } if (! has_SQ && fp->fn_aux) { char line[2048]; FILE *f = fopen(fp->fn_aux, "r"); if (f == NULL) return NULL; while (fgets(line, sizeof line, f)) { const char *name = strtok(line, "\t"); const char *length = strtok(NULL, "\t"); ksprintf(&str, "@SQ\tSN:%s\tLN:%s\n", name, length); } fclose(f); } if (str.l == 0) kputsn("", 0, &str); h = sam_hdr_parse(str.l, str.s); h->l_text = str.l; h->text = str.s; return h; } default: abort(); } } int sam_hdr_write(htsFile *fp, const bam_hdr_t *h) { switch (fp->format.format) { case binary_format: fp->format.category = sequence_data; fp->format.format = bam; /* fall-through */ case bam: if (bam_hdr_write(fp->fp.bgzf, h) < 0) return -1; break; case cram: { cram_fd *fd = fp->fp.cram; SAM_hdr *hdr = bam_header_to_cram((bam_hdr_t *)h); if (! hdr) return -1; if (cram_set_header(fd, hdr) < 0) return -1; if (fp->fn_aux) cram_load_reference(fd, fp->fn_aux); if (cram_write_SAM_hdr(fd, fd->header) < 0) return -1; } break; case text_format: fp->format.category = sequence_data; fp->format.format = sam; /* fall-through */ case sam: { char *p; hputs(h->text, fp->fp.hfile); p = strstr(h->text, "@SQ\t"); // FIXME: we need a loop to make sure "@SQ\t" does not match something unwanted!!! if (p == 0) { int i; for (i = 0; i < h->n_targets; ++i) { fp->line.l = 0; kputsn("@SQ\tSN:", 7, &fp->line); kputs(h->target_name[i], &fp->line); kputsn("\tLN:", 4, &fp->line); kputw(h->target_len[i], &fp->line); kputc('\n', &fp->line); if ( hwrite(fp->fp.hfile, fp->line.s, fp->line.l) != fp->line.l ) return -1; } } if ( hflush(fp->fp.hfile) != 0 ) return -1; } break; default: abort(); } return 0; } /********************** *** SAM record I/O *** **********************/ int sam_parse1(kstring_t *s, bam_hdr_t *h, bam1_t *b) { #define _read_token(_p) (_p); for (; *(_p) && *(_p) != '\t'; ++(_p)); if (*(_p) != '\t') goto err_ret; *(_p)++ = 0 #define _read_token_aux(_p) (_p); for (; *(_p) && *(_p) != '\t'; ++(_p)); *(_p)++ = 0 // this is different in that it does not test *(_p)=='\t' #define _get_mem(type_t, _x, _s, _l) ks_resize((_s), (_s)->l + (_l)); *(_x) = (type_t*)((_s)->s + (_s)->l); (_s)->l += (_l) #define _parse_err(cond, msg) do { if ((cond) && hts_verbose >= 1) { fprintf(stderr, "[E::%s] " msg "\n", __func__); goto err_ret; } } while (0) #define _parse_warn(cond, msg) if ((cond) && hts_verbose >= 2) fprintf(stderr, "[W::%s] " msg "\n", __func__) uint8_t *t; char *p = s->s, *q; int i; kstring_t str; bam1_core_t *c = &b->core; str.l = b->l_data = 0; str.s = (char*)b->data; str.m = b->m_data; memset(c, 0, 32); if (h->cigar_tab == 0) { h->cigar_tab = (int8_t*) malloc(128); for (i = 0; i < 128; ++i) h->cigar_tab[i] = -1; for (i = 0; BAM_CIGAR_STR[i]; ++i) h->cigar_tab[(int)BAM_CIGAR_STR[i]] = i; } // qname q = _read_token(p); _parse_warn(p - q <= 1, "empty query name"); _parse_err(p - q > 255, "query name too long"); kputsn_(q, p - q, &str); c->l_qname = p - q; // flag c->flag = strtol(p, &p, 0); if (*p++ != '\t') goto err_ret; // malformated flag // chr q = _read_token(p); if (strcmp(q, "*")) { _parse_err(h->n_targets == 0, "missing SAM header"); c->tid = bam_name2id(h, q); _parse_warn(c->tid < 0, "urecognized reference name; treated as unmapped"); } else c->tid = -1; // pos c->pos = strtol(p, &p, 10) - 1; if (*p++ != '\t') goto err_ret; if (c->pos < 0 && c->tid >= 0) { _parse_warn(1, "mapped query cannot have zero coordinate; treated as unmapped"); c->tid = -1; } if (c->tid < 0) c->flag |= BAM_FUNMAP; // mapq c->qual = strtol(p, &p, 10); if (*p++ != '\t') goto err_ret; // cigar if (*p != '*') { uint32_t *cigar; size_t n_cigar = 0; for (q = p; *p && *p != '\t'; ++p) if (!isdigit(*p)) ++n_cigar; if (*p++ != '\t') goto err_ret; _parse_err(n_cigar == 0, "no CIGAR operations"); _parse_err(n_cigar >= 65536, "too many CIGAR operations"); c->n_cigar = n_cigar; _get_mem(uint32_t, &cigar, &str, c->n_cigar * sizeof(uint32_t)); for (i = 0; i < c->n_cigar; ++i, ++q) { int op; cigar[i] = strtol(q, &q, 10)<= 128? -1 : h->cigar_tab[(int)*q]; _parse_err(op < 0, "unrecognized CIGAR operator"); cigar[i] |= op; } // can't use bam_endpos() directly as some fields not yet set up i = (!(c->flag&BAM_FUNMAP))? bam_cigar2rlen(c->n_cigar, cigar) : 1; } else { _parse_warn(!(c->flag&BAM_FUNMAP), "mapped query must have a CIGAR; treated as unmapped"); c->flag |= BAM_FUNMAP; q = _read_token(p); i = 1; } c->bin = hts_reg2bin(c->pos, c->pos + i, 14, 5); // mate chr q = _read_token(p); if (strcmp(q, "=") == 0) { c->mtid = c->tid; } else if (strcmp(q, "*") == 0) { c->mtid = -1; } else { c->mtid = bam_name2id(h, q); _parse_warn(c->mtid < 0, "urecognized mate reference name; treated as unmapped"); } // mpos c->mpos = strtol(p, &p, 10) - 1; if (*p++ != '\t') goto err_ret; if (c->mpos < 0 && c->mtid >= 0) { _parse_warn(1, "mapped mate cannot have zero coordinate; treated as unmapped"); c->mtid = -1; } // tlen c->isize = strtol(p, &p, 10); if (*p++ != '\t') goto err_ret; // seq q = _read_token(p); if (strcmp(q, "*")) { c->l_qseq = p - q - 1; i = bam_cigar2qlen(c->n_cigar, (uint32_t*)(str.s + c->l_qname)); _parse_err(c->n_cigar && i != c->l_qseq, "CIGAR and query sequence are of different length"); i = (c->l_qseq + 1) >> 1; _get_mem(uint8_t, &t, &str, i); memset(t, 0, i); for (i = 0; i < c->l_qseq; ++i) t[i>>1] |= seq_nt16_table[(int)q[i]] << ((~i&1)<<2); } else c->l_qseq = 0; // qual q = _read_token_aux(p); _get_mem(uint8_t, &t, &str, c->l_qseq); if (strcmp(q, "*")) { _parse_err(p - q - 1 != c->l_qseq, "SEQ and QUAL are of different length"); for (i = 0; i < c->l_qseq; ++i) t[i] = q[i] - 33; } else memset(t, 0xff, c->l_qseq); // aux // Note that (like the bam1_core_t fields) this aux data in b->data is // stored in host endianness; so there is no byte swapping needed here. while (p < s->s + s->l) { uint8_t type; q = _read_token_aux(p); // FIXME: can be accelerated for long 'B' arrays _parse_err(p - q - 1 < 6, "incomplete aux field"); kputsn_(q, 2, &str); q += 3; type = *q++; ++q; // q points to value if (type == 'A' || type == 'a' || type == 'c' || type == 'C') { kputc_('A', &str); kputc_(*q, &str); } else if (type == 'i' || type == 'I') { if (*q == '-') { long x = strtol(q, &q, 10); if (x >= INT8_MIN) { kputc_('c', &str); kputc_(x, &str); } else if (x >= INT16_MIN) { int16_t y = x; kputc_('s', &str); kputsn_((char*)&y, 2, &str); } else { int32_t y = x; kputc_('i', &str); kputsn_(&y, 4, &str); } } else { unsigned long x = strtoul(q, &q, 10); if (x <= UINT8_MAX) { kputc_('C', &str); kputc_(x, &str); } else if (x <= UINT16_MAX) { uint16_t y = x; kputc_('S', &str); kputsn_(&y, 2, &str); } else { uint32_t y = x; kputc_('I', &str); kputsn_(&y, 4, &str); } } } else if (type == 'f') { float x; x = strtod(q, &q); kputc_('f', &str); kputsn_(&x, 4, &str); } else if (type == 'd') { double x; x = strtod(q, &q); kputc_('d', &str); kputsn_(&x, 8, &str); } else if (type == 'Z' || type == 'H') { kputc_(type, &str);kputsn_(q, p - q, &str); // note that this include the trailing NULL } else if (type == 'B') { int32_t n; char *r; _parse_err(p - q - 1 < 3, "incomplete B-typed aux field"); type = *q++; // q points to the first ',' following the typing byte for (r = q, n = 0; *r; ++r) if (*r == ',') ++n; kputc_('B', &str); kputc_(type, &str); kputsn_(&n, 4, &str); // FIXME: to evaluate which is faster: a) aligned array and then memmove(); b) unaligned array; c) kputsn_() if (type == 'c') while (q + 1 < p) { int8_t x = strtol(q + 1, &q, 0); kputc_(x, &str); } else if (type == 'C') while (q + 1 < p) { uint8_t x = strtoul(q + 1, &q, 0); kputc_(x, &str); } else if (type == 's') while (q + 1 < p) { int16_t x = strtol(q + 1, &q, 0); kputsn_(&x, 2, &str); } else if (type == 'S') while (q + 1 < p) { uint16_t x = strtoul(q + 1, &q, 0); kputsn_(&x, 2, &str); } else if (type == 'i') while (q + 1 < p) { int32_t x = strtol(q + 1, &q, 0); kputsn_(&x, 4, &str); } else if (type == 'I') while (q + 1 < p) { uint32_t x = strtoul(q + 1, &q, 0); kputsn_(&x, 4, &str); } else if (type == 'f') while (q + 1 < p) { float x = strtod(q + 1, &q); kputsn_(&x, 4, &str); } else _parse_err(1, "unrecognized type"); } else _parse_err(1, "unrecognized type"); } b->data = (uint8_t*)str.s; b->l_data = str.l; b->m_data = str.m; return 0; #undef _parse_warn #undef _parse_err #undef _get_mem #undef _read_token_aux #undef _read_token err_ret: b->data = (uint8_t*)str.s; b->l_data = str.l; b->m_data = str.m; return -2; } int sam_read1(htsFile *fp, bam_hdr_t *h, bam1_t *b) { switch (fp->format.format) { case bam: { int r = bam_read1(fp->fp.bgzf, b); if (r >= 0) { if (b->core.tid >= h->n_targets || b->core.tid < -1 || b->core.mtid >= h->n_targets || b->core.mtid < -1) return -3; } return r; } case cram: { int ret = cram_get_bam_seq(fp->fp.cram, &b); return ret >= 0 ? ret : (cram_eof(fp->fp.cram) ? -1 : -2); } case sam: { int ret; err_recover: if (fp->line.l == 0) { ret = hts_getline(fp, KS_SEP_LINE, &fp->line); if (ret < 0) return -1; } ret = sam_parse1(&fp->line, h, b); fp->line.l = 0; if (ret < 0) { if (hts_verbose >= 1) fprintf(stderr, "[W::%s] parse error at line %lld\n", __func__, (long long)fp->lineno); if (h->ignore_sam_err) goto err_recover; } return ret; } default: abort(); } } int sam_format1(const bam_hdr_t *h, const bam1_t *b, kstring_t *str) { int i; uint8_t *s; const bam1_core_t *c = &b->core; str->l = 0; kputsn(bam_get_qname(b), c->l_qname-1, str); kputc('\t', str); // query name kputw(c->flag, str); kputc('\t', str); // flag if (c->tid >= 0) { // chr kputs(h->target_name[c->tid] , str); kputc('\t', str); } else kputsn("*\t", 2, str); kputw(c->pos + 1, str); kputc('\t', str); // pos kputw(c->qual, str); kputc('\t', str); // qual if (c->n_cigar) { // cigar uint32_t *cigar = bam_get_cigar(b); for (i = 0; i < c->n_cigar; ++i) { kputw(bam_cigar_oplen(cigar[i]), str); kputc(bam_cigar_opchr(cigar[i]), str); } } else kputc('*', str); kputc('\t', str); if (c->mtid < 0) kputsn("*\t", 2, str); // mate chr else if (c->mtid == c->tid) kputsn("=\t", 2, str); else { kputs(h->target_name[c->mtid], str); kputc('\t', str); } kputw(c->mpos + 1, str); kputc('\t', str); // mate pos kputw(c->isize, str); kputc('\t', str); // template len if (c->l_qseq) { // seq and qual uint8_t *s = bam_get_seq(b); for (i = 0; i < c->l_qseq; ++i) kputc("=ACMGRSVTWYHKDBN"[bam_seqi(s, i)], str); kputc('\t', str); s = bam_get_qual(b); if (s[0] == 0xff) kputc('*', str); else for (i = 0; i < c->l_qseq; ++i) kputc(s[i] + 33, str); } else kputsn("*\t*", 3, str); // FIXME change "s+N <= b->data+b->l_data" to "b->data+b->l_data - s >= N" // (or equivalent) everywhere to avoid looking past the end of the array s = bam_get_aux(b); // aux while (s+4 <= b->data + b->l_data) { uint8_t type, key[2]; key[0] = s[0]; key[1] = s[1]; s += 2; type = *s++; kputc('\t', str); kputsn((char*)key, 2, str); kputc(':', str); if (type == 'A') { kputsn("A:", 2, str); kputc(*s, str); ++s; } else if (type == 'C') { kputsn("i:", 2, str); kputw(*s, str); ++s; } else if (type == 'c') { kputsn("i:", 2, str); kputw(*(int8_t*)s, str); ++s; } else if (type == 'S') { if (s+2 <= b->data + b->l_data) { kputsn("i:", 2, str); kputw(*(uint16_t*)s, str); s += 2; } else return -1; } else if (type == 's') { if (s+2 <= b->data + b->l_data) { kputsn("i:", 2, str); kputw(*(int16_t*)s, str); s += 2; } else return -1; } else if (type == 'I') { if (s+4 <= b->data + b->l_data) { kputsn("i:", 2, str); kputuw(*(uint32_t*)s, str); s += 4; } else return -1; } else if (type == 'i') { if (s+4 <= b->data + b->l_data) { kputsn("i:", 2, str); kputw(*(int32_t*)s, str); s += 4; } else return -1; } else if (type == 'f') { if (s+4 <= b->data + b->l_data) { ksprintf(str, "f:%g", *(float*)s); s += 4; } else return -1; } else if (type == 'd') { if (s+8 <= b->data + b->l_data) { ksprintf(str, "d:%g", *(double*)s); s += 8; } else return -1; } else if (type == 'Z' || type == 'H') { kputc(type, str); kputc(':', str); while (s < b->data + b->l_data && *s) kputc(*s++, str); if (s >= b->data + b->l_data) return -1; ++s; } else if (type == 'B') { uint8_t sub_type = *(s++); int sub_type_size = aux_type2size(sub_type); uint32_t n; if (sub_type_size == 0 || b->data + b->l_data - s < 4) return -1; memcpy(&n, s, 4); s += 4; // now points to the start of the array if ((b->data + b->l_data - s) / sub_type_size < n) return -1; kputsn("B:", 2, str); kputc(sub_type, str); // write the typing for (i = 0; i < n; ++i) { // FIXME: for better performance, put the loop after "if" kputc(',', str); if ('c' == sub_type) { kputw(*(int8_t*)s, str); ++s; } else if ('C' == sub_type) { kputw(*(uint8_t*)s, str); ++s; } else if ('s' == sub_type) { kputw(*(int16_t*)s, str); s += 2; } else if ('S' == sub_type) { kputw(*(uint16_t*)s, str); s += 2; } else if ('i' == sub_type) { kputw(*(int32_t*)s, str); s += 4; } else if ('I' == sub_type) { kputuw(*(uint32_t*)s, str); s += 4; } else if ('f' == sub_type) { ksprintf(str, "%g", *(float*)s); s += 4; } else return -1; } } } return str->l; } int sam_write1(htsFile *fp, const bam_hdr_t *h, const bam1_t *b) { switch (fp->format.format) { case binary_format: fp->format.category = sequence_data; fp->format.format = bam; /* fall-through */ case bam: return bam_write1(fp->fp.bgzf, b); case cram: return cram_put_bam_seq(fp->fp.cram, (bam1_t *)b); case text_format: fp->format.category = sequence_data; fp->format.format = sam; /* fall-through */ case sam: if (sam_format1(h, b, &fp->line) < 0) return -1; kputc('\n', &fp->line); if ( hwrite(fp->fp.hfile, fp->line.s, fp->line.l) != fp->line.l ) return -1; return fp->line.l; default: abort(); } } /************************ *** Auxiliary fields *** ************************/ void bam_aux_append(bam1_t *b, const char tag[2], char type, int len, uint8_t *data) { int ori_len = b->l_data; b->l_data += 3 + len; if (b->m_data < b->l_data) { b->m_data = b->l_data; kroundup32(b->m_data); b->data = (uint8_t*)realloc(b->data, b->m_data); } b->data[ori_len] = tag[0]; b->data[ori_len + 1] = tag[1]; b->data[ori_len + 2] = type; memcpy(b->data + ori_len + 3, data, len); } static inline uint8_t *skip_aux(uint8_t *s) { int size = aux_type2size(*s); ++s; // skip type uint32_t n; switch (size) { case 'Z': case 'H': while (*s) ++s; return s + 1; case 'B': size = aux_type2size(*s); ++s; memcpy(&n, s, 4); s += 4; return s + size * n; case 0: abort(); break; default: return s + size; } } uint8_t *bam_aux_get(const bam1_t *b, const char tag[2]) { uint8_t *s; int y = tag[0]<<8 | tag[1]; s = bam_get_aux(b); while (s < b->data + b->l_data) { int x = (int)s[0]<<8 | s[1]; s += 2; if (x == y) return s; s = skip_aux(s); } return 0; } // s MUST BE returned by bam_aux_get() int bam_aux_del(bam1_t *b, uint8_t *s) { uint8_t *p, *aux; int l_aux = bam_get_l_aux(b); aux = bam_get_aux(b); p = s - 2; s = skip_aux(s); memmove(p, s, l_aux - (s - aux)); b->l_data -= s - p; return 0; } int32_t bam_aux2i(const uint8_t *s) { int type; type = *s++; if (type == 'c') return (int32_t)*(int8_t*)s; else if (type == 'C') return (int32_t)*(uint8_t*)s; else if (type == 's') return (int32_t)*(int16_t*)s; else if (type == 'S') return (int32_t)*(uint16_t*)s; else if (type == 'i' || type == 'I') return *(int32_t*)s; else return 0; } double bam_aux2f(const uint8_t *s) { int type; type = *s++; if (type == 'd') return *(double*)s; else if (type == 'f') return *(float*)s; else return 0.0; } char bam_aux2A(const uint8_t *s) { int type; type = *s++; if (type == 'A') return *(char*)s; else return 0; } char *bam_aux2Z(const uint8_t *s) { int type; type = *s++; if (type == 'Z' || type == 'H') return (char*)s; else return 0; } int sam_open_mode(char *mode, const char *fn, const char *format) { // TODO Parse "bam5" etc for compression level if (format == NULL) { // Try to pick a format based on the filename extension const char *ext = fn? strrchr(fn, '.') : NULL; if (ext == NULL || strchr(ext, '/')) return -1; return sam_open_mode(mode, fn, ext+1); } else if (strcmp(format, "bam") == 0) strcpy(mode, "b"); else if (strcmp(format, "cram") == 0) strcpy(mode, "c"); else if (strcmp(format, "sam") == 0) strcpy(mode, ""); else return -1; return 0; } // A version of sam_open_mode that can handle ,key=value options. // The format string is allocated and returned, to be freed by the caller. // Prefix should be "r" or "w", char *sam_open_mode_opts(const char *fn, const char *mode, const char *format) { char *mode_opts = malloc((format ? strlen(format) : 1) + (mode ? strlen(mode) : 1) + 12); char *opts, *cp; int format_len; if (!mode_opts) return NULL; strcpy(mode_opts, mode ? mode : "r"); cp = mode_opts + strlen(mode_opts); if (format == NULL) { // Try to pick a format based on the filename extension const char *ext = fn? strrchr(fn, '.') : NULL; if (ext == NULL || strchr(ext, '/')) { free(mode_opts); return NULL; } return sam_open_mode(cp, fn, ext+1) ? (free(mode_opts), NULL) : mode_opts; } if ((opts = strchr(format, ','))) { format_len = opts-format; } else { opts=""; format_len = strlen(format); } if (strncmp(format, "bam", format_len) == 0) { *cp++ = 'b'; } else if (strncmp(format, "cram", format_len) == 0) { *cp++ = 'c'; } else if (strncmp(format, "cram2", format_len) == 0) { *cp++ = 'c'; strcpy(cp, ",VERSION=2.1"); cp += 12; } else if (strncmp(format, "cram3", format_len) == 0) { *cp++ = 'c'; strcpy(cp, ",VERSION=3.0"); cp += 12; } else if (strncmp(format, "sam", format_len) == 0) { ; // format mode="" } else { free(mode_opts); return NULL; } strcpy(cp, opts); return mode_opts; } #define STRNCMP(a,b,n) (strncasecmp((a),(b),(n)) || strlen(a)!=(n)) int bam_str2flag(const char *str) { char *end, *beg = (char*) str; long int flag = strtol(str, &end, 0); if ( end!=str ) return flag; // the conversion was successful flag = 0; while ( *str ) { end = beg; while ( *end && *end!=',' ) end++; if ( !STRNCMP("PAIRED",beg,end-beg) ) flag |= BAM_FPAIRED; else if ( !STRNCMP("PROPER_PAIR",beg,end-beg) ) flag |= BAM_FPROPER_PAIR; else if ( !STRNCMP("UNMAP",beg,end-beg) ) flag |= BAM_FUNMAP; else if ( !STRNCMP("MUNMAP",beg,end-beg) ) flag |= BAM_FMUNMAP; else if ( !STRNCMP("REVERSE",beg,end-beg) ) flag |= BAM_FREVERSE; else if ( !STRNCMP("MREVERSE",beg,end-beg) ) flag |= BAM_FMREVERSE; else if ( !STRNCMP("READ1",beg,end-beg) ) flag |= BAM_FREAD1; else if ( !STRNCMP("READ2",beg,end-beg) ) flag |= BAM_FREAD2; else if ( !STRNCMP("SECONDARY",beg,end-beg) ) flag |= BAM_FSECONDARY; else if ( !STRNCMP("QCFAIL",beg,end-beg) ) flag |= BAM_FQCFAIL; else if ( !STRNCMP("DUP",beg,end-beg) ) flag |= BAM_FDUP; else if ( !STRNCMP("SUPPLEMENTARY",beg,end-beg) ) flag |= BAM_FSUPPLEMENTARY; else return -1; if ( !*end ) break; beg = end + 1; } return flag; } char *bam_flag2str(int flag) { kstring_t str = {0,0,0}; if ( flag&BAM_FPAIRED ) ksprintf(&str,"%s%s", str.l?",":"","PAIRED"); if ( flag&BAM_FPROPER_PAIR ) ksprintf(&str,"%s%s", str.l?",":"","PROPER_PAIR"); if ( flag&BAM_FUNMAP ) ksprintf(&str,"%s%s", str.l?",":"","UNMAP"); if ( flag&BAM_FMUNMAP ) ksprintf(&str,"%s%s", str.l?",":"","MUNMAP"); if ( flag&BAM_FREVERSE ) ksprintf(&str,"%s%s", str.l?",":"","REVERSE"); if ( flag&BAM_FMREVERSE ) ksprintf(&str,"%s%s", str.l?",":"","MREVERSE"); if ( flag&BAM_FREAD1 ) ksprintf(&str,"%s%s", str.l?",":"","READ1"); if ( flag&BAM_FREAD2 ) ksprintf(&str,"%s%s", str.l?",":"","READ2"); if ( flag&BAM_FSECONDARY ) ksprintf(&str,"%s%s", str.l?",":"","SECONDARY"); if ( flag&BAM_FQCFAIL ) ksprintf(&str,"%s%s", str.l?",":"","QCFAIL"); if ( flag&BAM_FDUP ) ksprintf(&str,"%s%s", str.l?",":"","DUP"); if ( flag&BAM_FSUPPLEMENTARY ) ksprintf(&str,"%s%s", str.l?",":"","SUPPLEMENTARY"); if ( str.l == 0 ) kputsn("", 0, &str); return str.s; } /************************** *** Pileup and Mpileup *** **************************/ #if !defined(BAM_NO_PILEUP) #include /******************* *** Memory pool *** *******************/ typedef struct { int k, x, y, end; } cstate_t; static cstate_t g_cstate_null = { -1, 0, 0, 0 }; typedef struct __linkbuf_t { bam1_t b; int32_t beg, end; cstate_t s; struct __linkbuf_t *next; } lbnode_t; typedef struct { int cnt, n, max; lbnode_t **buf; } mempool_t; static mempool_t *mp_init(void) { mempool_t *mp; mp = (mempool_t*)calloc(1, sizeof(mempool_t)); return mp; } static void mp_destroy(mempool_t *mp) { int k; for (k = 0; k < mp->n; ++k) { free(mp->buf[k]->b.data); free(mp->buf[k]); } free(mp->buf); free(mp); } static inline lbnode_t *mp_alloc(mempool_t *mp) { ++mp->cnt; if (mp->n == 0) return (lbnode_t*)calloc(1, sizeof(lbnode_t)); else return mp->buf[--mp->n]; } static inline void mp_free(mempool_t *mp, lbnode_t *p) { --mp->cnt; p->next = 0; // clear lbnode_t::next here if (mp->n == mp->max) { mp->max = mp->max? mp->max<<1 : 256; mp->buf = (lbnode_t**)realloc(mp->buf, sizeof(lbnode_t*) * mp->max); } mp->buf[mp->n++] = p; } /********************** *** CIGAR resolver *** **********************/ /* s->k: the index of the CIGAR operator that has just been processed. s->x: the reference coordinate of the start of s->k s->y: the query coordiante of the start of s->k */ static inline int resolve_cigar2(bam_pileup1_t *p, int32_t pos, cstate_t *s) { #define _cop(c) ((c)&BAM_CIGAR_MASK) #define _cln(c) ((c)>>BAM_CIGAR_SHIFT) bam1_t *b = p->b; bam1_core_t *c = &b->core; uint32_t *cigar = bam_get_cigar(b); int k; // determine the current CIGAR operation // fprintf(stderr, "%s\tpos=%d\tend=%d\t(%d,%d,%d)\n", bam_get_qname(b), pos, s->end, s->k, s->x, s->y); if (s->k == -1) { // never processed if (c->n_cigar == 1) { // just one operation, save a loop if (_cop(cigar[0]) == BAM_CMATCH || _cop(cigar[0]) == BAM_CEQUAL || _cop(cigar[0]) == BAM_CDIFF) s->k = 0, s->x = c->pos, s->y = 0; } else { // find the first match or deletion for (k = 0, s->x = c->pos, s->y = 0; k < c->n_cigar; ++k) { int op = _cop(cigar[k]); int l = _cln(cigar[k]); if (op == BAM_CMATCH || op == BAM_CDEL || op == BAM_CEQUAL || op == BAM_CDIFF) break; else if (op == BAM_CREF_SKIP) s->x += l; else if (op == BAM_CINS || op == BAM_CSOFT_CLIP) s->y += l; } assert(k < c->n_cigar); s->k = k; } } else { // the read has been processed before int op, l = _cln(cigar[s->k]); if (pos - s->x >= l) { // jump to the next operation assert(s->k < c->n_cigar); // otherwise a bug: this function should not be called in this case op = _cop(cigar[s->k+1]); if (op == BAM_CMATCH || op == BAM_CDEL || op == BAM_CREF_SKIP || op == BAM_CEQUAL || op == BAM_CDIFF) { // jump to the next without a loop if (_cop(cigar[s->k]) == BAM_CMATCH|| _cop(cigar[s->k]) == BAM_CEQUAL || _cop(cigar[s->k]) == BAM_CDIFF) s->y += l; s->x += l; ++s->k; } else { // find the next M/D/N/=/X if (_cop(cigar[s->k]) == BAM_CMATCH|| _cop(cigar[s->k]) == BAM_CEQUAL || _cop(cigar[s->k]) == BAM_CDIFF) s->y += l; s->x += l; for (k = s->k + 1; k < c->n_cigar; ++k) { op = _cop(cigar[k]), l = _cln(cigar[k]); if (op == BAM_CMATCH || op == BAM_CDEL || op == BAM_CREF_SKIP || op == BAM_CEQUAL || op == BAM_CDIFF) break; else if (op == BAM_CINS || op == BAM_CSOFT_CLIP) s->y += l; } s->k = k; } assert(s->k < c->n_cigar); // otherwise a bug } // else, do nothing } { // collect pileup information int op, l; op = _cop(cigar[s->k]); l = _cln(cigar[s->k]); p->is_del = p->indel = p->is_refskip = 0; if (s->x + l - 1 == pos && s->k + 1 < c->n_cigar) { // peek the next operation int op2 = _cop(cigar[s->k+1]); int l2 = _cln(cigar[s->k+1]); if (op2 == BAM_CDEL) p->indel = -(int)l2; else if (op2 == BAM_CINS) p->indel = l2; else if (op2 == BAM_CPAD && s->k + 2 < c->n_cigar) { // no working for adjacent padding int l3 = 0; for (k = s->k + 2; k < c->n_cigar; ++k) { op2 = _cop(cigar[k]); l2 = _cln(cigar[k]); if (op2 == BAM_CINS) l3 += l2; else if (op2 == BAM_CDEL || op2 == BAM_CMATCH || op2 == BAM_CREF_SKIP || op2 == BAM_CEQUAL || op2 == BAM_CDIFF) break; } if (l3 > 0) p->indel = l3; } } if (op == BAM_CMATCH || op == BAM_CEQUAL || op == BAM_CDIFF) { p->qpos = s->y + (pos - s->x); } else if (op == BAM_CDEL || op == BAM_CREF_SKIP) { p->is_del = 1; p->qpos = s->y; // FIXME: distinguish D and N!!!!! p->is_refskip = (op == BAM_CREF_SKIP); } // cannot be other operations; otherwise a bug p->is_head = (pos == c->pos); p->is_tail = (pos == s->end); } return 1; } /*********************** *** Pileup iterator *** ***********************/ // Dictionary of overlapping reads KHASH_MAP_INIT_STR(olap_hash, lbnode_t *) typedef khash_t(olap_hash) olap_hash_t; struct __bam_plp_t { mempool_t *mp; lbnode_t *head, *tail; int32_t tid, pos, max_tid, max_pos; int is_eof, max_plp, error, maxcnt; uint64_t id; bam_pileup1_t *plp; // for the "auto" interface only bam1_t *b; bam_plp_auto_f func; void *data; olap_hash_t *overlaps; }; bam_plp_t bam_plp_init(bam_plp_auto_f func, void *data) { bam_plp_t iter; iter = (bam_plp_t)calloc(1, sizeof(struct __bam_plp_t)); iter->mp = mp_init(); iter->head = iter->tail = mp_alloc(iter->mp); iter->max_tid = iter->max_pos = -1; iter->maxcnt = 8000; if (func) { iter->func = func; iter->data = data; iter->b = bam_init1(); } return iter; } void bam_plp_init_overlaps(bam_plp_t iter) { iter->overlaps = kh_init(olap_hash); // hash for tweaking quality of bases in overlapping reads } void bam_plp_destroy(bam_plp_t iter) { lbnode_t *p, *pnext; if ( iter->overlaps ) kh_destroy(olap_hash, iter->overlaps); for (p = iter->head; p != NULL; p = pnext) { pnext = p->next; mp_free(iter->mp, p); } mp_destroy(iter->mp); if (iter->b) bam_destroy1(iter->b); free(iter->plp); free(iter); } //--------------------------------- //--- Tweak overlapping reads //--------------------------------- /** * cigar_iref2iseq_set() - find the first CMATCH setting the ref and the read index * cigar_iref2iseq_next() - get the next CMATCH base * @cigar: pointer to current cigar block (rw) * @cigar_max: pointer just beyond the last cigar block * @icig: position within the current cigar block (rw) * @iseq: position in the sequence (rw) * @iref: position with respect to the beginning of the read (iref_pos - b->core.pos) (rw) * * Returns BAM_CMATCH or -1 when there is no more cigar to process or the requested position is not covered. */ static inline int cigar_iref2iseq_set(uint32_t **cigar, uint32_t *cigar_max, int *icig, int *iseq, int *iref) { int pos = *iref; if ( pos < 0 ) return -1; *icig = 0; *iseq = 0; *iref = 0; while ( *cigar> BAM_CIGAR_SHIFT; if ( cig==BAM_CSOFT_CLIP ) { (*cigar)++; *iseq += ncig; *icig = 0; continue; } if ( cig==BAM_CHARD_CLIP || cig==BAM_CPAD ) { (*cigar)++; *icig = 0; continue; } if ( cig==BAM_CMATCH || cig==BAM_CEQUAL || cig==BAM_CDIFF ) { pos -= ncig; if ( pos < 0 ) { *icig = ncig + pos; *iseq += *icig; *iref += *icig; return BAM_CMATCH; } (*cigar)++; *iseq += ncig; *icig = 0; *iref += ncig; continue; } if ( cig==BAM_CINS ) { (*cigar)++; *iseq += ncig; *icig = 0; continue; } if ( cig==BAM_CDEL || cig==BAM_CREF_SKIP ) { pos -= ncig; if ( pos<0 ) pos = 0; (*cigar)++; *icig = 0; *iref += ncig; continue; } fprintf(stderr,"todo: cigar %d\n", cig); assert(0); } *iseq = -1; return -1; } static inline int cigar_iref2iseq_next(uint32_t **cigar, uint32_t *cigar_max, int *icig, int *iseq, int *iref) { while ( *cigar < cigar_max ) { int cig = (**cigar) & BAM_CIGAR_MASK; int ncig = (**cigar) >> BAM_CIGAR_SHIFT; if ( cig==BAM_CMATCH || cig==BAM_CEQUAL || cig==BAM_CDIFF ) { if ( *icig >= ncig - 1 ) { *icig = 0; (*cigar)++; continue; } (*iseq)++; (*icig)++; (*iref)++; return BAM_CMATCH; } if ( cig==BAM_CDEL || cig==BAM_CREF_SKIP ) { (*cigar)++; (*iref) += ncig; *icig = 0; continue; } if ( cig==BAM_CINS ) { (*cigar)++; *iseq += ncig; *icig = 0; continue; } if ( cig==BAM_CSOFT_CLIP ) { (*cigar)++; *iseq += ncig; *icig = 0; continue; } if ( cig==BAM_CHARD_CLIP || cig==BAM_CPAD ) { (*cigar)++; *icig = 0; continue; } fprintf(stderr,"todo: cigar %d\n", cig); assert(0); } *iseq = -1; *iref = -1; return -1; } static void tweak_overlap_quality(bam1_t *a, bam1_t *b) { uint32_t *a_cigar = bam_get_cigar(a), *a_cigar_max = a_cigar + a->core.n_cigar; uint32_t *b_cigar = bam_get_cigar(b), *b_cigar_max = b_cigar + b->core.n_cigar; int a_icig = 0, a_iseq = 0; int b_icig = 0, b_iseq = 0; uint8_t *a_qual = bam_get_qual(a), *b_qual = bam_get_qual(b); uint8_t *a_seq = bam_get_seq(a), *b_seq = bam_get_seq(b); int iref = b->core.pos; int a_iref = iref - a->core.pos; int b_iref = iref - b->core.pos; int a_ret = cigar_iref2iseq_set(&a_cigar, a_cigar_max, &a_icig, &a_iseq, &a_iref); if ( a_ret<0 ) return; // no overlap int b_ret = cigar_iref2iseq_set(&b_cigar, b_cigar_max, &b_icig, &b_iseq, &b_iref); if ( b_ret<0 ) return; // no overlap #if DBG fprintf(stderr,"tweak %s n_cigar=%d %d .. %d-%d vs %d-%d\n", bam_get_qname(a), a->core.n_cigar, b->core.n_cigar, a->core.pos+1,a->core.pos+bam_cigar2rlen(a->core.n_cigar,bam_get_cigar(a)), b->core.pos+1, b->core.pos+bam_cigar2rlen(b->core.n_cigar,bam_get_cigar(b))); #endif while ( 1 ) { // Increment reference position while ( a_iref>=0 && a_iref < iref - a->core.pos ) a_ret = cigar_iref2iseq_next(&a_cigar, a_cigar_max, &a_icig, &a_iseq, &a_iref); if ( a_ret<0 ) break; // done if ( iref < a_iref + a->core.pos ) iref = a_iref + a->core.pos; while ( b_iref>=0 && b_iref < iref - b->core.pos ) b_ret = cigar_iref2iseq_next(&b_cigar, b_cigar_max, &b_icig, &b_iseq, &b_iref); if ( b_ret<0 ) break; // done if ( iref < b_iref + b->core.pos ) iref = b_iref + b->core.pos; iref++; if ( a_iref+a->core.pos != b_iref+b->core.pos ) continue; // only CMATCH positions, don't know what to do with indels if ( bam_seqi(a_seq,a_iseq) == bam_seqi(b_seq,b_iseq) ) { #if DBG fprintf(stderr,"%c",seq_nt16_str[bam_seqi(a_seq,a_iseq)]); #endif // we are very confident about this base int qual = a_qual[a_iseq] + b_qual[b_iseq]; a_qual[a_iseq] = qual>200 ? 200 : qual; b_qual[b_iseq] = 0; } else { if ( a_qual[a_iseq] >= b_qual[b_iseq] ) { #if DBG fprintf(stderr,"[%c/%c]",seq_nt16_str[bam_seqi(a_seq,a_iseq)],tolower(seq_nt16_str[bam_seqi(b_seq,b_iseq)])); #endif a_qual[a_iseq] = 0.8 * a_qual[a_iseq]; // not so confident about a_qual anymore given the mismatch b_qual[b_iseq] = 0; } else { #if DBG fprintf(stderr,"[%c/%c]",tolower(seq_nt16_str[bam_seqi(a_seq,a_iseq)]),seq_nt16_str[bam_seqi(b_seq,b_iseq)]); #endif b_qual[b_iseq] = 0.8 * b_qual[b_iseq]; a_qual[a_iseq] = 0; } } } #if DBG fprintf(stderr,"\n"); #endif } // Fix overlapping reads. Simple soft-clipping did not give good results. // Lowering qualities of unwanted bases is more selective and works better. // static void overlap_push(bam_plp_t iter, lbnode_t *node) { if ( !iter->overlaps ) return; // mapped mates and paired reads only if ( node->b.core.flag&BAM_FMUNMAP || !(node->b.core.flag&BAM_FPROPER_PAIR) ) return; // no overlap possible, unless some wild cigar if ( abs(node->b.core.isize) >= 2*node->b.core.l_qseq ) return; khiter_t kitr = kh_get(olap_hash, iter->overlaps, bam_get_qname(&node->b)); if ( kitr==kh_end(iter->overlaps) ) { int ret; kitr = kh_put(olap_hash, iter->overlaps, bam_get_qname(&node->b), &ret); kh_value(iter->overlaps, kitr) = node; } else { lbnode_t *a = kh_value(iter->overlaps, kitr); tweak_overlap_quality(&a->b, &node->b); kh_del(olap_hash, iter->overlaps, kitr); assert(a->end-1 == a->s.end); a->end = bam_endpos(&a->b); a->s.end = a->end - 1; } } static void overlap_remove(bam_plp_t iter, const bam1_t *b) { if ( !iter->overlaps ) return; khiter_t kitr; if ( b ) { kitr = kh_get(olap_hash, iter->overlaps, bam_get_qname(b)); if ( kitr!=kh_end(iter->overlaps) ) kh_del(olap_hash, iter->overlaps, kitr); } else { // remove all for (kitr = kh_begin(iter->overlaps); kitroverlaps); kitr++) if ( kh_exist(iter->overlaps, kitr) ) kh_del(olap_hash, iter->overlaps, kitr); } } // Prepares next pileup position in bam records collected by bam_plp_auto -> user func -> bam_plp_push. Returns // pointer to the piled records if next position is ready or NULL if there is not enough records in the // buffer yet (the current position is still the maximum position across all buffered reads). const bam_pileup1_t *bam_plp_next(bam_plp_t iter, int *_tid, int *_pos, int *_n_plp) { if (iter->error) { *_n_plp = -1; return 0; } *_n_plp = 0; if (iter->is_eof && iter->head == iter->tail) return 0; while (iter->is_eof || iter->max_tid > iter->tid || (iter->max_tid == iter->tid && iter->max_pos > iter->pos)) { int n_plp = 0; // write iter->plp at iter->pos lbnode_t **pptr = &iter->head; while (*pptr != iter->tail) { lbnode_t *p = *pptr; if (p->b.core.tid < iter->tid || (p->b.core.tid == iter->tid && p->end <= iter->pos)) { // then remove overlap_remove(iter, &p->b); *pptr = p->next; mp_free(iter->mp, p); } else { if (p->b.core.tid == iter->tid && p->beg <= iter->pos) { // here: p->end > pos; then add to pileup if (n_plp == iter->max_plp) { // then double the capacity iter->max_plp = iter->max_plp? iter->max_plp<<1 : 256; iter->plp = (bam_pileup1_t*)realloc(iter->plp, sizeof(bam_pileup1_t) * iter->max_plp); } iter->plp[n_plp].b = &p->b; if (resolve_cigar2(iter->plp + n_plp, iter->pos, &p->s)) ++n_plp; // actually always true... } pptr = &(*pptr)->next; } } *_n_plp = n_plp; *_tid = iter->tid; *_pos = iter->pos; // update iter->tid and iter->pos if (iter->head != iter->tail) { if (iter->tid > iter->head->b.core.tid) { fprintf(stderr, "[%s] unsorted input. Pileup aborts.\n", __func__); iter->error = 1; *_n_plp = -1; return 0; } } if (iter->tid < iter->head->b.core.tid) { // come to a new reference sequence iter->tid = iter->head->b.core.tid; iter->pos = iter->head->beg; // jump to the next reference } else if (iter->pos < iter->head->beg) { // here: tid == head->b.core.tid iter->pos = iter->head->beg; // jump to the next position } else ++iter->pos; // scan contiguously // return if (n_plp) return iter->plp; if (iter->is_eof && iter->head == iter->tail) break; } return 0; } int bam_plp_push(bam_plp_t iter, const bam1_t *b) { if (iter->error) return -1; if (b) { if (b->core.tid < 0) { overlap_remove(iter, b); return 0; } // Skip only unmapped reads here, any additional filtering must be done in iter->func if (b->core.flag & BAM_FUNMAP) { overlap_remove(iter, b); return 0; } if (iter->tid == b->core.tid && iter->pos == b->core.pos && iter->mp->cnt > iter->maxcnt) { overlap_remove(iter, b); return 0; } bam_copy1(&iter->tail->b, b); overlap_push(iter, iter->tail); #ifndef BAM_NO_ID iter->tail->b.id = iter->id++; #endif iter->tail->beg = b->core.pos; iter->tail->end = bam_endpos(b); iter->tail->s = g_cstate_null; iter->tail->s.end = iter->tail->end - 1; // initialize cstate_t if (b->core.tid < iter->max_tid) { fprintf(stderr, "[bam_pileup_core] the input is not sorted (chromosomes out of order)\n"); iter->error = 1; return -1; } if ((b->core.tid == iter->max_tid) && (iter->tail->beg < iter->max_pos)) { fprintf(stderr, "[bam_pileup_core] the input is not sorted (reads out of order)\n"); iter->error = 1; return -1; } iter->max_tid = b->core.tid; iter->max_pos = iter->tail->beg; if (iter->tail->end > iter->pos || iter->tail->b.core.tid > iter->tid) { iter->tail->next = mp_alloc(iter->mp); iter->tail = iter->tail->next; } } else iter->is_eof = 1; return 0; } const bam_pileup1_t *bam_plp_auto(bam_plp_t iter, int *_tid, int *_pos, int *_n_plp) { const bam_pileup1_t *plp; if (iter->func == 0 || iter->error) { *_n_plp = -1; return 0; } if ((plp = bam_plp_next(iter, _tid, _pos, _n_plp)) != 0) return plp; else { // no pileup line can be obtained; read alignments *_n_plp = 0; if (iter->is_eof) return 0; int ret; while ( (ret=iter->func(iter->data, iter->b)) >= 0) { if (bam_plp_push(iter, iter->b) < 0) { *_n_plp = -1; return 0; } if ((plp = bam_plp_next(iter, _tid, _pos, _n_plp)) != 0) return plp; // otherwise no pileup line can be returned; read the next alignment. } if ( ret < -1 ) { iter->error = ret; *_n_plp = -1; return 0; } bam_plp_push(iter, 0); if ((plp = bam_plp_next(iter, _tid, _pos, _n_plp)) != 0) return plp; return 0; } } void bam_plp_reset(bam_plp_t iter) { overlap_remove(iter, NULL); iter->max_tid = iter->max_pos = -1; iter->tid = iter->pos = 0; iter->is_eof = 0; while (iter->head != iter->tail) { lbnode_t *p = iter->head; iter->head = p->next; mp_free(iter->mp, p); } } void bam_plp_set_maxcnt(bam_plp_t iter, int maxcnt) { iter->maxcnt = maxcnt; } /************************ *** Mpileup iterator *** ************************/ struct __bam_mplp_t { int n; uint64_t min, *pos; bam_plp_t *iter; int *n_plp; const bam_pileup1_t **plp; }; bam_mplp_t bam_mplp_init(int n, bam_plp_auto_f func, void **data) { int i; bam_mplp_t iter; iter = (bam_mplp_t)calloc(1, sizeof(struct __bam_mplp_t)); iter->pos = (uint64_t*)calloc(n, sizeof(uint64_t)); iter->n_plp = (int*)calloc(n, sizeof(int)); iter->plp = (const bam_pileup1_t**)calloc(n, sizeof(bam_pileup1_t*)); iter->iter = (bam_plp_t*)calloc(n, sizeof(bam_plp_t)); iter->n = n; iter->min = (uint64_t)-1; for (i = 0; i < n; ++i) { iter->iter[i] = bam_plp_init(func, data[i]); iter->pos[i] = iter->min; } return iter; } void bam_mplp_init_overlaps(bam_mplp_t iter) { int i; for (i = 0; i < iter->n; ++i) bam_plp_init_overlaps(iter->iter[i]); } void bam_mplp_set_maxcnt(bam_mplp_t iter, int maxcnt) { int i; for (i = 0; i < iter->n; ++i) iter->iter[i]->maxcnt = maxcnt; } void bam_mplp_destroy(bam_mplp_t iter) { int i; for (i = 0; i < iter->n; ++i) bam_plp_destroy(iter->iter[i]); free(iter->iter); free(iter->pos); free(iter->n_plp); free(iter->plp); free(iter); } int bam_mplp_auto(bam_mplp_t iter, int *_tid, int *_pos, int *n_plp, const bam_pileup1_t **plp) { int i, ret = 0; uint64_t new_min = (uint64_t)-1; for (i = 0; i < iter->n; ++i) { if (iter->pos[i] == iter->min) { int tid, pos; iter->plp[i] = bam_plp_auto(iter->iter[i], &tid, &pos, &iter->n_plp[i]); if ( iter->iter[i]->error ) return -1; iter->pos[i] = iter->plp[i] ? (uint64_t)tid<<32 | pos : 0; } if (iter->plp[i] && iter->pos[i] < new_min) new_min = iter->pos[i]; } iter->min = new_min; if (new_min == (uint64_t)-1) return 0; *_tid = new_min>>32; *_pos = (uint32_t)new_min; for (i = 0; i < iter->n; ++i) { if (iter->pos[i] == iter->min) { // FIXME: valgrind reports "uninitialised value(s) at this line" n_plp[i] = iter->n_plp[i], plp[i] = iter->plp[i]; ++ret; } else n_plp[i] = 0, plp[i] = 0; } return ret; } #endif // ~!defined(BAM_NO_PILEUP) stacks-1.44/htslib/._synced_bcf_reader.c000644 000765 000024 00000000265 12750743246 021006 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/synced_bcf_reader.c000644 000765 000024 00000122076 12750743246 020576 0ustar00catchenstaff000000 000000 /* synced_bcf_reader.c -- stream through multiple VCF files. Copyright (C) 2012-2014 Genome Research Ltd. Author: Petr Danecek Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include #include #include #include #include #include "htslib/synced_bcf_reader.h" #include "htslib/kseq.h" #include "htslib/khash_str2int.h" #include "htslib/bgzf.h" #define MAX_CSI_COOR 0x7fffffff // maximum indexable coordinate of .csi typedef struct { uint32_t start, end; } region1_t; typedef struct _region_t { region1_t *regs; int nregs, mregs, creg; } region_t; static void _regions_add(bcf_sr_regions_t *reg, const char *chr, int start, int end); static bcf_sr_regions_t *_regions_init_string(const char *str); static int _regions_match_alleles(bcf_sr_regions_t *reg, int als_idx, bcf1_t *rec); char *bcf_sr_strerror(int errnum) { switch (errnum) { case open_failed: return strerror(errno); break; case not_bgzf: return "not compressed with bgzip"; break; case idx_load_failed: return "could not load index"; break; case file_type_error: return "unknown file type"; break; case api_usage_error: return "API usage error"; break; case header_error: return "could not parse header"; break; case no_eof: return "no BGZF EOF marker; file may be truncated"; break; default: return ""; } } static int *init_filters(bcf_hdr_t *hdr, const char *filters, int *nfilters) { kstring_t str = {0,0,0}; const char *tmp = filters, *prev = filters; int nout = 0, *out = NULL; while ( 1 ) { if ( *tmp==',' || !*tmp ) { out = (int*) realloc(out, (nout+1)*sizeof(int)); if ( tmp-prev==1 && *prev=='.' ) { out[nout] = -1; nout++; } else { str.l = 0; kputsn(prev, tmp-prev, &str); out[nout] = bcf_hdr_id2int(hdr, BCF_DT_ID, str.s); if ( out[nout]>=0 ) nout++; } if ( !*tmp ) break; prev = tmp+1; } tmp++; } if ( str.m ) free(str.s); *nfilters = nout; return out; } int bcf_sr_set_regions(bcf_srs_t *readers, const char *regions, int is_file) { assert( !readers->regions ); if ( readers->nreaders ) { fprintf(stderr,"[%s:%d %s] Error: bcf_sr_set_regions() must be called before bcf_sr_add_reader()\n", __FILE__,__LINE__,__FUNCTION__); return -1; } readers->regions = bcf_sr_regions_init(regions,is_file,0,1,-2); if ( !readers->regions ) return -1; readers->explicit_regs = 1; readers->require_index = 1; return 0; } int bcf_sr_set_targets(bcf_srs_t *readers, const char *targets, int is_file, int alleles) { assert( !readers->targets ); if ( targets[0]=='^' ) { readers->targets_exclude = 1; targets++; } readers->targets = bcf_sr_regions_init(targets,is_file,0,1,-2); if ( !readers->targets ) return -1; readers->targets_als = alleles; return 0; } int bcf_sr_add_reader(bcf_srs_t *files, const char *fname) { htsFile* file_ptr = hts_open(fname, "r"); if ( ! file_ptr ) { files->errnum = open_failed; return 0; } files->has_line = (int*) realloc(files->has_line, sizeof(int)*(files->nreaders+1)); files->has_line[files->nreaders] = 0; files->readers = (bcf_sr_t*) realloc(files->readers, sizeof(bcf_sr_t)*(files->nreaders+1)); bcf_sr_t *reader = &files->readers[files->nreaders++]; memset(reader,0,sizeof(bcf_sr_t)); reader->file = file_ptr; files->errnum = 0; if ( reader->file->format.compression==bgzf ) { BGZF *bgzf = hts_get_bgzfp(reader->file); if ( bgzf && bgzf_check_EOF(bgzf) == 0 ) { files->errnum = no_eof; fprintf(stderr,"[%s] Warning: no BGZF EOF marker; file may be truncated.\n", fname); } } if ( files->require_index ) { if ( reader->file->format.format==vcf ) { if ( reader->file->format.compression!=bgzf ) { files->errnum = not_bgzf; return 0; } reader->tbx_idx = tbx_index_load(fname); if ( !reader->tbx_idx ) { files->errnum = idx_load_failed; return 0; } reader->header = bcf_hdr_read(reader->file); } else if ( reader->file->format.format==bcf ) { if ( reader->file->format.compression!=bgzf ) { files->errnum = not_bgzf; return 0; } reader->header = bcf_hdr_read(reader->file); reader->bcf_idx = bcf_index_load(fname); if ( !reader->bcf_idx ) { files->errnum = idx_load_failed; return 0; } } else { files->errnum = file_type_error; return 0; } } else { if ( reader->file->format.format==bcf || reader->file->format.format==vcf ) { reader->header = bcf_hdr_read(reader->file); } else { files->errnum = file_type_error; return 0; } files->streaming = 1; } if ( files->streaming && files->nreaders>1 ) { files->errnum = api_usage_error; fprintf(stderr,"[%s:%d %s] Error: %d readers, yet require_index not set\n", __FILE__,__LINE__,__FUNCTION__,files->nreaders); return 0; } if ( files->streaming && files->regions ) { files->errnum = api_usage_error; fprintf(stderr,"[%s:%d %s] Error: cannot tabix-jump in streaming mode\n", __FILE__,__LINE__,__FUNCTION__); return 0; } if ( !reader->header ) { files->errnum = header_error; return 0; } reader->fname = strdup(fname); if ( files->apply_filters ) reader->filter_ids = init_filters(reader->header, files->apply_filters, &reader->nfilter_ids); // Update list of chromosomes if ( !files->explicit_regs && !files->streaming ) { int n,i; const char **names = reader->tbx_idx ? tbx_seqnames(reader->tbx_idx, &n) : bcf_hdr_seqnames(reader->header, &n); for (i=0; iregions ) files->regions = _regions_init_string(names[i]); else _regions_add(files->regions, names[i], -1, -1); } free(names); } return 1; } bcf_srs_t *bcf_sr_init(void) { bcf_srs_t *files = (bcf_srs_t*) calloc(1,sizeof(bcf_srs_t)); return files; } static void bcf_sr_destroy1(bcf_sr_t *reader) { free(reader->fname); if ( reader->tbx_idx ) tbx_destroy(reader->tbx_idx); if ( reader->bcf_idx ) hts_idx_destroy(reader->bcf_idx); bcf_hdr_destroy(reader->header); hts_close(reader->file); if ( reader->itr ) tbx_itr_destroy(reader->itr); int j; for (j=0; jmbuffer; j++) bcf_destroy1(reader->buffer[j]); free(reader->buffer); free(reader->samples); free(reader->filter_ids); } void bcf_sr_destroy(bcf_srs_t *files) { int i; for (i=0; inreaders; i++) bcf_sr_destroy1(&files->readers[i]); free(files->has_line); free(files->readers); for (i=0; in_smpl; i++) free(files->samples[i]); free(files->samples); if (files->targets) bcf_sr_regions_destroy(files->targets); if (files->regions) bcf_sr_regions_destroy(files->regions); if ( files->tmps.m ) free(files->tmps.s); free(files); } void bcf_sr_remove_reader(bcf_srs_t *files, int i) { assert( !files->samples ); // not ready for this yet bcf_sr_destroy1(&files->readers[i]); if ( i+1 < files->nreaders ) { memmove(&files->readers[i], &files->readers[i+1], (files->nreaders-i-1)*sizeof(bcf_sr_t)); memmove(&files->has_line[i], &files->has_line[i+1], (files->nreaders-i-1)*sizeof(int)); } files->nreaders--; } /* Removes duplicate records from the buffer. The meaning of "duplicate" is controlled by the $collapse variable, which can cause that from multiple lines only the first is considered and the rest is ignored. The removal is done by setting the redundant lines' positions to -1 and moving these lines at the end of the buffer. */ static void collapse_buffer(bcf_srs_t *files, bcf_sr_t *reader) { int irec,jrec, has_snp=0, has_indel=0, has_any=0; for (irec=1; irec<=reader->nbuffer; irec++) { bcf1_t *line = reader->buffer[irec]; if ( line->pos != reader->buffer[1]->pos ) break; if ( files->collapse&COLLAPSE_ANY ) { if ( !has_any ) has_any = 1; else line->pos = -1; } int line_type = bcf_get_variant_types(line); if ( files->collapse&COLLAPSE_SNPS && line_type&(VCF_SNP|VCF_MNP) ) { if ( !has_snp ) has_snp = 1; else line->pos = -1; } if ( files->collapse&COLLAPSE_INDELS && line_type&VCF_INDEL ) { if ( !has_indel ) has_indel = 1; else line->pos = -1; } } bcf1_t *tmp; irec = jrec = 1; while ( irec<=reader->nbuffer && jrec<=reader->nbuffer ) { if ( reader->buffer[irec]->pos != -1 ) { irec++; continue; } if ( jrec<=irec ) jrec = irec+1; while ( jrec<=reader->nbuffer && reader->buffer[jrec]->pos==-1 ) jrec++; if ( jrec<=reader->nbuffer ) { tmp = reader->buffer[irec]; reader->buffer[irec] = reader->buffer[jrec]; reader->buffer[jrec] = tmp; } } reader->nbuffer = irec - 1; } void debug_buffer(FILE *fp, bcf_sr_t *reader) { int j; for (j=0; j<=reader->nbuffer; j++) { bcf1_t *line = reader->buffer[j]; fprintf(fp,"\t%p\t%s%s\t%s:%d\t%s ", line,reader->fname,j==0?"*":" ",reader->header->id[BCF_DT_CTG][line->rid].key,line->pos+1,line->n_allele?line->d.allele[0]:""); int k; for (k=1; kn_allele; k++) fprintf(fp," %s", line->d.allele[k]); fprintf(fp,"\n"); } } void debug_buffers(FILE *fp, bcf_srs_t *files) { int i; for (i=0; inreaders; i++) { fprintf(fp, "has_line: %d\t%s\n", bcf_sr_has_line(files,i),files->readers[i].fname); debug_buffer(fp, &files->readers[i]); } fprintf(fp,"\n"); } static inline int has_filter(bcf_sr_t *reader, bcf1_t *line) { int i, j; if ( !line->d.n_flt ) { for (j=0; jnfilter_ids; j++) if ( reader->filter_ids[j]<0 ) return 1; return 0; } for (i=0; id.n_flt; i++) { for (j=0; jnfilter_ids; j++) if ( line->d.flt[i]==reader->filter_ids[j] ) return 1; } return 0; } static int _reader_seek(bcf_sr_t *reader, const char *seq, int start, int end) { if ( end>=MAX_CSI_COOR ) { fprintf(stderr,"The coordinate is out of csi index limit: %d\n", end+1); exit(1); } if ( reader->itr ) { hts_itr_destroy(reader->itr); reader->itr = NULL; } reader->nbuffer = 0; if ( reader->tbx_idx ) { int tid = tbx_name2id(reader->tbx_idx, seq); if ( tid==-1 ) return -1; // the sequence not present in this file reader->itr = tbx_itr_queryi(reader->tbx_idx,tid,start,end+1); } else { int tid = bcf_hdr_name2id(reader->header, seq); if ( tid==-1 ) return -1; // the sequence not present in this file reader->itr = bcf_itr_queryi(reader->bcf_idx,tid,start,end+1); } if ( !reader->itr ) fprintf(stderr,"Could not seek: %s:%d-%d\n",seq,start+1,end+1); assert(reader->itr); return 0; } /* * _readers_next_region() - jumps to next region if necessary * Returns 0 on success or -1 when there are no more regions left */ static int _readers_next_region(bcf_srs_t *files) { // Need to open new chromosome? Check number of lines in all readers' buffers int i, eos = 0; for (i=0; inreaders; i++) if ( !files->readers[i].itr && !files->readers[i].nbuffer ) eos++; if ( eos!=files->nreaders ) { // Some of the readers still has buffered lines return 0; } // No lines in the buffer, need to open new region or quit if ( bcf_sr_regions_next(files->regions)<0 ) return -1; for (i=0; inreaders; i++) _reader_seek(&files->readers[i],files->regions->seq_names[files->regions->iseq],files->regions->start,files->regions->end); return 0; } /* * _reader_fill_buffer() - buffers all records with the same coordinate */ static void _reader_fill_buffer(bcf_srs_t *files, bcf_sr_t *reader) { // Return if the buffer is full: the coordinate of the last buffered record differs if ( reader->nbuffer && reader->buffer[reader->nbuffer]->pos != reader->buffer[1]->pos ) return; // No iterator (sequence not present in this file) and not streaming if ( !reader->itr && !files->streaming ) return; // Fill the buffer with records starting at the same position int i, ret = 0; while (1) { if ( reader->nbuffer+1 >= reader->mbuffer ) { // Increase buffer size reader->mbuffer += 8; reader->buffer = (bcf1_t**) realloc(reader->buffer, sizeof(bcf1_t*)*reader->mbuffer); for (i=8; i>0; i--) // initialize { reader->buffer[reader->mbuffer-i] = bcf_init1(); reader->buffer[reader->mbuffer-i]->max_unpack = files->max_unpack; reader->buffer[reader->mbuffer-i]->pos = -1; // for rare cases when VCF starts from 1 } } if ( files->streaming ) { if ( reader->file->format.format==vcf ) { if ( (ret=hts_getline(reader->file, KS_SEP_LINE, &files->tmps)) < 0 ) break; // no more lines int ret = vcf_parse1(&files->tmps, reader->header, reader->buffer[reader->nbuffer+1]); if ( ret<0 ) break; } else if ( reader->file->format.format==bcf ) { if ( (ret=bcf_read1(reader->file, reader->header, reader->buffer[reader->nbuffer+1])) < 0 ) break; // no more lines } else { fprintf(stderr,"[%s:%d %s] fixme: not ready for this\n", __FILE__,__LINE__,__FUNCTION__); exit(1); } } else if ( reader->tbx_idx ) { if ( (ret=tbx_itr_next(reader->file, reader->tbx_idx, reader->itr, &files->tmps)) < 0 ) break; // no more lines vcf_parse1(&files->tmps, reader->header, reader->buffer[reader->nbuffer+1]); } else { if ( (ret=bcf_itr_next(reader->file, reader->itr, reader->buffer[reader->nbuffer+1])) < 0 ) break; // no more lines bcf_subset_format(reader->header,reader->buffer[reader->nbuffer+1]); } // apply filter if ( !reader->nfilter_ids ) bcf_unpack(reader->buffer[reader->nbuffer+1], BCF_UN_STR); else { bcf_unpack(reader->buffer[reader->nbuffer+1], BCF_UN_STR|BCF_UN_FLT); if ( !has_filter(reader, reader->buffer[reader->nbuffer+1]) ) continue; } reader->nbuffer++; if ( reader->buffer[reader->nbuffer]->pos != reader->buffer[1]->pos ) break; // the buffer is full } if ( ret<0 ) { // done for this region tbx_itr_destroy(reader->itr); reader->itr = NULL; } if ( files->collapse && reader->nbuffer>=2 && reader->buffer[1]->pos==reader->buffer[2]->pos ) collapse_buffer(files, reader); } /* * _readers_shift_buffer() - removes the first line and all subsequent lines with the same position */ static void _reader_shift_buffer(bcf_sr_t *reader) { int i; for (i=2; i<=reader->nbuffer; i++) if ( reader->buffer[i]->pos!=reader->buffer[1]->pos ) break; if ( i<=reader->nbuffer ) { // A record with a different position follows, swap it. Because of the reader's logic, // only one such line can be present. bcf1_t *tmp = reader->buffer[1]; reader->buffer[1] = reader->buffer[i]; reader->buffer[i] = tmp; reader->nbuffer = 1; } else reader->nbuffer = 0; // no other line } /* * _reader_match_alleles() - from multiple buffered lines selects the one which * corresponds best to the template line. The logic is controlled by COLLAPSE_* * Returns 0 on success or -1 when no good matching line is found. */ static int _reader_match_alleles(bcf_srs_t *files, bcf_sr_t *reader, bcf1_t *tmpl) { int i, irec = -1; // if no template given, use the first available record if ( !tmpl ) irec = 1; else { int tmpl_type = bcf_get_variant_types(tmpl); for (i=1; i<=reader->nbuffer; i++) { bcf1_t *line = reader->buffer[i]; if ( line->pos != reader->buffer[1]->pos ) break; // done with this reader // Easiest case: matching by position only if ( files->collapse&COLLAPSE_ANY ) { irec=i; break; } int line_type = bcf_get_variant_types(line); // No matter what the alleles are, as long as they are both SNPs if ( files->collapse&COLLAPSE_SNPS && tmpl_type&VCF_SNP && line_type&VCF_SNP ) { irec=i; break; } // ... or indels if ( files->collapse&COLLAPSE_INDELS && tmpl_type&VCF_INDEL && line_type&VCF_INDEL ) { irec=i; break; } // More thorough checking: REFs must match if ( tmpl->rlen != line->rlen ) continue; // different length if ( !tmpl->d.allele || !line->d.allele ) continue; // one of the lines is empty, someone is swapped buffered lines?! if ( strcmp(tmpl->d.allele[0], line->d.allele[0]) ) continue; // the strings do not match int ial,jal; if ( files->collapse==COLLAPSE_NONE ) { // Exact match, all alleles must be identical if ( tmpl->n_allele!=line->n_allele ) continue; // different number of alleles, skip int nmatch = 1; // REF has been already checked for (ial=1; ialn_allele; ial++) { for (jal=1; jaln_allele; jal++) if ( !strcmp(tmpl->d.allele[ial], line->d.allele[jal]) ) { nmatch++; break; } } if ( nmatch==tmpl->n_allele ) { irec=i; break; } // found: exact match continue; } if ( line->n_allele==1 && tmpl->n_allele==1 ) { irec=i; break; } // both sites are non-variant // COLLAPSE_SOME: at least some ALTs must match for (ial=1; ialn_allele; ial++) { for (jal=1; jaln_allele; jal++) if ( !strcmp(tmpl->d.allele[ial], line->d.allele[jal]) ) { irec=i; break; } if ( irec>=1 ) break; } if ( irec>=1 ) break; } if ( irec==-1 ) return -1; // no matching line was found } // Set the selected line (irec) as active: set it to buffer[0], move the remaining lines forward // and put the old bcf1_t record at the end. bcf1_t *tmp = reader->buffer[0]; reader->buffer[0] = reader->buffer[irec]; for (i=irec+1; i<=reader->nbuffer; i++) reader->buffer[i-1] = reader->buffer[i]; reader->buffer[ reader->nbuffer ] = tmp; reader->nbuffer--; return 0; } int _reader_next_line(bcf_srs_t *files) { int i, min_pos = INT_MAX; // Loop until next suitable line is found or all readers have finished while ( 1 ) { // Get all readers ready for the next region. if ( files->regions && _readers_next_region(files)<0 ) break; // Fill buffers const char *chr = NULL; for (i=0; inreaders; i++) { _reader_fill_buffer(files, &files->readers[i]); // Update the minimum coordinate if ( !files->readers[i].nbuffer ) continue; if ( min_pos > files->readers[i].buffer[1]->pos ) { min_pos = files->readers[i].buffer[1]->pos; chr = bcf_seqname(files->readers[i].header, files->readers[i].buffer[1]); } } if ( min_pos==INT_MAX ) { if ( !files->regions ) break; continue; } // Skip this position if not present in targets if ( files->targets ) { int ret = bcf_sr_regions_overlap(files->targets, chr, min_pos, min_pos); if ( (!files->targets_exclude && ret<0) || (files->targets_exclude && !ret) ) { // Remove all lines with this position from the buffer for (i=0; inreaders; i++) if ( files->readers[i].nbuffer && files->readers[i].buffer[1]->pos==min_pos ) _reader_shift_buffer(&files->readers[i]); min_pos = INT_MAX; continue; } } break; // done: min_pos is set } // There can be records with duplicate positions. Set the active line intelligently so that // the alleles match. int nret = 0; // number of readers sharing the position bcf1_t *first = NULL; // record which will be used for allele matching for (i=0; inreaders; i++) { files->has_line[i] = 0; // Skip readers with no records at this position if ( !files->readers[i].nbuffer || files->readers[i].buffer[1]->pos!=min_pos ) continue; // Until now buffer[0] of all reader was empty and the lines started at buffer[1]. // Now lines which are ready to be output will be moved to buffer[0]. if ( _reader_match_alleles(files, &files->readers[i], first) < 0 ) continue; if ( !first ) first = files->readers[i].buffer[0]; nret++; files->has_line[i] = 1; } return nret; } int bcf_sr_next_line(bcf_srs_t *files) { if ( !files->targets_als ) return _reader_next_line(files); while (1) { int i, ret = _reader_next_line(files); if ( !ret ) return ret; for (i=0; inreaders; i++) if ( files->has_line[i] ) break; if ( _regions_match_alleles(files->targets, files->targets_als-1, files->readers[i].buffer[0]) ) return ret; // Check if there are more duplicate lines in the buffers. If not, return this line as if it // matched the targets, even if there is a type mismatch for (i=0; inreaders; i++) { if ( !files->has_line[i] ) continue; if ( files->readers[i].nbuffer==0 || files->readers[i].buffer[1]->pos!=files->readers[i].buffer[0]->pos ) continue; break; } if ( i==files->nreaders ) return ret; // no more lines left, output even if target alleles are not of the same type } } static void bcf_sr_seek_start(bcf_srs_t *readers) { bcf_sr_regions_t *reg = readers->regions; int i; for (i=0; inseqs; i++) reg->regs[i].creg = -1; reg->iseq = 0; } int bcf_sr_seek(bcf_srs_t *readers, const char *seq, int pos) { if ( !readers->regions ) return 0; if ( !seq && !pos ) { // seek to start bcf_sr_seek_start(readers); return 0; } bcf_sr_regions_overlap(readers->regions, seq, pos, pos); int i, nret = 0; for (i=0; inreaders; i++) { nret += _reader_seek(&readers->readers[i],seq,pos,MAX_CSI_COOR-1); } return nret; } int bcf_sr_set_samples(bcf_srs_t *files, const char *fname, int is_file) { int i, j, nsmpl, free_smpl = 0; char **smpl = NULL; void *exclude = (fname[0]=='^') ? khash_str2int_init() : NULL; if ( exclude || strcmp("-",fname) ) // "-" stands for all samples { smpl = hts_readlist(fname, is_file, &nsmpl); if ( !smpl ) { fprintf(stderr,"Could not read the file: \"%s\"\n", fname); return 0; } if ( exclude ) { for (i=0; ireaders[0].header->samples; // intersection of all samples nsmpl = bcf_hdr_nsamples(files->readers[0].header); } files->samples = NULL; files->n_smpl = 0; for (i=0; inreaders; j++) { if ( bcf_hdr_id2int(files->readers[j].header, BCF_DT_SAMPLE, smpl[i])<0 ) break; n_isec++; } if ( n_isec!=files->nreaders ) { fprintf(stderr,"Warning: The sample \"%s\" was not found in %s, skipping\n", smpl[i], files->readers[n_isec].fname); continue; } files->samples = (char**) realloc(files->samples, (files->n_smpl+1)*sizeof(const char*)); files->samples[files->n_smpl++] = strdup(smpl[i]); } if ( exclude ) khash_str2int_destroy(exclude); if ( free_smpl ) { for (i=0; in_smpl ) { if ( files->nreaders>1 ) fprintf(stderr,"No samples in common.\n"); return 0; } for (i=0; inreaders; i++) { bcf_sr_t *reader = &files->readers[i]; reader->samples = (int*) malloc(sizeof(int)*files->n_smpl); reader->n_smpl = files->n_smpl; for (j=0; jn_smpl; j++) reader->samples[j] = bcf_hdr_id2int(reader->header, BCF_DT_SAMPLE, files->samples[j]); } return 1; } // Add a new region into a list sorted by start,end. On input the coordinates // are 1-based, stored 0-based, inclusive. static void _regions_add(bcf_sr_regions_t *reg, const char *chr, int start, int end) { if ( start==-1 && end==-1 ) { start = 0; end = MAX_CSI_COOR-1; } else { start--; end--; // store 0-based coordinates } if ( !reg->seq_hash ) reg->seq_hash = khash_str2int_init(); int iseq; if ( khash_str2int_get(reg->seq_hash, chr, &iseq)<0 ) { // the chromosome block does not exist iseq = reg->nseqs++; reg->seq_names = (char**) realloc(reg->seq_names,sizeof(char*)*reg->nseqs); reg->regs = (region_t*) realloc(reg->regs,sizeof(region_t)*reg->nseqs); memset(®->regs[reg->nseqs-1],0,sizeof(region_t)); reg->seq_names[iseq] = strdup(chr); reg->regs[iseq].creg = -1; khash_str2int_set(reg->seq_hash,reg->seq_names[iseq],iseq); } region_t *creg = ®->regs[iseq]; // the regions may not be sorted on input: binary search int i, min = 0, max = creg->nregs - 1; while ( min<=max ) { i = (max+min)/2; if ( start < creg->regs[i].start ) max = i - 1; else if ( start > creg->regs[i].start ) min = i + 1; else break; } if ( min>max || creg->regs[i].start!=start || creg->regs[i].end!=end ) { // no such region, insert a new one just after max hts_expand(region1_t,creg->nregs+1,creg->mregs,creg->regs); if ( ++max < creg->nregs ) memmove(&creg->regs[max+1],&creg->regs[max],(creg->nregs - max)*sizeof(region1_t)); creg->regs[max].start = start; creg->regs[max].end = end; creg->nregs++; } } // File name or a list of genomic locations. If file name, NULL is returned. static bcf_sr_regions_t *_regions_init_string(const char *str) { bcf_sr_regions_t *reg = (bcf_sr_regions_t *) calloc(1, sizeof(bcf_sr_regions_t)); reg->start = reg->end = -1; reg->prev_start = reg->prev_seq = -1; kstring_t tmp = {0,0,0}; const char *sp = str, *ep = str; int from, to; while ( 1 ) { while ( *ep && *ep!=',' && *ep!=':' ) ep++; tmp.l = 0; kputsn(sp,ep-sp,&tmp); if ( *ep==':' ) { sp = ep+1; from = hts_parse_decimal(sp,(char**)&ep,0); if ( sp==ep ) { fprintf(stderr,"[%s:%d %s] Could not parse the region(s): %s\n", __FILE__,__LINE__,__FUNCTION__,str); free(reg); free(tmp.s); return NULL; } if ( !*ep || *ep==',' ) { _regions_add(reg, tmp.s, from, from); sp = ep; continue; } if ( *ep!='-' ) { fprintf(stderr,"[%s:%d %s] Could not parse the region(s): %s\n", __FILE__,__LINE__,__FUNCTION__,str); free(reg); free(tmp.s); return NULL; } ep++; sp = ep; to = hts_parse_decimal(sp,(char**)&ep,0); if ( *ep && *ep!=',' ) { fprintf(stderr,"[%s:%d %s] Could not parse the region(s): %s\n", __FILE__,__LINE__,__FUNCTION__,str); free(reg); free(tmp.s); return NULL; } if ( sp==ep ) to = MAX_CSI_COOR-1; _regions_add(reg, tmp.s, from, to); if ( !*ep ) break; sp = ep; } else { if ( tmp.l ) _regions_add(reg, tmp.s, -1, -1); if ( !*ep ) break; sp = ++ep; } } free(tmp.s); return reg; } // ichr,ifrom,ito are 0-based; // returns -1 on error, 0 if the line is a comment line, 1 on success static int _regions_parse_line(char *line, int ichr,int ifrom,int ito, char **chr,char **chr_end,int *from,int *to) { *chr_end = NULL; if ( line[0]=='#' ) return 0; int k,l; // index of the start and end column of the tab-delimited file if ( ifrom <= ito ) k = ifrom, l = ito; else l = ifrom, k = ito; int i; char *se = line, *ss = NULL; // start and end char *tmp; for (i=0; i<=k && *se; i++) { ss = i==0 ? se++ : ++se; while (*se && *se!='\t') se++; } if ( i<=k ) return -1; if ( k==l ) { *from = *to = hts_parse_decimal(ss, &tmp, 0); if ( tmp==ss ) return -1; } else { if ( k==ifrom ) *from = hts_parse_decimal(ss, &tmp, 0); else *to = hts_parse_decimal(ss, &tmp, 0); if ( ss==tmp ) return -1; for (i=k; i0 ) ss = ++se; while (*se && *se!='\t') se++; } if ( i<=ichr ) return -1; *chr_end = se; *chr = ss; return 1; } bcf_sr_regions_t *bcf_sr_regions_init(const char *regions, int is_file, int ichr, int ifrom, int ito) { bcf_sr_regions_t *reg; if ( !is_file ) return _regions_init_string(regions); reg = (bcf_sr_regions_t *) calloc(1, sizeof(bcf_sr_regions_t)); reg->start = reg->end = -1; reg->prev_start = reg->prev_seq = -1; reg->file = hts_open(regions, "rb"); if ( !reg->file ) { fprintf(stderr,"[%s:%d %s] Could not open file: %s\n", __FILE__,__LINE__,__FUNCTION__,regions); free(reg); return NULL; } reg->tbx = tbx_index_load(regions); if ( !reg->tbx ) { int len = strlen(regions); int is_bed = strcasecmp(".bed",regions+len-4) ? 0 : 1; if ( !is_bed && !strcasecmp(".bed.gz",regions+len-7) ) is_bed = 1; if ( reg->file->format.format==vcf ) ito = 1; // read the whole file, tabix index is not present while ( hts_getline(reg->file, KS_SEP_LINE, ®->line) > 0 ) { char *chr, *chr_end; int from, to, ret; ret = _regions_parse_line(reg->line.s, ichr,ifrom,abs(ito), &chr,&chr_end,&from,&to); if ( ret < 0 ) { if ( ito<0 ) ret = _regions_parse_line(reg->line.s, ichr,ifrom,ifrom, &chr,&chr_end,&from,&to); if ( ret<0 ) { fprintf(stderr,"[%s:%d] Could not parse the file %s, using the columns %d,%d[,%d]\n", __FILE__,__LINE__,regions,ichr+1,ifrom+1,ito+1); hts_close(reg->file); reg->file = NULL; free(reg); return NULL; } } if ( !ret ) continue; if ( is_bed ) from++; *chr_end = 0; _regions_add(reg, chr, from, to); *chr_end = '\t'; } hts_close(reg->file); reg->file = NULL; if ( !reg->nseqs ) { free(reg); return NULL; } return reg; } reg->seq_names = (char**) tbx_seqnames(reg->tbx, ®->nseqs); if ( !reg->seq_hash ) reg->seq_hash = khash_str2int_init(); int i; for (i=0; inseqs; i++) { khash_str2int_set(reg->seq_hash,reg->seq_names[i],i); } reg->fname = strdup(regions); reg->is_bin = 1; return reg; } void bcf_sr_regions_destroy(bcf_sr_regions_t *reg) { int i; free(reg->fname); if ( reg->itr ) tbx_itr_destroy(reg->itr); if ( reg->tbx ) tbx_destroy(reg->tbx); if ( reg->file ) hts_close(reg->file); if ( reg->als ) free(reg->als); if ( reg->als_str.s ) free(reg->als_str.s); free(reg->line.s); if ( reg->regs ) { // free only in-memory names, tbx names are const for (i=0; inseqs; i++) { free(reg->seq_names[i]); free(reg->regs[i].regs); } } free(reg->regs); free(reg->seq_names); khash_str2int_destroy(reg->seq_hash); free(reg); } int bcf_sr_regions_seek(bcf_sr_regions_t *reg, const char *seq) { reg->iseq = reg->start = reg->end = -1; if ( khash_str2int_get(reg->seq_hash, seq, ®->iseq) < 0 ) return -1; // sequence seq not in regions // using in-memory regions if ( reg->regs ) { reg->regs[reg->iseq].creg = -1; return 0; } // reading regions from tabix if ( reg->itr ) tbx_itr_destroy(reg->itr); reg->itr = tbx_itr_querys(reg->tbx, seq); if ( reg->itr ) return 0; return -1; } int bcf_sr_regions_next(bcf_sr_regions_t *reg) { if ( reg->iseq<0 ) return -1; reg->start = reg->end = -1; reg->nals = 0; // using in-memory regions if ( reg->regs ) { while ( reg->iseq < reg->nseqs ) { reg->regs[reg->iseq].creg++; if ( reg->regs[reg->iseq].creg < reg->regs[reg->iseq].nregs ) break; reg->iseq++; } if ( reg->iseq >= reg->nseqs ) { reg->iseq = -1; return -1; } // no more regions left region1_t *creg = ®->regs[reg->iseq].regs[reg->regs[reg->iseq].creg]; reg->start = creg->start; reg->end = creg->end; return 0; } // reading from tabix char *chr, *chr_end; int ichr = 0, ifrom = 1, ito = 2, is_bed = 0, from, to; if ( reg->tbx ) { ichr = reg->tbx->conf.sc-1; ifrom = reg->tbx->conf.bc-1; ito = reg->tbx->conf.ec-1; if ( ito<0 ) ito = ifrom; is_bed = reg->tbx->conf.preset==TBX_UCSC ? 1 : 0; } int ret = 0; while ( !ret ) { if ( reg->itr ) { // tabix index present, reading a chromosome block ret = tbx_itr_next(reg->file, reg->tbx, reg->itr, ®->line); if ( ret<0 ) { reg->iseq = -1; return -1; } } else { if ( reg->is_bin ) { // Waited for seek which never came. Reopen in text mode and stream // through the regions, otherwise hts_getline would fail hts_close(reg->file); reg->file = hts_open(reg->fname, "r"); if ( !reg->file ) { fprintf(stderr,"[%s:%d %s] Could not open file: %s\n", __FILE__,__LINE__,__FUNCTION__,reg->fname); reg->file = NULL; bcf_sr_regions_destroy(reg); return -1; } reg->is_bin = 0; } // tabix index absent, reading the whole file ret = hts_getline(reg->file, KS_SEP_LINE, ®->line); if ( ret<0 ) { reg->iseq = -1; return -1; } } ret = _regions_parse_line(reg->line.s, ichr,ifrom,ito, &chr,&chr_end,&from,&to); if ( ret<0 ) { fprintf(stderr,"[%s:%d] Could not parse the file %s, using the columns %d,%d,%d\n", __FILE__,__LINE__,reg->fname,ichr+1,ifrom+1,ito+1); return -1; } } if ( is_bed ) from++; *chr_end = 0; if ( khash_str2int_get(reg->seq_hash, chr, ®->iseq)<0 ) { fprintf(stderr,"Broken tabix index? The sequence \"%s\" not in dictionary [%s]\n", chr,reg->line.s); exit(1); } *chr_end = '\t'; reg->start = from - 1; reg->end = to - 1; return 0; } static int _regions_match_alleles(bcf_sr_regions_t *reg, int als_idx, bcf1_t *rec) { if ( reg->regs ) { // payload is not supported for in-memory regions, switch to regidx instead in future fprintf(stderr,"Error: Compressed and indexed targets file is required\n"); exit(1); } int i = 0, max_len = 0; if ( !reg->nals ) { char *ss = reg->line.s; while ( inals = 1; while ( *se && *se!='\t' ) { if ( *se==',' ) reg->nals++; se++; } ks_resize(®->als_str, se-ss+1+reg->nals); reg->als_str.l = 0; hts_expand(char*,reg->nals,reg->mals,reg->als); reg->nals = 0; se = ss; while ( *(++se) ) { if ( *se=='\t' ) break; if ( *se!=',' ) continue; reg->als[reg->nals] = ®->als_str.s[reg->als_str.l]; kputsn(ss,se-ss,®->als_str); if ( ®->als_str.s[reg->als_str.l] - reg->als[reg->nals] > max_len ) max_len = ®->als_str.s[reg->als_str.l] - reg->als[reg->nals]; reg->als_str.l++; reg->nals++; ss = ++se; } reg->als[reg->nals] = ®->als_str.s[reg->als_str.l]; kputsn(ss,se-ss,®->als_str); if ( ®->als_str.s[reg->als_str.l] - reg->als[reg->nals] > max_len ) max_len = ®->als_str.s[reg->als_str.l] - reg->als[reg->nals]; reg->nals++; reg->als_type = max_len > 1 ? VCF_INDEL : VCF_SNP; // this is a simplified check, see vcf.c:bcf_set_variant_types } int type = bcf_get_variant_types(rec); if ( reg->als_type & VCF_INDEL ) return type & VCF_INDEL ? 1 : 0; return !(type & VCF_INDEL) ? 1 : 0; } int bcf_sr_regions_overlap(bcf_sr_regions_t *reg, const char *seq, int start, int end) { int iseq; if ( khash_str2int_get(reg->seq_hash, seq, &iseq)<0 ) return -1; // no such sequence if ( reg->prev_seq==-1 || iseq!=reg->prev_seq || reg->prev_start > start ) // new chromosome or after a seek { // flush regions left on previous chromosome if ( reg->missed_reg_handler && reg->prev_seq!=-1 && reg->iseq!=-1 ) bcf_sr_regions_flush(reg); bcf_sr_regions_seek(reg, seq); reg->start = reg->end = -1; } if ( reg->prev_seq==iseq && reg->iseq!=iseq ) return -2; // no more regions on this chromosome reg->prev_seq = reg->iseq; reg->prev_start = start; while ( iseq==reg->iseq && reg->end < start ) { if ( bcf_sr_regions_next(reg) < 0 ) return -2; // no more regions left if ( reg->iseq != iseq ) return -1; // does not overlap any regions if ( reg->missed_reg_handler && reg->end < start ) reg->missed_reg_handler(reg, reg->missed_reg_data); } if ( reg->start <= end ) return 0; // region overlap return -1; // no overlap } void bcf_sr_regions_flush(bcf_sr_regions_t *reg) { if ( !reg->missed_reg_handler || reg->prev_seq==-1 ) return; while ( !bcf_sr_regions_next(reg) ) reg->missed_reg_handler(reg, reg->missed_reg_data); return; } stacks-1.44/htslib/._tabix.c000644 000765 000024 00000000265 12750743246 016474 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/tabix.c000644 000765 000024 00000046703 12750743246 016266 0ustar00catchenstaff000000 000000 /* tabix.c -- Generic indexer for TAB-delimited genome position files. Copyright (C) 2009-2011 Broad Institute. Copyright (C) 2010-2012, 2014-2016 Genome Research Ltd. Author: Heng Li Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include #include #include #include #include #include #include "htslib/tbx.h" #include "htslib/sam.h" #include "htslib/vcf.h" #include "htslib/kseq.h" #include "htslib/bgzf.h" #include "htslib/hts.h" #include "htslib/regidx.h" typedef struct { char *regions_fname, *targets_fname; int print_header, header_only; } args_t; static void error(const char *format, ...) { va_list ap; va_start(ap, format); vfprintf(stderr, format, ap); va_end(ap); exit(EXIT_FAILURE); } #define IS_GFF (1<<0) #define IS_BED (1<<1) #define IS_SAM (1<<2) #define IS_VCF (1<<3) #define IS_BCF (1<<4) #define IS_BAM (1<<5) #define IS_CRAM (1<<6) #define IS_TXT (IS_GFF|IS_BED|IS_SAM|IS_VCF) int file_type(const char *fname) { int l = strlen(fname); int strcasecmp(const char *s1, const char *s2); if (l>=7 && strcasecmp(fname+l-7, ".gff.gz") == 0) return IS_GFF; else if (l>=7 && strcasecmp(fname+l-7, ".bed.gz") == 0) return IS_BED; else if (l>=7 && strcasecmp(fname+l-7, ".sam.gz") == 0) return IS_SAM; else if (l>=7 && strcasecmp(fname+l-7, ".vcf.gz") == 0) return IS_VCF; else if (l>=4 && strcasecmp(fname+l-4, ".bcf") == 0) return IS_BCF; else if (l>=4 && strcasecmp(fname+l-4, ".bam") == 0) return IS_BAM; else if (l>=4 && strcasecmp(fname+l-5, ".cram") == 0) return IS_CRAM; htsFile *fp = hts_open(fname,"r"); enum htsExactFormat format = fp->format.format; hts_close(fp); if ( format == bcf ) return IS_BCF; if ( format == bam ) return IS_BAM; if ( format == cram ) return IS_CRAM; if ( format == vcf ) return IS_VCF; return 0; } static char **parse_regions(char *regions_fname, char **argv, int argc, int *nregs) { kstring_t str = {0,0,0}; int iseq = 0, ireg = 0; char **regs = NULL; *nregs = argc; if ( regions_fname ) { // improve me: this is a too heavy machinery for parsing regions... regidx_t *idx = regidx_init(regions_fname, NULL, NULL, 0, NULL); if ( !idx ) error("Could not read %s\n", regions_fname); (*nregs) += regidx_nregs(idx); regs = (char**) malloc(sizeof(char*)*(*nregs)); int nseq; char **seqs = regidx_seq_names(idx, &nseq); for (iseq=0; iseqformat; regidx_t *reg_idx = NULL; if ( args->targets_fname ) { reg_idx = regidx_init(args->targets_fname, NULL, NULL, 0, NULL); if ( !reg_idx ) error("Could not read %s\n", args->targets_fname); } if ( format == bcf ) { htsFile *out = hts_open("-","w"); if ( !out ) error("Could not open stdout\n", fname); hts_idx_t *idx = bcf_index_load(fname); if ( !idx ) error("Could not load .csi index of %s\n", fname); bcf_hdr_t *hdr = bcf_hdr_read(fp); if ( !hdr ) error("Could not read the header: %s\n", fname); if ( args->print_header ) bcf_hdr_write(out,hdr); if ( !args->header_only ) { bcf1_t *rec = bcf_init(); for (i=0; i=0 ) { if ( reg_idx && !regidx_overlap(reg_idx, bcf_seqname(hdr,rec),rec->pos,rec->pos+rec->rlen-1, NULL) ) continue; bcf_write(out,hdr,rec); } tbx_itr_destroy(itr); } bcf_destroy(rec); } if ( hts_close(out) ) error("hts_close returned non-zero status for stdout\n"); bcf_hdr_destroy(hdr); hts_idx_destroy(idx); } else if ( format==vcf || format==sam || format==unknown_format ) { tbx_t *tbx = tbx_index_load(fname); if ( !tbx ) error("Could not load .tbi/.csi index of %s\n", fname); kstring_t str = {0,0,0}; if ( args->print_header ) { while ( hts_getline(fp, KS_SEP_LINE, &str) >= 0 ) { if ( !str.l || str.s[0]!=tbx->conf.meta_char ) break; puts(str.s); } } if ( !args->header_only ) { int nseq; const char **seq = NULL; if ( reg_idx ) seq = tbx_seqnames(tbx, &nseq); for (i=0; i= 0) { if ( reg_idx && !regidx_overlap(reg_idx,seq[itr->curr_tid],itr->curr_beg,itr->curr_end, NULL) ) continue; puts(str.s); } tbx_itr_destroy(itr); } free(seq); } free(str.s); tbx_destroy(tbx); } else if ( format==bam ) error("Please use \"samtools view\" for querying BAM files.\n"); if ( reg_idx ) regidx_destroy(reg_idx); if ( hts_close(fp) ) error("hts_close returned non-zero status: %s\n", fname); for (i=0; iblock_length ) return -1; char *buffer = fp->uncompressed_block; int skip_until = 0; // Skip the header: find out the position of the data block if ( buffer[0]==conf->meta_char ) { skip_until = 1; while (1) { if ( buffer[skip_until]=='\n' ) { skip_until++; if ( skip_until>=fp->block_length ) { if ( bgzf_read_block(fp) != 0 || !fp->block_length ) error("FIXME: No body in the file: %s\n", fname); skip_until = 0; } // The header has finished if ( buffer[skip_until]!=conf->meta_char ) break; } skip_until++; if ( skip_until>=fp->block_length ) { if (bgzf_read_block(fp) != 0 || !fp->block_length) error("FIXME: No body in the file: %s\n", fname); skip_until = 0; } } } // Output the new header FILE *hdr = fopen(header,"r"); if ( !hdr ) error("%s: %s", header,strerror(errno)); const size_t page_size = 32768; char *buf = malloc(page_size); BGZF *bgzf_out = bgzf_open("-", "w"); ssize_t nread; while ( (nread=fread(buf,1,page_size-1,hdr))>0 ) { if ( nreaderrcode); } if ( fclose(hdr) ) error("close failed: %s\n", header); // Output all remainig data read with the header block if ( fp->block_length - skip_until > 0 ) { if (bgzf_write(bgzf_out, buffer+skip_until, fp->block_length-skip_until) < 0) error("Error: %d\n",fp->errcode); } if (bgzf_flush(bgzf_out) < 0) error("Error: %d\n",bgzf_out->errcode); while (1) { nread = bgzf_raw_read(fp, buf, page_size); if ( nread<=0 ) break; int count = bgzf_raw_write(bgzf_out, buf, nread); if (count != nread) error("Write failed, wrote %d instead of %d bytes.\n", count,(int)nread); } if (bgzf_close(bgzf_out) < 0) error("Error: %d\n",bgzf_out->errcode); if (bgzf_close(fp) < 0) error("Error: %d\n",fp->errcode); free(buf); } else error("todo: reheader BCF, BAM\n"); // BCF is difficult, records contain pointers to the header. return 0; } static int usage(void) { fprintf(stderr, "\n"); fprintf(stderr, "Version: %s\n", hts_version()); fprintf(stderr, "Usage: tabix [OPTIONS] [FILE] [REGION [...]]\n"); fprintf(stderr, "\n"); fprintf(stderr, "Indexing Options:\n"); fprintf(stderr, " -0, --zero-based coordinates are zero-based\n"); fprintf(stderr, " -b, --begin INT column number for region start [4]\n"); fprintf(stderr, " -c, --comment CHAR skip comment lines starting with CHAR [null]\n"); fprintf(stderr, " -C, --csi generate CSI index for VCF (default is TBI)\n"); fprintf(stderr, " -e, --end INT column number for region end (if no end, set INT to -b) [5]\n"); fprintf(stderr, " -f, --force overwrite existing index without asking\n"); fprintf(stderr, " -m, --min-shift INT set minimal interval size for CSI indices to 2^INT [14]\n"); fprintf(stderr, " -p, --preset STR gff, bed, sam, vcf\n"); fprintf(stderr, " -s, --sequence INT column number for sequence names (suppressed by -p) [1]\n"); fprintf(stderr, " -S, --skip-lines INT skip first INT lines [0]\n"); fprintf(stderr, "\n"); fprintf(stderr, "Querying and other options:\n"); fprintf(stderr, " -h, --print-header print also the header lines\n"); fprintf(stderr, " -H, --only-header print only the header lines\n"); fprintf(stderr, " -l, --list-chroms list chromosome names\n"); fprintf(stderr, " -r, --reheader FILE replace the header with the content of FILE\n"); fprintf(stderr, " -R, --regions FILE restrict to regions listed in the file\n"); fprintf(stderr, " -T, --targets FILE similar to -R but streams rather than index-jumps\n"); fprintf(stderr, "\n"); return 1; } int main(int argc, char *argv[]) { int c, min_shift = 0, is_force = 0, list_chroms = 0, do_csi = 0; tbx_conf_t conf = tbx_conf_gff, *conf_ptr = NULL; char *reheader = NULL; args_t args; memset(&args,0,sizeof(args_t)); static const struct option loptions[] = { {"help", no_argument, NULL, 'h'}, {"regions", required_argument, NULL, 'R'}, {"targets", required_argument, NULL, 'T'}, {"csi", no_argument, NULL, 'C'}, {"zero-based", no_argument, NULL, '0'}, {"print-header", no_argument, NULL, 'h'}, {"only-header", no_argument, NULL, 'H'}, {"begin", required_argument, NULL, 'b'}, {"comment", required_argument, NULL, 'c'}, {"end", required_argument, NULL, 'e'}, {"force", no_argument, NULL, 'f'}, {"preset", required_argument, NULL, 'p'}, {"sequence", required_argument, NULL, 's'}, {"skip-lines", required_argument, NULL, 'S'}, {"list-chroms", no_argument, NULL, 'l'}, {"reheader", required_argument, NULL, 'r'}, {"version", no_argument, NULL, 1}, {NULL, 0, NULL, 0} }; char *tmp; while ((c = getopt_long(argc, argv, "hH?0b:c:e:fm:p:s:S:lr:CR:T:", loptions,NULL)) >= 0) { switch (c) { case 'R': args.regions_fname = optarg; break; case 'T': args.targets_fname = optarg; break; case 'C': do_csi = 1; break; case 'r': reheader = optarg; break; case 'h': args.print_header = 1; break; case 'H': args.print_header = 1; args.header_only = 1; break; case 'l': list_chroms = 1; break; case '0': conf.preset |= TBX_UCSC; break; case 'b': conf.bc = strtol(optarg,&tmp,10); if ( *tmp ) error("Could not parse argument: -b %s\n", optarg); break; case 'e': conf.ec = strtol(optarg,&tmp,10); if ( *tmp ) error("Could not parse argument: -e %s\n", optarg); break; case 'c': conf.meta_char = *optarg; break; case 'f': is_force = 1; break; case 'm': min_shift = strtol(optarg,&tmp,10); if ( *tmp ) error("Could not parse argument: -m %s\n", optarg); break; case 'p': if (strcmp(optarg, "gff") == 0) conf_ptr = &tbx_conf_gff; else if (strcmp(optarg, "bed") == 0) conf_ptr = &tbx_conf_bed; else if (strcmp(optarg, "sam") == 0) conf_ptr = &tbx_conf_sam; else if (strcmp(optarg, "vcf") == 0) conf_ptr = &tbx_conf_vcf; else if (strcmp(optarg, "bcf") == 0) ; // bcf is autodetected, preset is not needed else if (strcmp(optarg, "bam") == 0) ; // same as bcf else error("The preset string not recognised: '%s'\n", optarg); break; case 's': conf.sc = strtol(optarg,&tmp,10); if ( *tmp ) error("Could not parse argument: -s %s\n", optarg); break; case 'S': conf.line_skip = strtol(optarg,&tmp,10); if ( *tmp ) error("Could not parse argument: -S %s\n", optarg); break; case 1: printf( "tabix (htslib) %s\n" "Copyright (C) 2016 Genome Research Ltd.\n", hts_version()); return EXIT_SUCCESS; default: return usage(); } } if ( optind==argc ) return usage(); if ( list_chroms ) return query_chroms(argv[optind]); if ( argc > optind+1 || args.header_only || args.regions_fname || args.targets_fname ) { int nregs = 0; char **regs = NULL; if ( !args.header_only ) regs = parse_regions(args.regions_fname, argv+optind+1, argc-optind-1, &nregs); return query_regions(&args, argv[optind], regs, nregs); } char *fname = argv[optind]; int ftype = file_type(fname); if ( !conf_ptr ) // no preset given { if ( ftype==IS_GFF ) conf_ptr = &tbx_conf_gff; else if ( ftype==IS_BED ) conf_ptr = &tbx_conf_bed; else if ( ftype==IS_SAM ) conf_ptr = &tbx_conf_sam; else if ( ftype==IS_VCF ) { conf_ptr = &tbx_conf_vcf; if ( !min_shift && do_csi ) min_shift = 14; } else if ( ftype==IS_BCF ) { if ( !min_shift ) min_shift = 14; } else if ( ftype==IS_BAM ) { if ( !min_shift ) min_shift = 14; } } if ( do_csi ) { if ( !min_shift ) min_shift = 14; min_shift *= do_csi; // positive for CSIv2, negative for CSIv1 } if ( min_shift!=0 && !do_csi ) do_csi = 1; if ( reheader ) return reheader_file(fname, reheader, ftype, conf_ptr); if ( conf_ptr ) conf = *conf_ptr; char *suffix = ".tbi"; if ( do_csi ) suffix = ".csi"; else if ( ftype==IS_BAM ) suffix = ".bai"; else if ( ftype==IS_CRAM ) suffix = ".crai"; char *idx_fname = calloc(strlen(fname) + 5, 1); strcat(strcpy(idx_fname, fname), suffix); struct stat stat_tbi, stat_file; if ( !is_force && stat(idx_fname, &stat_tbi)==0 ) { // Before complaining about existing index, check if the VCF file isn't // newer. This is a common source of errors, people tend not to notice // that tabix failed stat(fname, &stat_file); if ( stat_file.st_mtime <= stat_tbi.st_mtime ) error("[tabix] the index file exists. Please use '-f' to overwrite.\n"); } free(idx_fname); if ( ftype==IS_CRAM ) { if ( bam_index_build(fname, min_shift)!=0 ) error("bam_index_build failed: %s\n", fname); return 0; } else if ( do_csi ) { if ( ftype==IS_BCF ) { if ( bcf_index_build(fname, min_shift)!=0 ) error("bcf_index_build failed: %s\n", fname); return 0; } if ( ftype==IS_BAM ) { if ( bam_index_build(fname, min_shift)!=0 ) error("bam_index_build failed: %s\n", fname); return 0; } if ( tbx_index_build(fname, min_shift, &conf)!=0 ) error("tbx_index_build failed: %s\n", fname); return 0; } else // TBI index { if ( tbx_index_build(fname, min_shift, &conf) ) error("tbx_index_build failed: %s\n", fname); return 0; } return 0; } stacks-1.44/htslib/._tbx.c000644 000765 000024 00000000265 12750743246 016162 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/tbx.c000644 000765 000024 00000024564 12750743246 015755 0ustar00catchenstaff000000 000000 /* tbx.c -- tabix API functions. Copyright (C) 2009, 2010, 2012-2015 Genome Research Ltd. Copyright (C) 2010-2012 Broad Institute. Author: Heng Li Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include #include #include #include "htslib/tbx.h" #include "htslib/bgzf.h" #include "htslib/khash.h" KHASH_DECLARE(s2i, kh_cstr_t, int64_t) tbx_conf_t tbx_conf_gff = { 0, 1, 4, 5, '#', 0 }; tbx_conf_t tbx_conf_bed = { TBX_UCSC, 1, 2, 3, '#', 0 }; tbx_conf_t tbx_conf_psltbl = { TBX_UCSC, 15, 17, 18, '#', 0 }; tbx_conf_t tbx_conf_sam = { TBX_SAM, 3, 4, 0, '@', 0 }; tbx_conf_t tbx_conf_vcf = { TBX_VCF, 1, 2, 0, '#', 0 }; typedef struct { int64_t beg, end; char *ss, *se; int tid; } tbx_intv_t; static inline int get_tid(tbx_t *tbx, const char *ss, int is_add) { khint_t k; khash_t(s2i) *d; if (tbx->dict == 0) tbx->dict = kh_init(s2i); d = (khash_t(s2i)*)tbx->dict; if (is_add) { int absent; k = kh_put(s2i, d, ss, &absent); if (absent) { kh_key(d, k) = strdup(ss); kh_val(d, k) = kh_size(d) - 1; } } else k = kh_get(s2i, d, ss); return k == kh_end(d)? -1 : kh_val(d, k); } int tbx_name2id(tbx_t *tbx, const char *ss) { return get_tid(tbx, ss, 0); } int tbx_parse1(const tbx_conf_t *conf, int len, char *line, tbx_intv_t *intv) { int i, b = 0, id = 1, ncols = 0; char *s; intv->ss = intv->se = 0; intv->beg = intv->end = -1; for (i = 0; i <= len; ++i) { if (line[i] == '\t' || line[i] == 0) { ++ncols; if (id == conf->sc) { intv->ss = line + b; intv->se = line + i; } else if (id == conf->bc) { // here ->beg is 0-based. intv->beg = intv->end = strtol(line + b, &s, 0); if ( s==line+b ) return -1; // expected int if (!(conf->preset&TBX_UCSC)) --intv->beg; else ++intv->end; if (intv->beg < 0) intv->beg = 0; if (intv->end < 1) intv->end = 1; } else { if ((conf->preset&0xffff) == TBX_GENERIC) { if (id == conf->ec) { intv->end = strtol(line + b, &s, 0); if ( s==line+b ) return -1; // expected int } } else if ((conf->preset&0xffff) == TBX_SAM) { if (id == 6) { // CIGAR int l = 0, op; char *t; for (s = line + b; s < line + i;) { long x = strtol(s, &t, 10); op = toupper(*t); if (op == 'M' || op == 'D' || op == 'N') l += x; s = t + 1; } if (l == 0) l = 1; intv->end = intv->beg + l; } } else if ((conf->preset&0xffff) == TBX_VCF) { if (id == 4) { if (b < i) intv->end = intv->beg + (i - b); } else if (id == 8) { // look for "END=" int c = line[i]; line[i] = 0; s = strstr(line + b, "END="); if (s == line + b) s += 4; else if (s) { s = strstr(line + b, ";END="); if (s) s += 5; } if (s) intv->end = strtol(s, &s, 0); line[i] = c; } } } b = i + 1; ++id; } } if (intv->ss == 0 || intv->se == 0 || intv->beg < 0 || intv->end < 0) return -1; return 0; } static inline int get_intv(tbx_t *tbx, kstring_t *str, tbx_intv_t *intv, int is_add) { if (tbx_parse1(&tbx->conf, str->l, str->s, intv) == 0) { int c = *intv->se; *intv->se = '\0'; intv->tid = get_tid(tbx, intv->ss, is_add); *intv->se = c; return (intv->tid >= 0 && intv->beg >= 0 && intv->end >= 0)? 0 : -1; } else { char *type = NULL; switch (tbx->conf.preset&0xffff) { case TBX_SAM: type = "TBX_SAM"; break; case TBX_VCF: type = "TBX_VCF"; break; case TBX_UCSC: type = "TBX_UCSC"; break; default: type = "TBX_GENERIC"; break; } fprintf(stderr, "[E::%s] failed to parse %s, was wrong -p [type] used?\nThe offending line was: \"%s\"\n", __func__, type, str->s); return -1; } } int tbx_readrec(BGZF *fp, void *tbxv, void *sv, int *tid, int *beg, int *end) { tbx_t *tbx = (tbx_t *) tbxv; kstring_t *s = (kstring_t *) sv; int ret; if ((ret = bgzf_getline(fp, '\n', s)) >= 0) { tbx_intv_t intv; get_intv(tbx, s, &intv, 0); *tid = intv.tid; *beg = intv.beg; *end = intv.end; } return ret; } void tbx_set_meta(tbx_t *tbx) { int i, l = 0, l_nm; uint32_t x[7]; char **name; uint8_t *meta; khint_t k; khash_t(s2i) *d = (khash_t(s2i)*)tbx->dict; memcpy(x, &tbx->conf, 24); name = (char**)malloc(sizeof(char*) * kh_size(d)); for (k = kh_begin(d), l = 0; k != kh_end(d); ++k) { if (!kh_exist(d, k)) continue; name[kh_val(d, k)] = (char*)kh_key(d, k); l += strlen(kh_key(d, k)) + 1; // +1 to include '\0' } l_nm = x[6] = l; meta = (uint8_t*)malloc(l_nm + 28); if (ed_is_big()) for (i = 0; i < 7; ++i) x[i] = ed_swap_4(x[i]); memcpy(meta, x, 28); for (l = 28, i = 0; i < (int)kh_size(d); ++i) { int x = strlen(name[i]) + 1; memcpy(meta + l, name[i], x); l += x; } free(name); hts_idx_set_meta(tbx->idx, l, meta, 0); } tbx_t *tbx_index(BGZF *fp, int min_shift, const tbx_conf_t *conf) { tbx_t *tbx; kstring_t str; int ret, first = 0, n_lvls, fmt; int64_t lineno = 0; uint64_t last_off = 0; tbx_intv_t intv; str.s = 0; str.l = str.m = 0; tbx = (tbx_t*)calloc(1, sizeof(tbx_t)); tbx->conf = *conf; if (min_shift > 0) n_lvls = (TBX_MAX_SHIFT - min_shift + 2) / 3, fmt = HTS_FMT_CSI; else min_shift = 14, n_lvls = 5, fmt = HTS_FMT_TBI; while ((ret = bgzf_getline(fp, '\n', &str)) >= 0) { ++lineno; if (lineno <= tbx->conf.line_skip || str.s[0] == tbx->conf.meta_char) { last_off = bgzf_tell(fp); continue; } if (first == 0) { tbx->idx = hts_idx_init(0, fmt, last_off, min_shift, n_lvls); first = 1; } get_intv(tbx, &str, &intv, 1); ret = hts_idx_push(tbx->idx, intv.tid, intv.beg, intv.end, bgzf_tell(fp), 1); if (ret < 0) { free(str.s); tbx_destroy(tbx); return NULL; } } if ( !tbx->idx ) tbx->idx = hts_idx_init(0, fmt, last_off, min_shift, n_lvls); // empty file if ( !tbx->dict ) tbx->dict = kh_init(s2i); hts_idx_finish(tbx->idx, bgzf_tell(fp)); tbx_set_meta(tbx); free(str.s); return tbx; } void tbx_destroy(tbx_t *tbx) { khash_t(s2i) *d = (khash_t(s2i)*)tbx->dict; if (d != NULL) { khint_t k; for (k = kh_begin(d); k != kh_end(d); ++k) if (kh_exist(d, k)) free((char*)kh_key(d, k)); } hts_idx_destroy(tbx->idx); kh_destroy(s2i, d); free(tbx); } int tbx_index_build2(const char *fn, const char *fnidx, int min_shift, const tbx_conf_t *conf) { tbx_t *tbx; BGZF *fp; int ret; if ( bgzf_is_bgzf(fn)!=1 ) { fprintf(stderr,"Not a BGZF file: %s\n", fn); return -1; } if ((fp = bgzf_open(fn, "r")) == 0) return -1; if ( !fp->is_compressed ) { bgzf_close(fp); return -1; } tbx = tbx_index(fp, min_shift, conf); bgzf_close(fp); if ( !tbx ) return -1; ret = hts_idx_save_as(tbx->idx, fn, fnidx, min_shift > 0? HTS_FMT_CSI : HTS_FMT_TBI); tbx_destroy(tbx); return ret; } int tbx_index_build(const char *fn, int min_shift, const tbx_conf_t *conf) { return tbx_index_build2(fn, NULL, min_shift, conf); } tbx_t *tbx_index_load2(const char *fn, const char *fnidx) { tbx_t *tbx; uint8_t *meta; char *nm, *p; uint32_t x[7]; int l_meta, l_nm; tbx = (tbx_t*)calloc(1, sizeof(tbx_t)); tbx->idx = fnidx? hts_idx_load2(fn, fnidx) : hts_idx_load(fn, HTS_FMT_TBI); if ( !tbx->idx ) { free(tbx); return NULL; } meta = hts_idx_get_meta(tbx->idx, &l_meta); if ( !meta ) { free(tbx); return NULL; } memcpy(x, meta, 28); memcpy(&tbx->conf, x, 24); p = nm = (char*)meta + 28; l_nm = x[6]; for (; p - nm < l_nm; p += strlen(p) + 1) get_tid(tbx, p, 1); return tbx; } tbx_t *tbx_index_load(const char *fn) { return tbx_index_load2(fn, NULL); } const char **tbx_seqnames(tbx_t *tbx, int *n) { khash_t(s2i) *d = (khash_t(s2i)*)tbx->dict; if (d == NULL) { *n = 0; return NULL; } int tid, m = kh_size(d); const char **names = (const char**) calloc(m,sizeof(const char*)); khint_t k; for (k=kh_begin(d); k Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include #include #include #include #include #include "htslib/kstring.h" #include "htslib/bgzf.h" #include "htslib/vcf.h" #include "htslib/tbx.h" #include "htslib/hfile.h" #include "htslib/khash_str2int.h" #include "hts_internal.h" #include "htslib/khash.h" KHASH_MAP_INIT_STR(vdict, bcf_idinfo_t) typedef khash_t(vdict) vdict_t; #include "htslib/kseq.h" KSTREAM_DECLARE(gzFile, gzread) uint32_t bcf_float_missing = 0x7F800001; uint32_t bcf_float_vector_end = 0x7F800002; uint8_t bcf_type_shift[] = { 0, 0, 1, 2, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 }; static bcf_idinfo_t bcf_idinfo_def = { .info = { 15, 15, 15 }, .hrec = { NULL, NULL, NULL}, .id = -1 }; static const char *dump_char(char *buffer, char c) { switch (c) { case '\n': strcpy(buffer, "\\n"); break; case '\r': strcpy(buffer, "\\r"); break; case '\t': strcpy(buffer, "\\t"); break; case '\'': case '\"': case '\\': sprintf(buffer, "\\%c", c); break; default: if (isprint_c(c)) sprintf(buffer, "%c", c); else sprintf(buffer, "\\x%02X", (unsigned char) c); break; } return buffer; } /************************* *** VCF header parser *** *************************/ int bcf_hdr_add_sample(bcf_hdr_t *h, const char *s) { if ( !s ) return 0; const char *ss = s; while ( !*ss && isspace(*ss) ) ss++; if ( !*ss ) { fprintf(stderr,"[E::%s] Empty sample name: trailing spaces/tabs in the header line?\n", __func__); abort(); } vdict_t *d = (vdict_t*)h->dict[BCF_DT_SAMPLE]; int ret; char *sdup = strdup(s); int k = kh_put(vdict, d, sdup, &ret); if (ret) { // absent kh_val(d, k) = bcf_idinfo_def; kh_val(d, k).id = kh_size(d) - 1; } else { if (hts_verbose >= 2) { fprintf(stderr, "[E::%s] Duplicated sample name '%s'\n", __func__, s); abort(); } free(sdup); return -1; } int n = kh_size(d); h->samples = (char**) realloc(h->samples,sizeof(char*)*n); h->samples[n-1] = sdup; h->dirty = 1; return 0; } int bcf_hdr_parse_sample_line(bcf_hdr_t *h, const char *str) { int ret = 0; int i = 0; const char *p, *q; // add samples for (p = q = str;; ++q) { if (*q != '\t' && *q != 0 && *q != '\n') continue; if (++i > 9) { char *s = (char*)malloc(q - p + 1); strncpy(s, p, q - p); s[q - p] = 0; if ( bcf_hdr_add_sample(h,s) < 0 ) ret = -1; free(s); } if (*q == 0 || *q == '\n') break; p = q + 1; } bcf_hdr_add_sample(h,NULL); return ret; } int bcf_hdr_sync(bcf_hdr_t *h) { int i; for (i = 0; i < 3; i++) { vdict_t *d = (vdict_t*)h->dict[i]; khint_t k; if ( h->n[i] < kh_size(d) ) { // this should be true only for i=2, BCF_DT_SAMPLE h->n[i] = kh_size(d); h->id[i] = (bcf_idpair_t*) realloc(h->id[i], kh_size(d)*sizeof(bcf_idpair_t)); } for (k=kh_begin(d); kid[i][kh_val(d,k).id].key = kh_key(d,k); h->id[i][kh_val(d,k).id].val = &kh_val(d,k); } } h->dirty = 0; return 0; } void bcf_hrec_destroy(bcf_hrec_t *hrec) { free(hrec->key); if ( hrec->value ) free(hrec->value); int i; for (i=0; inkeys; i++) { free(hrec->keys[i]); free(hrec->vals[i]); } free(hrec->keys); free(hrec->vals); free(hrec); } // Copies all fields except IDX. bcf_hrec_t *bcf_hrec_dup(bcf_hrec_t *hrec) { bcf_hrec_t *out = (bcf_hrec_t*) calloc(1,sizeof(bcf_hrec_t)); out->type = hrec->type; if ( hrec->key ) out->key = strdup(hrec->key); if ( hrec->value ) out->value = strdup(hrec->value); out->nkeys = hrec->nkeys; out->keys = (char**) malloc(sizeof(char*)*hrec->nkeys); out->vals = (char**) malloc(sizeof(char*)*hrec->nkeys); int i, j = 0; for (i=0; inkeys; i++) { if ( hrec->keys[i] && !strcmp("IDX",hrec->keys[i]) ) continue; if ( hrec->keys[i] ) out->keys[j] = strdup(hrec->keys[i]); if ( hrec->vals[i] ) out->vals[j] = strdup(hrec->vals[i]); j++; } if ( i!=j ) out->nkeys -= i-j; // IDX was omitted return out; } void bcf_hrec_debug(FILE *fp, bcf_hrec_t *hrec) { fprintf(fp, "key=[%s] value=[%s]", hrec->key, hrec->value?hrec->value:""); int i; for (i=0; inkeys; i++) fprintf(fp, "\t[%s]=[%s]", hrec->keys[i],hrec->vals[i]); fprintf(fp, "\n"); } void bcf_header_debug(bcf_hdr_t *hdr) { int i, j; for (i=0; inhrec; i++) { if ( !hdr->hrec[i]->value ) { fprintf(stderr, "##%s=<", hdr->hrec[i]->key); fprintf(stderr,"%s=%s", hdr->hrec[i]->keys[0], hdr->hrec[i]->vals[0]); for (j=1; jhrec[i]->nkeys; j++) fprintf(stderr,",%s=%s", hdr->hrec[i]->keys[j], hdr->hrec[i]->vals[j]); fprintf(stderr,">\n"); } else fprintf(stderr,"##%s=%s\n", hdr->hrec[i]->key,hdr->hrec[i]->value); } } void bcf_hrec_add_key(bcf_hrec_t *hrec, const char *str, int len) { int n = ++hrec->nkeys; hrec->keys = (char**) realloc(hrec->keys, sizeof(char*)*n); hrec->vals = (char**) realloc(hrec->vals, sizeof(char*)*n); assert( len ); hrec->keys[n-1] = (char*) malloc((len+1)*sizeof(char)); memcpy(hrec->keys[n-1],str,len); hrec->keys[n-1][len] = 0; hrec->vals[n-1] = NULL; } void bcf_hrec_set_val(bcf_hrec_t *hrec, int i, const char *str, int len, int is_quoted) { if ( !str ) { hrec->vals[i] = NULL; return; } if ( hrec->vals[i] ) free(hrec->vals[i]); if ( is_quoted ) { hrec->vals[i] = (char*) malloc((len+3)*sizeof(char)); hrec->vals[i][0] = '"'; memcpy(&hrec->vals[i][1],str,len); hrec->vals[i][len+1] = '"'; hrec->vals[i][len+2] = 0; } else { hrec->vals[i] = (char*) malloc((len+1)*sizeof(char)); memcpy(hrec->vals[i],str,len); hrec->vals[i][len] = 0; } } void hrec_add_idx(bcf_hrec_t *hrec, int idx) { int n = ++hrec->nkeys; hrec->keys = (char**) realloc(hrec->keys, sizeof(char*)*n); hrec->vals = (char**) realloc(hrec->vals, sizeof(char*)*n); hrec->keys[n-1] = strdup("IDX"); kstring_t str = {0,0,0}; kputw(idx, &str); hrec->vals[n-1] = str.s; } int bcf_hrec_find_key(bcf_hrec_t *hrec, const char *key) { int i; for (i=0; inkeys; i++) if ( !strcasecmp(key,hrec->keys[i]) ) return i; return -1; } static inline int is_escaped(const char *min, const char *str) { int n = 0; while ( --str>=min && *str=='\\' ) n++; return n%2; } bcf_hrec_t *bcf_hdr_parse_line(const bcf_hdr_t *h, const char *line, int *len) { const char *p = line; if (p[0] != '#' || p[1] != '#') { *len = 0; return NULL; } p += 2; const char *q = p; while ( *q && *q!='=' ) q++; int n = q-p; if ( *q!='=' || !n ) { *len = q-line+1; return NULL; } // wrong format bcf_hrec_t *hrec = (bcf_hrec_t*) calloc(1,sizeof(bcf_hrec_t)); hrec->key = (char*) malloc(sizeof(char)*(n+1)); memcpy(hrec->key,p,n); hrec->key[n] = 0; p = ++q; if ( *p!='<' ) // generic field, e.g. ##samtoolsVersion=0.1.18-r579 { while ( *q && *q!='\n' ) q++; hrec->value = (char*) malloc((q-p+1)*sizeof(char)); memcpy(hrec->value, p, q-p); hrec->value[q-p] = 0; *len = q-line+1; return hrec; } // structured line, e.g. // ##INFO= // ##PEDIGREE= int nopen = 1; while ( *q && *q!='\n' && nopen>0 ) { p = ++q; while ( *q && *q==' ' ) { p++; q++; } // ^[A-Za-z_][0-9A-Za-z_.]*$ if (p==q && *q && (isalpha(*q) || *q=='_')) { q++; while ( *q && (isalnum(*q) || *q=='_' || *q=='.') ) q++; } n = q-p; int m = 0; while ( *q && *q==' ' ) { q++; m++; } if ( *q!='=' || !n ) { // wrong format while ( *q && *q!='\n' ) q++; kstring_t tmp = {0,0,0}; kputsn(line,q-line,&tmp); fprintf(stderr,"Could not parse the header line: \"%s\"\n", tmp.s); free(tmp.s); *len = q-line+1; bcf_hrec_destroy(hrec); return NULL; } bcf_hrec_add_key(hrec, p, q-p-m); p = ++q; while ( *q && *q==' ' ) { p++; q++; } int quoted = *p=='"' ? 1 : 0; if ( quoted ) p++, q++; while ( *q && *q != '\n' ) { if ( quoted ) { if ( *q=='"' && !is_escaped(p,q) ) break; } else { if ( *q=='<' ) nopen++; if ( *q=='>' ) nopen--; if ( !nopen ) break; if ( *q==',' && nopen==1 ) break; } q++; } const char *r = q; while ( r > p && r[-1] == ' ' ) r--; bcf_hrec_set_val(hrec, hrec->nkeys-1, p, r-p, quoted); if ( quoted && *q=='"' ) q++; if ( *q=='>' ) { nopen--; q++; } } // Skip trailing spaces while ( *q && *q==' ' ) { q++; } *len = q-line+1; return hrec; } static int bcf_hdr_set_idx(bcf_hdr_t *hdr, int dict_type, const char *tag, bcf_idinfo_t *idinfo) { // If available, preserve existing IDX if ( idinfo->id==-1 ) idinfo->id = hdr->n[dict_type]++; else if ( idinfo->id < hdr->n[dict_type] && hdr->id[dict_type][idinfo->id].key ) { fprintf(stderr,"[%s:%d %s] Conflicting IDX=%d lines in the header dictionary, the new tag is %s\n", __FILE__,__LINE__,__FUNCTION__, idinfo->id, tag); exit(1); } if ( idinfo->id >= hdr->n[dict_type] ) hdr->n[dict_type] = idinfo->id+1; hts_expand0(bcf_idpair_t,hdr->n[dict_type],hdr->m[dict_type],hdr->id[dict_type]); // NB: the next kh_put call can invalidate the idinfo pointer, therefore // we leave it unassigned here. It myst be set explicitly in bcf_hdr_sync. hdr->id[dict_type][idinfo->id].key = tag; return 0; } // returns: 1 when hdr needs to be synced, 0 otherwise int bcf_hdr_register_hrec(bcf_hdr_t *hdr, bcf_hrec_t *hrec) { // contig int i,j, ret; khint_t k; char *str; if ( !strcmp(hrec->key, "contig") ) { hrec->type = BCF_HL_CTG; // Get the contig ID ($str) and length ($j) i = bcf_hrec_find_key(hrec,"length"); if ( i<0 ) j = 0; else if ( sscanf(hrec->vals[i],"%d",&j)!=1 ) return 0; i = bcf_hrec_find_key(hrec,"ID"); if ( i<0 ) return 0; str = strdup(hrec->vals[i]); // Register in the dictionary vdict_t *d = (vdict_t*)hdr->dict[BCF_DT_CTG]; khint_t k = kh_get(vdict, d, str); if ( k != kh_end(d) ) { free(str); return 0; } // already present k = kh_put(vdict, d, str, &ret); int idx = bcf_hrec_find_key(hrec,"IDX"); if ( idx!=-1 ) { char *tmp = hrec->vals[idx]; idx = strtol(hrec->vals[idx], &tmp, 10); if ( *tmp || idx < 0 ) { fprintf(stderr,"[%s:%d %s] Error parsing the IDX tag, skipping.\n", __FILE__,__LINE__,__FUNCTION__); return 0; } } kh_val(d, k) = bcf_idinfo_def; kh_val(d, k).id = idx; kh_val(d, k).info[0] = j; kh_val(d, k).hrec[0] = hrec; bcf_hdr_set_idx(hdr, BCF_DT_CTG, kh_key(d,k), &kh_val(d,k)); if ( idx==-1 ) hrec_add_idx(hrec, kh_val(d,k).id); return 1; } if ( !strcmp(hrec->key, "INFO") ) hrec->type = BCF_HL_INFO; else if ( !strcmp(hrec->key, "FILTER") ) hrec->type = BCF_HL_FLT; else if ( !strcmp(hrec->key, "FORMAT") ) hrec->type = BCF_HL_FMT; else if ( hrec->nkeys>0 ) { hrec->type = BCF_HL_STR; return 1; } else return 0; // INFO/FILTER/FORMAT char *id = NULL; int type = -1, num = -1, var = -1, idx = -1; for (i=0; inkeys; i++) { if ( !strcmp(hrec->keys[i], "ID") ) id = hrec->vals[i]; else if ( !strcmp(hrec->keys[i], "IDX") ) { char *tmp = hrec->vals[i]; idx = strtol(hrec->vals[i], &tmp, 10); if ( *tmp || idx < 0 ) { fprintf(stderr,"[%s:%d %s] Error parsing the IDX tag, skipping.\n", __FILE__,__LINE__,__FUNCTION__); return 0; } } else if ( !strcmp(hrec->keys[i], "Type") ) { if ( !strcmp(hrec->vals[i], "Integer") ) type = BCF_HT_INT; else if ( !strcmp(hrec->vals[i], "Float") ) type = BCF_HT_REAL; else if ( !strcmp(hrec->vals[i], "String") ) type = BCF_HT_STR; else if ( !strcmp(hrec->vals[i], "Character") ) type = BCF_HT_STR; else if ( !strcmp(hrec->vals[i], "Flag") ) type = BCF_HT_FLAG; else { if (hts_verbose >= 2) fprintf(stderr, "[E::%s] The type \"%s\" is not supported, assuming \"String\"\n", __func__, hrec->vals[i]); type = BCF_HT_STR; } } else if ( !strcmp(hrec->keys[i], "Number") ) { if ( !strcmp(hrec->vals[i],"A") ) var = BCF_VL_A; else if ( !strcmp(hrec->vals[i],"R") ) var = BCF_VL_R; else if ( !strcmp(hrec->vals[i],"G") ) var = BCF_VL_G; else if ( !strcmp(hrec->vals[i],".") ) var = BCF_VL_VAR; else { sscanf(hrec->vals[i],"%d",&num); var = BCF_VL_FIXED; } if (var != BCF_VL_FIXED) num = 0xfffff; } } uint32_t info = (uint32_t)num<<12 | var<<8 | type<<4 | hrec->type; if ( !id ) return 0; str = strdup(id); vdict_t *d = (vdict_t*)hdr->dict[BCF_DT_ID]; k = kh_get(vdict, d, str); if ( k != kh_end(d) ) { // already present free(str); if ( kh_val(d, k).hrec[info&0xf] ) return 0; kh_val(d, k).info[info&0xf] = info; kh_val(d, k).hrec[info&0xf] = hrec; if ( idx==-1 ) hrec_add_idx(hrec, kh_val(d, k).id); return 1; } k = kh_put(vdict, d, str, &ret); kh_val(d, k) = bcf_idinfo_def; kh_val(d, k).info[info&0xf] = info; kh_val(d, k).hrec[info&0xf] = hrec; kh_val(d, k).id = idx; bcf_hdr_set_idx(hdr, BCF_DT_ID, kh_key(d,k), &kh_val(d,k)); if ( idx==-1 ) hrec_add_idx(hrec, kh_val(d,k).id); return 1; } int bcf_hdr_add_hrec(bcf_hdr_t *hdr, bcf_hrec_t *hrec) { if ( !hrec ) return 0; hrec->type = BCF_HL_GEN; if ( !bcf_hdr_register_hrec(hdr,hrec) ) { // If one of the hashed field, then it is already present if ( hrec->type != BCF_HL_GEN ) { bcf_hrec_destroy(hrec); return 0; } // Is one of the generic fields and already present? int i; for (i=0; inhrec; i++) { if ( hdr->hrec[i]->type!=BCF_HL_GEN ) continue; if ( !strcmp(hdr->hrec[i]->key,hrec->key) && !strcmp(hrec->key,"fileformat") ) break; if ( !strcmp(hdr->hrec[i]->key,hrec->key) && !strcmp(hdr->hrec[i]->value,hrec->value) ) break; } if ( inhrec ) { bcf_hrec_destroy(hrec); return 0; } } // New record, needs to be added int n = ++hdr->nhrec; hdr->hrec = (bcf_hrec_t**) realloc(hdr->hrec, n*sizeof(bcf_hrec_t*)); hdr->hrec[n-1] = hrec; hdr->dirty = 1; return hrec->type==BCF_HL_GEN ? 0 : 1; } /* * Note that while querying of FLT,INFO,FMT,CTG lines is fast (the keys are hashed), * the STR,GEN lines are searched for linearly in a linked list of all header lines. * This may become a problem for VCFs with huge headers, we might need to build a * dictionary for these lines as well. */ bcf_hrec_t *bcf_hdr_get_hrec(const bcf_hdr_t *hdr, int type, const char *key, const char *value, const char *str_class) { int i; if ( type==BCF_HL_GEN ) { for (i=0; inhrec; i++) { if ( hdr->hrec[i]->type!=type ) continue; if ( strcmp(hdr->hrec[i]->key,key) ) continue; if ( !value || !strcmp(hdr->hrec[i]->value,value) ) return hdr->hrec[i]; } return NULL; } else if ( type==BCF_HL_STR ) { for (i=0; inhrec; i++) { if ( hdr->hrec[i]->type!=type ) continue; if ( strcmp(hdr->hrec[i]->key,str_class) ) continue; int j = bcf_hrec_find_key(hdr->hrec[i],key); if ( j>=0 && !strcmp(hdr->hrec[i]->vals[j],value) ) return hdr->hrec[i]; } return NULL; } vdict_t *d = type==BCF_HL_CTG ? (vdict_t*)hdr->dict[BCF_DT_CTG] : (vdict_t*)hdr->dict[BCF_DT_ID]; khint_t k = kh_get(vdict, d, value); if ( k == kh_end(d) ) return NULL; return kh_val(d, k).hrec[type==BCF_HL_CTG?0:type]; } void bcf_hdr_check_sanity(bcf_hdr_t *hdr) { static int PL_warned = 0, GL_warned = 0; if ( !PL_warned ) { int id = bcf_hdr_id2int(hdr, BCF_DT_ID, "PL"); if ( bcf_hdr_idinfo_exists(hdr,BCF_HL_FMT,id) && bcf_hdr_id2length(hdr,BCF_HL_FMT,id)!=BCF_VL_G ) { if (hts_verbose >= 2) fprintf(stderr,"[W::%s] PL should be declared as Number=G\n", __func__); PL_warned = 1; } } if ( !GL_warned ) { int id = bcf_hdr_id2int(hdr, BCF_DT_ID, "GL"); if ( bcf_hdr_idinfo_exists(hdr,BCF_HL_FMT,id) && bcf_hdr_id2length(hdr,BCF_HL_FMT,id)!=BCF_VL_G ) { if (hts_verbose >= 2) fprintf(stderr,"[W::%s] GL should be declared as Number=G\n", __func__); GL_warned = 1; } } } int bcf_hdr_parse(bcf_hdr_t *hdr, char *htxt) { int len, needs_sync = 0; char *p = htxt; // Check sanity: "fileformat" string must come as first bcf_hrec_t *hrec = bcf_hdr_parse_line(hdr,p,&len); if ( !hrec || !hrec->key || strcasecmp(hrec->key,"fileformat") ) fprintf(stderr, "[W::%s] The first line should be ##fileformat; is the VCF/BCF header broken?\n", __func__); needs_sync += bcf_hdr_add_hrec(hdr, hrec); // The filter PASS must appear first in the dictionary hrec = bcf_hdr_parse_line(hdr,"##FILTER=",&len); needs_sync += bcf_hdr_add_hrec(hdr, hrec); // Parse the whole header while ( (hrec=bcf_hdr_parse_line(hdr,p,&len)) ) { needs_sync += bcf_hdr_add_hrec(hdr, hrec); p += len; } int ret = bcf_hdr_parse_sample_line(hdr,p); bcf_hdr_sync(hdr); bcf_hdr_check_sanity(hdr); return ret; } int bcf_hdr_append(bcf_hdr_t *hdr, const char *line) { int len; bcf_hrec_t *hrec = bcf_hdr_parse_line(hdr, (char*) line, &len); if ( !hrec ) return -1; bcf_hdr_add_hrec(hdr, hrec); return 0; } void bcf_hdr_remove(bcf_hdr_t *hdr, int type, const char *key) { int i = 0; bcf_hrec_t *hrec; if ( !key ) { while ( inhrec ) { if ( hdr->hrec[i]->type!=type ) { i++; continue; } hrec = hdr->hrec[i]; if ( type==BCF_HL_FLT || type==BCF_HL_INFO || type==BCF_HL_FMT || type== BCF_HL_CTG ) { int j = bcf_hrec_find_key(hdr->hrec[i], "ID"); if ( j>0 ) { vdict_t *d = type==BCF_HL_CTG ? (vdict_t*)hdr->dict[BCF_DT_CTG] : (vdict_t*)hdr->dict[BCF_DT_ID]; khint_t k = kh_get(vdict, d, hdr->hrec[i]->vals[j]); kh_val(d, k).hrec[type==BCF_HL_CTG?0:type] = NULL; } } hdr->dirty = 1; hdr->nhrec--; if ( i < hdr->nhrec ) memmove(&hdr->hrec[i],&hdr->hrec[i+1],(hdr->nhrec-i)*sizeof(bcf_hrec_t*)); bcf_hrec_destroy(hrec); } return; } while (1) { if ( type==BCF_HL_FLT || type==BCF_HL_INFO || type==BCF_HL_FMT || type== BCF_HL_CTG ) { hrec = bcf_hdr_get_hrec(hdr, type, "ID", key, NULL); if ( !hrec ) return; for (i=0; inhrec; i++) if ( hdr->hrec[i]==hrec ) break; assert( inhrec ); vdict_t *d = type==BCF_HL_CTG ? (vdict_t*)hdr->dict[BCF_DT_CTG] : (vdict_t*)hdr->dict[BCF_DT_ID]; khint_t k = kh_get(vdict, d, key); kh_val(d, k).hrec[type==BCF_HL_CTG?0:type] = NULL; } else { for (i=0; inhrec; i++) { if ( hdr->hrec[i]->type!=type ) continue; if ( type==BCF_HL_GEN ) { if ( !strcmp(hdr->hrec[i]->key,key) ) break; } else { // not all structured lines have ID, we could be more sophisticated as in bcf_hdr_get_hrec() int j = bcf_hrec_find_key(hdr->hrec[i], "ID"); if ( j>=0 && !strcmp(hdr->hrec[i]->vals[j],key) ) break; } } if ( i==hdr->nhrec ) return; hrec = hdr->hrec[i]; } hdr->nhrec--; if ( i < hdr->nhrec ) memmove(&hdr->hrec[i],&hdr->hrec[i+1],(hdr->nhrec-i)*sizeof(bcf_hrec_t*)); bcf_hrec_destroy(hrec); hdr->dirty = 1; } } int bcf_hdr_printf(bcf_hdr_t *hdr, const char *fmt, ...) { va_list ap; va_start(ap, fmt); int n = vsnprintf(NULL, 0, fmt, ap) + 2; va_end(ap); char *line = (char*)malloc(n); va_start(ap, fmt); vsnprintf(line, n, fmt, ap); va_end(ap); int ret = bcf_hdr_append(hdr, line); free(line); return ret; } /********************** *** BCF header I/O *** **********************/ const char *bcf_hdr_get_version(const bcf_hdr_t *hdr) { bcf_hrec_t *hrec = bcf_hdr_get_hrec(hdr, BCF_HL_GEN, "fileformat", NULL, NULL); if ( !hrec ) { fprintf(stderr,"No version string found, assuming VCFv4.2\n"); return "VCFv4.2"; } return hrec->value; } void bcf_hdr_set_version(bcf_hdr_t *hdr, const char *version) { bcf_hrec_t *hrec = bcf_hdr_get_hrec(hdr, BCF_HL_GEN, "fileformat", NULL, NULL); if ( !hrec ) { int len; kstring_t str = {0,0,0}; ksprintf(&str,"##fileformat=%s", version); hrec = bcf_hdr_parse_line(hdr, str.s, &len); free(str.s); } else { free(hrec->value); hrec->value = strdup(version); } hdr->dirty = 1; } bcf_hdr_t *bcf_hdr_init(const char *mode) { int i; bcf_hdr_t *h; h = (bcf_hdr_t*)calloc(1, sizeof(bcf_hdr_t)); if (!h) return NULL; for (i = 0; i < 3; ++i) if ((h->dict[i] = kh_init(vdict)) == NULL) goto fail; if ( strchr(mode,'w') ) { bcf_hdr_append(h, "##fileformat=VCFv4.2"); // The filter PASS must appear first in the dictionary bcf_hdr_append(h, "##FILTER="); } return h; fail: for (i = 0; i < 3; ++i) kh_destroy(vdict, h->dict[i]); free(h); return NULL; } void bcf_hdr_destroy(bcf_hdr_t *h) { int i; khint_t k; for (i = 0; i < 3; ++i) { vdict_t *d = (vdict_t*)h->dict[i]; if (d == 0) continue; for (k = kh_begin(d); k != kh_end(d); ++k) if (kh_exist(d, k)) free((char*)kh_key(d, k)); kh_destroy(vdict, d); free(h->id[i]); } for (i=0; inhrec; i++) bcf_hrec_destroy(h->hrec[i]); if (h->nhrec) free(h->hrec); if (h->samples) free(h->samples); free(h->keep_samples); free(h->transl[0]); free(h->transl[1]); free(h->mem.s); free(h); } bcf_hdr_t *bcf_hdr_read(htsFile *hfp) { if (hfp->format.format == vcf) return vcf_hdr_read(hfp); BGZF *fp = hfp->fp.bgzf; uint8_t magic[5]; bcf_hdr_t *h; h = bcf_hdr_init("r"); if (!h) { fprintf(stderr, "[E::%s] failed to allocate bcf header\n", __func__); return NULL; } if (bgzf_read(fp, magic, 5) != 5) { fprintf(stderr,"[%s:%d %s] Failed to read the header (reading BCF in text mode?)\n", __FILE__,__LINE__,__FUNCTION__); bcf_hdr_destroy(h); return NULL; } if (strncmp((char*)magic, "BCF\2\2", 5) != 0) { if (!strncmp((char*)magic, "BCF", 3)) fprintf(stderr,"[%s:%d %s] invalid BCF2 magic string: only BCFv2.2 is supported.\n", __FILE__,__LINE__,__FUNCTION__); else if (hts_verbose >= 2) fprintf(stderr, "[E::%s] invalid BCF2 magic string\n", __func__); bcf_hdr_destroy(h); return NULL; } int hlen; char *htxt = NULL; if (bgzf_read(fp, &hlen, 4) != 4) goto fail; htxt = (char*)malloc(hlen); if (!htxt) goto fail; if (bgzf_read(fp, htxt, hlen) != hlen) goto fail; bcf_hdr_parse(h, htxt); // FIXME: Does this return anything meaningful? free(htxt); return h; fail: if (hts_verbose >= 2) { fprintf(stderr, "[E::%s] failed to read BCF header\n", __func__); } free(htxt); bcf_hdr_destroy(h); return NULL; } int bcf_hdr_write(htsFile *hfp, bcf_hdr_t *h) { if ( h->dirty ) bcf_hdr_sync(h); if (hfp->format.format == vcf || hfp->format.format == text_format) return vcf_hdr_write(hfp, h); int hlen; char *htxt = bcf_hdr_fmt_text(h, 1, &hlen); hlen++; // include the \0 byte BGZF *fp = hfp->fp.bgzf; if ( bgzf_write(fp, "BCF\2\2", 5) !=5 ) return -1; if ( bgzf_write(fp, &hlen, 4) !=4 ) return -1; if ( bgzf_write(fp, htxt, hlen) != hlen ) return -1; free(htxt); return 0; } /******************** *** BCF site I/O *** ********************/ bcf1_t *bcf_init() { bcf1_t *v; v = (bcf1_t*)calloc(1, sizeof(bcf1_t)); return v; } void bcf_clear(bcf1_t *v) { int i; for (i=0; id.m_info; i++) { if ( v->d.info[i].vptr_free ) { free(v->d.info[i].vptr - v->d.info[i].vptr_off); v->d.info[i].vptr_free = 0; } } for (i=0; id.m_fmt; i++) { if ( v->d.fmt[i].p_free ) { free(v->d.fmt[i].p - v->d.fmt[i].p_off); v->d.fmt[i].p_free = 0; } } v->rid = v->pos = v->rlen = v->unpacked = 0; bcf_float_set_missing(v->qual); v->n_info = v->n_allele = v->n_fmt = v->n_sample = 0; v->shared.l = v->indiv.l = 0; v->d.var_type = -1; v->d.shared_dirty = 0; v->d.indiv_dirty = 0; v->d.n_flt = 0; v->errcode = 0; if (v->d.m_als) v->d.als[0] = 0; if (v->d.m_id) v->d.id[0] = 0; } void bcf_empty(bcf1_t *v) { bcf_clear1(v); free(v->d.id); free(v->d.als); free(v->d.allele); free(v->d.flt); free(v->d.info); free(v->d.fmt); if (v->d.var ) free(v->d.var); free(v->shared.s); free(v->indiv.s); } void bcf_destroy(bcf1_t *v) { bcf_empty1(v); free(v); } static inline int bcf_read1_core(BGZF *fp, bcf1_t *v) { uint32_t x[8]; int ret; if ((ret = bgzf_read(fp, x, 32)) != 32) { if (ret == 0) return -1; return -2; } bcf_clear1(v); x[0] -= 24; // to exclude six 32-bit integers ks_resize(&v->shared, x[0]); ks_resize(&v->indiv, x[1]); memcpy(v, x + 2, 16); v->n_allele = x[6]>>16; v->n_info = x[6]&0xffff; v->n_fmt = x[7]>>24; v->n_sample = x[7]&0xffffff; v->shared.l = x[0], v->indiv.l = x[1]; // silent fix of broken BCFs produced by earlier versions of bcf_subset, prior to and including bd6ed8b4 if ( (!v->indiv.l || !v->n_sample) && v->n_fmt ) v->n_fmt = 0; if (bgzf_read(fp, v->shared.s, v->shared.l) != v->shared.l) return -1; if (bgzf_read(fp, v->indiv.s, v->indiv.l) != v->indiv.l) return -1; return 0; } #define bit_array_size(n) ((n)/8+1) #define bit_array_set(a,i) ((a)[(i)/8] |= 1 << ((i)%8)) #define bit_array_clear(a,i) ((a)[(i)/8] &= ~(1 << ((i)%8))) #define bit_array_test(a,i) ((a)[(i)/8] & (1 << ((i)%8))) static inline uint8_t *bcf_unpack_fmt_core1(uint8_t *ptr, int n_sample, bcf_fmt_t *fmt); int bcf_subset_format(const bcf_hdr_t *hdr, bcf1_t *rec) { if ( !hdr->keep_samples ) return 0; if ( !bcf_hdr_nsamples(hdr) ) { rec->indiv.l = rec->n_sample = 0; return 0; } int i, j; uint8_t *ptr = (uint8_t*)rec->indiv.s, *dst = NULL, *src; bcf_dec_t *dec = &rec->d; hts_expand(bcf_fmt_t, rec->n_fmt, dec->m_fmt, dec->fmt); for (i=0; im_fmt; ++i) dec->fmt[i].p_free = 0; for (i=0; in_fmt; i++) { ptr = bcf_unpack_fmt_core1(ptr, rec->n_sample, &dec->fmt[i]); src = dec->fmt[i].p - dec->fmt[i].size; if ( dst ) { memmove(dec->fmt[i-1].p + dec->fmt[i-1].p_len, dec->fmt[i].p - dec->fmt[i].p_off, dec->fmt[i].p_off); dec->fmt[i].p = dec->fmt[i-1].p + dec->fmt[i-1].p_len + dec->fmt[i].p_off; } dst = dec->fmt[i].p; for (j=0; jnsamples_ori; j++) { src += dec->fmt[i].size; if ( !bit_array_test(hdr->keep_samples,j) ) continue; memmove(dst, src, dec->fmt[i].size); dst += dec->fmt[i].size; } rec->indiv.l -= dec->fmt[i].p_len - (dst - dec->fmt[i].p); dec->fmt[i].p_len = dst - dec->fmt[i].p; } rec->unpacked |= BCF_UN_FMT; rec->n_sample = bcf_hdr_nsamples(hdr); return 0; } int bcf_read(htsFile *fp, const bcf_hdr_t *h, bcf1_t *v) { if (fp->format.format == vcf) return vcf_read(fp,h,v); int ret = bcf_read1_core(fp->fp.bgzf, v); if ( ret!=0 || !h->keep_samples ) return ret; return bcf_subset_format(h,v); } int bcf_readrec(BGZF *fp, void *null, void *vv, int *tid, int *beg, int *end) { bcf1_t *v = (bcf1_t *) vv; int ret; if ((ret = bcf_read1_core(fp, v)) >= 0) *tid = v->rid, *beg = v->pos, *end = v->pos + v->rlen; return ret; } static inline void bcf1_sync_id(bcf1_t *line, kstring_t *str) { // single typed string if ( line->d.id && strcmp(line->d.id, ".") ) bcf_enc_vchar(str, strlen(line->d.id), line->d.id); else bcf_enc_size(str, 0, BCF_BT_CHAR); } static inline void bcf1_sync_alleles(bcf1_t *line, kstring_t *str) { // list of typed strings int i; for (i=0; in_allele; i++) bcf_enc_vchar(str, strlen(line->d.allele[i]), line->d.allele[i]); if ( !line->rlen && line->n_allele ) line->rlen = strlen(line->d.allele[0]); } static inline void bcf1_sync_filter(bcf1_t *line, kstring_t *str) { // typed vector of integers if ( line->d.n_flt ) bcf_enc_vint(str, line->d.n_flt, line->d.flt, -1); else bcf_enc_vint(str, 0, 0, -1); } static inline void bcf1_sync_info(bcf1_t *line, kstring_t *str) { // pairs of typed vectors int i, irm = -1; for (i=0; in_info; i++) { bcf_info_t *info = &line->d.info[i]; if ( !info->vptr ) { // marked for removal if ( irm < 0 ) irm = i; continue; } kputsn_(info->vptr - info->vptr_off, info->vptr_len + info->vptr_off, str); if ( irm >=0 ) { bcf_info_t tmp = line->d.info[irm]; line->d.info[irm] = line->d.info[i]; line->d.info[i] = tmp; while ( irm<=i && line->d.info[irm].vptr ) irm++; } } if ( irm>=0 ) line->n_info = irm; } static int bcf1_sync(bcf1_t *line) { char *shared_ori = line->shared.s; size_t prev_len; kstring_t tmp = {0,0,0}; if ( !line->shared.l ) { // New line created via API, BCF data blocks do not exist. Get it ready for BCF output tmp = line->shared; bcf1_sync_id(line, &tmp); line->unpack_size[0] = tmp.l; prev_len = tmp.l; bcf1_sync_alleles(line, &tmp); line->unpack_size[1] = tmp.l - prev_len; prev_len = tmp.l; bcf1_sync_filter(line, &tmp); line->unpack_size[2] = tmp.l - prev_len; bcf1_sync_info(line, &tmp); line->shared = tmp; } else if ( line->d.shared_dirty ) { // The line was edited, update the BCF data block. if ( !(line->unpacked & BCF_UN_STR) ) bcf_unpack(line,BCF_UN_STR); // ptr_ori points to the original unchanged BCF data. uint8_t *ptr_ori = (uint8_t *) line->shared.s; // ID: single typed string if ( line->d.shared_dirty & BCF1_DIRTY_ID ) bcf1_sync_id(line, &tmp); else kputsn_(ptr_ori, line->unpack_size[0], &tmp); ptr_ori += line->unpack_size[0]; line->unpack_size[0] = tmp.l; prev_len = tmp.l; // REF+ALT: list of typed strings if ( line->d.shared_dirty & BCF1_DIRTY_ALS ) bcf1_sync_alleles(line, &tmp); else { kputsn_(ptr_ori, line->unpack_size[1], &tmp); if ( !line->rlen && line->n_allele ) line->rlen = strlen(line->d.allele[0]); } ptr_ori += line->unpack_size[1]; line->unpack_size[1] = tmp.l - prev_len; prev_len = tmp.l; if ( line->unpacked & BCF_UN_FLT ) { // FILTER: typed vector of integers if ( line->d.shared_dirty & BCF1_DIRTY_FLT ) bcf1_sync_filter(line, &tmp); else if ( line->d.n_flt ) kputsn_(ptr_ori, line->unpack_size[2], &tmp); else bcf_enc_vint(&tmp, 0, 0, -1); ptr_ori += line->unpack_size[2]; line->unpack_size[2] = tmp.l - prev_len; if ( line->unpacked & BCF_UN_INFO ) { // INFO: pairs of typed vectors if ( line->d.shared_dirty & BCF1_DIRTY_INF ) { bcf1_sync_info(line, &tmp); ptr_ori = (uint8_t*)line->shared.s + line->shared.l; } } } int size = line->shared.l - (size_t)ptr_ori + (size_t)line->shared.s; if ( size ) kputsn_(ptr_ori, size, &tmp); free(line->shared.s); line->shared = tmp; } if ( line->shared.s != shared_ori && line->unpacked & BCF_UN_INFO ) { // Reallocated line->shared.s block invalidated line->d.info[].vptr pointers size_t off_new = line->unpack_size[0] + line->unpack_size[1] + line->unpack_size[2]; int i; for (i=0; in_info; i++) { uint8_t *vptr_free = line->d.info[i].vptr_free ? line->d.info[i].vptr - line->d.info[i].vptr_off : NULL; line->d.info[i].vptr = (uint8_t*) line->shared.s + off_new + line->d.info[i].vptr_off; off_new += line->d.info[i].vptr_len + line->d.info[i].vptr_off; if ( vptr_free ) { free(vptr_free); line->d.info[i].vptr_free = 0; } } } if ( line->n_sample && line->n_fmt && (!line->indiv.l || line->d.indiv_dirty) ) { // The genotype fields changed or are not present tmp.l = tmp.m = 0; tmp.s = NULL; int i, irm = -1; for (i=0; in_fmt; i++) { bcf_fmt_t *fmt = &line->d.fmt[i]; if ( !fmt->p ) { // marked for removal if ( irm < 0 ) irm = i; continue; } kputsn_(fmt->p - fmt->p_off, fmt->p_len + fmt->p_off, &tmp); if ( irm >=0 ) { bcf_fmt_t tfmt = line->d.fmt[irm]; line->d.fmt[irm] = line->d.fmt[i]; line->d.fmt[i] = tfmt; while ( irm<=i && line->d.fmt[irm].p ) irm++; } } if ( irm>=0 ) line->n_fmt = irm; free(line->indiv.s); line->indiv = tmp; // Reallocated line->indiv.s block invalidated line->d.fmt[].p pointers size_t off_new = 0; for (i=0; in_fmt; i++) { uint8_t *p_free = line->d.fmt[i].p_free ? line->d.fmt[i].p - line->d.fmt[i].p_off : NULL; line->d.fmt[i].p = (uint8_t*) line->indiv.s + off_new + line->d.fmt[i].p_off; off_new += line->d.fmt[i].p_len + line->d.fmt[i].p_off; if ( p_free ) { free(p_free); line->d.fmt[i].p_free = 0; } } } if ( !line->n_sample ) line->n_fmt = 0; line->d.shared_dirty = line->d.indiv_dirty = 0; return 0; } bcf1_t *bcf_copy(bcf1_t *dst, bcf1_t *src) { bcf1_sync(src); bcf_clear(dst); dst->rid = src->rid; dst->pos = src->pos; dst->rlen = src->rlen; dst->qual = src->qual; dst->n_info = src->n_info; dst->n_allele = src->n_allele; dst->n_fmt = src->n_fmt; dst->n_sample = src->n_sample; dst->shared.m = dst->shared.l = src->shared.l; dst->shared.s = (char*) malloc(dst->shared.l); memcpy(dst->shared.s,src->shared.s,dst->shared.l); dst->indiv.m = dst->indiv.l = src->indiv.l; dst->indiv.s = (char*) malloc(dst->indiv.l); memcpy(dst->indiv.s,src->indiv.s,dst->indiv.l); return dst; } bcf1_t *bcf_dup(bcf1_t *src) { bcf1_t *out = bcf_init1(); return bcf_copy(out, src); } int bcf_write(htsFile *hfp, bcf_hdr_t *h, bcf1_t *v) { if ( h->dirty ) bcf_hdr_sync(h); if ( bcf_hdr_nsamples(h)!=v->n_sample ) { fprintf(stderr,"[%s:%d %s] Broken VCF record, the number of columns at %s:%d does not match the number of samples (%d vs %d).\n", __FILE__,__LINE__,__FUNCTION__,bcf_seqname(h,v),v->pos+1, v->n_sample,bcf_hdr_nsamples(h)); return -1; } if ( hfp->format.format == vcf || hfp->format.format == text_format ) return vcf_write(hfp,h,v); if ( v->errcode ) { // vcf_parse1() encountered a new contig or tag, undeclared in the // header. At this point, the header must have been printed, // proceeding would lead to a broken BCF file. Errors must be checked // and cleared by the caller before we can proceed. fprintf(stderr,"[%s:%d %s] Unchecked error (%d), exiting.\n", __FILE__,__LINE__,__FUNCTION__,v->errcode); exit(1); } bcf1_sync(v); // check if the BCF record was modified BGZF *fp = hfp->fp.bgzf; uint32_t x[8]; x[0] = v->shared.l + 24; // to include six 32-bit integers x[1] = v->indiv.l; memcpy(x + 2, v, 16); x[6] = (uint32_t)v->n_allele<<16 | v->n_info; x[7] = (uint32_t)v->n_fmt<<24 | v->n_sample; if ( bgzf_write(fp, x, 32) != 32 ) return -1; if ( bgzf_write(fp, v->shared.s, v->shared.l) != v->shared.l ) return -1; if ( bgzf_write(fp, v->indiv.s, v->indiv.l) != v->indiv.l ) return -1; return 0; } /********************** *** VCF header I/O *** **********************/ bcf_hdr_t *vcf_hdr_read(htsFile *fp) { kstring_t txt, *s = &fp->line; bcf_hdr_t *h; h = bcf_hdr_init("r"); if (!h) { fprintf(stderr, "[E::%s] failed to allocate bcf header\n", __func__); return NULL; } txt.l = txt.m = 0; txt.s = 0; while (hts_getline(fp, KS_SEP_LINE, s) >= 0) { if (s->l == 0) continue; if (s->s[0] != '#') { if (hts_verbose >= 2) fprintf(stderr, "[E::%s] no sample line\n", __func__); free(txt.s); bcf_hdr_destroy(h); return NULL; } if (s->s[1] != '#' && fp->fn_aux) { // insert contigs here int dret; gzFile f; kstream_t *ks; kstring_t tmp; tmp.l = tmp.m = 0; tmp.s = 0; f = gzopen(fp->fn_aux, "r"); ks = ks_init(f); while (ks_getuntil(ks, 0, &tmp, &dret) >= 0) { int c; kputs("##contig=\n", 2, &txt); if (dret != '\n') while ((c = ks_getc(ks)) != '\n' && c != -1); // skip the rest of the line } free(tmp.s); ks_destroy(ks); gzclose(f); } kputsn(s->s, s->l, &txt); kputc('\n', &txt); if (s->s[1] != '#') break; } if ( !txt.s ) { fprintf(stderr,"[%s:%d %s] Could not read the header\n", __FILE__,__LINE__,__FUNCTION__); return NULL; } bcf_hdr_parse(h, txt.s); // check tabix index, are all contigs listed in the header? add the missing ones tbx_t *idx = tbx_index_load(fp->fn); if ( idx ) { int i, n, need_sync = 0; const char **names = tbx_seqnames(idx, &n); for (i=0; ikey = strdup("contig"); bcf_hrec_add_key(hrec, "ID", strlen("ID")); bcf_hrec_set_val(hrec, hrec->nkeys-1, (char*) names[i], strlen(names[i]), 0); bcf_hdr_add_hrec(h, hrec); need_sync = 1; } free(names); tbx_destroy(idx); if ( need_sync ) bcf_hdr_sync(h); } free(txt.s); return h; } int bcf_hdr_set(bcf_hdr_t *hdr, const char *fname) { int i, n; char **lines = hts_readlines(fname, &n); if ( !lines ) return 1; for (i=0; ivalue ) { int j, nout = 0; ksprintf(str, "##%s=<", hrec->key); for (j=0; jnkeys; j++) { // do not output IDX if output is VCF if ( !is_bcf && !strcmp("IDX",hrec->keys[j]) ) continue; if ( nout ) kputc(',',str); ksprintf(str,"%s=%s", hrec->keys[j], hrec->vals[j]); nout++; } ksprintf(str,">\n"); } else ksprintf(str,"##%s=%s\n", hrec->key,hrec->value); } void bcf_hrec_format(const bcf_hrec_t *hrec, kstring_t *str) { _bcf_hrec_format(hrec,0,str); } char *bcf_hdr_fmt_text(const bcf_hdr_t *hdr, int is_bcf, int *len) { int i; kstring_t txt = {0,0,0}; for (i=0; inhrec; i++) _bcf_hrec_format(hdr->hrec[i], is_bcf, &txt); ksprintf(&txt,"#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO"); if ( bcf_hdr_nsamples(hdr) ) { ksprintf(&txt,"\tFORMAT"); for (i=0; isamples[i]); } ksprintf(&txt,"\n"); if ( len ) *len = txt.l; return txt.s; } const char **bcf_hdr_seqnames(const bcf_hdr_t *h, int *n) { vdict_t *d = (vdict_t*)h->dict[BCF_DT_CTG]; int tid, m = kh_size(d); const char **names = (const char**) calloc(m,sizeof(const char*)); khint_t k; for (k=kh_begin(d); kformat.compression!=no_compression ) ret = bgzf_write(fp->fp.bgzf, htxt, hlen); else ret = hwrite(fp->fp.hfile, htxt, hlen); free(htxt); return ret<0 ? -1 : 0; } /*********************** *** Typed value I/O *** ***********************/ void bcf_enc_vint(kstring_t *s, int n, int32_t *a, int wsize) { int32_t max = INT32_MIN + 1, min = INT32_MAX; int i; if (n == 0) bcf_enc_size(s, 0, BCF_BT_NULL); else if (n == 1) bcf_enc_int1(s, a[0]); else { if (wsize <= 0) wsize = n; for (i = 0; i < n; ++i) { if (a[i] == bcf_int32_missing || a[i] == bcf_int32_vector_end ) continue; if (max < a[i]) max = a[i]; if (min > a[i]) min = a[i]; } if (max <= INT8_MAX && min > bcf_int8_vector_end) { bcf_enc_size(s, wsize, BCF_BT_INT8); for (i = 0; i < n; ++i) if ( a[i]==bcf_int32_vector_end ) kputc(bcf_int8_vector_end, s); else if ( a[i]==bcf_int32_missing ) kputc(bcf_int8_missing, s); else kputc(a[i], s); } else if (max <= INT16_MAX && min > bcf_int16_vector_end) { bcf_enc_size(s, wsize, BCF_BT_INT16); for (i = 0; i < n; ++i) { int16_t x; if ( a[i]==bcf_int32_vector_end ) x = bcf_int16_vector_end; else if ( a[i]==bcf_int32_missing ) x = bcf_int16_missing; else x = a[i]; kputsn((char*)&x, 2, s); } } else { bcf_enc_size(s, wsize, BCF_BT_INT32); for (i = 0; i < n; ++i) { int32_t x = a[i]; kputsn((char*)&x, 4, s); } } } } void bcf_enc_vfloat(kstring_t *s, int n, float *a) { bcf_enc_size(s, n, BCF_BT_FLOAT); kputsn((char*)a, n << 2, s); } void bcf_enc_vchar(kstring_t *s, int l, const char *a) { bcf_enc_size(s, l, BCF_BT_CHAR); kputsn(a, l, s); } void bcf_fmt_array(kstring_t *s, int n, int type, void *data) { int j = 0; if (n == 0) { kputc('.', s); return; } if (type == BCF_BT_CHAR) { char *p = (char*)data; for (j = 0; j < n && *p; ++j, ++p) { if ( *p==bcf_str_missing ) kputc('.', s); else kputc(*p, s); } } else { #define BRANCH(type_t, is_missing, is_vector_end, kprint) { \ type_t *p = (type_t *) data; \ for (j=0; jl&7) { uint64_t zero = 0; int l = ((s->l + 7)>>3<<3) - s->l; kputsn((char*)&zero, l, s); } } // p,q is the start and the end of the FORMAT field #define MAX_N_FMT 255 /* Limited by size of bcf1_t n_fmt field */ static int vcf_parse_format(kstring_t *s, const bcf_hdr_t *h, bcf1_t *v, char *p, char *q) { if ( !bcf_hdr_nsamples(h) ) return 0; char *r, *t; int j, l, m, g; khint_t k; ks_tokaux_t aux1; vdict_t *d = (vdict_t*)h->dict[BCF_DT_ID]; kstring_t *mem = (kstring_t*)&h->mem; fmt_aux_t fmt[MAX_N_FMT]; mem->l = 0; char *end = s->s + s->l; if ( q>=end ) { fprintf(stderr,"[%s:%d %s] Error: FORMAT column with no sample columns starting at %s:%d\n", __FILE__,__LINE__,__FUNCTION__,s->s,v->pos+1); return -1; } // get format information from the dictionary v->n_fmt = 0; for (j = 0, t = kstrtok(p, ":", &aux1); t; t = kstrtok(0, 0, &aux1), ++j) { if (j >= MAX_N_FMT) { v->errcode |= BCF_ERR_LIMITS; fprintf(stderr,"[E::%s] Error: FORMAT column at %s:%d lists more identifiers than htslib can handle.\n", __func__, bcf_seqname(h,v), v->pos+1); return -1; } *(char*)aux1.p = 0; k = kh_get(vdict, d, t); if (k == kh_end(d) || kh_val(d, k).info[BCF_HL_FMT] == 15) { if (hts_verbose >= 2) fprintf(stderr, "[W::%s] FORMAT '%s' is not defined in the header, assuming Type=String\n", __func__, t); kstring_t tmp = {0,0,0}; int l; ksprintf(&tmp, "##FORMAT=", t); bcf_hrec_t *hrec = bcf_hdr_parse_line(h,tmp.s,&l); free(tmp.s); if ( bcf_hdr_add_hrec((bcf_hdr_t*)h, hrec) ) bcf_hdr_sync((bcf_hdr_t*)h); k = kh_get(vdict, d, t); v->errcode = BCF_ERR_TAG_UNDEF; if (k == kh_end(d)) { fprintf(stderr, "[E::%s] Could not add dummy header for FORMAT '%s'\n", __func__, t); v->errcode |= BCF_ERR_TAG_INVALID; return -1; } } fmt[j].max_l = fmt[j].max_m = fmt[j].max_g = 0; fmt[j].key = kh_val(d, k).id; fmt[j].is_gt = !strcmp(t, "GT"); fmt[j].y = h->id[0][fmt[j].key].val->info[BCF_HL_FMT]; v->n_fmt++; } // compute max int n_sample_ori = -1; r = q + 1; // r: position in the format string l = 0, m = g = 1, v->n_sample = 0; // m: max vector size, l: max field len, g: max number of alleles while ( rkeep_samples ) { n_sample_ori++; if ( !bit_array_test(h->keep_samples,n_sample_ori) ) { while ( *r!='\t' && r=v->n_fmt ) { fprintf(stderr,"Incorrect number of FORMAT fields at %s:%d\n", h->id[BCF_DT_CTG][v->rid].key,v->pos+1); exit(1); } } else break; } else if ( *r== ',' ) m++; else if ( fmt[j].is_gt && (*r == '|' || *r == '/') ) g++; if ( r>=end ) break; r++; l++; } v->n_sample++; if ( v->n_sample == bcf_hdr_nsamples(h) ) break; r++; } // allocate memory for arrays for (j = 0; j < v->n_fmt; ++j) { fmt_aux_t *f = &fmt[j]; if ( !f->max_m ) f->max_m = 1; // omitted trailing format field if ((f->y>>4&0xf) == BCF_HT_STR) { f->size = f->is_gt? f->max_g << 2 : f->max_l; } else if ((f->y>>4&0xf) == BCF_HT_REAL || (f->y>>4&0xf) == BCF_HT_INT) { f->size = f->max_m << 2; } else { fprintf(stderr, "[E::%s] the format type %d currently not supported\n", __func__, f->y>>4&0xf); abort(); // I do not know how to do with Flag in the genotype fields } align_mem(mem); f->offset = mem->l; ks_resize(mem, mem->l + v->n_sample * f->size); mem->l += v->n_sample * f->size; } for (j = 0; j < v->n_fmt; ++j) fmt[j].buf = (uint8_t*)mem->s + fmt[j].offset; // fill the sample fields; at beginning of the loop, t points to the first char of a format n_sample_ori = -1; t = q + 1; m = 0; // m: sample id while ( tkeep_samples ) { n_sample_ori++; if ( !bit_array_test(h->keep_samples,n_sample_ori) ) { while ( *t && ty>>4&0xf) == BCF_HT_STR) { if (z->is_gt) { // genotypes int32_t is_phased = 0, *x = (int32_t*)(z->buf + z->size * m); for (l = 0;; ++t) { if (*t == '.') ++t, x[l++] = is_phased; else x[l++] = (strtol(t, &t, 10) + 1) << 1 | is_phased; #if THOROUGH_SANITY_CHECKS assert( 0 ); // success of strtol,strtod not checked #endif is_phased = (*t == '|'); if (*t != '|' && *t != '/') break; } if ( !l ) x[l++] = 0; // An empty field, insert missing value for (; l < z->size>>2; ++l) x[l] = bcf_int32_vector_end; } else { char *x = (char*)z->buf + z->size * m; for (r = t, l = 0; *t != ':' && *t; ++t) x[l++] = *t; for (; l < z->size; ++l) x[l] = 0; } } else if ((z->y>>4&0xf) == BCF_HT_INT) { int32_t *x = (int32_t*)(z->buf + z->size * m); for (l = 0;; ++t) { if (*t == '.') x[l++] = bcf_int32_missing, ++t; // ++t to skip "." else x[l++] = strtol(t, &t, 10); if (*t != ',') break; } if ( !l ) x[l++] = bcf_int32_missing; for (; l < z->size>>2; ++l) x[l] = bcf_int32_vector_end; } else if ((z->y>>4&0xf) == BCF_HT_REAL) { float *x = (float*)(z->buf + z->size * m); for (l = 0;; ++t) { if (*t == '.' && !isdigit(t[1])) bcf_float_set_missing(x[l++]), ++t; // ++t to skip "." else x[l++] = strtod(t, &t); if (*t != ',') break; } if ( !l ) bcf_float_set_missing(x[l++]); // An empty field, insert missing value for (; l < z->size>>2; ++l) bcf_float_set_vector_end(x[l]); } else abort(); if (*t == '\0') { break; } else if (*t == ':') { t++; } else { char buffer[8]; fprintf(stderr,"[E::%s] Invalid character '%s' in '%s' FORMAT field at %s:%d\n", __FUNCTION__, dump_char(buffer, *t), h->id[BCF_DT_ID][z->key].key, bcf_seqname(h,v), v->pos+1); v->errcode |= BCF_ERR_CHAR; return -1; } } for (; j < v->n_fmt; ++j) { // fill end-of-vector values fmt_aux_t *z = &fmt[j]; if ((z->y>>4&0xf) == BCF_HT_STR) { if (z->is_gt) { int32_t *x = (int32_t*)(z->buf + z->size * m); if (z->size) x[0] = bcf_int32_missing; for (l = 1; l < z->size>>2; ++l) x[l] = bcf_int32_vector_end; } else { char *x = (char*)z->buf + z->size * m; if ( z->size ) x[0] = '.'; for (l = 1; l < z->size; ++l) x[l] = 0; } } else if ((z->y>>4&0xf) == BCF_HT_INT) { int32_t *x = (int32_t*)(z->buf + z->size * m); x[0] = bcf_int32_missing; for (l = 1; l < z->size>>2; ++l) x[l] = bcf_int32_vector_end; } else if ((z->y>>4&0xf) == BCF_HT_REAL) { float *x = (float*)(z->buf + z->size * m); bcf_float_set_missing(x[0]); for (l = 1; l < z->size>>2; ++l) bcf_float_set_vector_end(x[l]); } } m++; t++; } // write individual genotype information kstring_t *str = &v->indiv; int i; if (v->n_sample > 0) { for (i = 0; i < v->n_fmt; ++i) { fmt_aux_t *z = &fmt[i]; bcf_enc_int1(str, z->key); if ((z->y>>4&0xf) == BCF_HT_STR && !z->is_gt) { bcf_enc_size(str, z->size, BCF_BT_CHAR); kputsn((char*)z->buf, z->size * v->n_sample, str); } else if ((z->y>>4&0xf) == BCF_HT_INT || z->is_gt) { bcf_enc_vint(str, (z->size>>2) * v->n_sample, (int32_t*)z->buf, z->size>>2); } else { bcf_enc_size(str, z->size>>2, BCF_BT_FLOAT); kputsn((char*)z->buf, z->size * v->n_sample, str); } } } if ( v->n_sample!=bcf_hdr_nsamples(h) ) { fprintf(stderr,"[%s:%d %s] Number of columns at %s:%d does not match the number of samples (%d vs %d).\n", __FILE__,__LINE__,__FUNCTION__,bcf_seqname(h,v),v->pos+1, v->n_sample,bcf_hdr_nsamples(h)); v->errcode |= BCF_ERR_NCOLS; return -1; } if ( v->indiv.l > 0xffffffff ) { fprintf(stderr,"[%s:%d %s] The FORMAT at %s:%d is too long...\n", __FILE__,__LINE__,__FUNCTION__,bcf_seqname(h,v),v->pos+1); v->errcode |= BCF_ERR_LIMITS; // Error recovery: return -1 if this is a critical error or 0 if we want to ignore the FORMAT and proceed v->n_fmt = 0; return -1; } return 0; } int vcf_parse(kstring_t *s, const bcf_hdr_t *h, bcf1_t *v) { int i = 0; char *p, *q, *r, *t; kstring_t *str; khint_t k; ks_tokaux_t aux; bcf_clear1(v); str = &v->shared; memset(&aux, 0, sizeof(ks_tokaux_t)); for (p = kstrtok(s->s, "\t", &aux), i = 0; p; p = kstrtok(0, 0, &aux), ++i) { q = (char*)aux.p; *q = 0; if (i == 0) { // CHROM vdict_t *d = (vdict_t*)h->dict[BCF_DT_CTG]; k = kh_get(vdict, d, p); if (k == kh_end(d)) { // Simple error recovery for chromosomes not defined in the header. It will not help when VCF header has // been already printed, but will enable tools like vcfcheck to proceed. if (hts_verbose >= 2) fprintf(stderr, "[W::%s] contig '%s' is not defined in the header. (Quick workaround: index the file with tabix.)\n", __func__, p); kstring_t tmp = {0,0,0}; int l; ksprintf(&tmp, "##contig=", p); bcf_hrec_t *hrec = bcf_hdr_parse_line(h,tmp.s,&l); free(tmp.s); if ( bcf_hdr_add_hrec((bcf_hdr_t*)h, hrec) ) bcf_hdr_sync((bcf_hdr_t*)h); k = kh_get(vdict, d, p); v->errcode = BCF_ERR_CTG_UNDEF; if (k == kh_end(d)) { fprintf(stderr, "[E::%s] Could not add dummy header for contig '%s'\n", __func__, p); v->errcode |= BCF_ERR_CTG_INVALID; return -1; } } v->rid = kh_val(d, k).id; } else if (i == 1) { // POS v->pos = atoi(p) - 1; } else if (i == 2) { // ID if (strcmp(p, ".")) bcf_enc_vchar(str, q - p, p); else bcf_enc_size(str, 0, BCF_BT_CHAR); } else if (i == 3) { // REF bcf_enc_vchar(str, q - p, p); v->n_allele = 1, v->rlen = q - p; } else if (i == 4) { // ALT if (strcmp(p, ".")) { for (r = t = p;; ++r) { if (*r == ',' || *r == 0) { bcf_enc_vchar(str, r - t, t); t = r + 1; ++v->n_allele; } if (r == q) break; } } } else if (i == 5) { // QUAL if (strcmp(p, ".")) v->qual = atof(p); else memcpy(&v->qual, &bcf_float_missing, 4); if ( v->max_unpack && !(v->max_unpack>>1) ) return 0; // BCF_UN_STR } else if (i == 6) { // FILTER if (strcmp(p, ".")) { int32_t *a; int n_flt = 1, i; ks_tokaux_t aux1; vdict_t *d = (vdict_t*)h->dict[BCF_DT_ID]; // count the number of filters if (*(q-1) == ';') *(q-1) = 0; for (r = p; *r; ++r) if (*r == ';') ++n_flt; a = (int32_t*)alloca(n_flt * sizeof(int32_t)); // add filters for (t = kstrtok(p, ";", &aux1), i = 0; t; t = kstrtok(0, 0, &aux1)) { *(char*)aux1.p = 0; k = kh_get(vdict, d, t); if (k == kh_end(d)) { // Simple error recovery for FILTERs not defined in the header. It will not help when VCF header has // been already printed, but will enable tools like vcfcheck to proceed. if (hts_verbose >= 2) fprintf(stderr, "[W::%s] FILTER '%s' is not defined in the header\n", __func__, t); kstring_t tmp = {0,0,0}; int l; ksprintf(&tmp, "##FILTER=", t); bcf_hrec_t *hrec = bcf_hdr_parse_line(h,tmp.s,&l); free(tmp.s); if ( bcf_hdr_add_hrec((bcf_hdr_t*)h, hrec) ) bcf_hdr_sync((bcf_hdr_t*)h); k = kh_get(vdict, d, t); v->errcode = BCF_ERR_TAG_UNDEF; if (k == kh_end(d)) { fprintf(stderr, "[E::%s] Could not add dummy header for FILTER '%s'\n", __func__, t); v->errcode |= BCF_ERR_TAG_INVALID; return -1; } } a[i++] = kh_val(d, k).id; } n_flt = i; bcf_enc_vint(str, n_flt, a, -1); } else bcf_enc_vint(str, 0, 0, -1); if ( v->max_unpack && !(v->max_unpack>>2) ) return 0; // BCF_UN_FLT } else if (i == 7) { // INFO char *key; vdict_t *d = (vdict_t*)h->dict[BCF_DT_ID]; v->n_info = 0; if (strcmp(p, ".")) { if (*(q-1) == ';') *(q-1) = 0; for (r = key = p;; ++r) { int c; char *val, *end; if (*r != ';' && *r != '=' && *r != 0) continue; val = end = 0; c = *r; *r = 0; if (c == '=') { val = r + 1; for (end = val; *end != ';' && *end != 0; ++end); c = *end; *end = 0; } else end = r; if ( !*key ) { if (c==0) break; r = end; key = r + 1; continue; } // faulty VCF, ";;" in the INFO k = kh_get(vdict, d, key); if (k == kh_end(d) || kh_val(d, k).info[BCF_HL_INFO] == 15) { if (hts_verbose >= 2) fprintf(stderr, "[W::%s] INFO '%s' is not defined in the header, assuming Type=String\n", __func__, key); kstring_t tmp = {0,0,0}; int l; ksprintf(&tmp, "##INFO=", key); bcf_hrec_t *hrec = bcf_hdr_parse_line(h,tmp.s,&l); free(tmp.s); if ( bcf_hdr_add_hrec((bcf_hdr_t*)h, hrec) ) bcf_hdr_sync((bcf_hdr_t*)h); k = kh_get(vdict, d, key); v->errcode = BCF_ERR_TAG_UNDEF; if (k == kh_end(d)) { fprintf(stderr, "[E::%s] Could not add dummy header for INFO '%s'\n", __func__, key); v->errcode |= BCF_ERR_TAG_INVALID; return -1; } } uint32_t y = kh_val(d, k).info[BCF_HL_INFO]; ++v->n_info; bcf_enc_int1(str, kh_val(d, k).id); if (val == 0) { bcf_enc_size(str, 0, BCF_BT_NULL); } else if ((y>>4&0xf) == BCF_HT_FLAG || (y>>4&0xf) == BCF_HT_STR) { // if Flag has a value, treat it as a string bcf_enc_vchar(str, end - val, val); } else { // int/float value/array int i, n_val; char *t, *te; for (t = val, n_val = 1; *t; ++t) // count the number of values if (*t == ',') ++n_val; if ((y>>4&0xf) == BCF_HT_INT) { int32_t *z; z = (int32_t*)alloca(n_val * sizeof(int32_t)); for (i = 0, t = val; i < n_val; ++i, ++t) { z[i] = strtol(t, &te, 10); if ( te==t ) // conversion failed { z[i] = bcf_int32_missing; while ( *te && *te!=',' ) te++; } t = te; } bcf_enc_vint(str, n_val, z, -1); if (strcmp(key, "END") == 0) v->rlen = z[0] - v->pos; } else if ((y>>4&0xf) == BCF_HT_REAL) { float *z; z = (float*)alloca(n_val * sizeof(float)); for (i = 0, t = val; i < n_val; ++i, ++t) { z[i] = strtod(t, &te); if ( te==t ) // conversion failed { bcf_float_set_missing(z[i]); while ( *te && *te!=',' ) te++; } t = te; } bcf_enc_vfloat(str, n_val, z); } } if (c == 0) break; r = end; key = r + 1; } } if ( v->max_unpack && !(v->max_unpack>>3) ) return 0; } else if (i == 8) // FORMAT return vcf_parse_format(s, h, v, p, q); } return 0; } int vcf_read(htsFile *fp, const bcf_hdr_t *h, bcf1_t *v) { int ret; ret = hts_getline(fp, KS_SEP_LINE, &fp->line); if (ret < 0) return -1; return vcf_parse1(&fp->line, h, v); } static inline uint8_t *bcf_unpack_fmt_core1(uint8_t *ptr, int n_sample, bcf_fmt_t *fmt) { uint8_t *ptr_start = ptr; fmt->id = bcf_dec_typed_int1(ptr, &ptr); fmt->n = bcf_dec_size(ptr, &ptr, &fmt->type); fmt->size = fmt->n << bcf_type_shift[fmt->type]; fmt->p = ptr; fmt->p_off = ptr - ptr_start; fmt->p_free = 0; ptr += n_sample * fmt->size; fmt->p_len = ptr - fmt->p; return ptr; } static inline uint8_t *bcf_unpack_info_core1(uint8_t *ptr, bcf_info_t *info) { uint8_t *ptr_start = ptr; info->key = bcf_dec_typed_int1(ptr, &ptr); info->len = bcf_dec_size(ptr, &ptr, &info->type); info->vptr = ptr; info->vptr_off = ptr - ptr_start; info->vptr_free = 0; info->v1.i = 0; if (info->len == 1) { if (info->type == BCF_BT_INT8 || info->type == BCF_BT_CHAR) info->v1.i = *(int8_t*)ptr; else if (info->type == BCF_BT_INT32) info->v1.i = *(int32_t*)ptr; else if (info->type == BCF_BT_FLOAT) info->v1.f = *(float*)ptr; else if (info->type == BCF_BT_INT16) info->v1.i = *(int16_t*)ptr; } ptr += info->len << bcf_type_shift[info->type]; info->vptr_len = ptr - info->vptr; return ptr; } int bcf_unpack(bcf1_t *b, int which) { if ( !b->shared.l ) return 0; // Building a new BCF record from scratch uint8_t *ptr = (uint8_t*)b->shared.s, *ptr_ori; int *offset, i; bcf_dec_t *d = &b->d; if (which & BCF_UN_FLT) which |= BCF_UN_STR; if (which & BCF_UN_INFO) which |= BCF_UN_SHR; if ((which&BCF_UN_STR) && !(b->unpacked&BCF_UN_STR)) { kstring_t tmp; // ID tmp.l = 0; tmp.s = d->id; tmp.m = d->m_id; ptr_ori = ptr; ptr = bcf_fmt_sized_array(&tmp, ptr); b->unpack_size[0] = ptr - ptr_ori; kputc('\0', &tmp); d->id = tmp.s; d->m_id = tmp.m; // REF and ALT are in a single block (d->als) and d->alleles are pointers into this block tmp.l = 0; tmp.s = d->als; tmp.m = d->m_als; offset = (int*)alloca(b->n_allele * sizeof(int)); ptr_ori = ptr; for (i = 0; i < b->n_allele; ++i) { offset[i] = tmp.l; ptr = bcf_fmt_sized_array(&tmp, ptr); kputc('\0', &tmp); } b->unpack_size[1] = ptr - ptr_ori; d->als = tmp.s; d->m_als = tmp.m; hts_expand(char*, b->n_allele, d->m_allele, d->allele); // NM: hts_expand() is a macro for (i = 0; i < b->n_allele; ++i) d->allele[i] = d->als + offset[i]; b->unpacked |= BCF_UN_STR; } if ((which&BCF_UN_FLT) && !(b->unpacked&BCF_UN_FLT)) { // FILTER ptr = (uint8_t*)b->shared.s + b->unpack_size[0] + b->unpack_size[1]; ptr_ori = ptr; if (*ptr>>4) { int type; d->n_flt = bcf_dec_size(ptr, &ptr, &type); hts_expand(int, d->n_flt, d->m_flt, d->flt); for (i = 0; i < d->n_flt; ++i) d->flt[i] = bcf_dec_int1(ptr, type, &ptr); } else ++ptr, d->n_flt = 0; b->unpack_size[2] = ptr - ptr_ori; b->unpacked |= BCF_UN_FLT; } if ((which&BCF_UN_INFO) && !(b->unpacked&BCF_UN_INFO)) { // INFO ptr = (uint8_t*)b->shared.s + b->unpack_size[0] + b->unpack_size[1] + b->unpack_size[2]; hts_expand(bcf_info_t, b->n_info, d->m_info, d->info); for (i = 0; i < d->m_info; ++i) d->info[i].vptr_free = 0; for (i = 0; i < b->n_info; ++i) ptr = bcf_unpack_info_core1(ptr, &d->info[i]); b->unpacked |= BCF_UN_INFO; } if ((which&BCF_UN_FMT) && b->n_sample && !(b->unpacked&BCF_UN_FMT)) { // FORMAT ptr = (uint8_t*)b->indiv.s; hts_expand(bcf_fmt_t, b->n_fmt, d->m_fmt, d->fmt); for (i = 0; i < d->m_fmt; ++i) d->fmt[i].p_free = 0; for (i = 0; i < b->n_fmt; ++i) ptr = bcf_unpack_fmt_core1(ptr, b->n_sample, &d->fmt[i]); b->unpacked |= BCF_UN_FMT; } return 0; } int vcf_format(const bcf_hdr_t *h, const bcf1_t *v, kstring_t *s) { int i; bcf_unpack((bcf1_t*)v, BCF_UN_ALL); kputs(h->id[BCF_DT_CTG][v->rid].key, s); // CHROM kputc('\t', s); kputw(v->pos + 1, s); // POS kputc('\t', s); kputs(v->d.id ? v->d.id : ".", s); // ID kputc('\t', s); // REF if (v->n_allele > 0) kputs(v->d.allele[0], s); else kputc('.', s); kputc('\t', s); // ALT if (v->n_allele > 1) { for (i = 1; i < v->n_allele; ++i) { if (i > 1) kputc(',', s); kputs(v->d.allele[i], s); } } else kputc('.', s); kputc('\t', s); // QUAL if ( bcf_float_is_missing(v->qual) ) kputc('.', s); // QUAL else ksprintf(s, "%g", v->qual); kputc('\t', s); // FILTER if (v->d.n_flt) { for (i = 0; i < v->d.n_flt; ++i) { if (i) kputc(';', s); kputs(h->id[BCF_DT_ID][v->d.flt[i]].key, s); } } else kputc('.', s); kputc('\t', s); // INFO if (v->n_info) { int first = 1; for (i = 0; i < v->n_info; ++i) { bcf_info_t *z = &v->d.info[i]; if ( !z->vptr ) continue; if ( !first ) kputc(';', s); first = 0; kputs(h->id[BCF_DT_ID][z->key].key, s); if (z->len <= 0) continue; kputc('=', s); if (z->len == 1) { switch (z->type) { case BCF_BT_INT8: if ( z->v1.i==bcf_int8_missing ) kputc('.', s); else kputw(z->v1.i, s); break; case BCF_BT_INT16: if ( z->v1.i==bcf_int16_missing ) kputc('.', s); else kputw(z->v1.i, s); break; case BCF_BT_INT32: if ( z->v1.i==bcf_int32_missing ) kputc('.', s); else kputw(z->v1.i, s); break; case BCF_BT_FLOAT: if ( bcf_float_is_missing(z->v1.f) ) kputc('.', s); else ksprintf(s, "%g", z->v1.f); break; case BCF_BT_CHAR: kputc(z->v1.i, s); break; default: fprintf(stderr,"todo: type %d\n", z->type); exit(1); break; } } else bcf_fmt_array(s, z->len, z->type, z->vptr); } if ( first ) kputc('.', s); } else kputc('.', s); // FORMAT and individual information if (v->n_sample) { int i,j; if ( v->n_fmt) { int gt_i = -1; bcf_fmt_t *fmt = v->d.fmt; int first = 1; for (i = 0; i < (int)v->n_fmt; ++i) { if ( !fmt[i].p ) continue; kputc(!first ? ':' : '\t', s); first = 0; if ( fmt[i].id<0 ) //!bcf_hdr_idinfo_exists(h,BCF_HL_FMT,fmt[i].id) ) { fprintf(stderr, "[E::%s] invalid BCF, the FORMAT tag id=%d not present in the header.\n", __func__, fmt[i].id); abort(); } kputs(h->id[BCF_DT_ID][fmt[i].id].key, s); if (strcmp(h->id[BCF_DT_ID][fmt[i].id].key, "GT") == 0) gt_i = i; } if ( first ) kputs("\t.", s); for (j = 0; j < v->n_sample; ++j) { kputc('\t', s); first = 1; for (i = 0; i < (int)v->n_fmt; ++i) { bcf_fmt_t *f = &fmt[i]; if ( !f->p ) continue; if (!first) kputc(':', s); first = 0; if (gt_i == i) bcf_format_gt(f,j,s); else bcf_fmt_array(s, f->n, f->type, f->p + j * f->size); } if ( first ) kputc('.', s); } } else for (j=0; j<=v->n_sample; j++) kputs("\t.", s); } kputc('\n', s); return 0; } int vcf_write_line(htsFile *fp, kstring_t *line) { int ret; if ( line->s[line->l-1]!='\n' ) kputc('\n',line); if ( fp->format.compression!=no_compression ) ret = bgzf_write(fp->fp.bgzf, line->s, line->l); else ret = hwrite(fp->fp.hfile, line->s, line->l); return ret==line->l ? 0 : -1; } int vcf_write(htsFile *fp, const bcf_hdr_t *h, bcf1_t *v) { int ret; fp->line.l = 0; vcf_format1(h, v, &fp->line); if ( fp->format.compression!=no_compression ) ret = bgzf_write(fp->fp.bgzf, fp->line.s, fp->line.l); else ret = hwrite(fp->fp.hfile, fp->line.s, fp->line.l); return ret==fp->line.l ? 0 : -1; } /************************ * Data access routines * ************************/ int bcf_hdr_id2int(const bcf_hdr_t *h, int which, const char *id) { khint_t k; vdict_t *d = (vdict_t*)h->dict[which]; k = kh_get(vdict, d, id); return k == kh_end(d)? -1 : kh_val(d, k).id; } /******************** *** BCF indexing *** ********************/ hts_idx_t *bcf_index(htsFile *fp, int min_shift) { int n_lvls, i; bcf1_t *b; hts_idx_t *idx; bcf_hdr_t *h; int64_t max_len = 0, s; h = bcf_hdr_read(fp); if ( !h ) return NULL; int nids = 0; for (i = 0; i < h->n[BCF_DT_CTG]; ++i) { if ( !h->id[BCF_DT_CTG][i].val ) continue; if ( max_len < h->id[BCF_DT_CTG][i].val->info[0] ) max_len = h->id[BCF_DT_CTG][i].val->info[0]; nids++; } if ( !max_len ) max_len = ((int64_t)1<<31) - 1; // In case contig line is broken. max_len += 256; for (n_lvls = 0, s = 1< s; ++n_lvls, s <<= 3); idx = hts_idx_init(nids, HTS_FMT_CSI, bgzf_tell(fp->fp.bgzf), min_shift, n_lvls); b = bcf_init1(); while (bcf_read1(fp,h, b) >= 0) { int ret; ret = hts_idx_push(idx, b->rid, b->pos, b->pos + b->rlen, bgzf_tell(fp->fp.bgzf), 1); if (ret < 0) { bcf_destroy1(b); hts_idx_destroy(idx); return NULL; } } hts_idx_finish(idx, bgzf_tell(fp->fp.bgzf)); bcf_destroy1(b); bcf_hdr_destroy(h); return idx; } hts_idx_t *bcf_index_load2(const char *fn, const char *fnidx) { return fnidx? hts_idx_load2(fn, fnidx) : bcf_index_load(fn); } int bcf_index_build2(const char *fn, const char *fnidx, int min_shift) { htsFile *fp; hts_idx_t *idx; int ret; if ((fp = hts_open(fn, "rb")) == 0) return -1; if ( fp->format.compression!=bgzf ) { hts_close(fp); return -1; } idx = bcf_index(fp, min_shift); hts_close(fp); if ( !idx ) return -1; ret = hts_idx_save_as(idx, fn, fnidx, HTS_FMT_CSI); hts_idx_destroy(idx); return ret; } int bcf_index_build(const char *fn, int min_shift) { return bcf_index_build2(fn, NULL, min_shift); } /***************** *** Utilities *** *****************/ int bcf_hdr_combine(bcf_hdr_t *dst, const bcf_hdr_t *src) { int i, ndst_ori = dst->nhrec, need_sync = 0, ret = 0; for (i=0; inhrec; i++) { if ( src->hrec[i]->type==BCF_HL_GEN && src->hrec[i]->value ) { int j; for (j=0; jhrec[j]->type!=BCF_HL_GEN ) continue; // Checking only the key part of generic lines, otherwise // the VCFs are too verbose. Should we perhaps add a flag // to bcf_hdr_combine() and make this optional? if ( !strcmp(src->hrec[i]->key,dst->hrec[j]->key) ) break; } if ( j>=ndst_ori ) need_sync += bcf_hdr_add_hrec(dst, bcf_hrec_dup(src->hrec[i])); } else if ( src->hrec[i]->type==BCF_HL_STR ) { // NB: we are ignoring fields without ID int j = bcf_hrec_find_key(src->hrec[i],"ID"); if ( j>=0 ) { bcf_hrec_t *rec = bcf_hdr_get_hrec(dst, src->hrec[i]->type, "ID", src->hrec[i]->vals[j], src->hrec[i]->key); if ( !rec ) need_sync += bcf_hdr_add_hrec(dst, bcf_hrec_dup(src->hrec[i])); } } else { int j = bcf_hrec_find_key(src->hrec[i],"ID"); assert( j>=0 ); // this should always be true for valid VCFs bcf_hrec_t *rec = bcf_hdr_get_hrec(dst, src->hrec[i]->type, "ID", src->hrec[i]->vals[j], NULL); if ( !rec ) need_sync += bcf_hdr_add_hrec(dst, bcf_hrec_dup(src->hrec[i])); else if ( src->hrec[i]->type==BCF_HL_INFO || src->hrec[i]->type==BCF_HL_FMT ) { // Check that both records are of the same type. The bcf_hdr_id2length // macro cannot be used here because dst header is not synced yet. vdict_t *d_src = (vdict_t*)src->dict[BCF_DT_ID]; vdict_t *d_dst = (vdict_t*)dst->dict[BCF_DT_ID]; khint_t k_src = kh_get(vdict, d_src, src->hrec[i]->vals[0]); khint_t k_dst = kh_get(vdict, d_dst, src->hrec[i]->vals[0]); if ( (kh_val(d_src,k_src).info[rec->type]>>8 & 0xf) != (kh_val(d_dst,k_dst).info[rec->type]>>8 & 0xf) ) { fprintf(stderr,"Warning: trying to combine \"%s\" tag definitions of different lengths\n", src->hrec[i]->vals[0]); ret |= 1; } if ( (kh_val(d_src,k_src).info[rec->type]>>4 & 0xf) != (kh_val(d_dst,k_dst).info[rec->type]>>4 & 0xf) ) { fprintf(stderr,"Warning: trying to combine \"%s\" tag definitions of different types\n", src->hrec[i]->vals[0]); ret |= 1; } } } } if ( need_sync ) bcf_hdr_sync(dst); return ret; } bcf_hdr_t *bcf_hdr_merge(bcf_hdr_t *dst, const bcf_hdr_t *src) { if ( !dst ) { // this will effectively strip existing IDX attributes from src to become dst dst = bcf_hdr_init("r"); char *htxt = bcf_hdr_fmt_text(src, 0, NULL); bcf_hdr_parse(dst, htxt); free(htxt); return dst; } int i, ndst_ori = dst->nhrec, need_sync = 0, ret = 0; for (i=0; inhrec; i++) { if ( src->hrec[i]->type==BCF_HL_GEN && src->hrec[i]->value ) { int j; for (j=0; jhrec[j]->type!=BCF_HL_GEN ) continue; // Checking only the key part of generic lines, otherwise // the VCFs are too verbose. Should we perhaps add a flag // to bcf_hdr_combine() and make this optional? if ( !strcmp(src->hrec[i]->key,dst->hrec[j]->key) ) break; } if ( j>=ndst_ori ) need_sync += bcf_hdr_add_hrec(dst, bcf_hrec_dup(src->hrec[i])); } else if ( src->hrec[i]->type==BCF_HL_STR ) { // NB: we are ignoring fields without ID int j = bcf_hrec_find_key(src->hrec[i],"ID"); if ( j>=0 ) { bcf_hrec_t *rec = bcf_hdr_get_hrec(dst, src->hrec[i]->type, "ID", src->hrec[i]->vals[j], src->hrec[i]->key); if ( !rec ) need_sync += bcf_hdr_add_hrec(dst, bcf_hrec_dup(src->hrec[i])); } } else { int j = bcf_hrec_find_key(src->hrec[i],"ID"); assert( j>=0 ); // this should always be true for valid VCFs bcf_hrec_t *rec = bcf_hdr_get_hrec(dst, src->hrec[i]->type, "ID", src->hrec[i]->vals[j], NULL); if ( !rec ) need_sync += bcf_hdr_add_hrec(dst, bcf_hrec_dup(src->hrec[i])); else if ( src->hrec[i]->type==BCF_HL_INFO || src->hrec[i]->type==BCF_HL_FMT ) { // Check that both records are of the same type. The bcf_hdr_id2length // macro cannot be used here because dst header is not synced yet. vdict_t *d_src = (vdict_t*)src->dict[BCF_DT_ID]; vdict_t *d_dst = (vdict_t*)dst->dict[BCF_DT_ID]; khint_t k_src = kh_get(vdict, d_src, src->hrec[i]->vals[0]); khint_t k_dst = kh_get(vdict, d_dst, src->hrec[i]->vals[0]); if ( (kh_val(d_src,k_src).info[rec->type]>>8 & 0xf) != (kh_val(d_dst,k_dst).info[rec->type]>>8 & 0xf) ) { fprintf(stderr,"Warning: trying to combine \"%s\" tag definitions of different lengths\n", src->hrec[i]->vals[0]); ret |= 1; } if ( (kh_val(d_src,k_src).info[rec->type]>>4 & 0xf) != (kh_val(d_dst,k_dst).info[rec->type]>>4 & 0xf) ) { fprintf(stderr,"Warning: trying to combine \"%s\" tag definitions of different types\n", src->hrec[i]->vals[0]); ret |= 1; } } } } if ( need_sync ) bcf_hdr_sync(dst); return dst; } int bcf_translate(const bcf_hdr_t *dst_hdr, bcf_hdr_t *src_hdr, bcf1_t *line) { int i; if ( line->errcode ) { fprintf(stderr,"[%s:%d %s] Unchecked error (%d), exiting.\n", __FILE__,__LINE__,__FUNCTION__,line->errcode); exit(1); } if ( src_hdr->ntransl==-1 ) return 0; // no need to translate, all tags have the same id if ( !src_hdr->ntransl ) // called for the first time, see what needs translating { int dict; for (dict=0; dict<2; dict++) // BCF_DT_ID and BCF_DT_CTG { src_hdr->transl[dict] = (int*) malloc(src_hdr->n[dict]*sizeof(int)); for (i=0; in[dict]; i++) { if ( !src_hdr->id[dict][i].key ) // gap left after removed BCF header lines { src_hdr->transl[dict][i] = -1; continue; } src_hdr->transl[dict][i] = bcf_hdr_id2int(dst_hdr,dict,src_hdr->id[dict][i].key); if ( src_hdr->transl[dict][i]!=-1 && i!=src_hdr->transl[dict][i] ) src_hdr->ntransl++; } } if ( !src_hdr->ntransl ) { free(src_hdr->transl[0]); src_hdr->transl[0] = NULL; free(src_hdr->transl[1]); src_hdr->transl[1] = NULL; src_hdr->ntransl = -1; } if ( src_hdr->ntransl==-1 ) return 0; } bcf_unpack(line,BCF_UN_ALL); // CHROM if ( src_hdr->transl[BCF_DT_CTG][line->rid] >=0 ) line->rid = src_hdr->transl[BCF_DT_CTG][line->rid]; // FILTER for (i=0; id.n_flt; i++) { int src_id = line->d.flt[i]; if ( src_hdr->transl[BCF_DT_ID][src_id] >=0 ) line->d.flt[i] = src_hdr->transl[BCF_DT_ID][src_id]; line->d.shared_dirty |= BCF1_DIRTY_FLT; } // INFO for (i=0; in_info; i++) { int src_id = line->d.info[i].key; int dst_id = src_hdr->transl[BCF_DT_ID][src_id]; if ( dst_id<0 ) continue; int src_size = src_id>>7 ? ( src_id>>15 ? BCF_BT_INT32 : BCF_BT_INT16) : BCF_BT_INT8; int dst_size = dst_id>>7 ? ( dst_id>>15 ? BCF_BT_INT32 : BCF_BT_INT16) : BCF_BT_INT8; if ( src_size==dst_size ) // can overwrite { line->d.info[i].key = dst_id; uint8_t *vptr = line->d.info[i].vptr - line->d.info[i].vptr_off; if ( dst_size==BCF_BT_INT8 ) { vptr[1] = (uint8_t)dst_id; } else if ( dst_size==BCF_BT_INT16 ) { *(uint16_t*)vptr = (uint16_t)dst_id; } else { *(uint32_t*)vptr = (uint32_t)dst_id; } } else // must realloc { bcf_info_t *info = &line->d.info[i]; assert( !info->vptr_free ); kstring_t str = {0,0,0}; bcf_enc_int1(&str, dst_id); bcf_enc_size(&str, info->len,info->type); info->vptr_off = str.l; kputsn((char*)info->vptr, info->vptr_len, &str); info->vptr = (uint8_t*)str.s + info->vptr_off; info->vptr_free = 1; info->key = dst_id; line->d.shared_dirty |= BCF1_DIRTY_INF; } } // FORMAT for (i=0; in_fmt; i++) { int src_id = line->d.fmt[i].id; int dst_id = src_hdr->transl[BCF_DT_ID][src_id]; if ( dst_id<0 ) continue; int src_size = src_id>>7 ? ( src_id>>15 ? BCF_BT_INT32 : BCF_BT_INT16) : BCF_BT_INT8; int dst_size = dst_id>>7 ? ( dst_id>>15 ? BCF_BT_INT32 : BCF_BT_INT16) : BCF_BT_INT8; if ( src_size==dst_size ) // can overwrite { line->d.fmt[i].id = dst_id; uint8_t *p = line->d.fmt[i].p - line->d.fmt[i].p_off; // pointer to the vector size (4bits) and BT type (4bits) if ( dst_size==BCF_BT_INT8 ) { p[1] = dst_id; } else if ( dst_size==BCF_BT_INT16 ) { uint8_t *x = (uint8_t*) &dst_id; p[1] = x[0]; p[2] = x[1]; } else { uint8_t *x = (uint8_t*) &dst_id; p[1] = x[0]; p[2] = x[1]; p[3] = x[2]; p[4] = x[3]; } } else // must realloc { bcf_fmt_t *fmt = &line->d.fmt[i]; assert( !fmt->p_free ); kstring_t str = {0,0,0}; bcf_enc_int1(&str, dst_id); bcf_enc_size(&str, fmt->n, fmt->type); fmt->p_off = str.l; kputsn((char*)fmt->p, fmt->p_len, &str); fmt->p = (uint8_t*)str.s + fmt->p_off; fmt->p_free = 1; fmt->id = dst_id; line->d.indiv_dirty = 1; } } return 0; } bcf_hdr_t *bcf_hdr_dup(const bcf_hdr_t *hdr) { bcf_hdr_t *hout = bcf_hdr_init("r"); char *htxt = bcf_hdr_fmt_text(hdr, 1, NULL); if (!hout) { fprintf(stderr, "[E::%s] failed to allocate bcf header\n", __func__); free(htxt); return NULL; } bcf_hdr_parse(hout, htxt); free(htxt); return hout; } bcf_hdr_t *bcf_hdr_subset(const bcf_hdr_t *h0, int n, char *const* samples, int *imap) { int hlen; void *names_hash = khash_str2int_init(); char *htxt = bcf_hdr_fmt_text(h0, 1, &hlen); kstring_t str; bcf_hdr_t *h; str.l = str.m = 0; str.s = 0; h = bcf_hdr_init("w"); if (!h) { fprintf(stderr, "[E::%s] failed to allocate bcf header\n", __func__); free(htxt); return NULL; } bcf_hdr_set_version(h,bcf_hdr_get_version(h0)); int j; for (j=0; j 0) { char *p; int i = 0, end = n? 8 : 7; while ((p = strstr(htxt, "#CHROM\t")) != 0) if (p > htxt && *(p-1) == '\n') break; while ((p = strchr(p, '\t')) != 0 && i < end) ++i, ++p; if (i != end) { free(h); free(str.s); return 0; // malformated header } kputsn(htxt, p - htxt, &str); for (i = 0; i < n; ++i) { if ( khash_str2int_has_key(names_hash,samples[i]) ) { fprintf(stderr,"[E::bcf_hdr_subset] Duplicate sample name \"%s\".\n", samples[i]); free(str.s); free(htxt); khash_str2int_destroy(names_hash); bcf_hdr_destroy(h); return NULL; } imap[i] = bcf_hdr_id2int(h0, BCF_DT_SAMPLE, samples[i]); if (imap[i] < 0) continue; kputc('\t', &str); kputs(samples[i], &str); khash_str2int_inc(names_hash,samples[i]); } } else kputsn(htxt, hlen, &str); while (str.l && (!str.s[str.l-1] || str.s[str.l-1]=='\n') ) str.l--; // kill trailing zeros and newlines kputc('\n',&str); bcf_hdr_parse(h, str.s); free(str.s); free(htxt); khash_str2int_destroy(names_hash); return h; } int bcf_hdr_set_samples(bcf_hdr_t *hdr, const char *samples, int is_file) { if ( samples && !strcmp("-",samples) ) return 0; // keep all samples hdr->nsamples_ori = bcf_hdr_nsamples(hdr); if ( !samples ) { bcf_hdr_nsamples(hdr) = 0; return 0; } // exclude all samples int i, narr = bit_array_size(bcf_hdr_nsamples(hdr)); hdr->keep_samples = (uint8_t*) calloc(narr,1); if ( samples[0]=='^' ) for (i=0; ikeep_samples,i); int idx, n, ret = 0; char **smpls = hts_readlist(samples[0]=='^'?samples+1:samples, is_file, &n); if ( !smpls ) return -1; for (i=0; ikeep_samples, idx); else bit_array_set(hdr->keep_samples, idx); } for (i=0; insamples_ori; i++) if ( bit_array_test(hdr->keep_samples,i) ) bcf_hdr_nsamples(hdr)++; if ( !bcf_hdr_nsamples(hdr) ) { free(hdr->keep_samples); hdr->keep_samples=NULL; } else { char **samples = (char**) malloc(sizeof(char*)*bcf_hdr_nsamples(hdr)); idx = 0; for (i=0; insamples_ori; i++) if ( bit_array_test(hdr->keep_samples,i) ) samples[idx++] = strdup(hdr->samples[i]); free(hdr->samples); hdr->samples = samples; // delete original samples from the dictionary vdict_t *d = (vdict_t*)hdr->dict[BCF_DT_SAMPLE]; int k; for (k = kh_begin(d); k != kh_end(d); ++k) if (kh_exist(d, k)) free((char*)kh_key(d, k)); kh_destroy(vdict, d); // add the subset back hdr->dict[BCF_DT_SAMPLE] = d = kh_init(vdict); for (i=0; isamples[i], &ignore); kh_val(d, k) = bcf_idinfo_def; kh_val(d, k).id = kh_size(d) - 1; } bcf_hdr_sync(hdr); } return ret; } int bcf_subset(const bcf_hdr_t *h, bcf1_t *v, int n, int *imap) { kstring_t ind; ind.s = 0; ind.l = ind.m = 0; if (n) { bcf_fmt_t *fmt; int i, j; fmt = (bcf_fmt_t*)alloca(v->n_fmt * sizeof(bcf_fmt_t)); uint8_t *ptr = (uint8_t*)v->indiv.s; for (i = 0; i < v->n_fmt; ++i) ptr = bcf_unpack_fmt_core1(ptr, v->n_sample, &fmt[i]); for (i = 0; i < (int)v->n_fmt; ++i) { bcf_fmt_t *f = &fmt[i]; bcf_enc_int1(&ind, f->id); bcf_enc_size(&ind, f->n, f->type); for (j = 0; j < n; ++j) if (imap[j] >= 0) kputsn((char*)(f->p + imap[j] * f->size), f->size, &ind); } for (i = j = 0; j < n; ++j) if (imap[j] >= 0) ++i; v->n_sample = i; } else v->n_sample = 0; if ( !v->n_sample ) v->n_fmt = 0; free(v->indiv.s); v->indiv = ind; v->unpacked &= ~BCF_UN_FMT; // only BCF is ready for output, VCF will need to unpack again return 0; } int bcf_is_snp(bcf1_t *v) { int i; bcf_unpack(v, BCF_UN_STR); for (i = 0; i < v->n_allele; ++i) { if ( v->d.allele[i][1]==0 ) continue; // mpileup's allele, see also below. This is not completely satisfactory, // a general library is here narrowly tailored to fit samtools. if ( v->d.allele[i][0]=='<' && v->d.allele[i][1]=='X' && v->d.allele[i][2]=='>' ) continue; if ( v->d.allele[i][0]=='<' && v->d.allele[i][1]=='*' && v->d.allele[i][2]=='>' ) continue; break; } return i == v->n_allele; } static void bcf_set_variant_type(const char *ref, const char *alt, variant_t *var) { // The most frequent case if ( !ref[1] && !alt[1] ) { if ( *alt == '.' || *ref==*alt ) { var->n = 0; var->type = VCF_REF; return; } if ( *alt == 'X' ) { var->n = 0; var->type = VCF_REF; return; } // mpileup's X allele shouldn't be treated as variant if ( *alt == '*' ) { var->n = 0; var->type = VCF_REF; return; } var->n = 1; var->type = VCF_SNP; return; } if ( alt[0]=='<' ) { if ( alt[1]=='X' && alt[2]=='>' ) { var->n = 0; var->type = VCF_REF; return; } // mpileup's X allele shouldn't be treated as variant if ( alt[1]=='*' && alt[2]=='>' ) { var->n = 0; var->type = VCF_REF; return; } var->type = VCF_OTHER; return; } const char *r = ref, *a = alt; while (*r && *a && toupper(*r)==toupper(*a) ) { r++; a++; } // unfortunately, matching REF,ALT case is not guaranteed if ( *a && !*r ) { while ( *a ) a++; var->n = (a-alt)-(r-ref); var->type = VCF_INDEL; return; } else if ( *r && !*a ) { while ( *r ) r++; var->n = (a-alt)-(r-ref); var->type = VCF_INDEL; return; } else if ( !*r && !*a ) { var->n = 0; var->type = VCF_REF; return; } const char *re = r, *ae = a; while ( re[1] ) re++; while ( ae[1] ) ae++; while ( re>r && ae>a && toupper(*re)==toupper(*ae) ) { re--; ae--; } if ( ae==a ) { if ( re==r ) { var->n = 1; var->type = VCF_SNP; return; } var->n = -(re-r); if ( toupper(*re)==toupper(*ae) ) { var->type = VCF_INDEL; return; } var->type = VCF_OTHER; return; } else if ( re==r ) { var->n = ae-a; if ( toupper(*re)==toupper(*ae) ) { var->type = VCF_INDEL; return; } var->type = VCF_OTHER; return; } var->type = ( re-r == ae-a ) ? VCF_MNP : VCF_OTHER; var->n = ( re-r > ae-a ) ? -(re-r+1) : ae-a+1; // should do also complex events, SVs, etc... } static void bcf_set_variant_types(bcf1_t *b) { if ( !(b->unpacked & BCF_UN_STR) ) bcf_unpack(b, BCF_UN_STR); bcf_dec_t *d = &b->d; if ( d->n_var < b->n_allele ) { d->var = (variant_t *) realloc(d->var, sizeof(variant_t)*b->n_allele); d->n_var = b->n_allele; } int i; b->d.var_type = 0; for (i=1; in_allele; i++) { bcf_set_variant_type(d->allele[0],d->allele[i], &d->var[i]); b->d.var_type |= d->var[i].type; //fprintf(stderr,"[set_variant_type] %d %s %s -> %d %d .. %d\n", b->pos+1,d->allele[0],d->allele[i],d->var[i].type,d->var[i].n, b->d.var_type); } } int bcf_get_variant_types(bcf1_t *rec) { if ( rec->d.var_type==-1 ) bcf_set_variant_types(rec); return rec->d.var_type; } int bcf_get_variant_type(bcf1_t *rec, int ith_allele) { if ( rec->d.var_type==-1 ) bcf_set_variant_types(rec); return rec->d.var[ith_allele].type; } int bcf_update_info(const bcf_hdr_t *hdr, bcf1_t *line, const char *key, const void *values, int n, int type) { // Is the field already present? int i, inf_id = bcf_hdr_id2int(hdr,BCF_DT_ID,key); if ( !bcf_hdr_idinfo_exists(hdr,BCF_HL_INFO,inf_id) ) return -1; // No such INFO field in the header if ( !(line->unpacked & BCF_UN_INFO) ) bcf_unpack(line, BCF_UN_INFO); for (i=0; in_info; i++) if ( inf_id==line->d.info[i].key ) break; bcf_info_t *inf = i==line->n_info ? NULL : &line->d.info[i]; if ( !n || (type==BCF_HT_STR && !values) ) { if ( n==0 && !strcmp("END",key) ) line->rlen = line->n_allele ? strlen(line->d.allele[0]) : 0; if ( inf ) { // Mark the tag for removal, free existing memory if necessary if ( inf->vptr_free ) { free(inf->vptr - inf->vptr_off); inf->vptr_free = 0; } line->d.shared_dirty |= BCF1_DIRTY_INF; inf->vptr = NULL; } return 0; } // Encode the values and determine the size required to accommodate the values kstring_t str = {0,0,0}; bcf_enc_int1(&str, inf_id); if ( type==BCF_HT_INT ) bcf_enc_vint(&str, n, (int32_t*)values, -1); else if ( type==BCF_HT_REAL ) bcf_enc_vfloat(&str, n, (float*)values); else if ( type==BCF_HT_FLAG || type==BCF_HT_STR ) { if ( values==NULL ) bcf_enc_size(&str, 0, BCF_BT_NULL); else bcf_enc_vchar(&str, strlen((char*)values), (char*)values); } else { fprintf(stderr, "[E::%s] the type %d not implemented yet\n", __func__, type); abort(); } // Is the INFO tag already present if ( inf ) { // Is it big enough to accommodate new block? if ( str.l <= inf->vptr_len + inf->vptr_off ) { if ( str.l != inf->vptr_len + inf->vptr_off ) line->d.shared_dirty |= BCF1_DIRTY_INF; uint8_t *ptr = inf->vptr - inf->vptr_off; memcpy(ptr, str.s, str.l); free(str.s); int vptr_free = inf->vptr_free; bcf_unpack_info_core1(ptr, inf); inf->vptr_free = vptr_free; } else { assert( !inf->vptr_free ); // fix the caller or improve here: this has been modified before bcf_unpack_info_core1((uint8_t*)str.s, inf); inf->vptr_free = 1; line->d.shared_dirty |= BCF1_DIRTY_INF; } } else { // The tag is not present, create new one line->n_info++; hts_expand0(bcf_info_t, line->n_info, line->d.m_info , line->d.info); inf = &line->d.info[line->n_info-1]; bcf_unpack_info_core1((uint8_t*)str.s, inf); inf->vptr_free = 1; line->d.shared_dirty |= BCF1_DIRTY_INF; } line->unpacked |= BCF_UN_INFO; if ( n==1 && !strcmp("END",key) ) line->rlen = ((int32_t*)values)[0] - line->pos; return 0; } int bcf_update_format_string(const bcf_hdr_t *hdr, bcf1_t *line, const char *key, const char **values, int n) { if ( !n ) return bcf_update_format(hdr,line,key,NULL,0,BCF_HT_STR); int i, max_len = 0; for (i=0; i max_len ) max_len = len; } char *out = (char*) malloc(max_len*n); if ( !out ) return -2; for (i=0; iunpacked & BCF_UN_FMT) ) bcf_unpack(line, BCF_UN_FMT); for (i=0; in_fmt; i++) if ( line->d.fmt[i].id==fmt_id ) break; bcf_fmt_t *fmt = i==line->n_fmt ? NULL : &line->d.fmt[i]; if ( !n ) { if ( fmt ) { // Mark the tag for removal, free existing memory if necessary if ( fmt->p_free ) { free(fmt->p - fmt->p_off); fmt->p_free = 0; } line->d.indiv_dirty = 1; fmt->p = NULL; } return 0; } line->n_sample = bcf_hdr_nsamples(hdr); int nps = n / line->n_sample; // number of values per sample assert( nps && nps*line->n_sample==n ); // must be divisible by n_sample // Encode the values and determine the size required to accommodate the values kstring_t str = {0,0,0}; bcf_enc_int1(&str, fmt_id); if ( type==BCF_HT_INT ) bcf_enc_vint(&str, n, (int32_t*)values, nps); else if ( type==BCF_HT_REAL ) { bcf_enc_size(&str, nps, BCF_BT_FLOAT); kputsn((char*)values, nps*line->n_sample*sizeof(float), &str); } else if ( type==BCF_HT_STR ) { bcf_enc_size(&str, nps, BCF_BT_CHAR); kputsn((char*)values, nps*line->n_sample, &str); } else { fprintf(stderr, "[E::%s] the type %d not implemented yet\n", __func__, type); abort(); } if ( !fmt ) { // Not present, new format field line->n_fmt++; hts_expand0(bcf_fmt_t, line->n_fmt, line->d.m_fmt, line->d.fmt); // Special case: VCF specification requires that GT is always first if ( line->n_fmt > 1 && key[0]=='G' && key[1]=='T' && !key[2] ) { for (i=line->n_fmt-1; i>0; i--) line->d.fmt[i] = line->d.fmt[i-1]; fmt = &line->d.fmt[0]; } else fmt = &line->d.fmt[line->n_fmt-1]; bcf_unpack_fmt_core1((uint8_t*)str.s, line->n_sample, fmt); line->d.indiv_dirty = 1; fmt->p_free = 1; } else { // The tag is already present, check if it is big enough to accomodate the new block if ( str.l <= fmt->p_len + fmt->p_off ) { // good, the block is big enough if ( str.l != fmt->p_len + fmt->p_off ) line->d.indiv_dirty = 1; uint8_t *ptr = fmt->p - fmt->p_off; memcpy(ptr, str.s, str.l); free(str.s); int p_free = fmt->p_free; bcf_unpack_fmt_core1(ptr, line->n_sample, fmt); fmt->p_free = p_free; } else { assert( !fmt->p_free ); // fix the caller or improve here: this has been modified before bcf_unpack_fmt_core1((uint8_t*)str.s, line->n_sample, fmt); fmt->p_free = 1; line->d.indiv_dirty = 1; } } line->unpacked |= BCF_UN_FMT; return 0; } int bcf_update_filter(const bcf_hdr_t *hdr, bcf1_t *line, int *flt_ids, int n) { if ( !(line->unpacked & BCF_UN_FLT) ) bcf_unpack(line, BCF_UN_FLT); line->d.shared_dirty |= BCF1_DIRTY_FLT; line->d.n_flt = n; if ( !n ) return 0; hts_expand(int, line->d.n_flt, line->d.m_flt, line->d.flt); int i; for (i=0; id.flt[i] = flt_ids[i]; return 0; } int bcf_add_filter(const bcf_hdr_t *hdr, bcf1_t *line, int flt_id) { if ( !(line->unpacked & BCF_UN_FLT) ) bcf_unpack(line, BCF_UN_FLT); int i; for (i=0; id.n_flt; i++) if ( flt_id==line->d.flt[i] ) break; if ( id.n_flt ) return 0; // this filter is already set line->d.shared_dirty |= BCF1_DIRTY_FLT; if ( flt_id==0 ) // set to PASS line->d.n_flt = 1; else if ( line->d.n_flt==1 && line->d.flt[0]==0 ) line->d.n_flt = 1; else line->d.n_flt++; hts_expand(int, line->d.n_flt, line->d.m_flt, line->d.flt); line->d.flt[line->d.n_flt-1] = flt_id; return 1; } int bcf_remove_filter(const bcf_hdr_t *hdr, bcf1_t *line, int flt_id, int pass) { if ( !(line->unpacked & BCF_UN_FLT) ) bcf_unpack(line, BCF_UN_FLT); int i; for (i=0; id.n_flt; i++) if ( flt_id==line->d.flt[i] ) break; if ( i==line->d.n_flt ) return 0; // the filter is not present line->d.shared_dirty |= BCF1_DIRTY_FLT; if ( i!=line->d.n_flt-1 ) memmove(line->d.flt+i,line->d.flt+i+1,(line->d.n_flt-i-1)*sizeof(*line->d.flt)); line->d.n_flt--; if ( !line->d.n_flt && pass ) bcf_add_filter(hdr,line,0); return 0; } int bcf_has_filter(const bcf_hdr_t *hdr, bcf1_t *line, char *filter) { if ( filter[0]=='.' && !filter[1] ) filter = "PASS"; int id = bcf_hdr_id2int(hdr, BCF_DT_ID, filter); if ( !bcf_hdr_idinfo_exists(hdr,BCF_HL_FLT,id) ) return -1; // not defined in the header if ( !(line->unpacked & BCF_UN_FLT) ) bcf_unpack(line, BCF_UN_FLT); if ( id==0 && !line->d.n_flt) return 1; // PASS int i; for (i=0; id.n_flt; i++) if ( line->d.flt[i]==id ) return 1; return 0; } static inline int _bcf1_sync_alleles(const bcf_hdr_t *hdr, bcf1_t *line, int nals) { line->d.shared_dirty |= BCF1_DIRTY_ALS; line->n_allele = nals; hts_expand(char*, line->n_allele, line->d.m_allele, line->d.allele); char *als = line->d.als; int n = 0; while (nd.allele[n] = als; while ( *als ) als++; als++; n++; } // Update REF length bcf_info_t *end_info = bcf_get_info(hdr,line,"END"); line->rlen = end_info ? end_info->v1.i : strlen(line->d.allele[0]); return 0; } int bcf_update_alleles(const bcf_hdr_t *hdr, bcf1_t *line, const char **alleles, int nals) { kstring_t tmp = {0,0,0}; char *free_old = NULL; // If the supplied alleles are not pointers to line->d.als, the existing block can be reused. int i; for (i=0; i=line->d.als && alleles[i]d.als+line->d.m_als ) break; if ( i==nals ) { // all alleles point elsewhere, reuse the existing block tmp.l = 0; tmp.s = line->d.als; tmp.m = line->d.m_als; } else free_old = line->d.als; for (i=0; id.als = tmp.s; line->d.m_als = tmp.m; free(free_old); return _bcf1_sync_alleles(hdr,line,nals); } int bcf_update_alleles_str(const bcf_hdr_t *hdr, bcf1_t *line, const char *alleles_string) { kstring_t tmp; tmp.l = 0; tmp.s = line->d.als; tmp.m = line->d.m_als; kputs(alleles_string, &tmp); line->d.als = tmp.s; line->d.m_als = tmp.m; int nals = 1; char *t = line->d.als; while (*t) { if ( *t==',' ) { *t = 0; nals++; } t++; } return _bcf1_sync_alleles(hdr, line, nals); } int bcf_update_id(const bcf_hdr_t *hdr, bcf1_t *line, const char *id) { kstring_t tmp; tmp.l = 0; tmp.s = line->d.id; tmp.m = line->d.m_id; if ( id ) kputs(id, &tmp); else kputs(".", &tmp); line->d.id = tmp.s; line->d.m_id = tmp.m; line->d.shared_dirty |= BCF1_DIRTY_ID; return 0; } int bcf_add_id(const bcf_hdr_t *hdr, bcf1_t *line, const char *id) { if ( !id ) return 0; kstring_t tmp; tmp.l = 0; tmp.s = line->d.id; tmp.m = line->d.m_id; int len = strlen(id); char *dst = line->d.id; while ( *dst && (dst=strstr(dst,id)) ) { if ( dst[len]!=0 && dst[len]!=';' ) dst++; // a prefix, not a match else if ( dst==line->d.id || dst[-1]==';' ) return 0; // already present dst++; // a suffix, not a match } if ( line->d.id && (line->d.id[0]!='.' || line->d.id[1]) ) { tmp.l = strlen(line->d.id); kputc(';',&tmp); } kputs(id,&tmp); line->d.id = tmp.s; line->d.m_id = tmp.m; line->d.shared_dirty |= BCF1_DIRTY_ID; return 0; } bcf_fmt_t *bcf_get_fmt(const bcf_hdr_t *hdr, bcf1_t *line, const char *key) { int id = bcf_hdr_id2int(hdr, BCF_DT_ID, key); if ( !bcf_hdr_idinfo_exists(hdr,BCF_HL_FMT,id) ) return NULL; // no such FMT field in the header return bcf_get_fmt_id(line, id); } bcf_info_t *bcf_get_info(const bcf_hdr_t *hdr, bcf1_t *line, const char *key) { int id = bcf_hdr_id2int(hdr, BCF_DT_ID, key); if ( !bcf_hdr_idinfo_exists(hdr,BCF_HL_INFO,id) ) return NULL; // no such INFO field in the header return bcf_get_info_id(line, id); } bcf_fmt_t *bcf_get_fmt_id(bcf1_t *line, const int id) { int i; if ( !(line->unpacked & BCF_UN_FMT) ) bcf_unpack(line, BCF_UN_FMT); for (i=0; in_fmt; i++) { if ( line->d.fmt[i].id==id ) return &line->d.fmt[i]; } return NULL; } bcf_info_t *bcf_get_info_id(bcf1_t *line, const int id) { int i; if ( !(line->unpacked & BCF_UN_INFO) ) bcf_unpack(line, BCF_UN_INFO); for (i=0; in_info; i++) { if ( line->d.info[i].key==id ) return &line->d.info[i]; } return NULL; } int bcf_get_info_values(const bcf_hdr_t *hdr, bcf1_t *line, const char *tag, void **dst, int *ndst, int type) { int i,j, tag_id = bcf_hdr_id2int(hdr, BCF_DT_ID, tag); if ( !bcf_hdr_idinfo_exists(hdr,BCF_HL_INFO,tag_id) ) return -1; // no such INFO field in the header if ( bcf_hdr_id2type(hdr,BCF_HL_INFO,tag_id)!=type ) return -2; // expected different type if ( !(line->unpacked & BCF_UN_INFO) ) bcf_unpack(line, BCF_UN_INFO); for (i=0; in_info; i++) if ( line->d.info[i].key==tag_id ) break; if ( i==line->n_info ) return ( type==BCF_HT_FLAG ) ? 0 : -3; // the tag is not present in this record if ( type==BCF_HT_FLAG ) return 1; bcf_info_t *info = &line->d.info[i]; if ( !info->vptr ) return -3; // the tag was marked for removal if ( type==BCF_HT_STR ) { if ( *ndst < info->len+1 ) { *ndst = info->len + 1; *dst = realloc(*dst, *ndst); } memcpy(*dst,info->vptr,info->len); ((uint8_t*)*dst)[info->len] = 0; return info->len; } // Make sure the buffer is big enough int size1 = type==BCF_HT_INT ? sizeof(int32_t) : sizeof(float); if ( *ndst < info->len ) { *ndst = info->len; *dst = realloc(*dst, *ndst * size1); } if ( info->len == 1 ) { if ( info->type==BCF_BT_FLOAT ) *((float*)*dst) = info->v1.f; else { #define BRANCH(type_t, missing) { \ if ( info->v1.i==missing ) *((int32_t*)*dst) = bcf_int32_missing; \ else *((int32_t*)*dst) = info->v1.i; \ } switch (info->type) { case BCF_BT_INT8: BRANCH(int8_t, bcf_int8_missing ); break; case BCF_BT_INT16: BRANCH(int16_t, bcf_int16_missing); break; case BCF_BT_INT32: BRANCH(int32_t, bcf_int32_missing); break; } #undef BRANCH } return 1; } #define BRANCH(type_t, is_missing, is_vector_end, set_missing, out_type_t) { \ out_type_t *tmp = (out_type_t *) *dst; \ type_t *p = (type_t *) info->vptr; \ for (j=0; jlen; j++) \ { \ if ( is_vector_end ) return j; \ if ( is_missing ) set_missing; \ else *tmp = p[j]; \ tmp++; \ } \ return j; \ } switch (info->type) { case BCF_BT_INT8: BRANCH(int8_t, p[j]==bcf_int8_missing, p[j]==bcf_int8_vector_end, *tmp=bcf_int32_missing, int32_t); break; case BCF_BT_INT16: BRANCH(int16_t, p[j]==bcf_int16_missing, p[j]==bcf_int16_vector_end, *tmp=bcf_int32_missing, int32_t); break; case BCF_BT_INT32: BRANCH(int32_t, p[j]==bcf_int32_missing, p[j]==bcf_int32_vector_end, *tmp=bcf_int32_missing, int32_t); break; case BCF_BT_FLOAT: BRANCH(float, bcf_float_is_missing(p[j]), bcf_float_is_vector_end(p[j]), bcf_float_set_missing(*tmp), float); break; default: fprintf(stderr,"TODO: %s:%d .. info->type=%d\n", __FILE__,__LINE__, info->type); exit(1); } #undef BRANCH return -4; // this can never happen } int bcf_get_format_string(const bcf_hdr_t *hdr, bcf1_t *line, const char *tag, char ***dst, int *ndst) { int i,tag_id = bcf_hdr_id2int(hdr, BCF_DT_ID, tag); if ( !bcf_hdr_idinfo_exists(hdr,BCF_HL_FMT,tag_id) ) return -1; // no such FORMAT field in the header if ( bcf_hdr_id2type(hdr,BCF_HL_FMT,tag_id)!=BCF_HT_STR ) return -2; // expected different type if ( !(line->unpacked & BCF_UN_FMT) ) bcf_unpack(line, BCF_UN_FMT); for (i=0; in_fmt; i++) if ( line->d.fmt[i].id==tag_id ) break; if ( i==line->n_fmt ) return -3; // the tag is not present in this record bcf_fmt_t *fmt = &line->d.fmt[i]; if ( !fmt->p ) return -3; // the tag was marked for removal int nsmpl = bcf_hdr_nsamples(hdr); if ( !*dst ) { *dst = (char**) malloc(sizeof(char*)*nsmpl); if ( !*dst ) return -4; // could not alloc (*dst)[0] = NULL; } int n = (fmt->n+1)*nsmpl; if ( *ndst < n ) { (*dst)[0] = realloc((*dst)[0], n); if ( !(*dst)[0] ) return -4; // could not alloc *ndst = n; } for (i=0; ip + i*fmt->n; uint8_t *tmp = (uint8_t*)(*dst)[0] + i*(fmt->n+1); memcpy(tmp,src,fmt->n); tmp[fmt->n] = 0; (*dst)[i] = (char*) tmp; } return n; } int bcf_get_format_values(const bcf_hdr_t *hdr, bcf1_t *line, const char *tag, void **dst, int *ndst, int type) { int i,j, tag_id = bcf_hdr_id2int(hdr, BCF_DT_ID, tag); if ( !bcf_hdr_idinfo_exists(hdr,BCF_HL_FMT,tag_id) ) return -1; // no such FORMAT field in the header if ( tag[0]=='G' && tag[1]=='T' && tag[2]==0 ) { // Ugly: GT field is considered to be a string by the VCF header but BCF represents it as INT. if ( bcf_hdr_id2type(hdr,BCF_HL_FMT,tag_id)!=BCF_HT_STR ) return -2; } else if ( bcf_hdr_id2type(hdr,BCF_HL_FMT,tag_id)!=type ) return -2; // expected different type if ( !(line->unpacked & BCF_UN_FMT) ) bcf_unpack(line, BCF_UN_FMT); for (i=0; in_fmt; i++) if ( line->d.fmt[i].id==tag_id ) break; if ( i==line->n_fmt ) return -3; // the tag is not present in this record bcf_fmt_t *fmt = &line->d.fmt[i]; if ( !fmt->p ) return -3; // the tag was marked for removal if ( type==BCF_HT_STR ) { int n = fmt->n*bcf_hdr_nsamples(hdr); if ( *ndst < n ) { *dst = realloc(*dst, n); if ( !*dst ) return -4; // could not alloc *ndst = n; } memcpy(*dst,fmt->p,n); return n; } // Make sure the buffer is big enough int nsmpl = bcf_hdr_nsamples(hdr); int size1 = type==BCF_HT_INT ? sizeof(int32_t) : sizeof(float); if ( *ndst < fmt->n*nsmpl ) { *ndst = fmt->n*nsmpl; *dst = realloc(*dst, *ndst*size1); if ( !dst ) return -4; // could not alloc } #define BRANCH(type_t, is_missing, is_vector_end, set_missing, set_vector_end, out_type_t) { \ out_type_t *tmp = (out_type_t *) *dst; \ type_t *p = (type_t*) fmt->p; \ for (i=0; in; j++) \ { \ if ( is_missing ) set_missing; \ else if ( is_vector_end ) { set_vector_end; break; } \ else *tmp = p[j]; \ tmp++; \ } \ for (; jn; j++) { set_vector_end; tmp++; } \ p = (type_t *)((char *)p + fmt->size); \ } \ } switch (fmt->type) { case BCF_BT_INT8: BRANCH(int8_t, p[j]==bcf_int8_missing, p[j]==bcf_int8_vector_end, *tmp=bcf_int32_missing, *tmp=bcf_int32_vector_end, int32_t); break; case BCF_BT_INT16: BRANCH(int16_t, p[j]==bcf_int16_missing, p[j]==bcf_int16_vector_end, *tmp=bcf_int32_missing, *tmp=bcf_int32_vector_end, int32_t); break; case BCF_BT_INT32: BRANCH(int32_t, p[j]==bcf_int32_missing, p[j]==bcf_int32_vector_end, *tmp=bcf_int32_missing, *tmp=bcf_int32_vector_end, int32_t); break; case BCF_BT_FLOAT: BRANCH(float, bcf_float_is_missing(p[j]), bcf_float_is_vector_end(p[j]), bcf_float_set_missing(*tmp), bcf_float_set_vector_end(*tmp), float); break; default: fprintf(stderr,"TODO: %s:%d .. fmt->type=%d\n", __FILE__,__LINE__, fmt->type); exit(1); } #undef BRANCH return nsmpl*fmt->n; } stacks-1.44/htslib/._vcf_sweep.c000644 000765 000024 00000000265 12750743246 017346 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/vcf_sweep.c000644 000765 000024 00000012311 12750743246 017124 0ustar00catchenstaff000000 000000 /* vcf_sweep.c -- forward/reverse sweep API. Copyright (C) 2013 Genome Research Ltd. Author: Petr Danecek Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include "htslib/vcf_sweep.h" #include "htslib/bgzf.h" #define SW_FWD 0 #define SW_BWD 1 struct _bcf_sweep_t { htsFile *file; bcf_hdr_t *hdr; BGZF *fp; int direction; // to tell if the direction has changed int block_size; // the size of uncompressed data to hold in memory bcf1_t *rec; // bcf buffer int nrec, mrec; // number of used records; total size of the buffer int lrid, lpos, lnals, lals_len, mlals; // to check uniqueness of a record char *lals; uint64_t *idx; // uncompressed offsets of VCF/BCF records int iidx, nidx, midx; // i: current offset; n: used; m: allocated int idx_done; // the index is built during the first pass }; BGZF *hts_get_bgzfp(htsFile *fp); int hts_useek(htsFile *file, long uoffset, int where); long hts_utell(htsFile *file); static inline int sw_rec_equal(bcf_sweep_t *sw, bcf1_t *rec) { if ( sw->lrid!=rec->rid ) return 0; if ( sw->lpos!=rec->pos ) return 0; if ( sw->lnals!=rec->n_allele ) return 0; char *t = rec->d.allele[sw->lnals-1]; int len = t - rec->d.allele[0] + 1; while ( *t ) { t++; len++; } if ( sw->lals_len!=len ) return 0; if ( memcmp(sw->lals,rec->d.allele[0],len) ) return 0; return 1; } static void sw_rec_save(bcf_sweep_t *sw, bcf1_t *rec) { sw->lrid = rec->rid; sw->lpos = rec->pos; sw->lnals = rec->n_allele; char *t = rec->d.allele[sw->lnals-1]; int len = t - rec->d.allele[0] + 1; while ( *t ) { t++; len++; } sw->lals_len = len; hts_expand(char, len, sw->mlals, sw->lals); memcpy(sw->lals, rec->d.allele[0], len); } static void sw_fill_buffer(bcf_sweep_t *sw) { if ( !sw->iidx ) return; sw->iidx--; int ret = hts_useek(sw->file, sw->idx[sw->iidx], 0); assert( ret==0 ); sw->nrec = 0; bcf1_t *rec = &sw->rec[sw->nrec]; while ( (ret=bcf_read1(sw->file, sw->hdr, rec))==0 ) { bcf_unpack(rec, BCF_UN_STR); // if not in the last block, stop at the saved record if ( sw->iidx+1 < sw->nidx && sw_rec_equal(sw,rec) ) break; sw->nrec++; hts_expand0(bcf1_t, sw->nrec+1, sw->mrec, sw->rec); rec = &sw->rec[sw->nrec]; } sw_rec_save(sw, &sw->rec[0]); } bcf_sweep_t *bcf_sweep_init(const char *fname) { bcf_sweep_t *sw = (bcf_sweep_t*) calloc(1,sizeof(bcf_sweep_t)); sw->file = hts_open(fname, "r"); sw->fp = hts_get_bgzfp(sw->file); bgzf_index_build_init(sw->fp); sw->hdr = bcf_hdr_read(sw->file); sw->mrec = 1; sw->rec = (bcf1_t*) calloc(sw->mrec,(sizeof(bcf1_t))); sw->block_size = 1024*1024*3; sw->direction = SW_FWD; return sw; } void bcf_sweep_destroy(bcf_sweep_t *sw) { int i; for (i=0; imrec; i++) bcf_empty1(&sw->rec[i]); free(sw->idx); free(sw->rec); free(sw->lals); bcf_hdr_destroy(sw->hdr); hts_close(sw->file); free(sw); } static void sw_seek(bcf_sweep_t *sw, int direction) { sw->direction = direction; if ( direction==SW_FWD ) hts_useek(sw->file, sw->idx[0], 0); else { sw->iidx = sw->nidx; sw->nrec = 0; } } bcf1_t *bcf_sweep_fwd(bcf_sweep_t *sw) { if ( sw->direction==SW_BWD ) sw_seek(sw, SW_FWD); long pos = hts_utell(sw->file); bcf1_t *rec = &sw->rec[0]; int ret = bcf_read1(sw->file, sw->hdr, rec); if ( ret!=0 ) // last record, get ready for sweeping backwards { sw->idx_done = 1; sw->fp->idx_build_otf = 0; sw_seek(sw, SW_BWD); return NULL; } if ( !sw->idx_done ) { if ( !sw->nidx || pos - sw->idx[sw->nidx-1] > sw->block_size ) { sw->nidx++; hts_expand(uint64_t, sw->nidx, sw->midx, sw->idx); sw->idx[sw->nidx-1] = pos; } } return rec; } bcf1_t *bcf_sweep_bwd(bcf_sweep_t *sw) { if ( sw->direction==SW_FWD ) sw_seek(sw, SW_BWD); if ( !sw->nrec ) sw_fill_buffer(sw); if ( !sw->nrec ) return NULL; return &sw->rec[ --sw->nrec ]; } bcf_hdr_t *bcf_sweep_hdr(bcf_sweep_t *sw) { return sw->hdr; } stacks-1.44/htslib/._vcfutils.c000644 000765 000024 00000000265 12750743246 017224 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/vcfutils.c000644 000765 000024 00000064061 12750743246 017013 0ustar00catchenstaff000000 000000 /* vcfutils.c -- allele-related utility functions. Copyright (C) 2012-2015 Genome Research Ltd. Author: Petr Danecek Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include "htslib/vcfutils.h" #include "htslib/kbitset.h" int bcf_calc_ac(const bcf_hdr_t *header, bcf1_t *line, int *ac, int which) { int i; for (i=0; in_allele; i++) ac[i]=0; // Use INFO/AC,AN field only when asked if ( which&BCF_UN_INFO ) { bcf_unpack(line, BCF_UN_INFO); int an_id = bcf_hdr_id2int(header, BCF_DT_ID, "AN"); int ac_id = bcf_hdr_id2int(header, BCF_DT_ID, "AC"); int i, an=-1, ac_len=0, ac_type=0; uint8_t *ac_ptr=NULL; if ( an_id>=0 && ac_id>=0 ) { for (i=0; in_info; i++) { bcf_info_t *z = &line->d.info[i]; if ( z->key == an_id ) an = z->v1.i; else if ( z->key == ac_id ) { ac_ptr = z->vptr; ac_len = z->len; ac_type = z->type; } } } if ( an>=0 && ac_ptr ) { int nac = 0; #define BRANCH_INT(type_t) { \ type_t *p = (type_t *) ac_ptr; \ for (i=0; iid[BCF_DT_CTG][line->rid].key, line->pos+1); exit(1); break; } #undef BRANCH_INT if ( anid[BCF_DT_CTG][line->rid].key, line->pos+1); exit(1); } ac[0] = an - nac; return 1; } } // Split genotype fields only when asked if ( which&BCF_UN_FMT ) { int i, gt_id = bcf_hdr_id2int(header,BCF_DT_ID,"GT"); if ( gt_id<0 ) return 0; bcf_unpack(line, BCF_UN_FMT); bcf_fmt_t *fmt_gt = NULL; for (i=0; i<(int)line->n_fmt; i++) if ( line->d.fmt[i].id==gt_id ) { fmt_gt = &line->d.fmt[i]; break; } if ( !fmt_gt ) return 0; #define BRANCH_INT(type_t,vector_end) { \ for (i=0; in_sample; i++) \ { \ type_t *p = (type_t*) (fmt_gt->p + i*fmt_gt->size); \ int ial; \ for (ial=0; ialn; ial++) \ { \ if ( p[ial]==vector_end ) break; /* smaller ploidy */ \ if ( bcf_gt_is_missing(p[ial]) ) continue; /* missing allele */ \ if ( p[ial]>>1 > line->n_allele ) \ { \ fprintf(stderr,"[E::%s] Incorrect allele (\"%d\") in %s at %s:%d\n", __func__,(p[ial]>>1)-1, header->samples[i],header->id[BCF_DT_CTG][line->rid].key, line->pos+1); \ exit(1); \ } \ ac[(p[ial]>>1)-1]++; \ } \ } \ } switch (fmt_gt->type) { case BCF_BT_INT8: BRANCH_INT(int8_t, bcf_int8_vector_end); break; case BCF_BT_INT16: BRANCH_INT(int16_t, bcf_int16_vector_end); break; case BCF_BT_INT32: BRANCH_INT(int32_t, bcf_int32_vector_end); break; default: fprintf(stderr, "[E::%s] todo: %d at %s:%d\n", __func__, fmt_gt->type, header->id[BCF_DT_CTG][line->rid].key, line->pos+1); exit(1); break; } #undef BRANCH_INT return 1; } return 0; } int bcf_gt_type(bcf_fmt_t *fmt_ptr, int isample, int *_ial, int *_jal) { int i, nals = 0, has_ref = 0, has_alt = 0, ial = 0, jal = 0; #define BRANCH_INT(type_t,vector_end) { \ type_t *p = (type_t*) (fmt_ptr->p + isample*fmt_ptr->size); \ for (i=0; in; i++) \ { \ if ( p[i] == vector_end ) break; /* smaller ploidy */ \ if ( bcf_gt_is_missing(p[i]) ) return GT_UNKN; /* missing allele */ \ int tmp = p[i]>>1; \ if ( tmp>1 ) \ { \ if ( !ial ) { ial = tmp; has_alt = 1; } \ else if ( tmp!=ial ) \ { \ if ( tmptype) { case BCF_BT_INT8: BRANCH_INT(int8_t, bcf_int8_vector_end); break; case BCF_BT_INT16: BRANCH_INT(int16_t, bcf_int16_vector_end); break; case BCF_BT_INT32: BRANCH_INT(int32_t, bcf_int32_vector_end); break; default: fprintf(stderr, "[E::%s] todo: fmt_type %d\n", __func__, fmt_ptr->type); exit(1); break; } #undef BRANCH_INT if ( _ial ) *_ial = ial>0 ? ial-1 : ial; if ( _jal ) *_jal = jal>0 ? jal-1 : jal; if ( !nals ) return GT_UNKN; if ( nals==1 ) return has_ref ? GT_HAPL_R : GT_HAPL_A; if ( !has_ref ) return has_alt==1 ? GT_HOM_AA : GT_HET_AA; if ( !has_alt ) return GT_HOM_RR; return GT_HET_RA; } int bcf_trim_alleles(const bcf_hdr_t *header, bcf1_t *line) { int i; bcf_fmt_t *gt = bcf_get_fmt(header, line, "GT"); if ( !gt ) return 0; int *ac = (int*) calloc(line->n_allele,sizeof(int)); // check if all alleles are populated #define BRANCH(type_t,vector_end) { \ for (i=0; in_sample; i++) \ { \ type_t *p = (type_t*) (gt->p + i*gt->size); \ int ial; \ for (ial=0; ialn; ial++) \ { \ if ( p[ial]==vector_end ) break; /* smaller ploidy */ \ if ( bcf_gt_is_missing(p[ial]) ) continue; /* missing allele */ \ if ( (p[ial]>>1)-1 >= line->n_allele ) { free(ac); return -1; } \ ac[(p[ial]>>1)-1]++; \ } \ } \ } switch (gt->type) { case BCF_BT_INT8: BRANCH(int8_t, bcf_int8_vector_end); break; case BCF_BT_INT16: BRANCH(int16_t, bcf_int16_vector_end); break; case BCF_BT_INT32: BRANCH(int32_t, bcf_int32_vector_end); break; default: fprintf(stderr, "[E::%s] todo: %d at %s:%d\n", __func__, gt->type, header->id[BCF_DT_CTG][line->rid].key, line->pos+1); exit(1); break; } #undef BRANCH int nrm = 0; kbitset_t *rm_set = kbs_init(line->n_allele); for (i=1; in_allele; i++) { if ( !ac[i] ) { kbs_insert(rm_set, i); nrm++; } } free(ac); if ( nrm ) bcf_remove_allele_set(header, line, rm_set); kbs_destroy(rm_set); return nrm; } void bcf_remove_alleles(const bcf_hdr_t *header, bcf1_t *line, int rm_mask) { int i; kbitset_t *rm_set = kbs_init(line->n_allele); for (i=1; in_allele; i++) if ( rm_mask & 1<n_allele, sizeof(int)); // create map of indexes from old to new ALT numbering and modify ALT kstring_t str = {0,0,0}; kputs(line->d.allele[0], &str); int nrm = 0, i,j; // i: ori alleles, j: new alleles for (i=1, j=1; in_allele; i++) { if ( kbs_exists(rm_set, i) ) { // remove this allele line->d.allele[i] = NULL; nrm++; continue; } kputc(',', &str); kputs(line->d.allele[i], &str); map[i] = j; j++; } if ( !nrm ) { free(map); free(str.s); return; } int nR_ori = line->n_allele; int nR_new = line->n_allele-nrm; assert(nR_new > 0); // should not be able to remove reference allele int nA_ori = nR_ori-1; int nA_new = nR_new-1; int nG_ori = nR_ori*(nR_ori + 1)/2; int nG_new = nR_new*(nR_new + 1)/2; bcf_update_alleles_str(header, line, str.s); // remove from Number=G, Number=R and Number=A INFO fields. uint8_t *dat = NULL; int mdat = 0, ndat = 0, mdat_bytes = 0, nret; for (i=0; in_info; i++) { bcf_info_t *info = &line->d.info[i]; int vlen = bcf_hdr_id2length(header,BCF_HL_INFO,info->key); if ( vlen!=BCF_VL_A && vlen!=BCF_VL_G && vlen!=BCF_VL_R ) continue; // no need to change int type = bcf_hdr_id2type(header,BCF_HL_INFO,info->key); if ( type==BCF_HT_FLAG ) continue; int size = 1; if ( type==BCF_HT_REAL || type==BCF_HT_INT ) size = 4; mdat = mdat_bytes / size; nret = bcf_get_info_values(header, line, bcf_hdr_int2id(header,BCF_DT_ID,info->key), (void**)&dat, &mdat, type); mdat_bytes = mdat * size; if ( nret<0 ) { fprintf(stderr,"[%s:%d %s] Could not access INFO/%s at %s:%d [%d]\n", __FILE__,__LINE__,__FUNCTION__, bcf_hdr_int2id(header,BCF_DT_ID,info->key), bcf_seqname(header,line), line->pos+1, nret); exit(1); } if ( type==BCF_HT_STR ) { str.l = 0; char *ss = (char*) dat, *se = (char*) dat; if ( vlen==BCF_VL_A || vlen==BCF_VL_R ) { int nexp, inc = 0; if ( vlen==BCF_VL_A ) { nexp = nA_ori; inc = 1; } else nexp = nR_ori; for (j=0; jkey), (void*)str.s, str.l, type); if ( nret<0 ) { fprintf(stderr,"[%s:%d %s] Could not update INFO/%s at %s:%d [%d]\n", __FILE__,__LINE__,__FUNCTION__, bcf_hdr_int2id(header,BCF_DT_ID,info->key), bcf_seqname(header,line), line->pos+1, nret); exit(1); } continue; } if ( vlen==BCF_VL_A || vlen==BCF_VL_R ) { int inc = 0, ntop; if ( vlen==BCF_VL_A ) { assert( nret==nA_ori ); ntop = nA_ori; ndat = nA_new; inc = 1; } else { assert( nret==nR_ori ); ntop = nR_ori; ndat = nR_new; } int k = 0; #define BRANCH(type_t,is_vector_end) \ { \ type_t *ptr = (type_t*) dat; \ int size = sizeof(type_t); \ for (j=0; jkey), (void*)dat, ndat, type); if ( nret<0 ) { fprintf(stderr,"[%s:%d %s] Could not update INFO/%s at %s:%d [%d]\n", __FILE__,__LINE__,__FUNCTION__, bcf_hdr_int2id(header,BCF_DT_ID,info->key), bcf_seqname(header,line), line->pos+1, nret); exit(1); } } // Update GT fields, the allele indexes might have changed for (i=1; in_allele; i++) if ( map[i]!=i ) break; if ( in_allele ) { mdat = mdat_bytes / 4; // sizeof(int32_t) nret = bcf_get_genotypes(header,line,(void**)&dat,&mdat); mdat_bytes = mdat * 4; if ( nret>0 ) { nret /= line->n_sample; int32_t *ptr = (int32_t*) dat; for (i=0; in_sample; i++) { for (j=0; j=0 ); ptr[j] = (map[al]+1)<<1 | (ptr[j]&1); } ptr += nret; } bcf_update_genotypes(header, line, (void*)dat, nret*line->n_sample); } } // Remove from Number=G, Number=R and Number=A FORMAT fields. // Assuming haploid or diploid GTs for (i=0; in_fmt; i++) { bcf_fmt_t *fmt = &line->d.fmt[i]; int vlen = bcf_hdr_id2length(header,BCF_HL_FMT,fmt->id); if ( vlen!=BCF_VL_A && vlen!=BCF_VL_G && vlen!=BCF_VL_R ) continue; // no need to change int type = bcf_hdr_id2type(header,BCF_HL_FMT,fmt->id); if ( type==BCF_HT_FLAG ) continue; int size = 1; if ( type==BCF_HT_REAL || type==BCF_HT_INT ) size = 4; mdat = mdat_bytes / size; nret = bcf_get_format_values(header, line, bcf_hdr_int2id(header,BCF_DT_ID,fmt->id), (void**)&dat, &mdat, type); mdat_bytes = mdat * size; if ( nret<0 ) { fprintf(stderr,"[%s:%d %s] Could not access FORMAT/%s at %s:%d [%d]\n", __FILE__,__LINE__,__FUNCTION__, bcf_hdr_int2id(header,BCF_DT_ID,fmt->id), bcf_seqname(header,line), line->pos+1, nret); exit(1); } if ( type==BCF_HT_STR ) { int size = nret/line->n_sample; // number of bytes per sample str.l = 0; if ( vlen==BCF_VL_A || vlen==BCF_VL_R ) { int nexp, inc = 0; if ( vlen==BCF_VL_A ) { nexp = nA_ori; inc = 1; } else nexp = nR_ori; for (j=0; jn_sample; j++) { char *ss = ((char*)dat) + j*size, *se = ss + size, *ptr = ss; int k_src = 0, k_dst = 0, l = str.l; for (k_src=0; k_src=se || !*ptr) break; while ( ptrn_sample; j++) { char *ss = ((char*)dat) + j*size, *se = ss + size, *ptr = ss; int k_src = 0, k_dst = 0, l = str.l; int nexp = 0; // diploid or haploid? while ( ptr=se || !*ptr ) break; while ( ptr=se || !*ptr ) break; } } else // haploid { for (k_src=0; k_src=se || !*ptr ) break; while ( ptrid), (void*)str.s, str.l, type); if ( nret<0 ) { fprintf(stderr,"[%s:%d %s] Could not update FORMAT/%s at %s:%d [%d]\n", __FILE__,__LINE__,__FUNCTION__, bcf_hdr_int2id(header,BCF_DT_ID,fmt->id), bcf_seqname(header,line), line->pos+1, nret); exit(1); } continue; } int nori = nret / line->n_sample; if ( vlen==BCF_VL_A || vlen==BCF_VL_R || (vlen==BCF_VL_G && nori==nR_ori) ) // Number=A, R or haploid Number=G { int inc = 0, nnew; if ( vlen==BCF_VL_A ) { assert( nori==nA_ori ); // todo: will fail if all values are missing ndat = nA_new*line->n_sample; nnew = nA_new; inc = 1; } else { assert( nori==nR_ori ); // todo: will fail if all values are missing ndat = nR_new*line->n_sample; nnew = nR_new; } #define BRANCH(type_t,is_vector_end) \ { \ for (j=0; jn_sample; j++) \ { \ type_t *ptr_src = ((type_t*)dat) + j*nori; \ type_t *ptr_dst = ((type_t*)dat) + j*nnew; \ int size = sizeof(type_t); \ int k_src, k_dst = 0; \ for (k_src=0; k_srcn_sample; #define BRANCH(type_t,is_vector_end) \ { \ for (j=0; jn_sample; j++) \ { \ type_t *ptr_src = ((type_t*)dat) + j*nori; \ type_t *ptr_dst = ((type_t*)dat) + j*nG_new; \ int size = sizeof(type_t); \ int ia, ib, k_dst = 0, k_src; \ int nset = 0; /* haploid or diploid? */ \ for (k_src=0; k_srcid), (void*)dat, ndat, type); if ( nret<0 ) { fprintf(stderr,"[%s:%d %s] Could not update FORMAT/%s at %s:%d [%d]\n", __FILE__,__LINE__,__FUNCTION__, bcf_hdr_int2id(header,BCF_DT_ID,fmt->id), bcf_seqname(header,line), line->pos+1, nret); exit(1); } } free(dat); free(str.s); free(map); } stacks-1.44/htslib/htslib/._bgzf.h000644 000765 000024 00000000265 12750743246 017607 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/bgzf.h000644 000765 000024 00000026237 12750743246 017401 0ustar00catchenstaff000000 000000 /* The MIT License Copyright (c) 2008 Broad Institute / Massachusetts Institute of Technology 2011, 2012 Attractive Chaos Copyright (C) 2009, 2013, 2014 Genome Research Ltd Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ /* The BGZF library was originally written by Bob Handsaker from the Broad * Institute. It was later improved by the SAMtools developers. */ #ifndef HTSLIB_BGZF_H #define HTSLIB_BGZF_H #include #include #include #include #include "hts_defs.h" #ifdef __cplusplus extern "C" { #endif #define BGZF_BLOCK_SIZE 0xff00 // make sure compressBound(BGZF_BLOCK_SIZE) < BGZF_MAX_BLOCK_SIZE #define BGZF_MAX_BLOCK_SIZE 0x10000 #define BGZF_ERR_ZLIB 1 #define BGZF_ERR_HEADER 2 #define BGZF_ERR_IO 4 #define BGZF_ERR_MISUSE 8 struct hFILE; struct bgzf_mtaux_t; typedef struct __bgzidx_t bgzidx_t; struct BGZF { unsigned errcode:16, is_write:2, is_be:2; signed compress_level:9; unsigned is_compressed:2, is_gzip:1; int cache_size; int block_length, block_offset; int64_t block_address, uncompressed_address; void *uncompressed_block, *compressed_block; void *cache; // a pointer to a hash table struct hFILE *fp; // actual file handle struct bgzf_mtaux_t *mt; // only used for multi-threading bgzidx_t *idx; // BGZF index int idx_build_otf; // build index on the fly, set by bgzf_index_build_init() z_stream *gz_stream;// for gzip-compressed files }; #ifndef HTS_BGZF_TYPEDEF typedef struct BGZF BGZF; #define HTS_BGZF_TYPEDEF #endif #ifndef KSTRING_T #define KSTRING_T kstring_t typedef struct __kstring_t { size_t l, m; char *s; } kstring_t; #endif /****************** * Basic routines * ******************/ /** * Open an existing file descriptor for reading or writing. * * @param fd file descriptor * @param mode mode matching /[rwag][u0-9]+/: 'r' for reading, 'w' for * writing, 'a' for appending, 'g' for gzip rather than BGZF * compression (with 'w' only), and digit specifies the zlib * compression level. * Note that there is a distinction between 'u' and '0': the * first yields plain uncompressed output whereas the latter * outputs uncompressed data wrapped in the zlib format. * @return BGZF file handler; 0 on error */ BGZF* bgzf_dopen(int fd, const char *mode); #define bgzf_fdopen(fd, mode) bgzf_dopen((fd), (mode)) // for backward compatibility /** * Open the specified file for reading or writing. */ BGZF* bgzf_open(const char* path, const char *mode); /** * Open an existing hFILE stream for reading or writing. */ BGZF* bgzf_hopen(struct hFILE *fp, const char *mode); /** * Close the BGZF and free all associated resources. * * @param fp BGZF file handler * @return 0 on success and -1 on error */ int bgzf_close(BGZF *fp); /** * Read up to _length_ bytes from the file storing into _data_. * * @param fp BGZF file handler * @param data data array to read into * @param length size of data to read * @return number of bytes actually read; 0 on end-of-file and -1 on error */ ssize_t bgzf_read(BGZF *fp, void *data, size_t length) HTS_RESULT_USED; /** * Write _length_ bytes from _data_ to the file. If no I/O errors occur, * the complete _length_ bytes will be written (or queued for writing). * * @param fp BGZF file handler * @param data data array to write * @param length size of data to write * @return number of bytes written (i.e., _length_); negative on error */ ssize_t bgzf_write(BGZF *fp, const void *data, size_t length) HTS_RESULT_USED; /** * Read up to _length_ bytes directly from the underlying stream without * decompressing. Bypasses BGZF blocking, so must be used with care in * specialised circumstances only. * * @param fp BGZF file handler * @param data data array to read into * @param length number of raw bytes to read * @return number of bytes actually read; 0 on end-of-file and -1 on error */ ssize_t bgzf_raw_read(BGZF *fp, void *data, size_t length) HTS_RESULT_USED; /** * Write _length_ bytes directly to the underlying stream without * compressing. Bypasses BGZF blocking, so must be used with care * in specialised circumstances only. * * @param fp BGZF file handler * @param data data array to write * @param length number of raw bytes to write * @return number of bytes actually written; -1 on error */ ssize_t bgzf_raw_write(BGZF *fp, const void *data, size_t length) HTS_RESULT_USED; /** * Write the data in the buffer to the file. * * @param fp BGZF file handle * @return 0 on success and -1 on error */ int bgzf_flush(BGZF *fp) HTS_RESULT_USED; /** * Return a virtual file pointer to the current location in the file. * No interpetation of the value should be made, other than a subsequent * call to bgzf_seek can be used to position the file at the same point. * Return value is non-negative on success. */ #define bgzf_tell(fp) (((fp)->block_address << 16) | ((fp)->block_offset & 0xFFFF)) /** * Set the file to read from the location specified by _pos_. * * @param fp BGZF file handler * @param pos virtual file offset returned by bgzf_tell() * @param whence must be SEEK_SET * @return 0 on success and -1 on error */ int64_t bgzf_seek(BGZF *fp, int64_t pos, int whence) HTS_RESULT_USED; /** * Check if the BGZF end-of-file (EOF) marker is present * * @param fp BGZF file handler opened for reading * @return 1 if the EOF marker is present and correct; * 2 if it can't be checked, e.g., because fp isn't seekable; * 0 if the EOF marker is absent; * -1 (with errno set) on error */ int bgzf_check_EOF(BGZF *fp); /** * Check if a file is in the BGZF format * * @param fn file name * @return 1 if _fn_ is BGZF; 0 if not or on I/O error */ int bgzf_is_bgzf(const char *fn); /********************* * Advanced routines * *********************/ /** * Set the cache size. Only effective when compiled with -DBGZF_CACHE. * * @param fp BGZF file handler * @param size size of cache in bytes; 0 to disable caching (default) */ void bgzf_set_cache_size(BGZF *fp, int size); /** * Flush the file if the remaining buffer size is smaller than _size_ * @return 0 if flushing succeeded or was not needed; negative on error */ int bgzf_flush_try(BGZF *fp, ssize_t size) HTS_RESULT_USED; /** * Read one byte from a BGZF file. It is faster than bgzf_read() * @param fp BGZF file handler * @return byte read; -1 on end-of-file or error */ int bgzf_getc(BGZF *fp); /** * Read one line from a BGZF file. It is faster than bgzf_getc() * * @param fp BGZF file handler * @param delim delimitor * @param str string to write to; must be initialized * @return length of the string; 0 on end-of-file; negative on error */ int bgzf_getline(BGZF *fp, int delim, kstring_t *str); /** * Read the next BGZF block. */ int bgzf_read_block(BGZF *fp) HTS_RESULT_USED; /** * Enable multi-threading (only effective on writing and when the * library was compiled with -DBGZF_MT) * * @param fp BGZF file handler; must be opened for writing * @param n_threads #threads used for writing * @param n_sub_blks #blocks processed by each thread; a value 64-256 is recommended */ int bgzf_mt(BGZF *fp, int n_threads, int n_sub_blks); /** * Compress a single BGZF block. * * @param dst output buffer (must have size >= BGZF_MAX_BLOCK_SIZE) * @param dlen size of output buffer; updated on return to the number * of bytes actually written to dst * @param src buffer to be compressed * @param slen size of data to compress (must be <= BGZF_BLOCK_SIZE) * @param level compression level * @return 0 on success and negative on error */ int bgzf_compress(void *dst, size_t *dlen, const void *src, size_t slen, int level); /******************* * bgzidx routines * *******************/ /** * Position BGZF at the uncompressed offset * * @param fp BGZF file handler; must be opened for reading * @param uoffset file offset in the uncompressed data * @param where SEEK_SET supported atm * * Returns 0 on success and -1 on error. */ int bgzf_useek(BGZF *fp, long uoffset, int where) HTS_RESULT_USED; /** * Position in uncompressed BGZF * * @param fp BGZF file handler; must be opened for reading * * Returns the current offset on success and -1 on error. */ long bgzf_utell(BGZF *fp); /** * Tell BGZF to build index while compressing. * * @param fp BGZF file handler; can be opened for reading or writing. * * Returns 0 on success and -1 on error. */ int bgzf_index_build_init(BGZF *fp); /** * Load BGZF index * * @param fp BGZF file handler * @param bname base name * @param suffix suffix to add to bname (can be NULL) * * Returns 0 on success and -1 on error. */ int bgzf_index_load(BGZF *fp, const char *bname, const char *suffix); /** * Save BGZF index * * @param fp BGZF file handler * @param bname base name * @param suffix suffix to add to bname (can be NULL) * * Returns 0 on success and -1 on error. */ int bgzf_index_dump(BGZF *fp, const char *bname, const char *suffix) HTS_RESULT_USED; #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/htslib/._cram.h000644 000765 000024 00000000265 12750743246 017601 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/cram.h000644 000765 000024 00000034470 12750743246 017371 0ustar00catchenstaff000000 000000 /* cram.h -- public CRAM-specific API functions. Copyright (C) 2015 Genome Research Ltd. Author: James Bonfield Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ /*! \file * CRAM interface. * * Consider using the higher level hts_*() API for programs that wish to * be file format agnostic (see htslib/hts.h). * * This API should be used for CRAM specific code. The specifics of the * public API are implemented in cram_io.h, cram_encode.h and cram_decode.h * although these should not be included directly (use this file instead). */ #ifndef HTSLIB_CRAM_H #define HTSLIB_CRAM_H #include #include #include #include "hts.h" #ifdef __cplusplus extern "C" { #endif #ifndef _CRAM_STRUCTS_H_ enum cram_block_method { ERROR = -1, RAW = 0, GZIP = 1, BZIP2 = 2, LZMA = 3, RANS = 4, // Generic; either order RANS0 = 4, RANS1 = 10, // Not externalised; stored as RANS (generic) GZIP_RLE = 11, // NB: not externalised in CRAM }; enum cram_content_type { CT_ERROR = -1, FILE_HEADER = 0, COMPRESSION_HEADER = 1, MAPPED_SLICE = 2, UNMAPPED_SLICE = 3, // CRAM V1.0 only EXTERNAL = 4, CORE = 5, }; // Opaque data types, see cram_structs for the fully fledged versions. typedef struct SAM_hdr SAM_hdr; typedef struct cram_file_def cram_file_def; typedef struct cram_fd cram_fd; typedef struct cram_container cram_container; typedef struct cram_block cram_block; typedef struct cram_slice cram_slice; typedef struct cram_metrics cram_metrics; typedef struct cram_block_slice_hdr cram_block_slice_hdr; typedef struct cram_block_compression_hdr cram_block_compression_hdr; typedef struct refs_t refs_t; // need this? struct hFILE; #endif // Accessor functions /* *----------------------------------------------------------------------------- * cram_fd */ SAM_hdr *cram_fd_get_header(cram_fd *fd); void cram_fd_set_header(cram_fd *fd, SAM_hdr *hdr); int cram_fd_get_version(cram_fd *fd); void cram_fd_set_version(cram_fd *fd, int vers); int cram_major_vers(cram_fd *fd); int cram_minor_vers(cram_fd *fd); struct hFILE *cram_fd_get_fp(cram_fd *fd); void cram_fd_set_fp(cram_fd *fd, struct hFILE *fp); /* *----------------------------------------------------------------------------- * cram_container */ int32_t cram_container_get_length(cram_container *c); void cram_container_set_length(cram_container *c, int32_t length); int32_t cram_container_get_num_blocks(cram_container *c); void cram_container_set_num_blocks(cram_container *c, int32_t num_blocks); int32_t *cram_container_get_landmarks(cram_container *c, int32_t *num_landmarks); void cram_container_set_landmarks(cram_container *c, int32_t num_landmarks, int32_t *landmarks); /* Returns true if the container is empty (EOF marker) */ int cram_container_is_empty(cram_fd *fd); /* *----------------------------------------------------------------------------- * cram_block */ int32_t cram_block_get_content_id(cram_block *b); int32_t cram_block_get_comp_size(cram_block *b); int32_t cram_block_get_uncomp_size(cram_block *b); int32_t cram_block_get_crc32(cram_block *b); void * cram_block_get_data(cram_block *b); enum cram_content_type cram_block_get_content_type(cram_block *b); void cram_block_set_content_id(cram_block *b, int32_t id); void cram_block_set_comp_size(cram_block *b, int32_t size); void cram_block_set_uncomp_size(cram_block *b, int32_t size); void cram_block_set_crc32(cram_block *b, int32_t crc); void cram_block_set_data(cram_block *b, void *data); int cram_block_append(cram_block *b, void *data, int size); void cram_block_update_size(cram_block *b); // Offset is known as "size" internally, but it can be confusing. size_t cram_block_get_offset(cram_block *b); void cram_block_set_offset(cram_block *b, size_t offset); /* * Computes the size of a cram block, including the block * header itself. */ uint32_t cram_block_size(cram_block *b); /* * Renumbers RG numbers in a cram compression header. * * CRAM stores RG as the Nth number in the header, rather than a * string holding the ID: tag. This is smaller in space, but means * "samtools cat" to join files together that contain single but * different RG lines needs a way of renumbering them. * * The file descriptor is expected to be immediately after the * cram_container structure (ie before the cram compression header). * Due to the nature of the CRAM format, this needs to read and write * the blocks itself. Note that there may be multiple slices within * the container, meaning multiple compression headers to manipulate. * Changing RG may change the size of the compression header and * therefore the length field in the container. Hence we rewrite all * blocks just incase and also emit the adjusted container. * * The current implementation can only cope with renumbering a single * RG (and only then if it is using HUFFMAN or BETA codecs). In * theory it *may* be possible to renumber multiple RGs if they use * HUFFMAN to the CORE block or use an external block unshared by any * other data series. So we have an API that can be upgraded to * support this, but do not implement it for now. An example * implementation of RG as an EXTERNAL block would be to find that * block and rewrite it, returning the number of blocks consumed. * * Returns 0 on success; * -1 if unable to edit; * -2 on other errors (eg I/O). */ int cram_transcode_rg(cram_fd *in, cram_fd *out, cram_container *c, int nrg, int *in_rg, int *out_rg); /* * Copies the blocks representing the next num_slice slices from a * container from 'in' to 'out'. It is expected that the file pointer * is just after the read of the cram_container and cram compression * header. * * Returns 0 on success * -1 on failure */ int cram_copy_slice(cram_fd *in, cram_fd *out, int32_t num_slice); /* *----------------------------------------------------------------------------- * SAM_hdr */ /*! Tokenises a SAM header into a hash table. * * Also extracts a few bits on specific data types, such as @RG lines. * * @return * Returns a SAM_hdr struct on success (free with sam_hdr_free()); * NULL on failure */ SAM_hdr *sam_hdr_parse_(const char *hdr, int len); /* *----------------------------------------------------------------------------- * cram_io basics */ /**@{ ---------------------------------------------------------------------- * CRAM blocks - the dynamically growable data block. We have code to * create, update, (un)compress and read/write. * * These are derived from the deflate_interlaced.c blocks, but with the * CRAM extension of content types and IDs. */ /*! Allocates a new cram_block structure with a specified content_type and * id. * * @return * Returns block pointer on success; * NULL on failure */ cram_block *cram_new_block(enum cram_content_type content_type, int content_id); /*! Reads a block from a cram file. * * @return * Returns cram_block pointer on success; * NULL on failure */ cram_block *cram_read_block(cram_fd *fd); /*! Writes a CRAM block. * * @return * Returns 0 on success; * -1 on failure */ int cram_write_block(cram_fd *fd, cram_block *b); /*! Frees a CRAM block, deallocating internal data too. */ void cram_free_block(cram_block *b); /*! Uncompresses a CRAM block, if compressed. * * @return * Returns 0 on success; * -1 on failure */ int cram_uncompress_block(cram_block *b); /*! Compresses a block. * * Compresses a block using one of two different zlib strategies. If we only * want one choice set strat2 to be -1. * * The logic here is that sometimes Z_RLE does a better job than Z_FILTERED * or Z_DEFAULT_STRATEGY on quality data. If so, we'd rather use it as it is * significantly faster. * * @return * Returns 0 on success; * -1 on failure */ int cram_compress_block(cram_fd *fd, cram_block *b, cram_metrics *metrics, int method, int level); /**@}*/ /**@{ ---------------------------------------------------------------------- * Containers */ /*! Creates a new container, specifying the maximum number of slices * and records permitted. * * @return * Returns cram_container ptr on success; * NULL on failure */ cram_container *cram_new_container(int nrec, int nslice); void cram_free_container(cram_container *c); /*! Reads a container header. * * @return * Returns cram_container on success; * NULL on failure or no container left (fd->err == 0). */ cram_container *cram_read_container(cram_fd *fd); /*! Writes a container structure. * * @return * Returns 0 on success; * -1 on failure */ int cram_write_container(cram_fd *fd, cram_container *h); /* * Stores the container structure in dat and returns *size as the * number of bytes written to dat[]. The input size of dat is also * held in *size and should be initialised to cram_container_size(c). * * Returns 0 on success; * -1 on failure */ int cram_store_container(cram_fd *fd, cram_container *c, char *dat, int *size); int cram_container_size(cram_container *c); /**@}*/ /**@{ ---------------------------------------------------------------------- * The top-level cram opening, closing and option handling */ /*! Opens a CRAM file for read (mode "rb") or write ("wb"). * * The filename may be "-" to indicate stdin or stdout. * * @return * Returns file handle on success; * NULL on failure. */ cram_fd *cram_open(const char *filename, const char *mode); /*! Opens an existing stream for reading or writing. * * @return * Returns file handle on success; * NULL on failure. */ cram_fd *cram_dopen(struct hFILE *fp, const char *filename, const char *mode); /*! Closes a CRAM file. * * @return * Returns 0 on success; * -1 on failure */ int cram_close(cram_fd *fd); /* * Seek within a CRAM file. * * Returns 0 on success * -1 on failure */ int cram_seek(cram_fd *fd, off_t offset, int whence); /* * Flushes a CRAM file. * Useful for when writing to stdout without wishing to close the stream. * * Returns 0 on success * -1 on failure */ int cram_flush(cram_fd *fd); /*! Checks for end of file on a cram_fd stream. * * @return * Returns 0 if not at end of file * 1 if we hit an expected EOF (end of range or EOF block) * 2 for other EOF (end of stream without EOF block) */ int cram_eof(cram_fd *fd); /*! Sets options on the cram_fd. * * See CRAM_OPT_* definitions in hts.h. * Use this immediately after opening. * * @return * Returns 0 on success; * -1 on failure */ int cram_set_option(cram_fd *fd, enum hts_fmt_option opt, ...); /*! Sets options on the cram_fd. * * See CRAM_OPT_* definitions in hts.h. * Use this immediately after opening. * * @return * Returns 0 on success; * -1 on failure */ int cram_set_voption(cram_fd *fd, enum hts_fmt_option opt, va_list args); /*! * Attaches a header to a cram_fd. * * This should be used when creating a new cram_fd for writing where * we have an SAM_hdr already constructed (eg from a file we've read * in). * * @return * Returns 0 on success; * -1 on failure */ int cram_set_header(cram_fd *fd, SAM_hdr *hdr); /* As int32_decoded/encode, but from/to blocks instead of cram_fd */ int int32_put_blk(cram_block *b, int32_t val); /**@}*/ /**@{ -------------------------------------------------------------------*/ /*! Deallocates all storage used by a SAM_hdr struct. * * This also decrements the header reference count. If after decrementing * it is still non-zero then the header is assumed to be in use by another * caller and the free is not done. * * This is a synonym for sam_hdr_dec_ref(). */ void sam_hdr_free(SAM_hdr *hdr); /*! Returns the current length of the SAM_hdr in text form. * * Call sam_hdr_rebuild() first if editing has taken place. */ int sam_hdr_length(SAM_hdr *hdr); /*! Returns the string form of the SAM_hdr. * * Call sam_hdr_rebuild() first if editing has taken place. */ char *sam_hdr_str(SAM_hdr *hdr); /*! Appends a formatted line to an existing SAM header. * * Line is a full SAM header record, eg "@SQ\tSN:foo\tLN:100", with * optional new-line. If it contains more than 1 line then multiple lines * will be added in order. * * Len is the length of the text data, or 0 if unknown (in which case * it should be null terminated). * * @return * Returns 0 on success; * -1 on failure */ /*! Add an @PG line. * * If we wish complete control over this use sam_hdr_add() directly. This * function uses that, but attempts to do a lot of tedious house work for * you too. * * - It will generate a suitable ID if the supplied one clashes. * - It will generate multiple @PG records if we have multiple PG chains. * * Call it as per sam_hdr_add() with a series of key,value pairs ending * in NULL. * * @return * Returns 0 on success; * -1 on failure */ int sam_hdr_add_PG(SAM_hdr *sh, const char *name, ...); /*! * A function to help with construction of CL tags in @PG records. * Takes an argc, argv pair and returns a single space-separated string. * This string should be deallocated by the calling function. * * @return * Returns malloced char * on success; * NULL on failure */ char *stringify_argv(int argc, char *argv[]); /**@}*/ #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/htslib/._faidx.h000644 000765 000024 00000000265 12750743246 017752 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/faidx.h000644 000765 000024 00000010722 12750743246 017534 0ustar00catchenstaff000000 000000 /* faidx.h -- FASTA random access. Copyright (C) 2008, 2009, 2013, 2014 Genome Research Ltd. Author: Heng Li Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #ifndef HTSLIB_FAIDX_H #define HTSLIB_FAIDX_H #include "hts_defs.h" #ifdef __cplusplus extern "C" { #endif /*! @header Index FASTA files and extract subsequence. The fai file index columns are: - chromosome name - chromosome length: number of bases - offset: number of bytes to skip to get to the first base from the beginning of the file, including the length of the sequence description string (">chr ..\n") - line length: number of bases per line (excluding \n) - binary line length: number of bytes, including \n @copyright The Wellcome Trust Sanger Institute. */ struct __faidx_t; typedef struct __faidx_t faidx_t; /*! @abstract Build index for a FASTA or bgzip-compressed FASTA file. @param fn FASTA file name @return 0 on success; or -1 on failure @discussion File "fn.fai" will be generated. */ int fai_build(const char *fn) HTS_RESULT_USED; /*! @abstract Destroy a faidx_t struct. @param fai Pointer to the struct to be destroyed */ void fai_destroy(faidx_t *fai); /*! @abstract Load index from "fn.fai". @param fn File name of the FASTA file */ faidx_t *fai_load(const char *fn); /*! @abstract Fetch the sequence in a region. @param fai Pointer to the faidx_t struct @param reg Region in the format "chr2:20,000-30,000" @param len Length of the region; -2 if seq not present, -1 general error @return Pointer to the sequence; null on failure @discussion The returned sequence is allocated by malloc family and should be destroyed by end users by calling free() on it. */ char *fai_fetch(const faidx_t *fai, const char *reg, int *len); /*! @abstract Fetch the number of sequences. @param fai Pointer to the faidx_t struct @return The number of sequences */ int faidx_fetch_nseq(const faidx_t *fai) HTS_DEPRECATED("Please use faidx_nseq instead"); /*! @abstract Fetch the sequence in a region. @param fai Pointer to the faidx_t struct @param c_name Region name @param p_beg_i Beginning position number (zero-based) @param p_end_i End position number (zero-based) @param len Length of the region; -2 if c_name not present, -1 general error @return Pointer to the sequence; null on failure @discussion The returned sequence is allocated by malloc family and should be destroyed by end users by calling free() on it. */ char *faidx_fetch_seq(const faidx_t *fai, const char *c_name, int p_beg_i, int p_end_i, int *len); /*! @abstract Query if sequence is present @param fai Pointer to the faidx_t struct @param seq Sequence name @return 1 if present or 0 if absent */ int faidx_has_seq(const faidx_t *fai, const char *seq); /*! @abstract Return number of sequences in fai index */ int faidx_nseq(const faidx_t *fai); /*! @abstract Return name of i-th sequence */ const char *faidx_iseq(const faidx_t *fai, int i); /*! @abstract Return sequence length, -1 if not present */ int faidx_seq_len(const faidx_t *fai, const char *seq); #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/htslib/._hfile.h000644 000765 000024 00000000265 12750743246 017746 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/hfile.h000644 000765 000024 00000015707 12750743246 017540 0ustar00catchenstaff000000 000000 /* hfile.h -- buffered low-level input/output streams. Copyright (C) 2013-2015 Genome Research Ltd. Author: John Marshall Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #ifndef HTSLIB_HFILE_H #define HTSLIB_HFILE_H #include #include #include "hts_defs.h" #ifdef __cplusplus extern "C" { #endif /* These fields are declared here solely for the benefit of the inline functions below. They may change in future releases. User code should not use them directly; you should imagine that hFILE is an opaque incomplete type. */ struct hFILE_backend; typedef struct hFILE { char *buffer, *begin, *end, *limit; const struct hFILE_backend *backend; off_t offset; unsigned at_eof:1; int has_errno; } hFILE; /*! @abstract Open the named file or URL as a stream @return An hFILE pointer, or NULL (with errno set) if an error occurred. @notes The usual @c fopen(3) @a mode letters are supported: one of @e r (read), @e w (write), @e a (append), optionally followed by any of @e + (update), @e e (close on @c exec(2)), @e x (create exclusively). */ hFILE *hopen(const char *filename, const char *mode) HTS_RESULT_USED; /*! @abstract Associate a stream with an existing open file descriptor @return An hFILE pointer, or NULL (with errno set) if an error occurred. @notes For socket descriptors (on Windows), mode should contain 's'. */ hFILE *hdopen(int fd, const char *mode) HTS_RESULT_USED; /*! @abstract Report whether the file name or URL denotes remote storage @return 0 if local, 1 if remote. @notes "Remote" means involving e.g. explicit network access, with the implication that callers may wish to cache such files' contents locally. */ int hisremote(const char *filename) HTS_RESULT_USED; /*! @abstract Flush (for output streams) and close the stream @return 0 if successful, or EOF (with errno set) if an error occurred. */ int hclose(hFILE *fp) HTS_RESULT_USED; /*! @abstract Close the stream, without flushing or propagating errors @notes For use while cleaning up after an error only. Preserves errno. */ void hclose_abruptly(hFILE *fp); /*! @abstract Return the stream's error indicator @return Non-zero (in fact, an errno value) if an error has occurred. @notes This would be called herror() and return true/false to parallel ferror(3), but a networking-related herror(3) function already exists. */ static inline int herrno(hFILE *fp) { return fp->has_errno; } /*! @abstract Clear the stream's error indicator */ static inline void hclearerr(hFILE *fp) { fp->has_errno = 0; } /*! @abstract Reposition the read/write stream offset @return The resulting offset within the stream (as per lseek(2)), or negative if an error occurred. */ off_t hseek(hFILE *fp, off_t offset, int whence) HTS_RESULT_USED; /*! @abstract Report the current stream offset @return The offset within the stream, starting from zero. */ static inline off_t htell(hFILE *fp) { return fp->offset + (fp->begin - fp->buffer); } /*! @abstract Read one character from the stream @return The character read, or EOF on end-of-file or error */ static inline int hgetc(hFILE *fp) { extern int hgetc2(hFILE *); return (fp->end > fp->begin)? (unsigned char) *(fp->begin++) : hgetc2(fp); } /*! @abstract Peek at characters to be read without removing them from buffers @param fp The file stream @param buffer The buffer to which the peeked bytes will be written @param nbytes The number of bytes to peek at; limited by the size of the internal buffer, which could be as small as 4K. @return The number of bytes peeked, which may be less than nbytes if EOF is encountered; or negative, if there was an I/O error. @notes The characters peeked at remain in the stream's internal buffer, and will be returned by later hread() etc calls. */ ssize_t hpeek(hFILE *fp, void *buffer, size_t nbytes) HTS_RESULT_USED; /*! @abstract Read a block of characters from the file @return The number of bytes read, or negative if an error occurred. @notes The full nbytes requested will be returned, except as limited by EOF or I/O errors. */ static inline ssize_t HTS_RESULT_USED hread(hFILE *fp, void *buffer, size_t nbytes) { extern ssize_t hread2(hFILE *, void *, size_t, size_t); size_t n = fp->end - fp->begin; if (n > nbytes) n = nbytes; memcpy(buffer, fp->begin, n); fp->begin += n; return (n == nbytes)? (ssize_t) n : hread2(fp, buffer, nbytes, n); } /*! @abstract Write a character to the stream @return The character written, or EOF if an error occurred. */ static inline int hputc(int c, hFILE *fp) { extern int hputc2(int, hFILE *); if (fp->begin < fp->limit) *(fp->begin++) = c; else c = hputc2(c, fp); return c; } /*! @abstract Write a string to the stream @return 0 if successful, or EOF if an error occurred. */ static inline int hputs(const char *text, hFILE *fp) { extern int hputs2(const char *, size_t, size_t, hFILE *); size_t nbytes = strlen(text), n = fp->limit - fp->begin; if (n > nbytes) n = nbytes; memcpy(fp->begin, text, n); fp->begin += n; return (n == nbytes)? 0 : hputs2(text, nbytes, n, fp); } /*! @abstract Write a block of characters to the file @return Either nbytes, or negative if an error occurred. @notes In the absence of I/O errors, the full nbytes will be written. */ static inline ssize_t HTS_RESULT_USED hwrite(hFILE *fp, const void *buffer, size_t nbytes) { extern ssize_t hwrite2(hFILE *, const void *, size_t, size_t); size_t n = fp->limit - fp->begin; if (n > nbytes) n = nbytes; memcpy(fp->begin, buffer, n); fp->begin += n; return (n==nbytes)? (ssize_t) n : hwrite2(fp, buffer, nbytes, n); } /*! @abstract For writing streams, flush buffered output to the underlying stream @return 0 if successful, or EOF if an error occurred. */ int hflush(hFILE *fp) HTS_RESULT_USED; #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/htslib/._hts.h000644 000765 000024 00000000265 12750743246 017455 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/hts.h000644 000765 000024 00000052676 12750743246 017255 0ustar00catchenstaff000000 000000 /* hts.h -- format-neutral I/O, indexing, and iterator API functions. Copyright (C) 2012-2015 Genome Research Ltd. Copyright (C) 2012 Broad Institute. Author: Heng Li Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #ifndef HTSLIB_HTS_H #define HTSLIB_HTS_H #include #include #include "hts_defs.h" #ifdef __cplusplus extern "C" { #endif #ifndef HTS_BGZF_TYPEDEF typedef struct BGZF BGZF; #define HTS_BGZF_TYPEDEF #endif struct cram_fd; struct hFILE; #ifndef KSTRING_T #define KSTRING_T kstring_t typedef struct __kstring_t { size_t l, m; char *s; } kstring_t; #endif #ifndef kroundup32 #define kroundup32(x) (--(x), (x)|=(x)>>1, (x)|=(x)>>2, (x)|=(x)>>4, (x)|=(x)>>8, (x)|=(x)>>16, ++(x)) #endif /** * hts_expand() - expands memory block pointed to by $ptr; * hts_expand0() the latter sets the newly allocated part to 0. * * @param n requested number of elements of type type_t * @param m size of memory allocated */ #define hts_expand(type_t, n, m, ptr) if ((n) > (m)) { \ (m) = (n); kroundup32(m); \ (ptr) = (type_t*)realloc((ptr), (m) * sizeof(type_t)); \ } #define hts_expand0(type_t, n, m, ptr) if ((n) > (m)) { \ int t = (m); (m) = (n); kroundup32(m); \ (ptr) = (type_t*)realloc((ptr), (m) * sizeof(type_t)); \ memset(((type_t*)ptr)+t,0,sizeof(type_t)*((m)-t)); \ } /************ * File I/O * ************/ // Add new entries only at the end (but before the *_maximum entry) // of these enums, as their numbering is part of the htslib ABI. enum htsFormatCategory { unknown_category, sequence_data, // Sequence data -- SAM, BAM, CRAM, etc variant_data, // Variant calling data -- VCF, BCF, etc index_file, // Index file associated with some data file region_list, // Coordinate intervals or regions -- BED, etc category_maximum = 32767 }; enum htsExactFormat { unknown_format, binary_format, text_format, sam, bam, bai, cram, crai, vcf, bcf, csi, gzi, tbi, bed, format_maximum = 32767 }; enum htsCompression { no_compression, gzip, bgzf, custom, compression_maximum = 32767 }; typedef struct htsFormat { enum htsFormatCategory category; enum htsExactFormat format; struct { short major, minor; } version; enum htsCompression compression; short compression_level; // currently unused void *specific; // format specific options; see struct hts_opt. } htsFormat; // Maintainers note htsFile cannot be an opaque structure because some of its // fields are part of libhts.so's ABI (hence these fields must not be moved): // - fp is used in the public sam_itr_next()/etc macros // - is_bin is used directly in samtools <= 1.1 and bcftools <= 1.1 // - is_write and is_cram are used directly in samtools <= 1.1 // - fp is used directly in samtools (up to and including current develop) // - line is used directly in bcftools (up to and including current develop) typedef struct { uint32_t is_bin:1, is_write:1, is_be:1, is_cram:1, dummy:28; int64_t lineno; kstring_t line; char *fn, *fn_aux; union { BGZF *bgzf; struct cram_fd *cram; struct hFILE *hfile; void *voidp; } fp; htsFormat format; } htsFile; // REQUIRED_FIELDS enum sam_fields { SAM_QNAME = 0x00000001, SAM_FLAG = 0x00000002, SAM_RNAME = 0x00000004, SAM_POS = 0x00000008, SAM_MAPQ = 0x00000010, SAM_CIGAR = 0x00000020, SAM_RNEXT = 0x00000040, SAM_PNEXT = 0x00000080, SAM_TLEN = 0x00000100, SAM_SEQ = 0x00000200, SAM_QUAL = 0x00000400, SAM_AUX = 0x00000800, SAM_RGAUX = 0x00001000, }; // Mostly CRAM only, but this could also include other format options enum hts_fmt_option { // CRAM specific CRAM_OPT_DECODE_MD, CRAM_OPT_PREFIX, CRAM_OPT_VERBOSITY, // make general CRAM_OPT_SEQS_PER_SLICE, CRAM_OPT_SLICES_PER_CONTAINER, CRAM_OPT_RANGE, CRAM_OPT_VERSION, // rename to cram_version? CRAM_OPT_EMBED_REF, CRAM_OPT_IGNORE_MD5, CRAM_OPT_REFERENCE, // make general CRAM_OPT_MULTI_SEQ_PER_SLICE, CRAM_OPT_NO_REF, CRAM_OPT_USE_BZIP2, CRAM_OPT_SHARED_REF, CRAM_OPT_NTHREADS, // deprecated, use HTS_OPT_NTHREADS CRAM_OPT_THREAD_POOL,// make general CRAM_OPT_USE_LZMA, CRAM_OPT_USE_RANS, CRAM_OPT_REQUIRED_FIELDS, // General purpose HTS_OPT_COMPRESSION_LEVEL = 100, HTS_OPT_NTHREADS, }; // For backwards compatibility #define cram_option hts_fmt_option typedef struct hts_opt { char *arg; // string form, strdup()ed enum hts_fmt_option opt; // tokenised key union { // ... and value int i; char *s; } val; struct hts_opt *next; } hts_opt; #define HTS_FILE_OPTS_INIT {{0},0} /********************** * Exported functions * **********************/ /* * Parses arg and appends it to the option list. * * Returns 0 on success; * -1 on failure. */ int hts_opt_add(hts_opt **opts, const char *c_arg); /* * Applies an hts_opt option list to a given htsFile. * * Returns 0 on success * -1 on failure */ int hts_opt_apply(htsFile *fp, hts_opt *opts); /* * Frees an hts_opt list. */ void hts_opt_free(hts_opt *opts); /* * Accepts a string file format (sam, bam, cram, vcf, bam) optionally * followed by a comma separated list of key=value options and splits * these up into the fields of htsFormat struct. * * Returns 0 on success * -1 on failure. */ int hts_parse_format(htsFormat *opt, const char *str); /* * Tokenise options as (key(=value)?,)*(key(=value)?)? * NB: No provision for ',' appearing in the value! * Add backslashing rules? * * This could be used as part of a general command line option parser or * as a string concatenated onto the file open mode. * * Returns 0 on success * -1 on failure. */ int hts_parse_opt_list(htsFormat *opt, const char *str); extern int hts_verbose; /*! @abstract Table for converting a nucleotide character to 4-bit encoding. The input character may be either an IUPAC ambiguity code, '=' for 0, or '0'/'1'/'2'/'3' for a result of 1/2/4/8. The result is encoded as 1/2/4/8 for A/C/G/T or combinations of these bits for ambiguous bases. */ extern const unsigned char seq_nt16_table[256]; /*! @abstract Table for converting a 4-bit encoded nucleotide to an IUPAC ambiguity code letter (or '=' when given 0). */ extern const char seq_nt16_str[]; /*! @abstract Table for converting a 4-bit encoded nucleotide to about 2 bits. Returns 0/1/2/3 for 1/2/4/8 (i.e., A/C/G/T), or 4 otherwise (0 or ambiguous). */ extern const int seq_nt16_int[]; /*! @abstract Get the htslib version number @return For released versions, a string like "N.N[.N]"; or git describe output if using a library built within a Git repository. */ const char *hts_version(void); /*! @abstract Determine format by peeking at the start of a file @param fp File opened for reading, positioned at the beginning @param fmt Format structure that will be filled out on return @return 0 for success, or negative if an error occurred. */ int hts_detect_format(struct hFILE *fp, htsFormat *fmt); /*! @abstract Get a human-readable description of the file format @param fmt Format structure holding type, version, compression, etc. @return Description string, to be freed by the caller after use. */ char *hts_format_description(const htsFormat *format); /*! @abstract Open a SAM/BAM/CRAM/VCF/BCF/etc file @param fn The file name or "-" for stdin/stdout @param mode Mode matching / [rwa][bceguxz0-9]* / @discussion With 'r' opens for reading; any further format mode letters are ignored as the format is detected by checking the first few bytes or BGZF blocks of the file. With 'w' or 'a' opens for writing or appending, with format specifier letters: b binary format (BAM, BCF, etc) rather than text (SAM, VCF, etc) c CRAM format g gzip compressed u uncompressed z bgzf compressed [0-9] zlib compression level and with non-format option letters (for any of 'r'/'w'/'a'): e close the file on exec(2) (opens with O_CLOEXEC, where supported) x create the file exclusively (opens with O_EXCL, where supported) Note that there is a distinction between 'u' and '0': the first yields plain uncompressed output whereas the latter outputs uncompressed data wrapped in the zlib format. @example [rw]b .. compressed BCF, BAM, FAI [rw]bu .. uncompressed BCF [rw]z .. compressed VCF [rw] .. uncompressed VCF */ htsFile *hts_open(const char *fn, const char *mode); /*! @abstract Open a SAM/BAM/CRAM/VCF/BCF/etc file @param fn The file name or "-" for stdin/stdout @param mode Open mode, as per hts_open() @param fmt Optional format specific parameters @discussion See hts_open() for description of fn and mode. // TODO Update documentation for s/opts/fmt/ Opts contains a format string (sam, bam, cram, vcf, bcf) which will, if defined, override mode. Opts also contains a linked list of hts_opt structures to apply to the open file handle. These can contain things like pointers to the reference or information on compression levels, block sizes, etc. */ htsFile *hts_open_format(const char *fn, const char *mode, const htsFormat *fmt); /*! @abstract Open an existing stream as a SAM/BAM/CRAM/VCF/BCF/etc file @param fn The already-open file handle @param mode Open mode, as per hts_open() */ htsFile *hts_hopen(struct hFILE *fp, const char *fn, const char *mode); /*! @abstract Close a file handle, flushing buffered data for output streams @param fp The file handle to be closed @return 0 for success, or negative if an error occurred. */ int hts_close(htsFile *fp); /*! @abstract Returns the file's format information @param fp The file handle @return Read-only pointer to the file's htsFormat. */ const htsFormat *hts_get_format(htsFile *fp); /*! @ abstract Returns a string containing the file format extension. @ param format Format structure containing the file type. @ return A string ("sam", "bam", etc) or "?" for unknown formats. */ const char *hts_format_file_extension(const htsFormat *format); /*! @abstract Sets a specified CRAM option on the open file handle. @param fp The file handle open the open file. @param opt The CRAM_OPT_* option. @param ... Optional arguments, dependent on the option used. @return 0 for success, or negative if an error occurred. */ int hts_set_opt(htsFile *fp, enum hts_fmt_option opt, ...); int hts_getline(htsFile *fp, int delimiter, kstring_t *str); char **hts_readlines(const char *fn, int *_n); /*! @abstract Parse comma-separated list or read list from a file @param list File name or comma-separated list @param is_file @param _n Size of the output array (number of items read) @return NULL on failure or pointer to newly allocated array of strings */ char **hts_readlist(const char *fn, int is_file, int *_n); /*! @abstract Create extra threads to aid compress/decompression for this file @param fp The file handle @param n The number of worker threads to create @return 0 for success, or negative if an error occurred. @notes THIS THREADING API IS LIKELY TO CHANGE IN FUTURE. */ int hts_set_threads(htsFile *fp, int n); /*! @abstract Set .fai filename for a file opened for reading @return 0 for success, negative on failure @discussion Called before *_hdr_read(), this provides the name of a .fai file used to provide a reference list if the htsFile contains no @SQ headers. */ int hts_set_fai_filename(htsFile *fp, const char *fn_aux); /************ * Indexing * ************/ /*! These HTS_IDX_* macros are used as special tid values for hts_itr_query()/etc, producing iterators operating as follows: - HTS_IDX_NOCOOR iterates over unmapped reads sorted at the end of the file - HTS_IDX_START iterates over the entire file - HTS_IDX_REST iterates from the current position to the end of the file - HTS_IDX_NONE always returns "no more alignment records" When one of these special tid values is used, beg and end are ignored. When REST or NONE is used, idx is also ignored and may be NULL. */ #define HTS_IDX_NOCOOR (-2) #define HTS_IDX_START (-3) #define HTS_IDX_REST (-4) #define HTS_IDX_NONE (-5) #define HTS_FMT_CSI 0 #define HTS_FMT_BAI 1 #define HTS_FMT_TBI 2 #define HTS_FMT_CRAI 3 struct __hts_idx_t; typedef struct __hts_idx_t hts_idx_t; typedef struct { uint64_t u, v; } hts_pair64_t; typedef int hts_readrec_func(BGZF *fp, void *data, void *r, int *tid, int *beg, int *end); typedef struct { uint32_t read_rest:1, finished:1, dummy:29; int tid, beg, end, n_off, i; int curr_tid, curr_beg, curr_end; uint64_t curr_off; hts_pair64_t *off; hts_readrec_func *readrec; struct { int n, m; int *a; } bins; } hts_itr_t; #define hts_bin_first(l) (((1<<(((l)<<1) + (l))) - 1) / 7) #define hts_bin_parent(l) (((l) - 1) >> 3) hts_idx_t *hts_idx_init(int n, int fmt, uint64_t offset0, int min_shift, int n_lvls); void hts_idx_destroy(hts_idx_t *idx); int hts_idx_push(hts_idx_t *idx, int tid, int beg, int end, uint64_t offset, int is_mapped); void hts_idx_finish(hts_idx_t *idx, uint64_t final_offset); /// Save an index to a file /** @param idx Index to be written @param fn Input BAM/BCF/etc filename, to which .bai/.csi/etc will be added @param fmt One of the HTS_FMT_* index formats @return 0 if successful, or negative if an error occurred. */ int hts_idx_save(const hts_idx_t *idx, const char *fn, int fmt) HTS_RESULT_USED; /// Save an index to a specific file /** @param idx Index to be written @param fn Input BAM/BCF/etc filename @param fnidx Output filename, or NULL to add .bai/.csi/etc to @a fn @param fmt One of the HTS_FMT_* index formats @return 0 if successful, or negative if an error occurred. */ int hts_idx_save_as(const hts_idx_t *idx, const char *fn, const char *fnidx, int fmt) HTS_RESULT_USED; /// Load an index file /** @param fn BAM/BCF/etc filename, to which .bai/.csi/etc will be added or the extension substituted, to search for an existing index file @param fmt One of the HTS_FMT_* index formats @return The index, or NULL if an error occurred. */ hts_idx_t *hts_idx_load(const char *fn, int fmt); /// Load a specific index file /** @param fn Input BAM/BCF/etc filename @param fnidx The input index filename @return The index, or NULL if an error occurred. */ hts_idx_t *hts_idx_load2(const char *fn, const char *fnidx); uint8_t *hts_idx_get_meta(hts_idx_t *idx, int *l_meta); void hts_idx_set_meta(hts_idx_t *idx, int l_meta, uint8_t *meta, int is_copy); int hts_idx_get_stat(const hts_idx_t* idx, int tid, uint64_t* mapped, uint64_t* unmapped); uint64_t hts_idx_get_n_no_coor(const hts_idx_t* idx); #define HTS_PARSE_THOUSANDS_SEP 1 ///< Ignore ',' separators within numbers /// Parse a numeric string /** The number may be expressed in scientific notation, and optionally may contain commas in the integer part (before any decimal point or E notation). @param str String to be parsed @param strend If non-NULL, set on return to point to the first character in @a str after those forming the parsed number @param flags Or'ed-together combination of HTS_PARSE_* flags @return Converted value of the parsed number. When @a strend is NULL, a warning will be printed (if hts_verbose is 2 or more) if there are any trailing characters after the number. */ long long hts_parse_decimal(const char *str, char **strend, int flags); /// Parse a "CHR:START-END"-style region string /** @param str String to be parsed @param beg Set on return to the 0-based start of the region @param end Set on return to the 1-based end of the region @return Pointer to the colon or '\0' after the reference sequence name, or NULL if @a str could not be parsed. */ const char *hts_parse_reg(const char *str, int *beg, int *end); hts_itr_t *hts_itr_query(const hts_idx_t *idx, int tid, int beg, int end, hts_readrec_func *readrec); void hts_itr_destroy(hts_itr_t *iter); typedef int (*hts_name2id_f)(void*, const char*); typedef const char *(*hts_id2name_f)(void*, int); typedef hts_itr_t *hts_itr_query_func(const hts_idx_t *idx, int tid, int beg, int end, hts_readrec_func *readrec); hts_itr_t *hts_itr_querys(const hts_idx_t *idx, const char *reg, hts_name2id_f getid, void *hdr, hts_itr_query_func *itr_query, hts_readrec_func *readrec); int hts_itr_next(BGZF *fp, hts_itr_t *iter, void *r, void *data) HTS_RESULT_USED; const char **hts_idx_seqnames(const hts_idx_t *idx, int *n, hts_id2name_f getid, void *hdr); // free only the array, not the values /** * hts_file_type() - Convenience function to determine file type * DEPRECATED: This function has been replaced by hts_detect_format(). * It and these FT_* macros will be removed in a future HTSlib release. */ #define FT_UNKN 0 #define FT_GZ 1 #define FT_VCF 2 #define FT_VCF_GZ (FT_GZ|FT_VCF) #define FT_BCF (1<<2) #define FT_BCF_GZ (FT_GZ|FT_BCF) #define FT_STDIN (1<<3) int hts_file_type(const char *fname); /********************** * MD5 implementation * **********************/ struct hts_md5_context; typedef struct hts_md5_context hts_md5_context; /*! @abstract Intialises an MD5 context. * @discussion * The expected use is to allocate an hts_md5_context using * hts_md5_init(). This pointer is then passed into one or more calls * of hts_md5_update() to compute successive internal portions of the * MD5 sum, which can then be externalised as a full 16-byte MD5sum * calculation by calling hts_md5_final(). This can then be turned * into ASCII via hts_md5_hex(). * * To dealloate any resources created by hts_md5_init() call the * hts_md5_destroy() function. * * @return hts_md5_context pointer on success, NULL otherwise. */ hts_md5_context *hts_md5_init(void); /*! @abstract Updates the context with the MD5 of the data. */ void hts_md5_update(hts_md5_context *ctx, const void *data, unsigned long size); /*! @abstract Computes the final 128-bit MD5 hash from the given context */ void hts_md5_final(unsigned char *digest, hts_md5_context *ctx); /*! @abstract Resets an md5_context to the initial state, as returned * by hts_md5_init(). */ void hts_md5_reset(hts_md5_context *ctx); /*! @abstract Converts a 128-bit MD5 hash into a 33-byte nul-termninated * hex string. */ void hts_md5_hex(char *hex, const unsigned char *digest); /*! @abstract Deallocates any memory allocated by hts_md5_init. */ void hts_md5_destroy(hts_md5_context *ctx); static inline int hts_reg2bin(int64_t beg, int64_t end, int min_shift, int n_lvls) { int l, s = min_shift, t = ((1<<((n_lvls<<1) + n_lvls)) - 1) / 7; for (--end, l = n_lvls; l > 0; --l, s += 3, t -= 1<<((l<<1)+l)) if (beg>>s == end>>s) return t + (beg>>s); return 0; } static inline int hts_bin_bot(int bin, int n_lvls) { int l, b; for (l = 0, b = bin; b; ++l, b = hts_bin_parent(b)); // compute the level of bin return (bin - hts_bin_first(l)) << (n_lvls - l) * 3; } /************** * Endianness * **************/ static inline int ed_is_big(void) { long one= 1; return !(*((char *)(&one))); } static inline uint16_t ed_swap_2(uint16_t v) { return (uint16_t)(((v & 0x00FF00FFU) << 8) | ((v & 0xFF00FF00U) >> 8)); } static inline void *ed_swap_2p(void *x) { *(uint16_t*)x = ed_swap_2(*(uint16_t*)x); return x; } static inline uint32_t ed_swap_4(uint32_t v) { v = ((v & 0x0000FFFFU) << 16) | (v >> 16); return ((v & 0x00FF00FFU) << 8) | ((v & 0xFF00FF00U) >> 8); } static inline void *ed_swap_4p(void *x) { *(uint32_t*)x = ed_swap_4(*(uint32_t*)x); return x; } static inline uint64_t ed_swap_8(uint64_t v) { v = ((v & 0x00000000FFFFFFFFLLU) << 32) | (v >> 32); v = ((v & 0x0000FFFF0000FFFFLLU) << 16) | ((v & 0xFFFF0000FFFF0000LLU) >> 16); return ((v & 0x00FF00FF00FF00FFLLU) << 8) | ((v & 0xFF00FF00FF00FF00LLU) >> 8); } static inline void *ed_swap_8p(void *x) { *(uint64_t*)x = ed_swap_8(*(uint64_t*)x); return x; } #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/htslib/._hts_defs.h000644 000765 000024 00000000265 12750743246 020456 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/hts_defs.h000644 000765 000024 00000004623 12750743246 020243 0ustar00catchenstaff000000 000000 /* hts_defs.h -- Miscellaneous definitions. Copyright (C) 2013-2015 Genome Research Ltd. Author: John Marshall Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #ifndef HTSLIB_HTS_DEFS_H #define HTSLIB_HTS_DEFS_H #ifdef __clang__ #ifdef __has_attribute #define HTS_COMPILER_HAS(attribute) __has_attribute(attribute) #endif #elif defined __GNUC__ #define HTS_GCC_AT_LEAST(major, minor) \ (__GNUC__ > (major) || (__GNUC__ == (major) && __GNUC_MINOR__ >= (minor))) #endif #ifndef HTS_COMPILER_HAS #define HTS_COMPILER_HAS(attribute) 0 #endif #ifndef HTS_GCC_AT_LEAST #define HTS_GCC_AT_LEAST(major, minor) 0 #endif #if HTS_COMPILER_HAS(__noreturn__) || HTS_GCC_AT_LEAST(3,0) #define HTS_NORETURN __attribute__ ((__noreturn__)) #else #define HTS_NORETURN #endif // GCC introduced warn_unused_result in 3.4 but added -Wno-unused-result later #if HTS_COMPILER_HAS(__warn_unused_result__) || HTS_GCC_AT_LEAST(4,5) #define HTS_RESULT_USED __attribute__ ((__warn_unused_result__)) #else #define HTS_RESULT_USED #endif #if HTS_COMPILER_HAS(__unused__) || HTS_GCC_AT_LEAST(3,0) #define HTS_UNUSED __attribute__ ((__unused__)) #else #define HTS_UNUSED #endif #if HTS_COMPILER_HAS(__deprecated__) || HTS_GCC_AT_LEAST(4,5) #define HTS_DEPRECATED(message) __attribute__ ((__deprecated__ (message))) #elif HTS_GCC_AT_LEAST(3,1) #define HTS_DEPRECATED(message) __attribute__ ((__deprecated__)) #else #define HTS_DEPRECATED(message) #endif #endif stacks-1.44/htslib/htslib/._kbitset.h000644 000765 000024 00000000265 12750743246 020324 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/kbitset.h000644 000765 000024 00000010143 12750743246 020103 0ustar00catchenstaff000000 000000 /* The MIT License Copyright (C) 2015 Genome Research Ltd. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #ifndef KBITSET_H #define KBITSET_H /* Example of using kbitset_t, which represents a subset of {0,..., N-1}, where N is the size specified in kbs_init(). kbitset_t *bset = kbs_init(100); kbs_insert(bset, 5); kbs_insert(bset, 68); kbs_delete(bset, 37); // ... if (kbs_exists(bset, 68)) printf("68 present\n"); kbitset_iter itr; int i; kbs_start(&itr); while ((i = kbs_next(bset, &itr)) >= 0) printf("%d present\n", i); kbs_destroy(bset); Example of declaring a kbitset_t-using function in a header file, so that only source files that actually use process() need to include : struct kbitset_t; void process(struct kbitset_t *bset); */ #include #include #include #define KBS_ELTBITS (CHAR_BIT * sizeof (unsigned long)) #define KBS_ELT(i) ((i) / KBS_ELTBITS) #define KBS_MASK(i) (1UL << ((i) % KBS_ELTBITS)) typedef struct kbitset_t { size_t n; unsigned long b[1]; } kbitset_t; // Initialise a bit set capable of holding ni integers, 0 <= i < ni. // The set returned is empty if fill == 0, or all of [0,ni) otherwise. static inline kbitset_t *kbs_init2(size_t ni, int fill) { size_t n = (ni + KBS_ELTBITS-1) / KBS_ELTBITS; kbitset_t *bs = (kbitset_t *) malloc(sizeof(kbitset_t) + n * sizeof(unsigned long)); if (bs == NULL) return NULL; bs->n = n; memset(bs->b, fill? ~0 : 0, n * sizeof (unsigned long)); bs->b[n] = ~0UL; return bs; } // Initialise an empty bit set capable of holding ni integers, 0 <= i < ni. static inline kbitset_t *kbs_init(size_t ni) { return kbs_init2(ni, 0); } // Destroy a bit set. static inline void kbs_destroy(kbitset_t *bs) { free(bs); } // Reset the bit set to empty. static inline void kbs_clear(kbitset_t *bs) { memset(bs->b, 0, bs->n * sizeof (unsigned long)); } // Reset the bit set to all of [0,ni). static inline void kbs_insert_all(kbitset_t *bs) { memset(bs->b, ~0, bs->n * sizeof (unsigned long)); } // Insert an element into the bit set. static inline void kbs_insert(kbitset_t *bs, int i) { bs->b[KBS_ELT(i)] |= KBS_MASK(i); } // Remove an element from the bit set. static inline void kbs_delete(kbitset_t *bs, int i) { bs->b[KBS_ELT(i)] &= ~KBS_MASK(i); } // Test whether the bit set contains the element. static inline int kbs_exists(const kbitset_t *bs, int i) { return (bs->b[KBS_ELT(i)] & KBS_MASK(i)) != 0; } typedef struct kbitset_iter_t { unsigned long mask; size_t elt; int i; } kbitset_iter_t; // Initialise or reset a bit set iterator. static inline void kbs_start(kbitset_iter_t *itr) { itr->mask = 1; itr->elt = 0; itr->i = 0; } // Return the next element contained in the bit set, or -1 if there are no more. static inline int kbs_next(const kbitset_t *bs, kbitset_iter_t *itr) { unsigned long b = bs->b[itr->elt]; for (;;) { if (itr->mask == 0) { while ((b = bs->b[++itr->elt]) == 0) itr->i += KBS_ELTBITS; if (itr->elt == bs->n) return -1; itr->mask = 1; } if (b & itr->mask) break; itr->i++; itr->mask <<= 1; } itr->mask <<= 1; return itr->i++; } #endif stacks-1.44/htslib/htslib/._kfunc.h000644 000765 000024 00000000265 12750743246 017765 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/kfunc.h000644 000765 000024 00000005234 12750743246 017551 0ustar00catchenstaff000000 000000 /* The MIT License Copyright (C) 2010, 2013 Genome Research Ltd. Copyright (C) 2011 Attractive Chaos Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #ifndef HTSLIB_KFUNC_H #define HTSLIB_KFUNC_H #ifdef __cplusplus extern "C" { #endif /* Log gamma function * \log{\Gamma(z)} * AS245, 2nd algorithm, http://lib.stat.cmu.edu/apstat/245 */ double kf_lgamma(double z); /* complementary error function * \frac{2}{\sqrt{\pi}} \int_x^{\infty} e^{-t^2} dt * AS66, 2nd algorithm, http://lib.stat.cmu.edu/apstat/66 */ double kf_erfc(double x); /* The following computes regularized incomplete gamma functions. * Formulas are taken from Wiki, with additional input from Numerical * Recipes in C (for modified Lentz's algorithm) and AS245 * (http://lib.stat.cmu.edu/apstat/245). * * A good online calculator is available at: * * http://www.danielsoper.com/statcalc/calc23.aspx * * It calculates upper incomplete gamma function, which equals * kf_gammaq(s,z)*tgamma(s). */ double kf_gammap(double s, double z); double kf_gammaq(double s, double z); /* Regularized incomplete beta function. The method is taken from * Numerical Recipe in C, 2nd edition, section 6.4. The following web * page calculates the incomplete beta function, which equals * kf_betai(a,b,x) * gamma(a) * gamma(b) / gamma(a+b): * * http://www.danielsoper.com/statcalc/calc36.aspx */ double kf_betai(double a, double b, double x); /* * n11 n12 | n1_ * n21 n22 | n2_ * -----------+---- * n_1 n_2 | n */ double kt_fisher_exact(int n11, int n12, int n21, int n22, double *_left, double *_right, double *two); #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/htslib/._khash.h000644 000765 000024 00000000265 12750743246 017755 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/khash.h000644 000765 000024 00000052044 12750743246 017542 0ustar00catchenstaff000000 000000 /* The MIT License Copyright (c) 2008, 2009, 2011 by Attractive Chaos Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ /* An example: #include "khash.h" KHASH_MAP_INIT_INT(32, char) int main() { int ret, is_missing; khiter_t k; khash_t(32) *h = kh_init(32); k = kh_put(32, h, 5, &ret); kh_value(h, k) = 10; k = kh_get(32, h, 10); is_missing = (k == kh_end(h)); k = kh_get(32, h, 5); kh_del(32, h, k); for (k = kh_begin(h); k != kh_end(h); ++k) if (kh_exist(h, k)) kh_value(h, k) = 1; kh_destroy(32, h); return 0; } */ /* 2013-05-02 (0.2.8): * Use quadratic probing. When the capacity is power of 2, stepping function i*(i+1)/2 guarantees to traverse each bucket. It is better than double hashing on cache performance and is more robust than linear probing. In theory, double hashing should be more robust than quadratic probing. However, my implementation is probably not for large hash tables, because the second hash function is closely tied to the first hash function, which reduce the effectiveness of double hashing. Reference: http://research.cs.vt.edu/AVresearch/hashing/quadratic.php 2011-12-29 (0.2.7): * Minor code clean up; no actual effect. 2011-09-16 (0.2.6): * The capacity is a power of 2. This seems to dramatically improve the speed for simple keys. Thank Zilong Tan for the suggestion. Reference: - http://code.google.com/p/ulib/ - http://nothings.org/computer/judy/ * Allow to optionally use linear probing which usually has better performance for random input. Double hashing is still the default as it is more robust to certain non-random input. * Added Wang's integer hash function (not used by default). This hash function is more robust to certain non-random input. 2011-02-14 (0.2.5): * Allow to declare global functions. 2009-09-26 (0.2.4): * Improve portability 2008-09-19 (0.2.3): * Corrected the example * Improved interfaces 2008-09-11 (0.2.2): * Improved speed a little in kh_put() 2008-09-10 (0.2.1): * Added kh_clear() * Fixed a compiling error 2008-09-02 (0.2.0): * Changed to token concatenation which increases flexibility. 2008-08-31 (0.1.2): * Fixed a bug in kh_get(), which has not been tested previously. 2008-08-31 (0.1.1): * Added destructor */ #ifndef __AC_KHASH_H #define __AC_KHASH_H /*! @header Generic hash table library. */ #define AC_VERSION_KHASH_H "0.2.8" #include #include #include /* compiler specific configuration */ #if UINT_MAX == 0xffffffffu typedef unsigned int khint32_t; #elif ULONG_MAX == 0xffffffffu typedef unsigned long khint32_t; #endif #if ULONG_MAX == ULLONG_MAX typedef unsigned long khint64_t; #else typedef unsigned long long khint64_t; #endif #ifndef kh_inline #ifdef _MSC_VER #define kh_inline __inline #else #define kh_inline inline #endif #endif /* kh_inline */ #ifndef klib_unused #if (defined __clang__ && __clang_major__ >= 3) || (defined __GNUC__ && __GNUC__ >= 3) #define klib_unused __attribute__ ((__unused__)) #else #define klib_unused #endif #endif /* klib_unused */ typedef khint32_t khint_t; typedef khint_t khiter_t; #define __ac_isempty(flag, i) ((flag[i>>4]>>((i&0xfU)<<1))&2) #define __ac_isdel(flag, i) ((flag[i>>4]>>((i&0xfU)<<1))&1) #define __ac_iseither(flag, i) ((flag[i>>4]>>((i&0xfU)<<1))&3) #define __ac_set_isdel_false(flag, i) (flag[i>>4]&=~(1ul<<((i&0xfU)<<1))) #define __ac_set_isempty_false(flag, i) (flag[i>>4]&=~(2ul<<((i&0xfU)<<1))) #define __ac_set_isboth_false(flag, i) (flag[i>>4]&=~(3ul<<((i&0xfU)<<1))) #define __ac_set_isdel_true(flag, i) (flag[i>>4]|=1ul<<((i&0xfU)<<1)) #define __ac_fsize(m) ((m) < 16? 1 : (m)>>4) #ifndef kroundup32 #define kroundup32(x) (--(x), (x)|=(x)>>1, (x)|=(x)>>2, (x)|=(x)>>4, (x)|=(x)>>8, (x)|=(x)>>16, ++(x)) #endif #ifndef kcalloc #define kcalloc(N,Z) calloc(N,Z) #endif #ifndef kmalloc #define kmalloc(Z) malloc(Z) #endif #ifndef krealloc #define krealloc(P,Z) realloc(P,Z) #endif #ifndef kfree #define kfree(P) free(P) #endif static const double __ac_HASH_UPPER = 0.77; #define __KHASH_TYPE(name, khkey_t, khval_t) \ typedef struct kh_##name##_s { \ khint_t n_buckets, size, n_occupied, upper_bound; \ khint32_t *flags; \ khkey_t *keys; \ khval_t *vals; \ } kh_##name##_t; #define __KHASH_PROTOTYPES(name, khkey_t, khval_t) \ extern kh_##name##_t *kh_init_##name(void); \ extern void kh_destroy_##name(kh_##name##_t *h); \ extern void kh_clear_##name(kh_##name##_t *h); \ extern khint_t kh_get_##name(const kh_##name##_t *h, khkey_t key); \ extern int kh_resize_##name(kh_##name##_t *h, khint_t new_n_buckets); \ extern khint_t kh_put_##name(kh_##name##_t *h, khkey_t key, int *ret); \ extern void kh_del_##name(kh_##name##_t *h, khint_t x); #define __KHASH_IMPL(name, SCOPE, khkey_t, khval_t, kh_is_map, __hash_func, __hash_equal) \ SCOPE kh_##name##_t *kh_init_##name(void) { \ return (kh_##name##_t*)kcalloc(1, sizeof(kh_##name##_t)); \ } \ SCOPE void kh_destroy_##name(kh_##name##_t *h) \ { \ if (h) { \ kfree((void *)h->keys); kfree(h->flags); \ kfree((void *)h->vals); \ kfree(h); \ } \ } \ SCOPE void kh_clear_##name(kh_##name##_t *h) \ { \ if (h && h->flags) { \ memset(h->flags, 0xaa, __ac_fsize(h->n_buckets) * sizeof(khint32_t)); \ h->size = h->n_occupied = 0; \ } \ } \ SCOPE khint_t kh_get_##name(const kh_##name##_t *h, khkey_t key) \ { \ if (h->n_buckets) { \ khint_t k, i, last, mask, step = 0; \ mask = h->n_buckets - 1; \ k = __hash_func(key); i = k & mask; \ last = i; \ while (!__ac_isempty(h->flags, i) && (__ac_isdel(h->flags, i) || !__hash_equal(h->keys[i], key))) { \ i = (i + (++step)) & mask; \ if (i == last) return h->n_buckets; \ } \ return __ac_iseither(h->flags, i)? h->n_buckets : i; \ } else return 0; \ } \ SCOPE int kh_resize_##name(kh_##name##_t *h, khint_t new_n_buckets) \ { /* This function uses 0.25*n_buckets bytes of working space instead of [sizeof(key_t+val_t)+.25]*n_buckets. */ \ khint32_t *new_flags = 0; \ khint_t j = 1; \ { \ kroundup32(new_n_buckets); \ if (new_n_buckets < 4) new_n_buckets = 4; \ if (h->size >= (khint_t)(new_n_buckets * __ac_HASH_UPPER + 0.5)) j = 0; /* requested size is too small */ \ else { /* hash table size to be changed (shrink or expand); rehash */ \ new_flags = (khint32_t*)kmalloc(__ac_fsize(new_n_buckets) * sizeof(khint32_t)); \ if (!new_flags) return -1; \ memset(new_flags, 0xaa, __ac_fsize(new_n_buckets) * sizeof(khint32_t)); \ if (h->n_buckets < new_n_buckets) { /* expand */ \ khkey_t *new_keys = (khkey_t*)krealloc((void *)h->keys, new_n_buckets * sizeof(khkey_t)); \ if (!new_keys) { kfree(new_flags); return -1; } \ h->keys = new_keys; \ if (kh_is_map) { \ khval_t *new_vals = (khval_t*)krealloc((void *)h->vals, new_n_buckets * sizeof(khval_t)); \ if (!new_vals) { kfree(new_flags); return -1; } \ h->vals = new_vals; \ } \ } /* otherwise shrink */ \ } \ } \ if (j) { /* rehashing is needed */ \ for (j = 0; j != h->n_buckets; ++j) { \ if (__ac_iseither(h->flags, j) == 0) { \ khkey_t key = h->keys[j]; \ khval_t val; \ khint_t new_mask; \ new_mask = new_n_buckets - 1; \ if (kh_is_map) val = h->vals[j]; \ __ac_set_isdel_true(h->flags, j); \ while (1) { /* kick-out process; sort of like in Cuckoo hashing */ \ khint_t k, i, step = 0; \ k = __hash_func(key); \ i = k & new_mask; \ while (!__ac_isempty(new_flags, i)) i = (i + (++step)) & new_mask; \ __ac_set_isempty_false(new_flags, i); \ if (i < h->n_buckets && __ac_iseither(h->flags, i) == 0) { /* kick out the existing element */ \ { khkey_t tmp = h->keys[i]; h->keys[i] = key; key = tmp; } \ if (kh_is_map) { khval_t tmp = h->vals[i]; h->vals[i] = val; val = tmp; } \ __ac_set_isdel_true(h->flags, i); /* mark it as deleted in the old hash table */ \ } else { /* write the element and jump out of the loop */ \ h->keys[i] = key; \ if (kh_is_map) h->vals[i] = val; \ break; \ } \ } \ } \ } \ if (h->n_buckets > new_n_buckets) { /* shrink the hash table */ \ h->keys = (khkey_t*)krealloc((void *)h->keys, new_n_buckets * sizeof(khkey_t)); \ if (kh_is_map) h->vals = (khval_t*)krealloc((void *)h->vals, new_n_buckets * sizeof(khval_t)); \ } \ kfree(h->flags); /* free the working space */ \ h->flags = new_flags; \ h->n_buckets = new_n_buckets; \ h->n_occupied = h->size; \ h->upper_bound = (khint_t)(h->n_buckets * __ac_HASH_UPPER + 0.5); \ } \ return 0; \ } \ SCOPE khint_t kh_put_##name(kh_##name##_t *h, khkey_t key, int *ret) \ { \ khint_t x; \ if (h->n_occupied >= h->upper_bound) { /* update the hash table */ \ if (h->n_buckets > (h->size<<1)) { \ if (kh_resize_##name(h, h->n_buckets - 1) < 0) { /* clear "deleted" elements */ \ *ret = -1; return h->n_buckets; \ } \ } else if (kh_resize_##name(h, h->n_buckets + 1) < 0) { /* expand the hash table */ \ *ret = -1; return h->n_buckets; \ } \ } /* TODO: to implement automatically shrinking; resize() already support shrinking */ \ { \ khint_t k, i, site, last, mask = h->n_buckets - 1, step = 0; \ x = site = h->n_buckets; k = __hash_func(key); i = k & mask; \ if (__ac_isempty(h->flags, i)) x = i; /* for speed up */ \ else { \ last = i; \ while (!__ac_isempty(h->flags, i) && (__ac_isdel(h->flags, i) || !__hash_equal(h->keys[i], key))) { \ if (__ac_isdel(h->flags, i)) site = i; \ i = (i + (++step)) & mask; \ if (i == last) { x = site; break; } \ } \ if (x == h->n_buckets) { \ if (__ac_isempty(h->flags, i) && site != h->n_buckets) x = site; \ else x = i; \ } \ } \ } \ if (__ac_isempty(h->flags, x)) { /* not present at all */ \ h->keys[x] = key; \ __ac_set_isboth_false(h->flags, x); \ ++h->size; ++h->n_occupied; \ *ret = 1; \ } else if (__ac_isdel(h->flags, x)) { /* deleted */ \ h->keys[x] = key; \ __ac_set_isboth_false(h->flags, x); \ ++h->size; \ *ret = 2; \ } else *ret = 0; /* Don't touch h->keys[x] if present and not deleted */ \ return x; \ } \ SCOPE void kh_del_##name(kh_##name##_t *h, khint_t x) \ { \ if (x != h->n_buckets && !__ac_iseither(h->flags, x)) { \ __ac_set_isdel_true(h->flags, x); \ --h->size; \ } \ } #define KHASH_DECLARE(name, khkey_t, khval_t) \ __KHASH_TYPE(name, khkey_t, khval_t) \ __KHASH_PROTOTYPES(name, khkey_t, khval_t) #define KHASH_INIT2(name, SCOPE, khkey_t, khval_t, kh_is_map, __hash_func, __hash_equal) \ __KHASH_TYPE(name, khkey_t, khval_t) \ __KHASH_IMPL(name, SCOPE, khkey_t, khval_t, kh_is_map, __hash_func, __hash_equal) #define KHASH_INIT(name, khkey_t, khval_t, kh_is_map, __hash_func, __hash_equal) \ KHASH_INIT2(name, static kh_inline klib_unused, khkey_t, khval_t, kh_is_map, __hash_func, __hash_equal) /* --- BEGIN OF HASH FUNCTIONS --- */ /*! @function @abstract Integer hash function @param key The integer [khint32_t] @return The hash value [khint_t] */ #define kh_int_hash_func(key) (khint32_t)(key) /*! @function @abstract Integer comparison function */ #define kh_int_hash_equal(a, b) ((a) == (b)) /*! @function @abstract 64-bit integer hash function @param key The integer [khint64_t] @return The hash value [khint_t] */ #define kh_int64_hash_func(key) (khint32_t)((key)>>33^(key)^(key)<<11) /*! @function @abstract 64-bit integer comparison function */ #define kh_int64_hash_equal(a, b) ((a) == (b)) /*! @function @abstract const char* hash function @param s Pointer to a null terminated string @return The hash value */ static kh_inline khint_t __ac_X31_hash_string(const char *s) { khint_t h = (khint_t)*s; if (h) for (++s ; *s; ++s) h = (h << 5) - h + (khint_t)*s; return h; } /*! @function @abstract Another interface to const char* hash function @param key Pointer to a null terminated string [const char*] @return The hash value [khint_t] */ #define kh_str_hash_func(key) __ac_X31_hash_string(key) /*! @function @abstract Const char* comparison function */ #define kh_str_hash_equal(a, b) (strcmp(a, b) == 0) static kh_inline khint_t __ac_Wang_hash(khint_t key) { key += ~(key << 15); key ^= (key >> 10); key += (key << 3); key ^= (key >> 6); key += ~(key << 11); key ^= (key >> 16); return key; } #define kh_int_hash_func2(k) __ac_Wang_hash((khint_t)key) /* --- END OF HASH FUNCTIONS --- */ /* Other convenient macros... */ /*! @abstract Type of the hash table. @param name Name of the hash table [symbol] */ #define khash_t(name) kh_##name##_t /*! @function @abstract Initiate a hash table. @param name Name of the hash table [symbol] @return Pointer to the hash table [khash_t(name)*] */ #define kh_init(name) kh_init_##name() /*! @function @abstract Destroy a hash table. @param name Name of the hash table [symbol] @param h Pointer to the hash table [khash_t(name)*] */ #define kh_destroy(name, h) kh_destroy_##name(h) /*! @function @abstract Reset a hash table without deallocating memory. @param name Name of the hash table [symbol] @param h Pointer to the hash table [khash_t(name)*] */ #define kh_clear(name, h) kh_clear_##name(h) /*! @function @abstract Resize a hash table. @param name Name of the hash table [symbol] @param h Pointer to the hash table [khash_t(name)*] @param s New size [khint_t] */ #define kh_resize(name, h, s) kh_resize_##name(h, s) /*! @function @abstract Insert a key to the hash table. @param name Name of the hash table [symbol] @param h Pointer to the hash table [khash_t(name)*] @param k Key [type of keys] @param r Extra return code: -1 if the operation failed; 0 if the key is present in the hash table; 1 if the bucket is empty (never used); 2 if the element in the bucket has been deleted [int*] @return Iterator to the inserted element [khint_t] */ #define kh_put(name, h, k, r) kh_put_##name(h, k, r) /*! @function @abstract Retrieve a key from the hash table. @param name Name of the hash table [symbol] @param h Pointer to the hash table [khash_t(name)*] @param k Key [type of keys] @return Iterator to the found element, or kh_end(h) if the element is absent [khint_t] */ #define kh_get(name, h, k) kh_get_##name(h, k) /*! @function @abstract Remove a key from the hash table. @param name Name of the hash table [symbol] @param h Pointer to the hash table [khash_t(name)*] @param k Iterator to the element to be deleted [khint_t] */ #define kh_del(name, h, k) kh_del_##name(h, k) /*! @function @abstract Test whether a bucket contains data. @param h Pointer to the hash table [khash_t(name)*] @param x Iterator to the bucket [khint_t] @return 1 if containing data; 0 otherwise [int] */ #define kh_exist(h, x) (!__ac_iseither((h)->flags, (x))) /*! @function @abstract Get key given an iterator @param h Pointer to the hash table [khash_t(name)*] @param x Iterator to the bucket [khint_t] @return Key [type of keys] */ #define kh_key(h, x) ((h)->keys[x]) /*! @function @abstract Get value given an iterator @param h Pointer to the hash table [khash_t(name)*] @param x Iterator to the bucket [khint_t] @return Value [type of values] @discussion For hash sets, calling this results in segfault. */ #define kh_val(h, x) ((h)->vals[x]) /*! @function @abstract Alias of kh_val() */ #define kh_value(h, x) ((h)->vals[x]) /*! @function @abstract Get the start iterator @param h Pointer to the hash table [khash_t(name)*] @return The start iterator [khint_t] */ #define kh_begin(h) (khint_t)(0) /*! @function @abstract Get the end iterator @param h Pointer to the hash table [khash_t(name)*] @return The end iterator [khint_t] */ #define kh_end(h) ((h)->n_buckets) /*! @function @abstract Get the number of elements in the hash table @param h Pointer to the hash table [khash_t(name)*] @return Number of elements in the hash table [khint_t] */ #define kh_size(h) ((h)->size) /*! @function @abstract Get the number of buckets in the hash table @param h Pointer to the hash table [khash_t(name)*] @return Number of buckets in the hash table [khint_t] */ #define kh_n_buckets(h) ((h)->n_buckets) /*! @function @abstract Iterate over the entries in the hash table @param h Pointer to the hash table [khash_t(name)*] @param kvar Variable to which key will be assigned @param vvar Variable to which value will be assigned @param code Block of code to execute */ #define kh_foreach(h, kvar, vvar, code) { khint_t __i; \ for (__i = kh_begin(h); __i != kh_end(h); ++__i) { \ if (!kh_exist(h,__i)) continue; \ (kvar) = kh_key(h,__i); \ (vvar) = kh_val(h,__i); \ code; \ } } /*! @function @abstract Iterate over the values in the hash table @param h Pointer to the hash table [khash_t(name)*] @param vvar Variable to which value will be assigned @param code Block of code to execute */ #define kh_foreach_value(h, vvar, code) { khint_t __i; \ for (__i = kh_begin(h); __i != kh_end(h); ++__i) { \ if (!kh_exist(h,__i)) continue; \ (vvar) = kh_val(h,__i); \ code; \ } } /* More conenient interfaces */ /*! @function @abstract Instantiate a hash set containing integer keys @param name Name of the hash table [symbol] */ #define KHASH_SET_INIT_INT(name) \ KHASH_INIT(name, khint32_t, char, 0, kh_int_hash_func, kh_int_hash_equal) /*! @function @abstract Instantiate a hash map containing integer keys @param name Name of the hash table [symbol] @param khval_t Type of values [type] */ #define KHASH_MAP_INIT_INT(name, khval_t) \ KHASH_INIT(name, khint32_t, khval_t, 1, kh_int_hash_func, kh_int_hash_equal) /*! @function @abstract Instantiate a hash map containing 64-bit integer keys @param name Name of the hash table [symbol] */ #define KHASH_SET_INIT_INT64(name) \ KHASH_INIT(name, khint64_t, char, 0, kh_int64_hash_func, kh_int64_hash_equal) /*! @function @abstract Instantiate a hash map containing 64-bit integer keys @param name Name of the hash table [symbol] @param khval_t Type of values [type] */ #define KHASH_MAP_INIT_INT64(name, khval_t) \ KHASH_INIT(name, khint64_t, khval_t, 1, kh_int64_hash_func, kh_int64_hash_equal) typedef const char *kh_cstr_t; /*! @function @abstract Instantiate a hash map containing const char* keys @param name Name of the hash table [symbol] */ #define KHASH_SET_INIT_STR(name) \ KHASH_INIT(name, kh_cstr_t, char, 0, kh_str_hash_func, kh_str_hash_equal) /*! @function @abstract Instantiate a hash map containing const char* keys @param name Name of the hash table [symbol] @param khval_t Type of values [type] */ #define KHASH_MAP_INIT_STR(name, khval_t) \ KHASH_INIT(name, kh_cstr_t, khval_t, 1, kh_str_hash_func, kh_str_hash_equal) #endif /* __AC_KHASH_H */ stacks-1.44/htslib/htslib/._khash_str2int.h000644 000765 000024 00000000265 12750743246 021442 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/khash_str2int.h000644 000765 000024 00000007626 12750743246 021235 0ustar00catchenstaff000000 000000 /* khash_str2int.h -- C-string to integer hash table. Copyright (C) 2013 Genome Research Ltd. Author: Petr Danecek Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #ifndef HTSLIB_KHASH_STR2INT_H #define HTSLIB_KHASH_STR2INT_H #include KHASH_MAP_INIT_STR(str2int, int) /* * Wrappers for khash dictionaries used by mpileup. */ static inline void *khash_str2int_init(void) { return kh_init(str2int); } /* * Destroy the hash structure, but not the keys */ static inline void khash_str2int_destroy(void *_hash) { khash_t(str2int) *hash = (khash_t(str2int)*)_hash; if (hash) kh_destroy(str2int, hash); // Note that strings are not freed. } /* * Destroys both the hash structure and the keys */ static inline void khash_str2int_destroy_free(void *_hash) { khash_t(str2int) *hash = (khash_t(str2int)*)_hash; khint_t k; if (hash == 0) return; for (k = 0; k < kh_end(hash); ++k) if (kh_exist(hash, k)) free((char*)kh_key(hash, k)); kh_destroy(str2int, hash); } /* * Returns 1 if key exists or 0 if not */ static inline int khash_str2int_has_key(void *_hash, const char *str) { khash_t(str2int) *hash = (khash_t(str2int)*)_hash; khint_t k = kh_get(str2int, hash, str); if ( k == kh_end(hash) ) return 0; return 1; } /* * Returns 0 on success and -1 when the key is not present. On success, * *value is set, unless NULL is passed. */ static inline int khash_str2int_get(void *_hash, const char *str, int *value) { khash_t(str2int) *hash = (khash_t(str2int)*)_hash; khint_t k; if ( !hash ) return -1; k = kh_get(str2int, hash, str); if ( k == kh_end(hash) ) return -1; if ( !value ) return 0; *value = kh_val(hash, k); return 0; } /* * Add a new string to the dictionary, auto-incrementing the value. * On success returns the newly inserted integer id, on error -1 * is returned. Note that the key must continue to exist throughout * the whole life of _hash. */ static inline int khash_str2int_inc(void *_hash, const char *str) { khint_t k; int ret; khash_t(str2int) *hash = (khash_t(str2int)*)_hash; if ( !hash ) return -1; k = kh_put(str2int, hash, str, &ret); if (ret == 0) return kh_val(hash, k); kh_val(hash, k) = kh_size(hash) - 1; return kh_val(hash, k); } /* * Set a new key,value pair. On success returns the bin index, on * error -1 is returned. Note that the key must contnue to exist * throughout the whole life of _hash. */ static inline int khash_str2int_set(void *_hash, const char *str, int value) { khint_t k; int ret; khash_t(str2int) *hash = (khash_t(str2int)*)_hash; if ( !hash ) return -1; k = kh_put(str2int, hash, str, &ret); kh_val(hash,k) = value; return k; } /* * Return the number of keys in the hash table. */ static inline int khash_str2int_size(void *_hash) { khash_t(str2int) *hash = (khash_t(str2int)*)_hash; return kh_size(hash); } #endif stacks-1.44/htslib/htslib/._klist.h000644 000765 000024 00000000265 12750743246 020005 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/klist.h000644 000765 000024 00000011736 12750743246 017575 0ustar00catchenstaff000000 000000 /* The MIT License Copyright (c) 2008-2009, by Attractive Chaos Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #ifndef _AC_KLIST_H #define _AC_KLIST_H #include #ifndef klib_unused #if (defined __clang__ && __clang_major__ >= 3) || (defined __GNUC__ && __GNUC__ >= 3) #define klib_unused __attribute__ ((__unused__)) #else #define klib_unused #endif #endif /* klib_unused */ #define KMEMPOOL_INIT2(SCOPE, name, kmptype_t, kmpfree_f) \ typedef struct { \ size_t cnt, n, max; \ kmptype_t **buf; \ } kmp_##name##_t; \ SCOPE kmp_##name##_t *kmp_init_##name(void) { \ return calloc(1, sizeof(kmp_##name##_t)); \ } \ SCOPE void kmp_destroy_##name(kmp_##name##_t *mp) { \ size_t k; \ for (k = 0; k < mp->n; ++k) { \ kmpfree_f(mp->buf[k]); free(mp->buf[k]); \ } \ free(mp->buf); free(mp); \ } \ SCOPE kmptype_t *kmp_alloc_##name(kmp_##name##_t *mp) { \ ++mp->cnt; \ if (mp->n == 0) return calloc(1, sizeof(kmptype_t)); \ return mp->buf[--mp->n]; \ } \ SCOPE void kmp_free_##name(kmp_##name##_t *mp, kmptype_t *p) { \ --mp->cnt; \ if (mp->n == mp->max) { \ mp->max = mp->max? mp->max<<1 : 16; \ mp->buf = realloc(mp->buf, sizeof(kmptype_t *) * mp->max); \ } \ mp->buf[mp->n++] = p; \ } #define KMEMPOOL_INIT(name, kmptype_t, kmpfree_f) \ KMEMPOOL_INIT2(static inline klib_unused, name, kmptype_t, kmpfree_f) #define kmempool_t(name) kmp_##name##_t #define kmp_init(name) kmp_init_##name() #define kmp_destroy(name, mp) kmp_destroy_##name(mp) #define kmp_alloc(name, mp) kmp_alloc_##name(mp) #define kmp_free(name, mp, p) kmp_free_##name(mp, p) #define KLIST_INIT2(SCOPE, name, kltype_t, kmpfree_t) \ struct __kl1_##name { \ kltype_t data; \ struct __kl1_##name *next; \ }; \ typedef struct __kl1_##name kl1_##name; \ KMEMPOOL_INIT2(SCOPE, name, kl1_##name, kmpfree_t) \ typedef struct { \ kl1_##name *head, *tail; \ kmp_##name##_t *mp; \ size_t size; \ } kl_##name##_t; \ SCOPE kl_##name##_t *kl_init_##name(void) { \ kl_##name##_t *kl = calloc(1, sizeof(kl_##name##_t)); \ kl->mp = kmp_init(name); \ kl->head = kl->tail = kmp_alloc(name, kl->mp); \ kl->head->next = 0; \ return kl; \ } \ SCOPE void kl_destroy_##name(kl_##name##_t *kl) { \ kl1_##name *p; \ for (p = kl->head; p != kl->tail; p = p->next) \ kmp_free(name, kl->mp, p); \ kmp_free(name, kl->mp, p); \ kmp_destroy(name, kl->mp); \ free(kl); \ } \ SCOPE kltype_t *kl_pushp_##name(kl_##name##_t *kl) { \ kl1_##name *q, *p = kmp_alloc(name, kl->mp); \ q = kl->tail; p->next = 0; kl->tail->next = p; kl->tail = p; \ ++kl->size; \ return &q->data; \ } \ SCOPE int kl_shift_##name(kl_##name##_t *kl, kltype_t *d) { \ kl1_##name *p; \ if (kl->head->next == 0) return -1; \ --kl->size; \ p = kl->head; kl->head = kl->head->next; \ if (d) *d = p->data; \ kmp_free(name, kl->mp, p); \ return 0; \ } #define KLIST_INIT(name, kltype_t, kmpfree_t) \ KLIST_INIT2(static inline klib_unused, name, kltype_t, kmpfree_t) #define kliter_t(name) kl1_##name #define klist_t(name) kl_##name##_t #define kl_val(iter) ((iter)->data) #define kl_next(iter) ((iter)->next) #define kl_begin(kl) ((kl)->head) #define kl_end(kl) ((kl)->tail) #define kl_init(name) kl_init_##name() #define kl_destroy(name, kl) kl_destroy_##name(kl) #define kl_pushp(name, kl) kl_pushp_##name(kl) #define kl_shift(name, kl, d) kl_shift_##name(kl, d) #endif stacks-1.44/htslib/htslib/._knetfile.h000644 000765 000024 00000000265 12750743246 020460 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/knetfile.h000644 000765 000024 00000005414 12750743246 020244 0ustar00catchenstaff000000 000000 /* The MIT License Copyright (c) 2008 by Genome Research Ltd (GRL). 2010 by Attractive Chaos Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #ifndef KNETFILE_H #define KNETFILE_H #include #include #ifndef _WIN32 #define netread(fd, ptr, len) read(fd, ptr, len) #define netwrite(fd, ptr, len) write(fd, ptr, len) #define netclose(fd) close(fd) #else #include #define netread(fd, ptr, len) recv(fd, ptr, len, 0) #define netwrite(fd, ptr, len) send(fd, ptr, len, 0) #define netclose(fd) closesocket(fd) #endif // FIXME: currently I/O is unbuffered #define KNF_TYPE_LOCAL 1 #define KNF_TYPE_FTP 2 #define KNF_TYPE_HTTP 3 typedef struct knetFile_s { int type, fd; int64_t offset; char *host, *port; // the following are for FTP only int ctrl_fd, pasv_ip[4], pasv_port, max_response, no_reconnect, is_ready; char *response, *retr, *size_cmd; int64_t seek_offset; // for lazy seek int64_t file_size; // the following are for HTTP only char *path, *http_host; } knetFile; #define knet_tell(fp) ((fp)->offset) #define knet_fileno(fp) ((fp)->fd) #ifdef __cplusplus extern "C" { #endif #ifdef _WIN32 int knet_win32_init(); void knet_win32_destroy(); #endif knetFile *knet_open(const char *fn, const char *mode); /* This only works with local files. */ knetFile *knet_dopen(int fd, const char *mode); /* If ->is_ready==0, this routine updates ->fd; otherwise, it simply reads from ->fd. */ ssize_t knet_read(knetFile *fp, void *buf, size_t len); /* This routine only sets ->offset and ->is_ready=0. It does not communicate with the FTP server. */ off_t knet_seek(knetFile *fp, off_t off, int whence); int knet_close(knetFile *fp); #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/htslib/._kseq.h000644 000765 000024 00000000265 12750743246 017622 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/kseq.h000644 000765 000024 00000021304 12750743246 017402 0ustar00catchenstaff000000 000000 /* The MIT License Copyright (c) 2008, 2009, 2011 Attractive Chaos Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ /* Last Modified: 05MAR2012 */ #ifndef AC_KSEQ_H #define AC_KSEQ_H #include #include #include #define KS_SEP_SPACE 0 // isspace(): \t, \n, \v, \f, \r #define KS_SEP_TAB 1 // isspace() && !' ' #define KS_SEP_LINE 2 // line separator: "\n" (Unix) or "\r\n" (Windows) #define KS_SEP_MAX 2 #define __KS_TYPE(type_t) \ typedef struct __kstream_t { \ int begin, end; \ int is_eof:2, bufsize:30; \ uint64_t seek_pos; \ type_t f; \ unsigned char *buf; \ } kstream_t; #define ks_eof(ks) ((ks)->is_eof && (ks)->begin >= (ks)->end) #define ks_rewind(ks) ((ks)->is_eof = (ks)->begin = (ks)->end = 0) #define __KS_BASIC(SCOPE, type_t, __bufsize) \ SCOPE kstream_t *ks_init(type_t f) \ { \ kstream_t *ks = (kstream_t*)calloc(1, sizeof(kstream_t)); \ ks->f = f; ks->bufsize = __bufsize; \ ks->buf = (unsigned char*)malloc(__bufsize); \ return ks; \ } \ SCOPE void ks_destroy(kstream_t *ks) \ { \ if (!ks) return; \ free(ks->buf); \ free(ks); \ } #define __KS_INLINED(__read) \ static inline int ks_getc(kstream_t *ks) \ { \ if (ks->is_eof && ks->begin >= ks->end) return -1; \ if (ks->begin >= ks->end) { \ ks->begin = 0; \ ks->end = __read(ks->f, ks->buf, ks->bufsize); \ if (ks->end == 0) { ks->is_eof = 1; return -1; } \ } \ ks->seek_pos++; \ return (int)ks->buf[ks->begin++]; \ } \ static inline int ks_getuntil(kstream_t *ks, int delimiter, kstring_t *str, int *dret) \ { return ks_getuntil2(ks, delimiter, str, dret, 0); } #ifndef KSTRING_T #define KSTRING_T kstring_t typedef struct __kstring_t { size_t l, m; char *s; } kstring_t; #endif #ifndef kroundup32 #define kroundup32(x) (--(x), (x)|=(x)>>1, (x)|=(x)>>2, (x)|=(x)>>4, (x)|=(x)>>8, (x)|=(x)>>16, ++(x)) #endif #define __KS_GETUNTIL(SCOPE, __read) \ SCOPE int ks_getuntil2(kstream_t *ks, int delimiter, kstring_t *str, int *dret, int append) \ { \ int gotany = 0; \ if (dret) *dret = 0; \ str->l = append? str->l : 0; \ uint64_t seek_pos = str->l; \ for (;;) { \ int i; \ if (ks->begin >= ks->end) { \ if (!ks->is_eof) { \ ks->begin = 0; \ ks->end = __read(ks->f, ks->buf, ks->bufsize); \ if (ks->end == 0) { ks->is_eof = 1; break; } \ } else break; \ } \ if (delimiter == KS_SEP_LINE) { \ for (i = ks->begin; i < ks->end; ++i) \ if (ks->buf[i] == '\n') break; \ } else if (delimiter > KS_SEP_MAX) { \ for (i = ks->begin; i < ks->end; ++i) \ if (ks->buf[i] == delimiter) break; \ } else if (delimiter == KS_SEP_SPACE) { \ for (i = ks->begin; i < ks->end; ++i) \ if (isspace(ks->buf[i])) break; \ } else if (delimiter == KS_SEP_TAB) { \ for (i = ks->begin; i < ks->end; ++i) \ if (isspace(ks->buf[i]) && ks->buf[i] != ' ') break; \ } else i = 0; /* never come to here! */ \ if (str->m - str->l < (size_t)(i - ks->begin + 1)) { \ str->m = str->l + (i - ks->begin) + 1; \ kroundup32(str->m); \ str->s = (char*)realloc(str->s, str->m); \ } \ seek_pos += i - ks->begin; if ( i < ks->end ) seek_pos++; \ gotany = 1; \ memcpy(str->s + str->l, ks->buf + ks->begin, i - ks->begin); \ str->l = str->l + (i - ks->begin); \ ks->begin = i + 1; \ if (i < ks->end) { \ if (dret) *dret = ks->buf[i]; \ break; \ } \ } \ if (!gotany && ks_eof(ks)) return -1; \ ks->seek_pos += seek_pos; \ if (str->s == 0) { \ str->m = 1; \ str->s = (char*)calloc(1, 1); \ } else if (delimiter == KS_SEP_LINE && str->l > 1 && str->s[str->l-1] == '\r') --str->l; \ str->s[str->l] = '\0'; \ return str->l; \ } #define KSTREAM_INIT2(SCOPE, type_t, __read, __bufsize) \ __KS_TYPE(type_t) \ __KS_BASIC(SCOPE, type_t, __bufsize) \ __KS_GETUNTIL(SCOPE, __read) \ __KS_INLINED(__read) #define KSTREAM_INIT(type_t, __read, __bufsize) KSTREAM_INIT2(static, type_t, __read, __bufsize) #define KSTREAM_DECLARE(type_t, __read) \ __KS_TYPE(type_t) \ extern int ks_getuntil2(kstream_t *ks, int delimiter, kstring_t *str, int *dret, int append); \ extern kstream_t *ks_init(type_t f); \ extern void ks_destroy(kstream_t *ks); \ __KS_INLINED(__read) /****************** * FASTA/Q parser * ******************/ #define kseq_rewind(ks) ((ks)->last_char = (ks)->f->is_eof = (ks)->f->begin = (ks)->f->end = 0) #define __KSEQ_BASIC(SCOPE, type_t) \ SCOPE kseq_t *kseq_init(type_t fd) \ { \ kseq_t *s = (kseq_t*)calloc(1, sizeof(kseq_t)); \ s->f = ks_init(fd); \ return s; \ } \ SCOPE void kseq_destroy(kseq_t *ks) \ { \ if (!ks) return; \ free(ks->name.s); free(ks->comment.s); free(ks->seq.s); free(ks->qual.s); \ ks_destroy(ks->f); \ free(ks); \ } /* Return value: >=0 length of the sequence (normal) -1 end-of-file -2 truncated quality string */ #define __KSEQ_READ(SCOPE) \ SCOPE int kseq_read(kseq_t *seq) \ { \ int c; \ kstream_t *ks = seq->f; \ if (seq->last_char == 0) { /* then jump to the next header line */ \ while ((c = ks_getc(ks)) != -1 && c != '>' && c != '@'); \ if (c == -1) return -1; /* end of file */ \ seq->last_char = c; \ } /* else: the first header char has been read in the previous call */ \ seq->comment.l = seq->seq.l = seq->qual.l = 0; /* reset all members */ \ if (ks_getuntil(ks, 0, &seq->name, &c) < 0) return -1; /* normal exit: EOF */ \ if (c != '\n') ks_getuntil(ks, KS_SEP_LINE, &seq->comment, 0); /* read FASTA/Q comment */ \ if (seq->seq.s == 0) { /* we can do this in the loop below, but that is slower */ \ seq->seq.m = 256; \ seq->seq.s = (char*)malloc(seq->seq.m); \ } \ while ((c = ks_getc(ks)) != -1 && c != '>' && c != '+' && c != '@') { \ if (c == '\n') continue; /* skip empty lines */ \ seq->seq.s[seq->seq.l++] = c; /* this is safe: we always have enough space for 1 char */ \ ks_getuntil2(ks, KS_SEP_LINE, &seq->seq, 0, 1); /* read the rest of the line */ \ } \ if (c == '>' || c == '@') seq->last_char = c; /* the first header char has been read */ \ if (seq->seq.l + 1 >= seq->seq.m) { /* seq->seq.s[seq->seq.l] below may be out of boundary */ \ seq->seq.m = seq->seq.l + 2; \ kroundup32(seq->seq.m); /* rounded to the next closest 2^k */ \ seq->seq.s = (char*)realloc(seq->seq.s, seq->seq.m); \ } \ seq->seq.s[seq->seq.l] = 0; /* null terminated string */ \ if (c != '+') return seq->seq.l; /* FASTA */ \ if (seq->qual.m < seq->seq.m) { /* allocate memory for qual in case insufficient */ \ seq->qual.m = seq->seq.m; \ seq->qual.s = (char*)realloc(seq->qual.s, seq->qual.m); \ } \ while ((c = ks_getc(ks)) != -1 && c != '\n'); /* skip the rest of '+' line */ \ if (c == -1) return -2; /* error: no quality string */ \ while (ks_getuntil2(ks, KS_SEP_LINE, &seq->qual, 0, 1) >= 0 && seq->qual.l < seq->seq.l); \ seq->last_char = 0; /* we have not come to the next header line */ \ if (seq->seq.l != seq->qual.l) return -2; /* error: qual string is of a different length */ \ return seq->seq.l; \ } #define __KSEQ_TYPE(type_t) \ typedef struct { \ kstring_t name, comment, seq, qual; \ int last_char; \ kstream_t *f; \ } kseq_t; #define KSEQ_INIT2(SCOPE, type_t, __read) \ KSTREAM_INIT(type_t, __read, 16384) \ __KSEQ_TYPE(type_t) \ __KSEQ_BASIC(SCOPE, type_t) \ __KSEQ_READ(SCOPE) #define KSEQ_INIT(type_t, __read) KSEQ_INIT2(static, type_t, __read) #define KSEQ_DECLARE(type_t) \ __KS_TYPE(type_t) \ __KSEQ_TYPE(type_t) \ extern kseq_t *kseq_init(type_t fd); \ void kseq_destroy(kseq_t *ks); \ int kseq_read(kseq_t *seq); #endif stacks-1.44/htslib/htslib/._ksort.h000644 000765 000024 00000000265 12750743246 020021 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/ksort.h000644 000765 000024 00000023615 12750743246 017610 0ustar00catchenstaff000000 000000 /* The MIT License Copyright (c) 2008 Genome Research Ltd (GRL). Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ /* Contact: Heng Li */ /* 2012-12-11 (0.1.4): * Defined __ks_insertsort_##name as static to compile with C99. 2008-11-16 (0.1.4): * Fixed a bug in introsort() that happens in rare cases. 2008-11-05 (0.1.3): * Fixed a bug in introsort() for complex comparisons. * Fixed a bug in mergesort(). The previous version is not stable. 2008-09-15 (0.1.2): * Accelerated introsort. On my Mac (not on another Linux machine), my implementation is as fast as std::sort on random input. * Added combsort and in introsort, switch to combsort if the recursion is too deep. 2008-09-13 (0.1.1): * Added k-small algorithm 2008-09-05 (0.1.0): * Initial version */ #ifndef AC_KSORT_H #define AC_KSORT_H #include #include typedef struct { void *left, *right; int depth; } ks_isort_stack_t; #define KSORT_SWAP(type_t, a, b) { register type_t t=(a); (a)=(b); (b)=t; } #define KSORT_INIT(name, type_t, __sort_lt) \ void ks_mergesort_##name(size_t n, type_t array[], type_t temp[]) \ { \ type_t *a2[2], *a, *b; \ int curr, shift; \ \ a2[0] = array; \ a2[1] = temp? temp : (type_t*)malloc(sizeof(type_t) * n); \ for (curr = 0, shift = 0; (1ul<> 1) - 1; i != (size_t)(-1); --i) \ ks_heapadjust_##name(i, lsize, l); \ } \ void ks_heapsort_##name(size_t lsize, type_t l[]) \ { \ size_t i; \ for (i = lsize - 1; i > 0; --i) { \ type_t tmp; \ tmp = *l; *l = l[i]; l[i] = tmp; ks_heapadjust_##name(0, i, l); \ } \ } \ static inline void __ks_insertsort_##name(type_t *s, type_t *t) \ { \ type_t *i, *j, swap_tmp; \ for (i = s + 1; i < t; ++i) \ for (j = i; j > s && __sort_lt(*j, *(j-1)); --j) { \ swap_tmp = *j; *j = *(j-1); *(j-1) = swap_tmp; \ } \ } \ void ks_combsort_##name(size_t n, type_t a[]) \ { \ const double shrink_factor = 1.2473309501039786540366528676643; \ int do_swap; \ size_t gap = n; \ type_t tmp, *i, *j; \ do { \ if (gap > 2) { \ gap = (size_t)(gap / shrink_factor); \ if (gap == 9 || gap == 10) gap = 11; \ } \ do_swap = 0; \ for (i = a; i < a + n - gap; ++i) { \ j = i + gap; \ if (__sort_lt(*j, *i)) { \ tmp = *i; *i = *j; *j = tmp; \ do_swap = 1; \ } \ } \ } while (do_swap || gap > 2); \ if (gap != 1) __ks_insertsort_##name(a, a + n); \ } \ void ks_introsort_##name(size_t n, type_t a[]) \ { \ int d; \ ks_isort_stack_t *top, *stack; \ type_t rp, swap_tmp; \ type_t *s, *t, *i, *j, *k; \ \ if (n < 1) return; \ else if (n == 2) { \ if (__sort_lt(a[1], a[0])) { swap_tmp = a[0]; a[0] = a[1]; a[1] = swap_tmp; } \ return; \ } \ for (d = 2; 1ul<>1) + 1; \ if (__sort_lt(*k, *i)) { \ if (__sort_lt(*k, *j)) k = j; \ } else k = __sort_lt(*j, *i)? i : j; \ rp = *k; \ if (k != t) { swap_tmp = *k; *k = *t; *t = swap_tmp; } \ for (;;) { \ do ++i; while (__sort_lt(*i, rp)); \ do --j; while (i <= j && __sort_lt(rp, *j)); \ if (j <= i) break; \ swap_tmp = *i; *i = *j; *j = swap_tmp; \ } \ swap_tmp = *i; *i = *t; *t = swap_tmp; \ if (i-s > t-i) { \ if (i-s > 16) { top->left = s; top->right = i-1; top->depth = d; ++top; } \ s = t-i > 16? i+1 : t; \ } else { \ if (t-i > 16) { top->left = i+1; top->right = t; top->depth = d; ++top; } \ t = i-s > 16? i-1 : s; \ } \ } else { \ if (top == stack) { \ free(stack); \ __ks_insertsort_##name(a, a+n); \ return; \ } else { --top; s = (type_t*)top->left; t = (type_t*)top->right; d = top->depth; } \ } \ } \ } \ /* This function is adapted from: http://ndevilla.free.fr/median/ */ \ /* 0 <= kk < n */ \ type_t ks_ksmall_##name(size_t n, type_t arr[], size_t kk) \ { \ type_t *low, *high, *k, *ll, *hh, *mid; \ low = arr; high = arr + n - 1; k = arr + kk; \ for (;;) { \ if (high <= low) return *k; \ if (high == low + 1) { \ if (__sort_lt(*high, *low)) KSORT_SWAP(type_t, *low, *high); \ return *k; \ } \ mid = low + (high - low) / 2; \ if (__sort_lt(*high, *mid)) KSORT_SWAP(type_t, *mid, *high); \ if (__sort_lt(*high, *low)) KSORT_SWAP(type_t, *low, *high); \ if (__sort_lt(*low, *mid)) KSORT_SWAP(type_t, *mid, *low); \ KSORT_SWAP(type_t, *mid, *(low+1)); \ ll = low + 1; hh = high; \ for (;;) { \ do ++ll; while (__sort_lt(*ll, *low)); \ do --hh; while (__sort_lt(*low, *hh)); \ if (hh < ll) break; \ KSORT_SWAP(type_t, *ll, *hh); \ } \ KSORT_SWAP(type_t, *low, *hh); \ if (hh <= k) low = ll; \ if (hh >= k) high = hh - 1; \ } \ } \ void ks_shuffle_##name(size_t n, type_t a[]) \ { \ int i, j; \ for (i = n; i > 1; --i) { \ type_t tmp; \ j = (int)(drand48() * i); \ tmp = a[j]; a[j] = a[i-1]; a[i-1] = tmp; \ } \ } #define ks_mergesort(name, n, a, t) ks_mergesort_##name(n, a, t) #define ks_introsort(name, n, a) ks_introsort_##name(n, a) #define ks_combsort(name, n, a) ks_combsort_##name(n, a) #define ks_heapsort(name, n, a) ks_heapsort_##name(n, a) #define ks_heapmake(name, n, a) ks_heapmake_##name(n, a) #define ks_heapadjust(name, i, n, a) ks_heapadjust_##name(i, n, a) #define ks_ksmall(name, n, a, k) ks_ksmall_##name(n, a, k) #define ks_shuffle(name, n, a) ks_shuffle_##name(n, a) #define ks_lt_generic(a, b) ((a) < (b)) #define ks_lt_str(a, b) (strcmp((a), (b)) < 0) typedef const char *ksstr_t; #define KSORT_INIT_GENERIC(type_t) KSORT_INIT(type_t, type_t, ks_lt_generic) #define KSORT_INIT_STR KSORT_INIT(str, ksstr_t, ks_lt_str) #endif stacks-1.44/htslib/htslib/._kstring.h000644 000765 000024 00000000265 12750743246 020340 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/kstring.h000644 000765 000024 00000015637 12750743246 020134 0ustar00catchenstaff000000 000000 /* The MIT License Copyright (C) 2011 by Attractive Chaos Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #ifndef KSTRING_H #define KSTRING_H #include #include #include #include #include #ifndef kroundup32 #define kroundup32(x) (--(x), (x)|=(x)>>1, (x)|=(x)>>2, (x)|=(x)>>4, (x)|=(x)>>8, (x)|=(x)>>16, ++(x)) #endif #if defined __GNUC__ && (__GNUC__ > 2 || (__GNUC__ == 2 && __GNUC_MINOR__ > 4)) #define KS_ATTR_PRINTF(fmt, arg) __attribute__((__format__ (__printf__, fmt, arg))) #else #define KS_ATTR_PRINTF(fmt, arg) #endif /* kstring_t is a simple non-opaque type whose fields are likely to be * used directly by user code (but see also ks_str() and ks_len() below). * A kstring_t object is initialised by either of * kstring_t str = { 0, 0, NULL }; * kstring_t str; ...; str.l = str.m = 0; str.s = NULL; * and either ownership of the underlying buffer should be given away before * the object disappears (see ks_release() below) or the kstring_t should be * destroyed with free(str.s); */ #ifndef KSTRING_T #define KSTRING_T kstring_t typedef struct __kstring_t { size_t l, m; char *s; } kstring_t; #endif typedef struct { uint64_t tab[4]; int sep, finished; const char *p; // end of the current token } ks_tokaux_t; #ifdef __cplusplus extern "C" { #endif int kvsprintf(kstring_t *s, const char *fmt, va_list ap) KS_ATTR_PRINTF(2,0); int ksprintf(kstring_t *s, const char *fmt, ...) KS_ATTR_PRINTF(2,3); int ksplit_core(char *s, int delimiter, int *_max, int **_offsets); char *kstrstr(const char *str, const char *pat, int **_prep); char *kstrnstr(const char *str, const char *pat, int n, int **_prep); void *kmemmem(const void *_str, int n, const void *_pat, int m, int **_prep); /* kstrtok() is similar to strtok_r() except that str is not * modified and both str and sep can be NULL. For efficiency, it is * actually recommended to set both to NULL in the subsequent calls * if sep is not changed. */ char *kstrtok(const char *str, const char *sep, ks_tokaux_t *aux); /* kgetline() uses the supplied fgets()-like function to read a "\n"- * or "\r\n"-terminated line from fp. The line read is appended to the * kstring without its terminator and 0 is returned; EOF is returned at * EOF or on error (determined by querying fp, as per fgets()). */ typedef char *kgets_func(char *, int, void *); int kgetline(kstring_t *s, kgets_func *fgets, void *fp); #ifdef __cplusplus } #endif static inline int ks_resize(kstring_t *s, size_t size) { if (s->m < size) { char *tmp; s->m = size; kroundup32(s->m); if ((tmp = (char*)realloc(s->s, s->m))) s->s = tmp; else return -1; } return 0; } static inline char *ks_str(kstring_t *s) { return s->s; } static inline size_t ks_len(kstring_t *s) { return s->l; } // Give ownership of the underlying buffer away to something else (making // that something else responsible for freeing it), leaving the kstring_t // empty and ready to be used again, or ready to go out of scope without // needing free(str.s) to prevent a memory leak. static inline char *ks_release(kstring_t *s) { char *ss = s->s; s->l = s->m = 0; s->s = NULL; return ss; } static inline int kputsn(const char *p, int l, kstring_t *s) { if (s->l + l + 1 >= s->m) { char *tmp; s->m = s->l + l + 2; kroundup32(s->m); if ((tmp = (char*)realloc(s->s, s->m))) s->s = tmp; else return EOF; } memcpy(s->s + s->l, p, l); s->l += l; s->s[s->l] = 0; return l; } static inline int kputs(const char *p, kstring_t *s) { return kputsn(p, strlen(p), s); } static inline int kputc(int c, kstring_t *s) { if (s->l + 1 >= s->m) { char *tmp; s->m = s->l + 2; kroundup32(s->m); if ((tmp = (char*)realloc(s->s, s->m))) s->s = tmp; else return EOF; } s->s[s->l++] = c; s->s[s->l] = 0; return c; } static inline int kputc_(int c, kstring_t *s) { if (s->l + 1 > s->m) { char *tmp; s->m = s->l + 1; kroundup32(s->m); if ((tmp = (char*)realloc(s->s, s->m))) s->s = tmp; else return EOF; } s->s[s->l++] = c; return 1; } static inline int kputsn_(const void *p, int l, kstring_t *s) { if (s->l + l > s->m) { char *tmp; s->m = s->l + l; kroundup32(s->m); if ((tmp = (char*)realloc(s->s, s->m))) s->s = tmp; else return EOF; } memcpy(s->s + s->l, p, l); s->l += l; return l; } static inline int kputw(int c, kstring_t *s) { char buf[16]; int i, l = 0; unsigned int x = c; if (c < 0) x = -x; do { buf[l++] = x%10 + '0'; x /= 10; } while (x > 0); if (c < 0) buf[l++] = '-'; if (s->l + l + 1 >= s->m) { char *tmp; s->m = s->l + l + 2; kroundup32(s->m); if ((tmp = (char*)realloc(s->s, s->m))) s->s = tmp; else return EOF; } for (i = l - 1; i >= 0; --i) s->s[s->l++] = buf[i]; s->s[s->l] = 0; return 0; } static inline int kputuw(unsigned c, kstring_t *s) { char buf[16]; int l, i; unsigned x; if (c == 0) return kputc('0', s); for (l = 0, x = c; x > 0; x /= 10) buf[l++] = x%10 + '0'; if (s->l + l + 1 >= s->m) { char *tmp; s->m = s->l + l + 2; kroundup32(s->m); if ((tmp = (char*)realloc(s->s, s->m))) s->s = tmp; else return EOF; } for (i = l - 1; i >= 0; --i) s->s[s->l++] = buf[i]; s->s[s->l] = 0; return 0; } static inline int kputl(long c, kstring_t *s) { char buf[32]; int i, l = 0; unsigned long x = c; if (c < 0) x = -x; do { buf[l++] = x%10 + '0'; x /= 10; } while (x > 0); if (c < 0) buf[l++] = '-'; if (s->l + l + 1 >= s->m) { char *tmp; s->m = s->l + l + 2; kroundup32(s->m); if ((tmp = (char*)realloc(s->s, s->m))) s->s = tmp; else return EOF; } for (i = l - 1; i >= 0; --i) s->s[s->l++] = buf[i]; s->s[s->l] = 0; return 0; } /* * Returns 's' split by delimiter, with *n being the number of components; * NULL on failue. */ static inline int *ksplit(kstring_t *s, int delimiter, int *n) { int max = 0, *offsets = 0; *n = ksplit_core(s->s, delimiter, &max, &offsets); return offsets; } #endif stacks-1.44/htslib/htslib/._regidx.h000644 000765 000024 00000000265 12750743246 020141 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/regidx.h000644 000765 000024 00000013246 12750743246 017727 0ustar00catchenstaff000000 000000 /* Copyright (C) 2014 Genome Research Ltd. Author: Petr Danecek Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ /* Regions indexing with an optional payload. Inspired by samtools/bedidx.c. This code is intended as future replacement of bcf_sr_regions_t. Example of usage: // Init the parser and print regions. In this example the payload is a // pointer to a string. For the description of parse_custom and // free_custom functions, see regidx_parse_f and regidx_free_f below, // and for working example see test/test-regidx.c. regidx_t *idx = regidx_init(in_fname,parse_custom,free_custom,sizeof(char*),NULL); // Query overlap with chr:from-to regitr_t itr; if ( regidx_overlap(idx, chr,from,to, &itr) ) printf("There is an overlap!\n"); while ( REGITR_OVERLAP(itr,from,to) ) { printf("[%d,%d] overlaps with [%d,%d], payload=%s\n", from,to, REGITR_START(itr), REGITR_END(itr), REGITR_PAYLOAD(itr,char*)); itr.i++; } regidx_destroy(regs); */ #ifndef HTSLIB_REGIDX_H #define HTSLIB_REGIDX_H #include #include #ifdef __cplusplus extern "C" { #endif typedef struct _regidx_t regidx_t; typedef struct { uint32_t start, end; } reg_t; typedef struct { int i, n; reg_t *reg; void *payload; } regitr_t; #define REGITR_START(itr) (itr).reg[(itr).i].start #define REGITR_END(itr) (itr).reg[(itr).i].end #define REGITR_PAYLOAD(itr,type_t) ((type_t*)(itr).payload)[(itr).i] #define REGITR_OVERLAP(itr,from,to) (itr.i < itr.n && REGITR_START(itr)<=to && REGITR_END(itr)>=from ) /* * regidx_parse_f - Function to parse one input line, such as regidx_parse_bed * or regidx_parse_tab below. The function is expected to set `chr_from` and * `chr_to` to point to first and last character of chromosome name and set * coordinates `reg->start` and `reg->end` (0-based, inclusive). If * regidx_init() was called with non-zero payload_size, the `payload` points * to a memory location of the payload_size and `usr` is data passed to * regidx_init(). Any memory allocated by the function will be freed by * regidx_free_f on regidx_destroy(). * * Return value: 0 on success, -1 to skip a record, -2 on fatal error. */ typedef int (*regidx_parse_f)(const char *line, char **chr_beg, char **chr_end, reg_t *reg, void *payload, void *usr); typedef void (*regidx_free_f)(void *payload); int regidx_parse_bed(const char*,char**,char**,reg_t*,void*,void*); // CHROM,FROM,TO (0-based,right-open) int regidx_parse_tab(const char*,char**,char**,reg_t*,void*,void*); // CHROM,POS (1-based, inclusive) /* * regidx_init() - creates new index * @param fname: input file name or NULL if regions will be added one-by-one via regidx_insert() * @param parsef: regidx_parse_bed, regidx_parse_tab or see description of regidx_parse_f. If NULL, * the format will be autodected, currently either regidx_parse_tab (the default) or * regidx_parse_bed (file must be named 'bed' or 'bed.gz') will be used. Note that * the exact autodetection algorithm will change. * @param freef: NULL or see description of regidx_parse_f * @param payload_size: 0 with regidx_parse_bed, regidx_parse_tab or see regidx_parse_f * @param usr: optional user data passed to regidx_parse_f * * Returns index on success or NULL on error. */ regidx_t *regidx_init(const char *fname, regidx_parse_f parsef, regidx_free_f freef, size_t payload_size, void *usr); /* * regidx_destroy() - free memory allocated by regidx_init */ void regidx_destroy(regidx_t *idx); /* * regidx_overlap() - check overlap of the location chr:from-to with regions * @param start,end: 0-based start, end coordinate (inclusive) * @param itr: pointer to iterator, can be NULL if not needed * * Returns 0 if there is no overlap or 1 if overlap is found. The overlapping * regions can be iterated as shown in the example above. */ int regidx_overlap(regidx_t *idx, const char *chr, uint32_t start, uint32_t end, regitr_t *itr); /* * regidx_insert() - add a new region. * * After last region has been added, call regidx_insert(idx,NULL) to * build the index. * * Returns 0 on success or -1 on error. */ int regidx_insert(regidx_t *idx, char *line); /* * regidx_seq_names() - return list of all sequence names */ char **regidx_seq_names(regidx_t *idx, int *n); /* * regidx_seq_nregs() - number of regions * regidx_nregs() - total number of regions */ int regidx_seq_nregs(regidx_t *idx, const char *seq); int regidx_nregs(regidx_t *idx); #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/htslib/._sam.h000644 000765 000024 00000000265 12750743246 017437 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/sam.h000644 000765 000024 00000041301 12750743246 017216 0ustar00catchenstaff000000 000000 /* sam.h -- SAM and BAM file I/O and manipulation. Copyright (C) 2008, 2009, 2013-2014 Genome Research Ltd. Copyright (C) 2010, 2012, 2013 Broad Institute. Author: Heng Li Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #ifndef HTSLIB_SAM_H #define HTSLIB_SAM_H #include #include "hts.h" #ifdef __cplusplus extern "C" { #endif /********************** *** SAM/BAM header *** **********************/ /*! @typedef @abstract Structure for the alignment header. @field n_targets number of reference sequences @field l_text length of the plain text in the header @field target_len lengths of the reference sequences @field target_name names of the reference sequences @field text plain text @field sdict header dictionary */ typedef struct { int32_t n_targets, ignore_sam_err; uint32_t l_text; uint32_t *target_len; int8_t *cigar_tab; char **target_name; char *text; void *sdict; } bam_hdr_t; /**************************** *** CIGAR related macros *** ****************************/ #define BAM_CMATCH 0 #define BAM_CINS 1 #define BAM_CDEL 2 #define BAM_CREF_SKIP 3 #define BAM_CSOFT_CLIP 4 #define BAM_CHARD_CLIP 5 #define BAM_CPAD 6 #define BAM_CEQUAL 7 #define BAM_CDIFF 8 #define BAM_CBACK 9 #define BAM_CIGAR_STR "MIDNSHP=XB" #define BAM_CIGAR_SHIFT 4 #define BAM_CIGAR_MASK 0xf #define BAM_CIGAR_TYPE 0x3C1A7 #define bam_cigar_op(c) ((c)&BAM_CIGAR_MASK) #define bam_cigar_oplen(c) ((c)>>BAM_CIGAR_SHIFT) #define bam_cigar_opchr(c) (BAM_CIGAR_STR[bam_cigar_op(c)]) #define bam_cigar_gen(l, o) ((l)<>((o)<<1)&3) // bit 1: consume query; bit 2: consume reference /*! @abstract the read is paired in sequencing, no matter whether it is mapped in a pair */ #define BAM_FPAIRED 1 /*! @abstract the read is mapped in a proper pair */ #define BAM_FPROPER_PAIR 2 /*! @abstract the read itself is unmapped; conflictive with BAM_FPROPER_PAIR */ #define BAM_FUNMAP 4 /*! @abstract the mate is unmapped */ #define BAM_FMUNMAP 8 /*! @abstract the read is mapped to the reverse strand */ #define BAM_FREVERSE 16 /*! @abstract the mate is mapped to the reverse strand */ #define BAM_FMREVERSE 32 /*! @abstract this is read1 */ #define BAM_FREAD1 64 /*! @abstract this is read2 */ #define BAM_FREAD2 128 /*! @abstract not primary alignment */ #define BAM_FSECONDARY 256 /*! @abstract QC failure */ #define BAM_FQCFAIL 512 /*! @abstract optical or PCR duplicate */ #define BAM_FDUP 1024 /*! @abstract supplementary alignment */ #define BAM_FSUPPLEMENTARY 2048 /************************* *** Alignment records *** *************************/ /*! @typedef @abstract Structure for core alignment information. @field tid chromosome ID, defined by bam_hdr_t @field pos 0-based leftmost coordinate @field bin bin calculated by bam_reg2bin() @field qual mapping quality @field l_qname length of the query name @field flag bitwise flag @field n_cigar number of CIGAR operations @field l_qseq length of the query sequence (read) @field mtid chromosome ID of next read in template, defined by bam_hdr_t @field mpos 0-based leftmost coordinate of next read in template */ typedef struct { int32_t tid; int32_t pos; uint32_t bin:16, qual:8, l_qname:8; uint32_t flag:16, n_cigar:16; int32_t l_qseq; int32_t mtid; int32_t mpos; int32_t isize; } bam1_core_t; /*! @typedef @abstract Structure for one alignment. @field core core information about the alignment @field l_data current length of bam1_t::data @field m_data maximum length of bam1_t::data @field data all variable-length data, concatenated; structure: qname-cigar-seq-qual-aux @discussion Notes: 1. qname is zero tailing and core.l_qname includes the tailing '\0'. 2. l_qseq is calculated from the total length of an alignment block on reading or from CIGAR. 3. cigar data is encoded 4 bytes per CIGAR operation. 4. seq is nybble-encoded according to bam_nt16_table. */ typedef struct { bam1_core_t core; int l_data, m_data; uint8_t *data; #ifndef BAM_NO_ID uint64_t id; #endif } bam1_t; /*! @function @abstract Get whether the query is on the reverse strand @param b pointer to an alignment @return boolean true if query is on the reverse strand */ #define bam_is_rev(b) (((b)->core.flag&BAM_FREVERSE) != 0) /*! @function @abstract Get whether the query's mate is on the reverse strand @param b pointer to an alignment @return boolean true if query's mate on the reverse strand */ #define bam_is_mrev(b) (((b)->core.flag&BAM_FMREVERSE) != 0) /*! @function @abstract Get the name of the query @param b pointer to an alignment @return pointer to the name string, null terminated */ #define bam_get_qname(b) ((char*)(b)->data) /*! @function @abstract Get the CIGAR array @param b pointer to an alignment @return pointer to the CIGAR array @discussion In the CIGAR array, each element is a 32-bit integer. The lower 4 bits gives a CIGAR operation and the higher 28 bits keep the length of a CIGAR. */ #define bam_get_cigar(b) ((uint32_t*)((b)->data + (b)->core.l_qname)) /*! @function @abstract Get query sequence @param b pointer to an alignment @return pointer to sequence @discussion Each base is encoded in 4 bits: 1 for A, 2 for C, 4 for G, 8 for T and 15 for N. Two bases are packed in one byte with the base at the higher 4 bits having smaller coordinate on the read. It is recommended to use bam_seqi() macro to get the base. */ #define bam_get_seq(b) ((b)->data + ((b)->core.n_cigar<<2) + (b)->core.l_qname) /*! @function @abstract Get query quality @param b pointer to an alignment @return pointer to quality string */ #define bam_get_qual(b) ((b)->data + ((b)->core.n_cigar<<2) + (b)->core.l_qname + (((b)->core.l_qseq + 1)>>1)) /*! @function @abstract Get auxiliary data @param b pointer to an alignment @return pointer to the concatenated auxiliary data */ #define bam_get_aux(b) ((b)->data + ((b)->core.n_cigar<<2) + (b)->core.l_qname + (((b)->core.l_qseq + 1)>>1) + (b)->core.l_qseq) /*! @function @abstract Get length of auxiliary data @param b pointer to an alignment @return length of the concatenated auxiliary data */ #define bam_get_l_aux(b) ((b)->l_data - ((b)->core.n_cigar<<2) - (b)->core.l_qname - (b)->core.l_qseq - (((b)->core.l_qseq + 1)>>1)) /*! @function @abstract Get a base on read @param s Query sequence returned by bam_get_seq() @param i The i-th position, 0-based @return 4-bit integer representing the base. */ #define bam_seqi(s, i) ((s)[(i)>>1] >> ((~(i)&1)<<2) & 0xf) /************************** *** Exported functions *** **************************/ /*************** *** BAM I/O *** ***************/ bam_hdr_t *bam_hdr_init(void); bam_hdr_t *bam_hdr_read(BGZF *fp); int bam_hdr_write(BGZF *fp, const bam_hdr_t *h) HTS_RESULT_USED; void bam_hdr_destroy(bam_hdr_t *h); int bam_name2id(bam_hdr_t *h, const char *ref); bam_hdr_t* bam_hdr_dup(const bam_hdr_t *h0); bam1_t *bam_init1(void); void bam_destroy1(bam1_t *b); int bam_read1(BGZF *fp, bam1_t *b) HTS_RESULT_USED; int bam_write1(BGZF *fp, const bam1_t *b) HTS_RESULT_USED; bam1_t *bam_copy1(bam1_t *bdst, const bam1_t *bsrc); bam1_t *bam_dup1(const bam1_t *bsrc); int bam_cigar2qlen(int n_cigar, const uint32_t *cigar); int bam_cigar2rlen(int n_cigar, const uint32_t *cigar); /*! @abstract Calculate the rightmost base position of an alignment on the reference genome. @param b pointer to an alignment @return the coordinate of the first base after the alignment, 0-based @discussion For a mapped read, this is just b->core.pos + bam_cigar2rlen. For an unmapped read (either according to its flags or if it has no cigar string), we return b->core.pos + 1 by convention. */ int32_t bam_endpos(const bam1_t *b); int bam_str2flag(const char *str); /** returns negative value on error */ char *bam_flag2str(int flag); /** The string must be freed by the user */ /************************* *** BAM/CRAM indexing *** *************************/ // These BAM iterator functions work only on BAM files. To work with either // BAM or CRAM files use the sam_index_load() & sam_itr_*() functions. #define bam_itr_destroy(iter) hts_itr_destroy(iter) #define bam_itr_queryi(idx, tid, beg, end) sam_itr_queryi(idx, tid, beg, end) #define bam_itr_querys(idx, hdr, region) sam_itr_querys(idx, hdr, region) #define bam_itr_next(htsfp, itr, r) hts_itr_next((htsfp)->fp.bgzf, (itr), (r), 0) // Load/build .csi or .bai BAM index file. Does not work with CRAM. // It is recommended to use the sam_index_* functions below instead. #define bam_index_load(fn) hts_idx_load((fn), HTS_FMT_BAI) #define bam_index_build(fn, min_shift) (sam_index_build((fn), (min_shift))) /// Load a BAM (.csi or .bai) or CRAM (.crai) index file /** @param fp File handle of the data file whose index is being opened @param fn BAM/CRAM/etc filename to search alongside for the index file @return The index, or NULL if an error occurred. */ hts_idx_t *sam_index_load(htsFile *fp, const char *fn); /// Load a specific BAM (.csi or .bai) or CRAM (.crai) index file /** @param fp File handle of the data file whose index is being opened @param fn BAM/CRAM/etc data file filename @param fnidx Index filename, or NULL to search alongside @a fn @return The index, or NULL if an error occurred. */ hts_idx_t *sam_index_load2(htsFile *fp, const char *fn, const char *fnidx); /// Generate and save an index file /** @param fn Input BAM/etc filename, to which .csi/etc will be added @param min_shift Positive to generate CSI, or 0 to generate BAI @return 0 if successful, or negative if an error occurred (usually -1; or -2: opening fn failed; -3: format not indexable) */ int sam_index_build(const char *fn, int min_shift) HTS_RESULT_USED; /// Generate and save an index to a specific file /** @param fn Input BAM/CRAM/etc filename @param fnidx Output filename, or NULL to add .bai/.csi/etc to @a fn @param min_shift Positive to generate CSI, or 0 to generate BAI @return 0 if successful, or negative if an error occurred. */ int sam_index_build2(const char *fn, const char *fnidx, int min_shift) HTS_RESULT_USED; #define sam_itr_destroy(iter) hts_itr_destroy(iter) hts_itr_t *sam_itr_queryi(const hts_idx_t *idx, int tid, int beg, int end); hts_itr_t *sam_itr_querys(const hts_idx_t *idx, bam_hdr_t *hdr, const char *region); #define sam_itr_next(htsfp, itr, r) hts_itr_next((htsfp)->fp.bgzf, (itr), (r), (htsfp)) /*************** *** SAM I/O *** ***************/ #define sam_open(fn, mode) (hts_open((fn), (mode))) #define sam_open_format(fn, mode, fmt) (hts_open_format((fn), (mode), (fmt))) #define sam_close(fp) hts_close(fp) int sam_open_mode(char *mode, const char *fn, const char *format); // A version of sam_open_mode that can handle ,key=value options. // The format string is allocated and returned, to be freed by the caller. // Prefix should be "r" or "w", char *sam_open_mode_opts(const char *fn, const char *mode, const char *format); typedef htsFile samFile; bam_hdr_t *sam_hdr_parse(int l_text, const char *text); bam_hdr_t *sam_hdr_read(samFile *fp); int sam_hdr_write(samFile *fp, const bam_hdr_t *h) HTS_RESULT_USED; int sam_parse1(kstring_t *s, bam_hdr_t *h, bam1_t *b) HTS_RESULT_USED; int sam_format1(const bam_hdr_t *h, const bam1_t *b, kstring_t *str) HTS_RESULT_USED; int sam_read1(samFile *fp, bam_hdr_t *h, bam1_t *b) HTS_RESULT_USED; int sam_write1(samFile *fp, const bam_hdr_t *h, const bam1_t *b) HTS_RESULT_USED; /************************************* *** Manipulating auxiliary fields *** *************************************/ uint8_t *bam_aux_get(const bam1_t *b, const char tag[2]); int32_t bam_aux2i(const uint8_t *s); double bam_aux2f(const uint8_t *s); char bam_aux2A(const uint8_t *s); char *bam_aux2Z(const uint8_t *s); void bam_aux_append(bam1_t *b, const char tag[2], char type, int len, uint8_t *data); int bam_aux_del(bam1_t *b, uint8_t *s); /************************** *** Pileup and Mpileup *** **************************/ #if !defined(BAM_NO_PILEUP) /*! @typedef @abstract Structure for one alignment covering the pileup position. @field b pointer to the alignment @field qpos position of the read base at the pileup site, 0-based @field indel indel length; 0 for no indel, positive for ins and negative for del @field level the level of the read in the "viewer" mode @field is_del 1 iff the base on the padded read is a deletion @field is_head ??? @field is_tail ??? @field is_refskip ??? @field aux ??? @discussion See also bam_plbuf_push() and bam_lplbuf_push(). The difference between the two functions is that the former does not set bam_pileup1_t::level, while the later does. Level helps the implementation of alignment viewers, but calculating this has some overhead. */ typedef struct { bam1_t *b; int32_t qpos; int indel, level; uint32_t is_del:1, is_head:1, is_tail:1, is_refskip:1, aux:28; } bam_pileup1_t; typedef int (*bam_plp_auto_f)(void *data, bam1_t *b); struct __bam_plp_t; typedef struct __bam_plp_t *bam_plp_t; struct __bam_mplp_t; typedef struct __bam_mplp_t *bam_mplp_t; /** * bam_plp_init() - sets an iterator over multiple * @func: see mplp_func in bam_plcmd.c in samtools for an example. Expected return * status: 0 on success, -1 on end, < -1 on non-recoverable errors * @data: user data to pass to @func */ bam_plp_t bam_plp_init(bam_plp_auto_f func, void *data); void bam_plp_destroy(bam_plp_t iter); int bam_plp_push(bam_plp_t iter, const bam1_t *b); const bam_pileup1_t *bam_plp_next(bam_plp_t iter, int *_tid, int *_pos, int *_n_plp); const bam_pileup1_t *bam_plp_auto(bam_plp_t iter, int *_tid, int *_pos, int *_n_plp); void bam_plp_set_maxcnt(bam_plp_t iter, int maxcnt); void bam_plp_reset(bam_plp_t iter); bam_mplp_t bam_mplp_init(int n, bam_plp_auto_f func, void **data); /** * bam_mplp_init_overlaps() - if called, mpileup will detect overlapping * read pairs and for each base pair set the base quality of the * lower-quality base to zero, thus effectively discarding it from * calling. If the two bases are identical, the quality of the other base * is increased to the sum of their qualities (capped at 200), otherwise * it is multiplied by 0.8. */ void bam_mplp_init_overlaps(bam_mplp_t iter); void bam_mplp_destroy(bam_mplp_t iter); void bam_mplp_set_maxcnt(bam_mplp_t iter, int maxcnt); int bam_mplp_auto(bam_mplp_t iter, int *_tid, int *_pos, int *n_plp, const bam_pileup1_t **plp); #endif // ~!defined(BAM_NO_PILEUP) #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/htslib/._synced_bcf_reader.h000644 000765 000024 00000000265 12750743246 022300 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/synced_bcf_reader.h000644 000765 000024 00000030572 12750743246 022067 0ustar00catchenstaff000000 000000 /* synced_bcf_reader.h -- stream through multiple VCF files. Copyright (C) 2012-2014 Genome Research Ltd. Author: Petr Danecek Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ /* The synced_bcf_reader allows to keep multiple VCFs open and stream them using the next_line iterator in a seamless matter without worrying about chromosomes and synchronizing the sites. This is used by vcfcheck to compare multiple VCFs simultaneously and is used also for merging, creating intersections, etc. The synced_bcf_reader also provides API for reading indexed BCF/VCF, hiding differences in BCF/VCF opening, indexing and reading. Example of usage: bcf_srs_t *sr = bcf_sr_init(); for (i=0; ihas_line[i] #define bcf_sr_get_line(_readers, i) ((_readers)->has_line[i] ? ((_readers)->readers[i].buffer[0]) : NULL) #define bcf_sr_swap_line(_readers, i, lieu) { bcf1_t *tmp = lieu; lieu = (_readers)->readers[i].buffer[0]; (_readers)->readers[i].buffer[0] = tmp; } #define bcf_sr_region_done(_readers,i) (!(_readers)->has_line[i] && !(_readers)->readers[i].nbuffer ? 1 : 0) #define bcf_sr_get_header(_readers, i) (_readers)->readers[i].header #define bcf_sr_get_reader(_readers, i) &((_readers)->readers[i]) /** * bcf_sr_seek() - set all readers to selected position * @seq: sequence name; NULL to seek to start * @pos: 0-based coordinate */ int bcf_sr_seek(bcf_srs_t *readers, const char *seq, int pos); /** * bcf_sr_set_samples() - sets active samples * @readers: holder of the open readers * @samples: this can be one of: file name with one sample per line; * or column-separated list of samples; or '-' for a list of * samples shared by all files. If first character is the * exclamation mark, all but the listed samples are included. * @is_file: 0: list of samples; 1: file with sample names * * Returns 1 if the call succeeded, or 0 on error. */ int bcf_sr_set_samples(bcf_srs_t *readers, const char *samples, int is_file); /** * bcf_sr_set_targets(), bcf_sr_set_regions() - init targets/regions * @readers: holder of the open readers * @targets: list of regions, one-based and inclusive. * @is_fname: 0: targets is a comma-separated list of regions (chr,chr:from-to) * 1: targets is a tabix indexed file with a list of regions * ( or ) * * Returns 0 if the call succeeded, or -1 on error. * * Both functions behave the same way, unlisted positions will be skipped by * bcf_sr_next_line(). However, there is an important difference: regions use * index to jump to desired positions while targets streams the whole files * and merely skip unlisted positions. * * Moreover, bcf_sr_set_targets() accepts an optional parameter $alleles which * is intepreted as a 1-based column index in the tab-delimited file where * alleles are listed. This in principle enables to perform the COLLAPSE_* * logic also with tab-delimited files. However, the current implementation * considers the alleles merely as a suggestion for prioritizing one of possibly * duplicate VCF lines. It is up to the caller to examine targets->als if * perfect match is sought after. Note that the duplicate positions in targets * file are currently not supported. * Targets (but not regions) can be prefixed with "^" to request logical complement, * for example "^X,Y,MT" indicates that sequences X, Y and MT should be skipped. */ int bcf_sr_set_targets(bcf_srs_t *readers, const char *targets, int is_file, int alleles); int bcf_sr_set_regions(bcf_srs_t *readers, const char *regions, int is_file); /* * bcf_sr_regions_init() * @regions: regions can be either a comma-separated list of regions * (chr|chr:pos|chr:from-to|chr:from-) or VCF, BED, or * tab-delimited file (the default). Uncompressed files * are stored in memory while bgzip-compressed and tabix-indexed * region files are streamed. * @is_file: 0: regions is a comma-separated list of regions * (chr|chr:pos|chr:from-to|chr:from-) * 1: VCF, BED or tab-delimited file * @chr, from, to: * Column indexes of chromosome, start position and end position * in the tab-delimited file. The positions are 1-based and * inclusive. * These parameters are ignored when reading from VCF, BED or * tabix-indexed files. When end position column is not present, * supply 'from' in place of 'to'. When 'to' is negative, first * abs(to) will be attempted and if that fails, 'from' will be used * instead. */ bcf_sr_regions_t *bcf_sr_regions_init(const char *regions, int is_file, int chr, int from, int to); void bcf_sr_regions_destroy(bcf_sr_regions_t *regions); /* * bcf_sr_regions_seek() - seek to the chromosome block * * Returns 0 on success or -1 on failure. Sets reg->seq appropriately and * reg->start,reg->end to -1. */ int bcf_sr_regions_seek(bcf_sr_regions_t *regions, const char *chr); /* * bcf_sr_regions_next() - retrieves next region. Returns 0 on success and -1 * when all regions have been read. The fields reg->seq, reg->start and * reg->end are filled with the genomic coordinates on succes or with * NULL,-1,-1 when no region is available. The coordinates are 0-based, * inclusive. */ int bcf_sr_regions_next(bcf_sr_regions_t *reg); /* * bcf_sr_regions_overlap() - checks if the interval overlaps any of * the regions, the coordinates are 0-based, inclusive. The coordinate queries * must come in ascending order. * * Returns 0 if the position is in regions; -1 if the position is not in the * regions and more regions exist; -2 if not in the regions and there are no more * regions left. */ int bcf_sr_regions_overlap(bcf_sr_regions_t *reg, const char *seq, int start, int end); /* * bcf_sr_regions_flush() - calls repeatedly regs->missed_reg_handler() until * all remaining records are processed. */ void bcf_sr_regions_flush(bcf_sr_regions_t *regs); #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/htslib/._tbx.h000644 000765 000024 00000000265 12750743246 017454 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/tbx.h000644 000765 000024 00000005536 12750743246 017245 0ustar00catchenstaff000000 000000 /* tbx.h -- tabix API functions. Copyright (C) 2009, 2012-2015 Genome Research Ltd. Copyright (C) 2010, 2012 Broad Institute. Author: Heng Li Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #ifndef HTSLIB_TBX_H #define HTSLIB_TBX_H #include "hts.h" #ifdef __cplusplus extern "C" { #endif #define TBX_MAX_SHIFT 31 #define TBX_GENERIC 0 #define TBX_SAM 1 #define TBX_VCF 2 #define TBX_UCSC 0x10000 typedef struct { int32_t preset; int32_t sc, bc, ec; // seq col., beg col. and end col. int32_t meta_char, line_skip; } tbx_conf_t; typedef struct { tbx_conf_t conf; hts_idx_t *idx; void *dict; } tbx_t; extern tbx_conf_t tbx_conf_gff, tbx_conf_bed, tbx_conf_psltbl, tbx_conf_sam, tbx_conf_vcf; #define tbx_itr_destroy(iter) hts_itr_destroy(iter) #define tbx_itr_queryi(tbx, tid, beg, end) hts_itr_query((tbx)->idx, (tid), (beg), (end), tbx_readrec) #define tbx_itr_querys(tbx, s) hts_itr_querys((tbx)->idx, (s), (hts_name2id_f)(tbx_name2id), (tbx), hts_itr_query, tbx_readrec) #define tbx_itr_next(htsfp, tbx, itr, r) hts_itr_next(hts_get_bgzfp(htsfp), (itr), (r), (tbx)) #define tbx_bgzf_itr_next(bgzfp, tbx, itr, r) hts_itr_next((bgzfp), (itr), (r), (tbx)) int tbx_name2id(tbx_t *tbx, const char *ss); /* Internal helper function used by tbx_itr_next() */ BGZF *hts_get_bgzfp(htsFile *fp); int tbx_readrec(BGZF *fp, void *tbxv, void *sv, int *tid, int *beg, int *end); int tbx_index_build(const char *fn, int min_shift, const tbx_conf_t *conf); int tbx_index_build2(const char *fn, const char *fnidx, int min_shift, const tbx_conf_t *conf); tbx_t *tbx_index_load(const char *fn); tbx_t *tbx_index_load2(const char *fn, const char *fnidx); const char **tbx_seqnames(tbx_t *tbx, int *n); // free the array but not the values void tbx_destroy(tbx_t *tbx); #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/htslib/._vcf.h000644 000765 000024 00000000265 12750743246 017435 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/vcf.h000644 000765 000024 00000115714 12750743246 017226 0ustar00catchenstaff000000 000000 /* vcf.h -- VCF/BCF API functions. Copyright (C) 2012, 2013 Broad Institute. Copyright (C) 2012-2014 Genome Research Ltd. Author: Heng Li Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ /* todo: - make the function names consistent - provide calls to abstract away structs as much as possible */ #ifndef HTSLIB_VCF_H #define HTSLIB_VCF_H #include #include #include #include "hts.h" #include "kstring.h" #include "hts_defs.h" #ifdef __cplusplus extern "C" { #endif /***************** * Header struct * *****************/ #define BCF_HL_FLT 0 // header line #define BCF_HL_INFO 1 #define BCF_HL_FMT 2 #define BCF_HL_CTG 3 #define BCF_HL_STR 4 // structured header line TAG= #define BCF_HL_GEN 5 // generic header line #define BCF_HT_FLAG 0 // header type #define BCF_HT_INT 1 #define BCF_HT_REAL 2 #define BCF_HT_STR 3 #define BCF_VL_FIXED 0 // variable length #define BCF_VL_VAR 1 #define BCF_VL_A 2 #define BCF_VL_G 3 #define BCF_VL_R 4 /* === Dictionary === The header keeps three dictonaries. The first keeps IDs in the "FILTER/INFO/FORMAT" lines, the second keeps the sequence names and lengths in the "contig" lines and the last keeps the sample names. bcf_hdr_t::dict[] is the actual hash table, which is opaque to the end users. In the hash table, the key is the ID or sample name as a C string and the value is a bcf_idinfo_t struct. bcf_hdr_t::id[] points to key-value pairs in the hash table in the order that they appear in the VCF header. bcf_hdr_t::n[] is the size of the hash table or, equivalently, the length of the id[] arrays. */ #define BCF_DT_ID 0 // dictionary type #define BCF_DT_CTG 1 #define BCF_DT_SAMPLE 2 // Complete textual representation of a header line typedef struct { int type; // One of the BCF_HL_* type char *key; // The part before '=', i.e. FILTER/INFO/FORMAT/contig/fileformat etc. char *value; // Set only for generic lines, NULL for FILTER/INFO, etc. int nkeys; // Number of structured fields char **keys, **vals; // The key=value pairs } bcf_hrec_t; typedef struct { uint32_t info[3]; // stores Number:20, var:4, Type:4, ColType:4 in info[0..2] // for BCF_HL_FLT,INFO,FMT and contig length in info[0] for BCF_HL_CTG bcf_hrec_t *hrec[3]; int id; } bcf_idinfo_t; typedef struct { const char *key; const bcf_idinfo_t *val; } bcf_idpair_t; // Note that bcf_hdr_t structs must always be created via bcf_hdr_init() typedef struct { int32_t n[3]; // n:the size of the dictionary block in use, (allocated size, m, is below to preserve ABI) bcf_idpair_t *id[3]; void *dict[3]; // ID dictionary, contig dict and sample dict char **samples; bcf_hrec_t **hrec; int nhrec, dirty; int ntransl, *transl[2]; // for bcf_translate() int nsamples_ori; // for bcf_hdr_set_samples() uint8_t *keep_samples; kstring_t mem; int32_t m[3]; // m: allocated size of the dictionary block in use (see n above) } bcf_hdr_t; extern uint8_t bcf_type_shift[]; /************** * VCF record * **************/ #define BCF_BT_NULL 0 #define BCF_BT_INT8 1 #define BCF_BT_INT16 2 #define BCF_BT_INT32 3 #define BCF_BT_FLOAT 5 #define BCF_BT_CHAR 7 #define VCF_REF 0 #define VCF_SNP 1 #define VCF_MNP 2 #define VCF_INDEL 4 #define VCF_OTHER 8 typedef struct { int type, n; // variant type and the number of bases affected, negative for deletions } variant_t; typedef struct { int id; // id: numeric tag id, the corresponding string is bcf_hdr_t::id[BCF_DT_ID][$id].key int n, size, type; // n: number of values per-sample; size: number of bytes per-sample; type: one of BCF_BT_* types uint8_t *p; // same as vptr and vptr_* in bcf_info_t below uint32_t p_len; uint32_t p_off:31, p_free:1; } bcf_fmt_t; typedef struct { int key; // key: numeric tag id, the corresponding string is bcf_hdr_t::id[BCF_DT_ID][$key].key int type, len; // type: one of BCF_BT_* types; len: vector length, 1 for scalars union { int32_t i; // integer value float f; // float value } v1; // only set if $len==1; for easier access uint8_t *vptr; // pointer to data array in bcf1_t->shared.s, excluding the size+type and tag id bytes uint32_t vptr_len; // length of the vptr block or, when set, of the vptr_mod block, excluding offset uint32_t vptr_off:31, // vptr offset, i.e., the size of the INFO key plus size+type bytes vptr_free:1; // indicates that vptr-vptr_off must be freed; set only when modified and the new // data block is bigger than the original } bcf_info_t; #define BCF1_DIRTY_ID 1 #define BCF1_DIRTY_ALS 2 #define BCF1_DIRTY_FLT 4 #define BCF1_DIRTY_INF 8 typedef struct { int m_fmt, m_info, m_id, m_als, m_allele, m_flt; // allocated size (high-water mark); do not change int n_flt; // Number of FILTER fields int *flt; // FILTER keys in the dictionary char *id, *als; // ID and REF+ALT block (\0-seperated) char **allele; // allele[0] is the REF (allele[] pointers to the als block); all null terminated bcf_info_t *info; // INFO bcf_fmt_t *fmt; // FORMAT and individual sample variant_t *var; // $var and $var_type set only when set_variant_types called int n_var, var_type; int shared_dirty; // if set, shared.s must be recreated on BCF output int indiv_dirty; // if set, indiv.s must be recreated on BCF output } bcf_dec_t; #define BCF_ERR_CTG_UNDEF 1 #define BCF_ERR_TAG_UNDEF 2 #define BCF_ERR_NCOLS 4 #define BCF_ERR_LIMITS 8 #define BCF_ERR_CHAR 16 #define BCF_ERR_CTG_INVALID 32 #define BCF_ERR_TAG_INVALID 64 /* The bcf1_t structure corresponds to one VCF/BCF line. Reading from VCF file is slower because the string is first to be parsed, packed into BCF line (done in vcf_parse), then unpacked into internal bcf1_t structure. If it is known in advance that some of the fields will not be required (notably the sample columns), parsing of these can be skipped by setting max_unpack appropriately. Similarly, it is fast to output a BCF line because the columns (kept in shared.s, indiv.s, etc.) are written directly by bcf_write, whereas a VCF line must be formatted in vcf_format. */ typedef struct { int32_t rid; // CHROM int32_t pos; // POS int32_t rlen; // length of REF float qual; // QUAL uint32_t n_info:16, n_allele:16; uint32_t n_fmt:8, n_sample:24; kstring_t shared, indiv; bcf_dec_t d; // lazy evaluation: $d is not generated by bcf_read(), but by explicitly calling bcf_unpack() int max_unpack; // Set to BCF_UN_STR, BCF_UN_FLT, or BCF_UN_INFO to boost performance of vcf_parse when some of the fields won't be needed int unpacked; // remember what has been unpacked to allow calling bcf_unpack() repeatedly without redoing the work int unpack_size[3]; // the original block size of ID, REF+ALT and FILTER int errcode; // one of BCF_ERR_* codes } bcf1_t; /******* * API * *******/ /*********************************************************************** * BCF and VCF I/O * * A note about naming conventions: htslib internally represents VCF * records as bcf1_t data structures, therefore most functions are * prefixed with bcf_. There are a few exceptions where the functions must * be aware of both BCF and VCF worlds, such as bcf_parse vs vcf_parse. In * these cases, functions prefixed with bcf_ are more general and work * with both BCF and VCF. * ***********************************************************************/ /** These macros are defined only for consistency with other parts of htslib */ #define bcf_init1() bcf_init() #define bcf_read1(fp,h,v) bcf_read((fp),(h),(v)) #define vcf_read1(fp,h,v) vcf_read((fp),(h),(v)) #define bcf_write1(fp,h,v) bcf_write((fp),(h),(v)) #define vcf_write1(fp,h,v) vcf_write((fp),(h),(v)) #define bcf_destroy1(v) bcf_destroy(v) #define bcf_empty1(v) bcf_empty(v) #define vcf_parse1(s,h,v) vcf_parse((s),(h),(v)) #define bcf_clear1(v) bcf_clear(v) #define vcf_format1(h,v,s) vcf_format((h),(v),(s)) /** * bcf_hdr_init() - create an empty BCF header. * @param mode "r" or "w" * * When opened for writing, the mandatory fileFormat and * FILTER=PASS lines are added automatically. */ bcf_hdr_t *bcf_hdr_init(const char *mode); /** Destroy a BCF header struct */ void bcf_hdr_destroy(bcf_hdr_t *h); /** Initialize a bcf1_t object; equivalent to calloc(1, sizeof(bcf1_t)) */ bcf1_t *bcf_init(void); /** Deallocate a bcf1_t object */ void bcf_destroy(bcf1_t *v); /** * Same as bcf_destroy() but frees only the memory allocated by bcf1_t, * not the bcf1_t object itself. */ void bcf_empty(bcf1_t *v); /** * Make the bcf1_t object ready for next read. Intended mostly for * internal use, the user should rarely need to call this function * directly. */ void bcf_clear(bcf1_t *v); /** bcf_open and vcf_open mode: please see hts_open() in hts.h */ typedef htsFile vcfFile; #define bcf_open(fn, mode) hts_open((fn), (mode)) #define vcf_open(fn, mode) hts_open((fn), (mode)) #define bcf_close(fp) hts_close(fp) #define vcf_close(fp) hts_close(fp) /** Reads VCF or BCF header */ bcf_hdr_t *bcf_hdr_read(htsFile *fp); /** * bcf_hdr_set_samples() - for more efficient VCF parsing when only one/few samples are needed * @samples: samples to include or exclude from file or as a comma-separated string. * LIST|FILE .. select samples in list/file * ^LIST|FILE .. exclude samples from list/file * - .. include all samples * NULL .. exclude all samples * @is_file: @samples is a file (1) or a comma-separated list (0) * * The bottleneck of VCF reading is parsing of genotype fields. If the * reader knows in advance that only subset of samples is needed (possibly * no samples at all), the performance of bcf_read() can be significantly * improved by calling bcf_hdr_set_samples after bcf_hdr_read(). * The function bcf_read() will subset the VCF/BCF records automatically * with the notable exception when reading records via bcf_itr_next(). * In this case, bcf_subset_format() must be called explicitly, because * bcf_readrec() does not see the header. * * Returns 0 on success, -1 on error or a positive integer if the list * contains samples not present in the VCF header. In such a case, the * return value is the index of the offending sample. */ int bcf_hdr_set_samples(bcf_hdr_t *hdr, const char *samples, int is_file); int bcf_subset_format(const bcf_hdr_t *hdr, bcf1_t *rec); /** Writes VCF or BCF header */ int bcf_hdr_write(htsFile *fp, bcf_hdr_t *h); /** * Parse VCF line contained in kstring and populate the bcf1_t struct * The line must not end with \n or \r characters. */ int vcf_parse(kstring_t *s, const bcf_hdr_t *h, bcf1_t *v); /** The opposite of vcf_parse. It should rarely be called directly, see vcf_write */ int vcf_format(const bcf_hdr_t *h, const bcf1_t *v, kstring_t *s); /** * bcf_read() - read next VCF or BCF record * * Returns -1 on critical errors, 0 otherwise. On errors which are not * critical for reading, such as missing header definitions, v->errcode is * set to one of BCF_ERR* code and must be checked before calling * vcf_write(). */ int bcf_read(htsFile *fp, const bcf_hdr_t *h, bcf1_t *v); /** * bcf_unpack() - unpack/decode a BCF record (fills the bcf1_t::d field) * * Note that bcf_unpack() must be called even when reading VCF. It is safe * to call the function repeatedly, it will not unpack the same field * twice. */ #define BCF_UN_STR 1 // up to ALT inclusive #define BCF_UN_FLT 2 // up to FILTER #define BCF_UN_INFO 4 // up to INFO #define BCF_UN_SHR (BCF_UN_STR|BCF_UN_FLT|BCF_UN_INFO) // all shared information #define BCF_UN_FMT 8 // unpack format and each sample #define BCF_UN_IND BCF_UN_FMT // a synonymo of BCF_UN_FMT #define BCF_UN_ALL (BCF_UN_SHR|BCF_UN_FMT) // everything int bcf_unpack(bcf1_t *b, int which); /* * bcf_dup() - create a copy of BCF record. * * Note that bcf_unpack() must be called on the returned copy as if it was * obtained from bcf_read(). Also note that bcf_dup() calls bcf_sync1(src) * internally to reflect any changes made by bcf_update_* functions. */ bcf1_t *bcf_dup(bcf1_t *src); bcf1_t *bcf_copy(bcf1_t *dst, bcf1_t *src); /** * bcf_write() - write one VCF or BCF record. The type is determined at the open() call. */ int bcf_write(htsFile *fp, bcf_hdr_t *h, bcf1_t *v); /** * The following functions work only with VCFs and should rarely be called * directly. Usually one wants to use their bcf_* alternatives, which work * transparently with both VCFs and BCFs. */ bcf_hdr_t *vcf_hdr_read(htsFile *fp); int vcf_hdr_write(htsFile *fp, const bcf_hdr_t *h); int vcf_read(htsFile *fp, const bcf_hdr_t *h, bcf1_t *v); int vcf_write(htsFile *fp, const bcf_hdr_t *h, bcf1_t *v); /** Helper function for the bcf_itr_next() macro; internal use, ignore it */ int bcf_readrec(BGZF *fp, void *null, void *v, int *tid, int *beg, int *end); /************************************************************************** * Header querying and manipulation routines **************************************************************************/ /** Create a new header using the supplied template */ bcf_hdr_t *bcf_hdr_dup(const bcf_hdr_t *hdr); /** * Copy header lines from src to dst if not already present in dst. See also bcf_translate(). * Returns 0 on success or sets a bit on error: * 1 .. conflicting definitions of tag length * // todo */ int bcf_hdr_combine(bcf_hdr_t *dst, const bcf_hdr_t *src) HTS_DEPRECATED("Please use bcf_hdr_merge instead"); /** * bcf_hdr_merge() - copy header lines from src to dst, see also bcf_translate() * @param dst: the destination header to be merged into, NULL on the first pass * @param src: the source header * * Notes: * - use as: * bcf_hdr_t *dst = NULL; * for (i=0; in[BCF_DT_SAMPLE] /** The following functions are for internal use and should rarely be called directly */ int bcf_hdr_parse(bcf_hdr_t *hdr, char *htxt); int bcf_hdr_sync(bcf_hdr_t *h); bcf_hrec_t *bcf_hdr_parse_line(const bcf_hdr_t *h, const char *line, int *len); void bcf_hrec_format(const bcf_hrec_t *hrec, kstring_t *str); int bcf_hdr_add_hrec(bcf_hdr_t *hdr, bcf_hrec_t *hrec); /** * bcf_hdr_get_hrec() - get header line info * @param type: one of the BCF_HL_* types: FLT,INFO,FMT,CTG,STR,GEN * @param key: the header key for generic lines (e.g. "fileformat"), any field * for structured lines, typically "ID". * @param value: the value which pairs with key. Can be be NULL for BCF_HL_GEN * @param str_class: the class of BCF_HL_STR line (e.g. "ALT" or "SAMPLE"), otherwise NULL */ bcf_hrec_t *bcf_hdr_get_hrec(const bcf_hdr_t *hdr, int type, const char *key, const char *value, const char *str_class); bcf_hrec_t *bcf_hrec_dup(bcf_hrec_t *hrec); void bcf_hrec_add_key(bcf_hrec_t *hrec, const char *str, int len); void bcf_hrec_set_val(bcf_hrec_t *hrec, int i, const char *str, int len, int is_quoted); int bcf_hrec_find_key(bcf_hrec_t *hrec, const char *key); void hrec_add_idx(bcf_hrec_t *hrec, int idx); void bcf_hrec_destroy(bcf_hrec_t *hrec); /************************************************************************** * Individual record querying and manipulation routines **************************************************************************/ /** See the description of bcf_hdr_subset() */ int bcf_subset(const bcf_hdr_t *h, bcf1_t *v, int n, int *imap); /** * bcf_translate() - translate tags ids to be consistent with different header. This function * is useful when lines from multiple VCF need to be combined. * @dst_hdr: the destination header, to be used in bcf_write(), see also bcf_hdr_combine() * @src_hdr: the source header, used in bcf_read() * @src_line: line obtained by bcf_read() */ int bcf_translate(const bcf_hdr_t *dst_hdr, bcf_hdr_t *src_hdr, bcf1_t *src_line); /** * bcf_get_variant_type[s]() - returns one of VCF_REF, VCF_SNP, etc */ int bcf_get_variant_types(bcf1_t *rec); int bcf_get_variant_type(bcf1_t *rec, int ith_allele); int bcf_is_snp(bcf1_t *v); /** * bcf_update_filter() - sets the FILTER column * @flt_ids: The filter IDs to set, numeric IDs returned by bcf_hdr_id2int(hdr, BCF_DT_ID, "PASS") * @n: Number of filters. If n==0, all filters are removed */ int bcf_update_filter(const bcf_hdr_t *hdr, bcf1_t *line, int *flt_ids, int n); /** * bcf_add_filter() - adds to the FILTER column * @flt_id: filter ID to add, numeric ID returned by bcf_hdr_id2int(hdr, BCF_DT_ID, "PASS") * * If flt_id is PASS, all existing filters are removed first. If other than PASS, existing PASS is removed. */ int bcf_add_filter(const bcf_hdr_t *hdr, bcf1_t *line, int flt_id); /** * bcf_remove_filter() - removes from the FILTER column * @flt_id: filter ID to remove, numeric ID returned by bcf_hdr_id2int(hdr, BCF_DT_ID, "PASS") * @pass: when set to 1 and no filters are present, set to PASS */ int bcf_remove_filter(const bcf_hdr_t *hdr, bcf1_t *line, int flt_id, int pass); /** * Returns 1 if present, 0 if absent, or -1 if filter does not exist. "PASS" and "." can be used interchangeably. */ int bcf_has_filter(const bcf_hdr_t *hdr, bcf1_t *line, char *filter); /** * bcf_update_alleles() and bcf_update_alleles_str() - update REF and ALLT column * @alleles: Array of alleles * @nals: Number of alleles * @alleles_string: Comma-separated alleles, starting with the REF allele */ int bcf_update_alleles(const bcf_hdr_t *hdr, bcf1_t *line, const char **alleles, int nals); int bcf_update_alleles_str(const bcf_hdr_t *hdr, bcf1_t *line, const char *alleles_string); /** * bcf_update_id() - sets new ID string * bcf_add_id() - adds to the ID string checking for duplicates */ int bcf_update_id(const bcf_hdr_t *hdr, bcf1_t *line, const char *id); int bcf_add_id(const bcf_hdr_t *hdr, bcf1_t *line, const char *id); /* * bcf_update_info_*() - functions for updating INFO fields * @hdr: the BCF header * @line: VCF line to be edited * @key: the INFO tag to be updated * @values: pointer to the array of values. Pass NULL to remove the tag. * @n: number of values in the array. When set to 0, the INFO tag is removed * * The @string in bcf_update_info_flag() is optional, @n indicates whether * the flag is set or removed. * * Returns 0 on success or negative value on error. */ #define bcf_update_info_int32(hdr,line,key,values,n) bcf_update_info((hdr),(line),(key),(values),(n),BCF_HT_INT) #define bcf_update_info_float(hdr,line,key,values,n) bcf_update_info((hdr),(line),(key),(values),(n),BCF_HT_REAL) #define bcf_update_info_flag(hdr,line,key,string,n) bcf_update_info((hdr),(line),(key),(string),(n),BCF_HT_FLAG) #define bcf_update_info_string(hdr,line,key,string) bcf_update_info((hdr),(line),(key),(string),1,BCF_HT_STR) int bcf_update_info(const bcf_hdr_t *hdr, bcf1_t *line, const char *key, const void *values, int n, int type); /* * bcf_update_format_*() - functions for updating FORMAT fields * @values: pointer to the array of values, the same number of elements * is expected for each sample. Missing values must be padded * with bcf_*_missing or bcf_*_vector_end values. * @n: number of values in the array. If n==0, existing tag is removed. * * The function bcf_update_format_string() is a higher-level (slower) variant of * bcf_update_format_char(). The former accepts array of \0-terminated strings * whereas the latter requires that the strings are collapsed into a single array * of fixed-length strings. In case of strings with variable length, shorter strings * can be \0-padded. Note that the collapsed strings passed to bcf_update_format_char() * are not \0-terminated. * * Returns 0 on success or negative value on error. */ #define bcf_update_format_int32(hdr,line,key,values,n) bcf_update_format((hdr),(line),(key),(values),(n),BCF_HT_INT) #define bcf_update_format_float(hdr,line,key,values,n) bcf_update_format((hdr),(line),(key),(values),(n),BCF_HT_REAL) #define bcf_update_format_char(hdr,line,key,values,n) bcf_update_format((hdr),(line),(key),(values),(n),BCF_HT_STR) #define bcf_update_genotypes(hdr,line,gts,n) bcf_update_format((hdr),(line),"GT",(gts),(n),BCF_HT_INT) // See bcf_gt_ macros below int bcf_update_format_string(const bcf_hdr_t *hdr, bcf1_t *line, const char *key, const char **values, int n); int bcf_update_format(const bcf_hdr_t *hdr, bcf1_t *line, const char *key, const void *values, int n, int type); // Macros for setting genotypes correctly, for use with bcf_update_genotypes only; idx corresponds // to VCF's GT (1-based index to ALT or 0 for the reference allele) and val is the opposite, obtained // from bcf_get_genotypes() below. #define bcf_gt_phased(idx) (((idx)+1)<<1|1) #define bcf_gt_unphased(idx) (((idx)+1)<<1) #define bcf_gt_missing 0 #define bcf_gt_is_missing(val) ((val)>>1 ? 0 : 1) #define bcf_gt_is_phased(idx) ((idx)&1) #define bcf_gt_allele(val) (((val)>>1)-1) /** Conversion between alleles indexes to Number=G genotype index (assuming diploid, all 0-based) */ #define bcf_alleles2gt(a,b) ((a)>(b)?((a)*((a)+1)/2+(b)):((b)*((b)+1)/2+(a))) static inline void bcf_gt2alleles(int igt, int *a, int *b) { int k = 0, dk = 1; while ( k 0 ) * for (i=0; iid[type][int_id].key) /** * bcf_hdr_name2id() - Translates sequence names (chromosomes) into numeric ID * bcf_hdr_id2name() - Translates numeric ID to sequence name */ static inline int bcf_hdr_name2id(const bcf_hdr_t *hdr, const char *id) { return bcf_hdr_id2int(hdr, BCF_DT_CTG, id); } static inline const char *bcf_hdr_id2name(const bcf_hdr_t *hdr, int rid) { return hdr->id[BCF_DT_CTG][rid].key; } static inline const char *bcf_seqname(const bcf_hdr_t *hdr, bcf1_t *rec) { return hdr->id[BCF_DT_CTG][rec->rid].key; } /** * bcf_hdr_id2*() - Macros for accessing bcf_idinfo_t * @type: one of BCF_HL_FLT, BCF_HL_INFO, BCF_HL_FMT * @int_id: return value of bcf_hdr_id2int, must be >=0 * * The returned values are: * bcf_hdr_id2length .. whether the number of values is fixed or variable, one of BCF_VL_* * bcf_hdr_id2number .. the number of values, 0xfffff for variable length fields * bcf_hdr_id2type .. the field type, one of BCF_HT_* * bcf_hdr_id2coltype .. the column type, one of BCF_HL_* * * Notes: Prior to using the macros, the presence of the info should be * tested with bcf_hdr_idinfo_exists(). */ #define bcf_hdr_id2length(hdr,type,int_id) ((hdr)->id[BCF_DT_ID][int_id].val->info[type]>>8 & 0xf) #define bcf_hdr_id2number(hdr,type,int_id) ((hdr)->id[BCF_DT_ID][int_id].val->info[type]>>12) #define bcf_hdr_id2type(hdr,type,int_id) ((hdr)->id[BCF_DT_ID][int_id].val->info[type]>>4 & 0xf) #define bcf_hdr_id2coltype(hdr,type,int_id) ((hdr)->id[BCF_DT_ID][int_id].val->info[type] & 0xf) #define bcf_hdr_idinfo_exists(hdr,type,int_id) ((int_id<0 || bcf_hdr_id2coltype(hdr,type,int_id)==0xf) ? 0 : 1) #define bcf_hdr_id2hrec(hdr,dict_type,col_type,int_id) ((hdr)->id[(dict_type)==BCF_DT_CTG?BCF_DT_CTG:BCF_DT_ID][int_id].val->hrec[(dict_type)==BCF_DT_CTG?0:(col_type)]) void bcf_fmt_array(kstring_t *s, int n, int type, void *data); uint8_t *bcf_fmt_sized_array(kstring_t *s, uint8_t *ptr); void bcf_enc_vchar(kstring_t *s, int l, const char *a); void bcf_enc_vint(kstring_t *s, int n, int32_t *a, int wsize); void bcf_enc_vfloat(kstring_t *s, int n, float *a); /************************************************************************** * BCF index * * Note that these functions work with BCFs only. See synced_bcf_reader.h * which provides (amongst other things) an API to work transparently with * both indexed BCFs and VCFs. **************************************************************************/ #define bcf_itr_destroy(iter) hts_itr_destroy(iter) #define bcf_itr_queryi(idx, tid, beg, end) hts_itr_query((idx), (tid), (beg), (end), bcf_readrec) #define bcf_itr_querys(idx, hdr, s) hts_itr_querys((idx), (s), (hts_name2id_f)(bcf_hdr_name2id), (hdr), hts_itr_query, bcf_readrec) #define bcf_itr_next(htsfp, itr, r) hts_itr_next((htsfp)->fp.bgzf, (itr), (r), 0) #define bcf_index_load(fn) hts_idx_load(fn, HTS_FMT_CSI) #define bcf_index_seqnames(idx, hdr, nptr) hts_idx_seqnames((idx),(nptr),(hts_id2name_f)(bcf_hdr_id2name),(hdr)) hts_idx_t *bcf_index_load2(const char *fn, const char *fnidx); int bcf_index_build(const char *fn, int min_shift); int bcf_index_build2(const char *fn, const char *fnidx, int min_shift); /******************* * Typed value I/O * *******************/ /* Note that in contrast with BCFv2.1 specification, HTSlib implementation allows missing values in vectors. For integer types, the values 0x80, 0x8000, 0x80000000 are interpreted as missing values and 0x81, 0x8001, 0x80000001 as end-of-vector indicators. Similarly for floats, the value of 0x7F800001 is interpreted as a missing value and 0x7F800002 as an end-of-vector indicator. Note that the end-of-vector byte is not part of the vector. This trial BCF version (v2.2) is compatible with the VCF specification and enables to handle correctly vectors with different ploidy in presence of missing values. */ #define bcf_int8_vector_end (INT8_MIN+1) #define bcf_int16_vector_end (INT16_MIN+1) #define bcf_int32_vector_end (INT32_MIN+1) #define bcf_str_vector_end 0 #define bcf_int8_missing INT8_MIN #define bcf_int16_missing INT16_MIN #define bcf_int32_missing INT32_MIN #define bcf_str_missing 0x07 extern uint32_t bcf_float_vector_end; extern uint32_t bcf_float_missing; static inline void bcf_float_set(float *ptr, uint32_t value) { union { uint32_t i; float f; } u; u.i = value; *ptr = u.f; } #define bcf_float_set_vector_end(x) bcf_float_set(&(x),bcf_float_vector_end) #define bcf_float_set_missing(x) bcf_float_set(&(x),bcf_float_missing) static inline int bcf_float_is_missing(float f) { union { uint32_t i; float f; } u; u.f = f; return u.i==bcf_float_missing ? 1 : 0; } static inline int bcf_float_is_vector_end(float f) { union { uint32_t i; float f; } u; u.f = f; return u.i==bcf_float_vector_end ? 1 : 0; } static inline void bcf_format_gt(bcf_fmt_t *fmt, int isample, kstring_t *str) { #define BRANCH(type_t, missing, vector_end) { \ type_t *ptr = (type_t*) (fmt->p + isample*fmt->size); \ int i; \ for (i=0; in && ptr[i]!=vector_end; i++) \ { \ if ( i ) kputc("/|"[ptr[i]&1], str); \ if ( !(ptr[i]>>1) ) kputc('.', str); \ else kputw((ptr[i]>>1) - 1, str); \ } \ if (i == 0) kputc('.', str); \ } switch (fmt->type) { case BCF_BT_INT8: BRANCH(int8_t, bcf_int8_missing, bcf_int8_vector_end); break; case BCF_BT_INT16: BRANCH(int16_t, bcf_int16_missing, bcf_int16_vector_end); break; case BCF_BT_INT32: BRANCH(int32_t, bcf_int32_missing, bcf_int32_vector_end); break; case BCF_BT_NULL: kputc('.', str); break; default: fprintf(stderr,"FIXME: type %d in bcf_format_gt?\n", fmt->type); abort(); break; } #undef BRANCH } static inline void bcf_enc_size(kstring_t *s, int size, int type) { if (size >= 15) { kputc(15<<4|type, s); if (size >= 128) { if (size >= 32768) { int32_t x = size; kputc(1<<4|BCF_BT_INT32, s); kputsn((char*)&x, 4, s); } else { int16_t x = size; kputc(1<<4|BCF_BT_INT16, s); kputsn((char*)&x, 2, s); } } else { kputc(1<<4|BCF_BT_INT8, s); kputc(size, s); } } else kputc(size<<4|type, s); } static inline int bcf_enc_inttype(long x) { if (x <= INT8_MAX && x > bcf_int8_missing) return BCF_BT_INT8; if (x <= INT16_MAX && x > bcf_int16_missing) return BCF_BT_INT16; return BCF_BT_INT32; } static inline void bcf_enc_int1(kstring_t *s, int32_t x) { if (x == bcf_int32_vector_end) { bcf_enc_size(s, 1, BCF_BT_INT8); kputc(bcf_int8_vector_end, s); } else if (x == bcf_int32_missing) { bcf_enc_size(s, 1, BCF_BT_INT8); kputc(bcf_int8_missing, s); } else if (x <= INT8_MAX && x > bcf_int8_missing) { bcf_enc_size(s, 1, BCF_BT_INT8); kputc(x, s); } else if (x <= INT16_MAX && x > bcf_int16_missing) { int16_t z = x; bcf_enc_size(s, 1, BCF_BT_INT16); kputsn((char*)&z, 2, s); } else { int32_t z = x; bcf_enc_size(s, 1, BCF_BT_INT32); kputsn((char*)&z, 4, s); } } static inline int32_t bcf_dec_int1(const uint8_t *p, int type, uint8_t **q) { if (type == BCF_BT_INT8) { *q = (uint8_t*)p + 1; return *(int8_t*)p; } else if (type == BCF_BT_INT16) { *q = (uint8_t*)p + 2; return *(int16_t*)p; } else { *q = (uint8_t*)p + 4; return *(int32_t*)p; } } static inline int32_t bcf_dec_typed_int1(const uint8_t *p, uint8_t **q) { return bcf_dec_int1(p + 1, *p&0xf, q); } static inline int32_t bcf_dec_size(const uint8_t *p, uint8_t **q, int *type) { *type = *p & 0xf; if (*p>>4 != 15) { *q = (uint8_t*)p + 1; return *p>>4; } else return bcf_dec_typed_int1(p + 1, q); } #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/htslib/._vcf_sweep.h000644 000765 000024 00000000265 12750743246 020640 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/vcf_sweep.h000644 000765 000024 00000003062 12750743246 020421 0ustar00catchenstaff000000 000000 /* vcf_sweep.h -- forward/reverse sweep API. Copyright (C) 2013 Genome Research Ltd. Author: Petr Danecek Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #ifndef HTSLIB_VCF_SWEEP_H #define HTSLIB_VCF_SWEEP_H #include "hts.h" #include "vcf.h" #ifdef __cplusplus extern "C" { #endif typedef struct _bcf_sweep_t bcf_sweep_t; bcf_sweep_t *bcf_sweep_init(const char *fname); void bcf_sweep_destroy(bcf_sweep_t *sw); bcf_hdr_t *bcf_sweep_hdr(bcf_sweep_t *sw); bcf1_t *bcf_sweep_fwd(bcf_sweep_t *sw); bcf1_t *bcf_sweep_bwd(bcf_sweep_t *sw); #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/htslib/._vcfutils.h000644 000765 000024 00000000265 12750743246 020516 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/htslib/vcfutils.h000644 000765 000024 00000011063 12750743246 020277 0ustar00catchenstaff000000 000000 /* vcfutils.h -- allele-related utility functions. Copyright (C) 2012, 2013, 2015 Genome Research Ltd. Author: Petr Danecek Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #ifndef HTSLIB_VCFUTILS_H #define HTSLIB_VCFUTILS_H #include "vcf.h" #ifdef __cplusplus extern "C" { #endif struct kbitset_t; /** * bcf_trim_alleles() - remove ALT alleles unused in genotype fields * @header: for access to BCF_DT_ID dictionary * @line: VCF line obtain from vcf_parse1 * * Returns the number of removed alleles on success or negative * on error: * -1 .. some allele index is out of bounds */ int bcf_trim_alleles(const bcf_hdr_t *header, bcf1_t *line); /** * bcf_remove_alleles() - remove ALT alleles according to bitmask @mask * @header: for access to BCF_DT_ID dictionary * @line: VCF line obtained from vcf_parse1 * @mask: alleles to remove * * If you have more than 31 alleles, then the integer bit mask will * overflow, so use bcf_remove_allele_set instead */ void bcf_remove_alleles(const bcf_hdr_t *header, bcf1_t *line, int mask); /** * bcf_remove_allele_set() - remove ALT alleles according to bitset @rm_set * @header: for access to BCF_DT_ID dictionary * @line: VCF line obtained from vcf_parse1 * @rm_set: pointer to kbitset_t object with bits set for allele * indexes to remove * * Number=A,R,G INFO and FORMAT fields will be updated accordingly. */ void bcf_remove_allele_set(const bcf_hdr_t *header, bcf1_t *line, const struct kbitset_t *rm_set); /** * bcf_calc_ac() - calculate the number of REF and ALT alleles * @header: for access to BCF_DT_ID dictionary * @line: VCF line obtained from vcf_parse1 * @ac: array of length line->n_allele * @which: determine if INFO/AN,AC and indv fields be used * * Returns 1 if the call succeeded, or 0 if the value could not * be determined. * * The value of @which determines if existing INFO/AC,AN can be * used (BCF_UN_INFO) and and if indv fields can be splitted * (BCF_UN_FMT). */ int bcf_calc_ac(const bcf_hdr_t *header, bcf1_t *line, int *ac, int which); /** * bcf_gt_type() - determines type of the genotype * @fmt_ptr: the GT format field as set for example by set_fmt_ptr * @isample: sample index (starting from 0) * @ial: index of the 1st non-reference allele (starting from 1) * @jal: index of the 2nd non-reference allele (starting from 1) * * Returns the type of the genotype (one of GT_HOM_RR, GT_HET_RA, * GT_HOM_AA, GT_HET_AA, GT_HAPL_R, GT_HAPL_A or GT_UNKN). If $ial * is not NULL and the genotype has one or more non-reference * alleles, $ial will be set. In case of GT_HET_AA, $ial is the * position of the allele which appeared first in ALT. If $jal is * not null and the genotype is GT_HET_AA, $jal will be set and is * the position of the second allele in ALT. */ #define GT_HOM_RR 0 // note: the actual value of GT_* matters, used in dosage r2 calculation #define GT_HOM_AA 1 #define GT_HET_RA 2 #define GT_HET_AA 3 #define GT_HAPL_R 4 #define GT_HAPL_A 5 #define GT_UNKN 6 int bcf_gt_type(bcf_fmt_t *fmt_ptr, int isample, int *ial, int *jal); static inline int bcf_acgt2int(char c) { if ( (int)c>96 ) c -= 32; if ( c=='A' ) return 0; if ( c=='C' ) return 1; if ( c=='G' ) return 2; if ( c=='T' ) return 3; return -1; } #define bcf_int2acgt(i) "ACGT"[i] /** * bcf_ij2G() - common task: allele indexes to Number=G index (diploid) * @i,j: allele indexes, 0-based, i<=j * * Returns index to the Number=G diploid array */ #define bcf_ij2G(i, j) ((j)*((j)+1)/2+(i)) #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/cram/._cram.h000644 000765 000024 00000000265 12750743246 017236 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/cram.h000644 000765 000024 00000004563 12750743246 017026 0ustar00catchenstaff000000 000000 /* Copyright (c) 2012-2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /*! \file * CRAM interface. * * Consider using the higher level hts_*() API for programs that wish to * be file format agnostic (see htslib/hts.h). * * This API should be used for CRAM specific code. The specifics of the * public API are implemented in cram_io.h, cram_encode.h and cram_decode.h * although these should not be included directly (use this file instead). */ #ifndef _CRAM_H_ #define _CRAM_H_ #include "cram/cram_samtools.h" #include "cram/sam_header.h" #include "cram_structs.h" #include "cram_io.h" #include "cram_encode.h" #include "cram_decode.h" #include "cram_stats.h" #include "cram_codecs.h" #include "cram_index.h" // Validate against the external cram.h, // // This contains duplicated portions from cram_io.h and cram_structs.h, // so we want to ensure that the prototypes match. #include "htslib/cram.h" #endif stacks-1.44/htslib/cram/._cram_codecs.c000644 000765 000024 00000000265 12750743246 020551 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/cram_codecs.c000644 000765 000024 00000136466 12750743246 020351 0ustar00catchenstaff000000 000000 /* Copyright (c) 2012-2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /* * FIXME: add checking of cram_external_type to return NULL on unsupported * {codec,type} tuples. */ #include #include #include #include #include #include "cram/cram.h" /* * --------------------------------------------------------------------------- * Block bit-level I/O functions. * All defined static here to promote easy inlining by the compiler. */ #if 0 /* Get a single bit, MSB first */ static signed int get_bit_MSB(cram_block *block) { unsigned int val; if (block->byte > block->alloc) return -1; val = block->data[block->byte] >> block->bit; if (--block->bit == -1) { block->bit = 7; block->byte++; //printf("(%02X)", block->data[block->byte]); } //printf("-B%d-", val&1); return val & 1; } #endif /* * Count number of successive 0 and 1 bits */ static int get_one_bits_MSB(cram_block *block) { int n = 0, b; if (block->byte >= block->uncomp_size) return -1; do { b = block->data[block->byte] >> block->bit; if (--block->bit == -1) { block->bit = 7; block->byte++; if (block->byte == block->uncomp_size && (b&1)) return -1; } n++; } while (b&1); return n-1; } static int get_zero_bits_MSB(cram_block *block) { int n = 0, b; if (block->byte >= block->uncomp_size) return -1; do { b = block->data[block->byte] >> block->bit; if (--block->bit == -1) { block->bit = 7; block->byte++; if (block->byte == block->uncomp_size && !(b&1)) return -1; } n++; } while (!(b&1)); return n-1; } #if 0 /* Stores a single bit */ static void store_bit_MSB(cram_block *block, unsigned int bit) { if (block->byte >= block->alloc) { block->alloc = block->alloc ? block->alloc*2 : 1024; block->data = realloc(block->data, block->alloc); } if (bit) block->data[block->byte] |= (1 << block->bit); if (--block->bit == -1) { block->bit = 7; block->byte++; block->data[block->byte] = 0; } } #endif #if 0 /* Rounds to the next whole byte boundary first */ static void store_bytes_MSB(cram_block *block, char *bytes, int len) { if (block->bit != 7) { block->bit = 7; block->byte++; } while (block->byte + len >= block->alloc) { block->alloc = block->alloc ? block->alloc*2 : 1024; block->data = realloc(block->data, block->alloc); } memcpy(&block->data[block->byte], bytes, len); block->byte += len; } #endif /* Local optimised copy for inlining */ static inline unsigned int get_bits_MSB(cram_block *block, int nbits) { unsigned int val = 0; int i; #if 0 // Fits within the current byte */ if (nbits <= block->bit+1) { val = (block->data[block->byte]>>(block->bit-(nbits-1))) & ((1<bit -= nbits) == -1) { block->bit = 7; block->byte++; } return val; } // partial first byte val = block->data[block->byte] & ((1<<(block->bit+1))-1); nbits -= block->bit+1; block->bit = 7; block->byte++; // whole middle bytes while (nbits >= 8) { val = (val << 8) | block->data[block->byte++]; nbits -= 8; } val <<= nbits; val |= (block->data[block->byte]>>(block->bit-(nbits-1))) & ((1<bit -= nbits; return val; #endif #if 0 /* Inefficient implementation! */ //printf("{"); for (i = 0; i < nbits; i++) //val = (val << 1) | get_bit_MSB(block); GET_BIT_MSB(block, val); #endif #if 1 /* Combination of 1st two methods */ if (nbits <= block->bit+1) { val = (block->data[block->byte]>>(block->bit-(nbits-1))) & ((1<bit -= nbits) == -1) { block->bit = 7; block->byte++; } return val; } switch(nbits) { // case 15: GET_BIT_MSB(block, val); // case 14: GET_BIT_MSB(block, val); // case 13: GET_BIT_MSB(block, val); // case 12: GET_BIT_MSB(block, val); // case 11: GET_BIT_MSB(block, val); // case 10: GET_BIT_MSB(block, val); // case 9: GET_BIT_MSB(block, val); case 8: GET_BIT_MSB(block, val); case 7: GET_BIT_MSB(block, val); case 6: GET_BIT_MSB(block, val); case 5: GET_BIT_MSB(block, val); case 4: GET_BIT_MSB(block, val); case 3: GET_BIT_MSB(block, val); case 2: GET_BIT_MSB(block, val); case 1: GET_BIT_MSB(block, val); break; default: for (i = 0; i < nbits; i++) //val = (val << 1) | get_bit_MSB(block); GET_BIT_MSB(block, val); } #endif //printf("=0x%x}", val); return val; } /* * Can store up to 24-bits worth of data encoded in an integer value * Possibly we'd want to have a less optimal store_bits function when dealing * with nbits > 24, but for now we assume the codes generated are never * that big. (Given this is only possible with 121392 or more * characters with exactly the correct frequency distribution we check * for it elsewhere.) */ static int store_bits_MSB(cram_block *block, unsigned int val, int nbits) { /* fprintf(stderr, " store_bits: %02x %d\n", val, nbits); */ /* * Use slow mode until we tweak the huffman generator to never generate * codes longer than 24-bits. */ unsigned int mask; if (block->byte+4 >= block->alloc) { if (block->byte) { block->alloc *= 2; block->data = realloc(block->data, block->alloc + 4); if (!block->data) return -1; } else { block->alloc = 1024; block->data = realloc(block->data, block->alloc + 4); if (!block->data) return -1; block->data[0] = 0; // initialise first byte of buffer } } /* fits in current bit-field */ if (nbits <= block->bit+1) { block->data[block->byte] |= (val << (block->bit+1-nbits)); if ((block->bit-=nbits) == -1) { block->bit = 7; block->byte++; block->data[block->byte] = 0; } return 0; } block->data[block->byte] |= (val >> (nbits -= block->bit+1)); block->bit = 7; block->byte++; block->data[block->byte] = 0; mask = 1<<(nbits-1); do { if (val & mask) block->data[block->byte] |= (1 << block->bit); if (--block->bit == -1) { block->bit = 7; block->byte++; block->data[block->byte] = 0; } mask >>= 1; } while(--nbits); return 0; } /* * Returns the next 'size' bytes from a block, or NULL if insufficient * data left.This is just a pointer into the block data and not an * allocated object, so do not free the result. */ static char *cram_extract_block(cram_block *b, int size) { char *cp = (char *)b->data + b->idx; b->idx += size; if (b->idx > b->uncomp_size) return NULL; return cp; } /* * --------------------------------------------------------------------------- * EXTERNAL */ int cram_external_decode_int(cram_slice *slice, cram_codec *c, cram_block *in, char *out, int *out_size) { int l; char *cp; cram_block *b; /* Find the external block */ b = cram_get_block_by_id(slice, c->external.content_id); if (!b) return *out_size?-1:0; cp = (char *)b->data + b->idx; // E_INT and E_LONG are guaranteed single item queries l = safe_itf8_get(cp, (char *)b->data + b->uncomp_size, (int32_t *)out); b->idx += l; *out_size = 1; return l > 0 ? 0 : -1; } int cram_external_decode_char(cram_slice *slice, cram_codec *c, cram_block *in, char *out, int *out_size) { char *cp; cram_block *b; /* Find the external block */ b = cram_get_block_by_id(slice, c->external.content_id); if (!b) return *out_size?-1:0; cp = cram_extract_block(b, *out_size); if (!cp) return -1; if (out) memcpy(out, cp, *out_size); return 0; } static int cram_external_decode_block(cram_slice *slice, cram_codec *c, cram_block *in, char *out_, int *out_size) { char *cp; cram_block *b = NULL; cram_block *out = (cram_block *)out_; /* Find the external block */ b = cram_get_block_by_id(slice, c->external.content_id); if (!b) return *out_size?-1:0; cp = cram_extract_block(b, *out_size); if (!cp) return -1; BLOCK_APPEND(out, cp, *out_size); return 0; } void cram_external_decode_free(cram_codec *c) { if (c) free(c); } cram_codec *cram_external_decode_init(char *data, int size, enum cram_external_type option, int version) { cram_codec *c; char *cp = data; if (!(c = malloc(sizeof(*c)))) return NULL; c->codec = E_EXTERNAL; if (option == E_INT || option == E_LONG) c->decode = cram_external_decode_int; else if (option == E_BYTE_ARRAY || option == E_BYTE) c->decode = cram_external_decode_char; else c->decode = cram_external_decode_block; c->free = cram_external_decode_free; cp += itf8_get(cp, &c->external.content_id); if (cp - data != size) { fprintf(stderr, "Malformed external header stream\n"); free(c); return NULL; } c->external.type = option; return c; } int cram_external_encode_int(cram_slice *slice, cram_codec *c, char *in, int in_size) { uint32_t *i32 = (uint32_t *)in; itf8_put_blk(c->out, *i32); return 0; } int cram_external_encode_char(cram_slice *slice, cram_codec *c, char *in, int in_size) { BLOCK_APPEND(c->out, in, in_size); return 0; } void cram_external_encode_free(cram_codec *c) { if (!c) return; free(c); } int cram_external_encode_store(cram_codec *c, cram_block *b, char *prefix, int version) { char tmp[99], *tp = tmp; int len = 0; if (prefix) { size_t l = strlen(prefix); BLOCK_APPEND(b, prefix, l); len += l; } tp += itf8_put(tp, c->e_external.content_id); len += itf8_put_blk(b, c->codec); len += itf8_put_blk(b, tp-tmp); BLOCK_APPEND(b, tmp, tp-tmp); len += tp-tmp; return len; } cram_codec *cram_external_encode_init(cram_stats *st, enum cram_external_type option, void *dat, int version) { cram_codec *c; c = malloc(sizeof(*c)); if (!c) return NULL; c->codec = E_EXTERNAL; c->free = cram_external_encode_free; if (option == E_INT || option == E_LONG) c->encode = cram_external_encode_int; else if (option == E_BYTE_ARRAY || option == E_BYTE) c->encode = cram_external_encode_char; else abort(); c->store = cram_external_encode_store; c->e_external.content_id = (size_t)dat; return c; } /* * --------------------------------------------------------------------------- * BETA */ int cram_beta_decode_int(cram_slice *slice, cram_codec *c, cram_block *in, char *out, int *out_size) { int32_t *out_i = (int32_t *)out; int i, n; if (c->beta.nbits) { if (cram_not_enough_bits(in, c->beta.nbits)) return -1; for (i = 0, n = *out_size; i < n; i++) out_i[i] = get_bits_MSB(in, c->beta.nbits) - c->beta.offset; } else { for (i = 0, n = *out_size; i < n; i++) out_i[i] = -c->beta.offset; } return 0; } int cram_beta_decode_char(cram_slice *slice, cram_codec *c, cram_block *in, char *out, int *out_size) { int i, n; if (c->beta.nbits) { if (cram_not_enough_bits(in, c->beta.nbits)) return -1; if (out) for (i = 0, n = *out_size; i < n; i++) out[i] = get_bits_MSB(in, c->beta.nbits) - c->beta.offset; else for (i = 0, n = *out_size; i < n; i++) get_bits_MSB(in, c->beta.nbits); } else { if (out) for (i = 0, n = *out_size; i < n; i++) out[i] = -c->beta.offset; } return 0; } void cram_beta_decode_free(cram_codec *c) { if (c) free(c); } cram_codec *cram_beta_decode_init(char *data, int size, enum cram_external_type option, int version) { cram_codec *c; char *cp = data; if (!(c = malloc(sizeof(*c)))) return NULL; c->codec = E_BETA; if (option == E_INT || option == E_LONG) c->decode = cram_beta_decode_int; else if (option == E_BYTE_ARRAY || option == E_BYTE) c->decode = cram_beta_decode_char; else abort(); c->free = cram_beta_decode_free; cp += itf8_get(cp, &c->beta.offset); cp += itf8_get(cp, &c->beta.nbits); if (cp - data != size || c->beta.nbits < 0 || c->beta.nbits > 8 * sizeof(int)) { fprintf(stderr, "Malformed beta header stream\n"); free(c); return NULL; } return c; } int cram_beta_encode_store(cram_codec *c, cram_block *b, char *prefix, int version) { int len = 0; if (prefix) { size_t l = strlen(prefix); BLOCK_APPEND(b, prefix, l); len += l; } len += itf8_put_blk(b, c->codec); len += itf8_put_blk(b, itf8_size(c->e_beta.offset) + itf8_size(c->e_beta.nbits)); // codec length len += itf8_put_blk(b, c->e_beta.offset); len += itf8_put_blk(b, c->e_beta.nbits); return len; } int cram_beta_encode_int(cram_slice *slice, cram_codec *c, char *in, int in_size) { int *syms = (int *)in; int i, r = 0; for (i = 0; i < in_size; i++) r |= store_bits_MSB(c->out, syms[i] + c->e_beta.offset, c->e_beta.nbits); return r; } int cram_beta_encode_char(cram_slice *slice, cram_codec *c, char *in, int in_size) { unsigned char *syms = (unsigned char *)in; int i, r = 0; for (i = 0; i < in_size; i++) r |= store_bits_MSB(c->out, syms[i] + c->e_beta.offset, c->e_beta.nbits); return r; } void cram_beta_encode_free(cram_codec *c) { if (c) free(c); } cram_codec *cram_beta_encode_init(cram_stats *st, enum cram_external_type option, void *dat, int version) { cram_codec *c; int min_val, max_val, len = 0; c = malloc(sizeof(*c)); if (!c) return NULL; c->codec = E_BETA; c->free = cram_beta_encode_free; if (option == E_INT) c->encode = cram_beta_encode_int; else c->encode = cram_beta_encode_char; c->store = cram_beta_encode_store; if (dat) { min_val = ((int *)dat)[0]; max_val = ((int *)dat)[1]; } else { min_val = INT_MAX; max_val = INT_MIN; int i; for (i = 0; i < MAX_STAT_VAL; i++) { if (!st->freqs[i]) continue; if (min_val > i) min_val = i; max_val = i; } if (st->h) { khint_t k; for (k = kh_begin(st->h); k != kh_end(st->h); k++) { if (!kh_exist(st->h, k)) continue; i = kh_key(st->h, k); if (min_val > i) min_val = i; if (max_val < i) max_val = i; } } } assert(max_val >= min_val); c->e_beta.offset = -min_val; max_val -= min_val; while (max_val) { len++; max_val >>= 1; } c->e_beta.nbits = len; return c; } /* * --------------------------------------------------------------------------- * SUBEXP */ int cram_subexp_decode(cram_slice *slice, cram_codec *c, cram_block *in, char *out, int *out_size) { int32_t *out_i = (int32_t *)out; int n, count; int k = c->subexp.k; for (count = 0, n = *out_size; count < n; count++) { int i = 0, tail; int val; /* Get number of 1s */ //while (get_bit_MSB(in) == 1) i++; i = get_one_bits_MSB(in); if (i < 0 || cram_not_enough_bits(in, i > 0 ? i + k - 1 : k)) return -1; /* * Val is * i > 0: 2^(k+i-1) + k+i-1 bits * i = 0: k bits */ if (i) { tail = i + k-1; val = 0; while (tail) { //val = val<<1; val |= get_bit_MSB(in); GET_BIT_MSB(in, val); tail--; } val += 1 << (i + k-1); } else { tail = k; val = 0; while (tail) { //val = val<<1; val |= get_bit_MSB(in); GET_BIT_MSB(in, val); tail--; } } out_i[count] = val - c->subexp.offset; } return 0; } void cram_subexp_decode_free(cram_codec *c) { if (c) free(c); } cram_codec *cram_subexp_decode_init(char *data, int size, enum cram_external_type option, int version) { cram_codec *c; char *cp = data; if (!(c = malloc(sizeof(*c)))) return NULL; c->codec = E_SUBEXP; c->decode = cram_subexp_decode; c->free = cram_subexp_decode_free; c->subexp.k = -1; cp += safe_itf8_get(cp, data + size, &c->subexp.offset); cp += safe_itf8_get(cp, data + size, &c->subexp.k); if (cp - data != size || c->subexp.k < 0) { fprintf(stderr, "Malformed subexp header stream\n"); free(c); return NULL; } return c; } /* * --------------------------------------------------------------------------- * GAMMA */ int cram_gamma_decode(cram_slice *slice, cram_codec *c, cram_block *in, char *out, int *out_size) { int32_t *out_i = (int32_t *)out; int i, n; for (i = 0, n = *out_size; i < n; i++) { int nz = 0; int val; //while (get_bit_MSB(in) == 0) nz++; nz = get_zero_bits_MSB(in); if (cram_not_enough_bits(in, nz)) return -1; val = 1; while (nz > 0) { //val <<= 1; val |= get_bit_MSB(in); GET_BIT_MSB(in, val); nz--; } out_i[i] = val - c->gamma.offset; } return 0; } void cram_gamma_decode_free(cram_codec *c) { if (c) free(c); } cram_codec *cram_gamma_decode_init(char *data, int size, enum cram_external_type option, int version) { cram_codec *c; char *cp = data; if (!(c = malloc(sizeof(*c)))) return NULL; c->codec = E_GAMMA; c->decode = cram_gamma_decode; c->free = cram_gamma_decode_free; cp += itf8_get(cp, &c->gamma.offset); if (cp - data != size) { fprintf(stderr, "Malformed gamma header stream\n"); free(c); return NULL; } return c; } /* * --------------------------------------------------------------------------- * HUFFMAN */ static int code_sort(const void *vp1, const void *vp2) { const cram_huffman_code *c1 = (const cram_huffman_code *)vp1; const cram_huffman_code *c2 = (const cram_huffman_code *)vp2; if (c1->len != c2->len) return c1->len - c2->len; else return c1->symbol - c2->symbol; } void cram_huffman_decode_free(cram_codec *c) { if (!c) return; if (c->huffman.codes) free(c->huffman.codes); free(c); } int cram_huffman_decode_null(cram_slice *slice, cram_codec *c, cram_block *in, char *out, int *out_size) { return -1; } int cram_huffman_decode_char0(cram_slice *slice, cram_codec *c, cram_block *in, char *out, int *out_size) { int i, n; if (!out) return 0; /* Special case of 0 length codes */ for (i = 0, n = *out_size; i < n; i++) { out[i] = c->huffman.codes[0].symbol; } return 0; } int cram_huffman_decode_char(cram_slice *slice, cram_codec *c, cram_block *in, char *out, int *out_size) { int i, n, ncodes = c->huffman.ncodes; const cram_huffman_code * const codes = c->huffman.codes; for (i = 0, n = *out_size; i < n; i++) { int idx = 0; int val = 0, len = 0, last_len = 0; for (;;) { int dlen = codes[idx].len - last_len; if (cram_not_enough_bits(in, dlen)) return -1; //val <<= dlen; //val |= get_bits_MSB(in, dlen); //last_len = (len += dlen); last_len = (len += dlen); for (; dlen; dlen--) GET_BIT_MSB(in, val); idx = val - codes[idx].p; if (idx >= ncodes || idx < 0) return -1; if (codes[idx].code == val && codes[idx].len == len) { out[i] = codes[idx].symbol; break; } } } return 0; } int cram_huffman_decode_int0(cram_slice *slice, cram_codec *c, cram_block *in, char *out, int *out_size) { int32_t *out_i = (int32_t *)out; int i, n; const cram_huffman_code * const codes = c->huffman.codes; /* Special case of 0 length codes */ for (i = 0, n = *out_size; i < n; i++) { out_i[i] = codes[0].symbol; } return 0; } int cram_huffman_decode_int(cram_slice *slice, cram_codec *c, cram_block *in, char *out, int *out_size) { int32_t *out_i = (int32_t *)out; int i, n, ncodes = c->huffman.ncodes; const cram_huffman_code * const codes = c->huffman.codes; for (i = 0, n = *out_size; i < n; i++) { int idx = 0; int val = 0, len = 0, last_len = 0; // Now one bit at a time for remaining checks for (;;) { int dlen = codes[idx].len - last_len; if (cram_not_enough_bits(in, dlen)) return -1; //val <<= dlen; //val |= get_bits_MSB(in, dlen); //last_len = (len += dlen); last_len = (len += dlen); for (; dlen; dlen--) GET_BIT_MSB(in, val); idx = val - codes[idx].p; if (idx >= ncodes || idx < 0) return -1; if (codes[idx].code == val && codes[idx].len == len) { out_i[i] = codes[idx].symbol; break; } } } return 0; } /* * Initialises a huffman decoder from an encoding data stream. */ cram_codec *cram_huffman_decode_init(char *data, int size, enum cram_external_type option, int version) { int32_t ncodes = 0, i, j; char *cp = data, *data_end = &data[size]; cram_codec *h; cram_huffman_code *codes; int32_t val, last_len, max_len = 0; int l; cp += safe_itf8_get(cp, data_end, &ncodes); h = calloc(1, sizeof(*h)); if (!h) return NULL; h->codec = E_HUFFMAN; h->free = cram_huffman_decode_free; h->huffman.ncodes = ncodes; codes = h->huffman.codes = malloc(ncodes * sizeof(*codes)); if (!codes) { free(h); return NULL; } /* Read symbols and bit-lengths */ for (i = 0, l = 1; i < ncodes && l > 0; i++, cp += l) { l = safe_itf8_get(cp, data_end, &codes[i].symbol); } if (l < 1) { fprintf(stderr, "Malformed huffman header stream\n"); free(h); return NULL; } cp += safe_itf8_get(cp, data_end, &i); if (i != ncodes) { fprintf(stderr, "Malformed huffman header stream\n"); free(h); return NULL; } if (ncodes == 0) { /* NULL huffman stream. Ensure it returns an error if anything tries to use it. */ h->decode = cram_huffman_decode_null; return h; } for (i = 0, l = 1; i < ncodes; i++, cp += l) { l = safe_itf8_get(cp, data_end, &codes[i].len); if (l < 1) break; if (max_len < codes[i].len) max_len = codes[i].len; } if (l < 1 || cp - data != size || max_len >= ncodes) { fprintf(stderr, "Malformed huffman header stream\n"); free(h); return NULL; } /* Sort by bit length and then by symbol value */ qsort(codes, ncodes, sizeof(*codes), code_sort); /* Assign canonical codes */ val = -1, last_len = 0; for (i = 0; i < ncodes; i++) { val++; if (codes[i].len > last_len) { while (codes[i].len > last_len) { val <<= 1; last_len++; } } codes[i].code = val; } /* * Compute the next starting point, offset by the i'th value. * For example if codes 10, 11, 12, 13 are 30, 31, 32, 33 then * codes[10..13].p = 30 - 10. */ last_len = 0; for (i = j = 0; i < ncodes; i++) { if (codes[i].len > last_len) { j = codes[i].code - i; last_len = codes[i].len; } codes[i].p = j; } // puts("==HUFF LEN=="); // for (i = 0; i <= last_len+1; i++) { // printf("len %d=%d prefix %d\n", i, h->huffman.lengths[i], h->huffman.prefix[i]); // } // puts("===HUFFMAN CODES==="); // for (i = 0; i < ncodes; i++) { // int j; // printf("%d: %d %d %d ", i, codes[i].symbol, codes[i].len, codes[i].code); // j = codes[i].len; // while (j) { // putchar(codes[i].code & (1 << --j) ? '1' : '0'); // } // printf(" %d\n", codes[i].code); // } if (option == E_BYTE || option == E_BYTE_ARRAY) { if (h->huffman.codes[0].len == 0) h->decode = cram_huffman_decode_char0; else h->decode = cram_huffman_decode_char; } else if (option == E_BYTE_ARRAY_BLOCK) { abort(); } else { if (h->huffman.codes[0].len == 0) h->decode = cram_huffman_decode_int0; else h->decode = cram_huffman_decode_int; } return (cram_codec *)h; } int cram_huffman_encode_char0(cram_slice *slice, cram_codec *c, char *in, int in_size) { return 0; } int cram_huffman_encode_char(cram_slice *slice, cram_codec *c, char *in, int in_size) { int i, code, len, r = 0; unsigned char *syms = (unsigned char *)in; while (in_size--) { int sym = *syms++; if (sym >= -1 && sym < MAX_HUFF) { i = c->e_huffman.val2code[sym+1]; assert(c->e_huffman.codes[i].symbol == sym); code = c->e_huffman.codes[i].code; len = c->e_huffman.codes[i].len; } else { /* Slow - use a lookup table for when sym < MAX_HUFF? */ for (i = 0; i < c->e_huffman.nvals; i++) { if (c->e_huffman.codes[i].symbol == sym) break; } if (i == c->e_huffman.nvals) return -1; code = c->e_huffman.codes[i].code; len = c->e_huffman.codes[i].len; } r |= store_bits_MSB(c->out, code, len); } return r; } int cram_huffman_encode_int0(cram_slice *slice, cram_codec *c, char *in, int in_size) { return 0; } int cram_huffman_encode_int(cram_slice *slice, cram_codec *c, char *in, int in_size) { int i, code, len, r = 0; int *syms = (int *)in; while (in_size--) { int sym = *syms++; if (sym >= -1 && sym < MAX_HUFF) { i = c->e_huffman.val2code[sym+1]; assert(c->e_huffman.codes[i].symbol == sym); code = c->e_huffman.codes[i].code; len = c->e_huffman.codes[i].len; } else { /* Slow - use a lookup table for when sym < MAX_HUFFMAN_SYM? */ for (i = 0; i < c->e_huffman.nvals; i++) { if (c->e_huffman.codes[i].symbol == sym) break; } if (i == c->e_huffman.nvals) return -1; code = c->e_huffman.codes[i].code; len = c->e_huffman.codes[i].len; } r |= store_bits_MSB(c->out, code, len); } return r; } void cram_huffman_encode_free(cram_codec *c) { if (!c) return; if (c->e_huffman.codes) free(c->e_huffman.codes); free(c); } /* * Encodes a huffman tree. * Returns number of bytes written. */ int cram_huffman_encode_store(cram_codec *c, cram_block *b, char *prefix, int version) { int i, len = 0; cram_huffman_code *codes = c->e_huffman.codes; /* * Up to code length 127 means 2.5e+26 bytes of data required (worst * case huffman tree needs symbols with freqs matching the Fibonacci * series). So guaranteed 1 byte per code. * * Symbols themselves could be 5 bytes (eg -1 is 5 bytes in itf8). * * Therefore 6*ncodes + 5 + 5 + 1 + 5 is max memory */ char *tmp = malloc(6*c->e_huffman.nvals+16); char *tp = tmp; if (!tmp) return -1; if (prefix) { size_t l = strlen(prefix); BLOCK_APPEND(b, prefix, l); len += l; } tp += itf8_put(tp, c->e_huffman.nvals); for (i = 0; i < c->e_huffman.nvals; i++) { tp += itf8_put(tp, codes[i].symbol); } tp += itf8_put(tp, c->e_huffman.nvals); for (i = 0; i < c->e_huffman.nvals; i++) { tp += itf8_put(tp, codes[i].len); } len += itf8_put_blk(b, c->codec); len += itf8_put_blk(b, tp-tmp); BLOCK_APPEND(b, tmp, tp-tmp); len += tp-tmp; free(tmp); return len; } cram_codec *cram_huffman_encode_init(cram_stats *st, enum cram_external_type option, void *dat, int version) { int *vals = NULL, *freqs = NULL, vals_alloc = 0, *lens, code, len; int nvals, i, ntot = 0, max_val = 0, min_val = INT_MAX, k; cram_codec *c; cram_huffman_code *codes; c = malloc(sizeof(*c)); if (!c) return NULL; c->codec = E_HUFFMAN; /* Count number of unique symbols */ for (nvals = i = 0; i < MAX_STAT_VAL; i++) { if (!st->freqs[i]) continue; if (nvals >= vals_alloc) { vals_alloc = vals_alloc ? vals_alloc*2 : 1024; vals = realloc(vals, vals_alloc * sizeof(int)); freqs = realloc(freqs, vals_alloc * sizeof(int)); if (!vals || !freqs) { if (vals) free(vals); if (freqs) free(freqs); free(c); return NULL; } } vals[nvals] = i; freqs[nvals] = st->freqs[i]; assert(st->freqs[i] > 0); ntot += freqs[nvals]; if (max_val < i) max_val = i; if (min_val > i) min_val = i; nvals++; } if (st->h) { khint_t k; for (k = kh_begin(st->h); k != kh_end(st->h); k++) { if (!kh_exist(st->h, k)) continue; if (nvals >= vals_alloc) { vals_alloc = vals_alloc ? vals_alloc*2 : 1024; vals = realloc(vals, vals_alloc * sizeof(int)); freqs = realloc(freqs, vals_alloc * sizeof(int)); if (!vals || !freqs) return NULL; } vals[nvals]= kh_key(st->h, k); freqs[nvals] = kh_val(st->h, k); assert(freqs[nvals] > 0); ntot += freqs[nvals]; if (max_val < i) max_val = i; if (min_val > i) min_val = i; nvals++; } } assert(nvals > 0); freqs = realloc(freqs, 2*nvals*sizeof(*freqs)); lens = calloc(2*nvals, sizeof(*lens)); if (!lens || !freqs) return NULL; /* Inefficient, use pointers to form chain so we can insert and maintain * a sorted list? This is currently O(nvals^2) complexity. */ for (;;) { int low1 = INT_MAX, low2 = INT_MAX; int ind1 = 0, ind2 = 0; for (i = 0; i < nvals; i++) { if (freqs[i] < 0) continue; if (low1 > freqs[i]) low2 = low1, ind2 = ind1, low1 = freqs[i], ind1 = i; else if (low2 > freqs[i]) low2 = freqs[i], ind2 = i; } if (low2 == INT_MAX) break; freqs[nvals] = low1 + low2; lens[ind1] = nvals; lens[ind2] = nvals; freqs[ind1] *= -1; freqs[ind2] *= -1; nvals++; } nvals = nvals/2+1; /* Assign lengths */ for (i = 0; i < nvals; i++) { int code_len = 0; for (k = lens[i]; k; k = lens[k]) code_len++; lens[i] = code_len; freqs[i] *= -1; //fprintf(stderr, "%d / %d => %d\n", vals[i], freqs[i], lens[i]); } /* Sort, need in a struct */ if (!(codes = malloc(nvals * sizeof(*codes)))) return NULL; for (i = 0; i < nvals; i++) { codes[i].symbol = vals[i]; codes[i].len = lens[i]; } qsort(codes, nvals, sizeof(*codes), code_sort); /* * Generate canonical codes from lengths. * Sort by length. * Start with 0. * Every new code of same length is +1. * Every new code of new length is +1 then <<1 per extra length. * * /\ * a/\ * /\/\ * bcd/\ * ef * * a 1 0 * b 3 4 (0+1)<<2 * c 3 5 * d 3 6 * e 4 14 (6+1)<<1 * f 5 15 */ code = 0; len = codes[0].len; for (i = 0; i < nvals; i++) { while (len != codes[i].len) { code<<=1; len++; } codes[i].code = code++; if (codes[i].symbol >= -1 && codes[i].symbol < MAX_HUFF) c->e_huffman.val2code[codes[i].symbol+1] = i; //fprintf(stderr, "sym %d, code %d, len %d\n", // codes[i].symbol, codes[i].code, codes[i].len); } free(lens); free(vals); free(freqs); c->e_huffman.codes = codes; c->e_huffman.nvals = nvals; c->free = cram_huffman_encode_free; if (option == E_BYTE || option == E_BYTE_ARRAY) { if (c->e_huffman.codes[0].len == 0) c->encode = cram_huffman_encode_char0; else c->encode = cram_huffman_encode_char; } else { if (c->e_huffman.codes[0].len == 0) c->encode = cram_huffman_encode_int0; else c->encode = cram_huffman_encode_int; } c->store = cram_huffman_encode_store; return c; } /* * --------------------------------------------------------------------------- * BYTE_ARRAY_LEN */ int cram_byte_array_len_decode(cram_slice *slice, cram_codec *c, cram_block *in, char *out, int *out_size) { /* Fetch length */ int32_t len = 0, one = 1; int r; r = c->byte_array_len.len_codec->decode(slice, c->byte_array_len.len_codec, in, (char *)&len, &one); //printf("ByteArray Len=%d\n", len); if (!r && c->byte_array_len.val_codec && len >= 0) { r = c->byte_array_len.val_codec->decode(slice, c->byte_array_len.val_codec, in, out, &len); } else { return -1; } *out_size = len; return r; } void cram_byte_array_len_decode_free(cram_codec *c) { if (!c) return; if (c->byte_array_len.len_codec) c->byte_array_len.len_codec->free(c->byte_array_len.len_codec); if (c->byte_array_len.val_codec) c->byte_array_len.val_codec->free(c->byte_array_len.val_codec); free(c); } cram_codec *cram_byte_array_len_decode_init(char *data, int size, enum cram_external_type option, int version) { cram_codec *c; char *cp = data; char *endp = data + size; int32_t encoding = 0; int32_t sub_size = -1; if (!(c = malloc(sizeof(*c)))) return NULL; c->codec = E_BYTE_ARRAY_LEN; c->decode = cram_byte_array_len_decode; c->free = cram_byte_array_len_decode_free; cp += safe_itf8_get(cp, endp, &encoding); cp += safe_itf8_get(cp, endp, &sub_size); if (sub_size < 0 || endp - cp < sub_size) goto malformed; c->byte_array_len.len_codec = cram_decoder_init(encoding, cp, sub_size, E_INT, version); if (c->byte_array_len.len_codec == NULL) goto no_codec; cp += sub_size; sub_size = -1; cp += safe_itf8_get(cp, endp, &encoding); cp += safe_itf8_get(cp, endp, &sub_size); if (sub_size < 0 || endp - cp < sub_size) goto malformed; c->byte_array_len.val_codec = cram_decoder_init(encoding, cp, sub_size, option, version); if (c->byte_array_len.val_codec == NULL) goto no_codec; cp += sub_size; if (cp - data != size) goto malformed; return c; malformed: fprintf(stderr, "Malformed byte_array_len header stream\n"); no_codec: free(c); return NULL; } int cram_byte_array_len_encode(cram_slice *slice, cram_codec *c, char *in, int in_size) { int32_t i32 = in_size; int r = 0; r |= c->e_byte_array_len.len_codec->encode(slice, c->e_byte_array_len.len_codec, (char *)&i32, 1); r |= c->e_byte_array_len.val_codec->encode(slice, c->e_byte_array_len.val_codec, in, in_size); return r; } void cram_byte_array_len_encode_free(cram_codec *c) { if (!c) return; if (c->e_byte_array_len.len_codec) c->e_byte_array_len.len_codec->free(c->e_byte_array_len.len_codec); if (c->e_byte_array_len.val_codec) c->e_byte_array_len.val_codec->free(c->e_byte_array_len.val_codec); free(c); } int cram_byte_array_len_encode_store(cram_codec *c, cram_block *b, char *prefix, int version) { int len = 0, len2, len3; cram_codec *tc; cram_block *b_len, *b_val; if (prefix) { size_t l = strlen(prefix); BLOCK_APPEND(b, prefix, l); len += l; } tc = c->e_byte_array_len.len_codec; b_len = cram_new_block(0, 0); len2 = tc->store(tc, b_len, NULL, version); tc = c->e_byte_array_len.val_codec; b_val = cram_new_block(0, 0); len3 = tc->store(tc, b_val, NULL, version); len += itf8_put_blk(b, c->codec); len += itf8_put_blk(b, len2+len3); BLOCK_APPEND(b, BLOCK_DATA(b_len), BLOCK_SIZE(b_len)); BLOCK_APPEND(b, BLOCK_DATA(b_val), BLOCK_SIZE(b_val)); cram_free_block(b_len); cram_free_block(b_val); return len + len2 + len3; } cram_codec *cram_byte_array_len_encode_init(cram_stats *st, enum cram_external_type option, void *dat, int version) { cram_codec *c; cram_byte_array_len_encoder *e = (cram_byte_array_len_encoder *)dat; c = malloc(sizeof(*c)); if (!c) return NULL; c->codec = E_BYTE_ARRAY_LEN; c->free = cram_byte_array_len_encode_free; c->encode = cram_byte_array_len_encode; c->store = cram_byte_array_len_encode_store; c->e_byte_array_len.len_codec = cram_encoder_init(e->len_encoding, NULL, E_INT, e->len_dat, version); c->e_byte_array_len.val_codec = cram_encoder_init(e->val_encoding, NULL, E_BYTE_ARRAY, e->val_dat, version); return c; } /* * --------------------------------------------------------------------------- * BYTE_ARRAY_STOP */ static int cram_byte_array_stop_decode_char(cram_slice *slice, cram_codec *c, cram_block *in, char *out, int *out_size) { cram_block *b = NULL; char *cp, ch; b = cram_get_block_by_id(slice, c->byte_array_stop.content_id); if (!b) return *out_size?-1:0; if (b->idx >= b->uncomp_size) return -1; cp = (char *)b->data + b->idx; if (out) { while ((ch = *cp) != (char)c->byte_array_stop.stop) { if (cp - (char *)b->data >= b->uncomp_size) return -1; *out++ = ch; cp++; } } else { // Consume input, but produce no output while ((ch = *cp) != (char)c->byte_array_stop.stop) { if (cp - (char *)b->data >= b->uncomp_size) return -1; cp++; } } *out_size = cp - (char *)(b->data + b->idx); b->idx = cp - (char *)b->data + 1; return 0; } int cram_byte_array_stop_decode_block(cram_slice *slice, cram_codec *c, cram_block *in, char *out_, int *out_size) { cram_block *b; cram_block *out = (cram_block *)out_; char *cp, *out_cp, *cp_end; char stop; b = cram_get_block_by_id(slice, c->byte_array_stop.content_id); if (!b) return *out_size?-1:0; if (b->idx >= b->uncomp_size) return -1; cp = (char *)b->data + b->idx; cp_end = (char *)b->data + b->uncomp_size; out_cp = (char *)BLOCK_END(out); stop = c->byte_array_stop.stop; if (cp_end - cp < out->alloc - out->byte) { while (cp != cp_end && *cp != stop) *out_cp++ = *cp++; BLOCK_SIZE(out) = out_cp - (char *)BLOCK_DATA(out); } else { char *cp_start; for (cp_start = cp; cp != cp_end && *cp != stop; cp++) ; BLOCK_APPEND(out, cp_start, cp - cp_start); BLOCK_GROW(out, cp - cp_start); } *out_size = cp - (char *)(b->data + b->idx); b->idx = cp - (char *)b->data + 1; return 0; } void cram_byte_array_stop_decode_free(cram_codec *c) { if (!c) return; free(c); } cram_codec *cram_byte_array_stop_decode_init(char *data, int size, enum cram_external_type option, int version) { cram_codec *c; unsigned char *cp = (unsigned char *)data; if (!(c = malloc(sizeof(*c)))) return NULL; c->codec = E_BYTE_ARRAY_STOP; switch (option) { case E_BYTE_ARRAY_BLOCK: c->decode = cram_byte_array_stop_decode_block; break; case E_BYTE_ARRAY: c->decode = cram_byte_array_stop_decode_char; break; default: fprintf(stderr, "byte_array_stop codec only supports BYTE_ARRAYs.\n"); free(c); return NULL; } c->free = cram_byte_array_stop_decode_free; c->byte_array_stop.stop = *cp++; if (CRAM_MAJOR_VERS(version) == 1) { c->byte_array_stop.content_id = cp[0] + (cp[1]<<8) + (cp[2]<<16) + (cp[3]<<24); cp += 4; } else { cp += itf8_get(cp, &c->byte_array_stop.content_id); } if ((char *)cp - data != size) { fprintf(stderr, "Malformed byte_array_stop header stream\n"); free(c); return NULL; } return c; } int cram_byte_array_stop_encode(cram_slice *slice, cram_codec *c, char *in, int in_size) { BLOCK_APPEND(c->out, in, in_size); BLOCK_APPEND_CHAR(c->out, c->e_byte_array_stop.stop); return 0; } void cram_byte_array_stop_encode_free(cram_codec *c) { if (!c) return; free(c); } int cram_byte_array_stop_encode_store(cram_codec *c, cram_block *b, char *prefix, int version) { int len = 0; char buf[20], *cp = buf; if (prefix) { size_t l = strlen(prefix); BLOCK_APPEND(b, prefix, l); len += l; } cp += itf8_put(cp, c->codec); if (CRAM_MAJOR_VERS(version) == 1) { cp += itf8_put(cp, 5); *cp++ = c->e_byte_array_stop.stop; *cp++ = (c->e_byte_array_stop.content_id >> 0) & 0xff; *cp++ = (c->e_byte_array_stop.content_id >> 8) & 0xff; *cp++ = (c->e_byte_array_stop.content_id >> 16) & 0xff; *cp++ = (c->e_byte_array_stop.content_id >> 24) & 0xff; } else { cp += itf8_put(cp, 1 + itf8_size(c->e_byte_array_stop.content_id)); *cp++ = c->e_byte_array_stop.stop; cp += itf8_put(cp, c->e_byte_array_stop.content_id); } BLOCK_APPEND(b, buf, cp-buf); len += cp-buf; return len; } cram_codec *cram_byte_array_stop_encode_init(cram_stats *st, enum cram_external_type option, void *dat, int version) { cram_codec *c; c = malloc(sizeof(*c)); if (!c) return NULL; c->codec = E_BYTE_ARRAY_STOP; c->free = cram_byte_array_stop_encode_free; c->encode = cram_byte_array_stop_encode; c->store = cram_byte_array_stop_encode_store; c->e_byte_array_stop.stop = ((int *)dat)[0]; c->e_byte_array_stop.content_id = ((int *)dat)[1]; return c; } /* * --------------------------------------------------------------------------- */ const char *cram_encoding2str(enum cram_encoding t) { switch (t) { case E_NULL: return "NULL"; case E_EXTERNAL: return "EXTERNAL"; case E_GOLOMB: return "GOLOMB"; case E_HUFFMAN: return "HUFFMAN"; case E_BYTE_ARRAY_LEN: return "BYTE_ARRAY_LEN"; case E_BYTE_ARRAY_STOP: return "BYTE_ARRAY_STOP"; case E_BETA: return "BETA"; case E_SUBEXP: return "SUBEXP"; case E_GOLOMB_RICE: return "GOLOMB_RICE"; case E_GAMMA: return "GAMMA"; case E_NUM_CODECS: default: return "?"; } } static cram_codec *(*decode_init[])(char *data, int size, enum cram_external_type option, int version) = { NULL, cram_external_decode_init, NULL, cram_huffman_decode_init, cram_byte_array_len_decode_init, cram_byte_array_stop_decode_init, cram_beta_decode_init, cram_subexp_decode_init, NULL, cram_gamma_decode_init, }; cram_codec *cram_decoder_init(enum cram_encoding codec, char *data, int size, enum cram_external_type option, int version) { if (codec >= E_NULL && codec < E_NUM_CODECS && decode_init[codec]) { return decode_init[codec](data, size, option, version); } else { fprintf(stderr, "Unimplemented codec of type %s\n", cram_encoding2str(codec)); return NULL; } } static cram_codec *(*encode_init[])(cram_stats *stx, enum cram_external_type option, void *opt, int version) = { NULL, cram_external_encode_init, NULL, cram_huffman_encode_init, cram_byte_array_len_encode_init, cram_byte_array_stop_encode_init, cram_beta_encode_init, NULL, //cram_subexp_encode_init, NULL, NULL, //cram_gamma_encode_init, }; cram_codec *cram_encoder_init(enum cram_encoding codec, cram_stats *st, enum cram_external_type option, void *dat, int version) { if (st && !st->nvals) return NULL; if (encode_init[codec]) { cram_codec *r; if ((r = encode_init[codec](st, option, dat, version))) r->out = NULL; return r; } else { fprintf(stderr, "Unimplemented codec of type %s\n", cram_encoding2str(codec)); abort(); } } /* * Returns the content_id used by this codec, also in id2 if byte_array_len. * Returns -1 for the CORE block and -2 for unneeded. * id2 is only filled out for BYTE_ARRAY_LEN which uses 2 codecs. */ int cram_codec_to_id(cram_codec *c, int *id2) { int bnum1, bnum2 = -2; switch (c->codec) { case E_HUFFMAN: bnum1 = c->huffman.ncodes == 1 ? -2 : -1; break; case E_GOLOMB: case E_BETA: case E_SUBEXP: case E_GOLOMB_RICE: case E_GAMMA: bnum1 = -1; break; case E_EXTERNAL: bnum1 = c->external.content_id; break; case E_BYTE_ARRAY_LEN: bnum1 = cram_codec_to_id(c->byte_array_len.len_codec, NULL); bnum2 = cram_codec_to_id(c->byte_array_len.val_codec, NULL); break; case E_BYTE_ARRAY_STOP: bnum1 = c->byte_array_stop.content_id; break; case E_NULL: bnum1 = -2; break; default: fprintf(stderr, "Unknown codec type %d\n", c->codec); bnum1 = -1; } if (id2) *id2 = bnum2; return bnum1; } /* * cram_codec structures are specialised for decoding or encoding. * Unfortunately this makes turning a decoder into an encoder (such as * when transcoding files) problematic. * * This function converts a cram decoder codec into an encoder version * in-place (ie it modifiers the codec itself). * * Returns 0 on success; * -1 on failure. */ int cram_codec_decoder2encoder(cram_fd *fd, cram_codec *c) { int j; switch (c->codec) { case E_EXTERNAL: // shares struct with decode c->free = cram_external_encode_free; c->store = cram_external_encode_store; if (c->decode == cram_external_decode_int) c->encode = cram_external_encode_int; else if (c->decode == cram_external_decode_char) c->encode = cram_external_encode_char; else return -1; break; case E_HUFFMAN: { // New structure, so switch. // FIXME: we huffman and e_huffman structs amended, we could // unify this. cram_codec *t = malloc(sizeof(*t)); t->codec = E_HUFFMAN; t->free = cram_huffman_encode_free; t->store = cram_huffman_encode_store; t->e_huffman.codes = c->huffman.codes; t->e_huffman.nvals = c->huffman.ncodes; for (j = 0; j < t->e_huffman.nvals; j++) { int32_t sym = t->e_huffman.codes[j].symbol; if (sym >= -1 && sym < MAX_HUFF) t->e_huffman.val2code[sym+1] = j; } if (c->decode == cram_huffman_decode_char0) t->encode = cram_huffman_encode_char0; else if (c->decode == cram_huffman_decode_char) t->encode = cram_huffman_encode_char; else if (c->decode == cram_huffman_decode_int0) t->encode = cram_huffman_encode_int0; else if (c->decode == cram_huffman_decode_int) t->encode = cram_huffman_encode_int; else { free(t); return -1; } *c = *t; free(t); break; } case E_BETA: // shares struct with decode c->free = cram_beta_encode_free; c->store = cram_beta_encode_store; if (c->decode == cram_beta_decode_int) c->encode = cram_beta_encode_int; else if (c->decode == cram_beta_decode_char) c->encode = cram_beta_encode_char; else return -1; break; case E_BYTE_ARRAY_LEN: { cram_codec *t = malloc(sizeof(*t)); t->codec = E_BYTE_ARRAY_LEN; t->free = cram_byte_array_len_encode_free; t->store = cram_byte_array_len_encode_store; t->encode = cram_byte_array_len_encode; t->e_byte_array_len.len_codec = c->byte_array_len.len_codec; t->e_byte_array_len.val_codec = c->byte_array_len.val_codec; if (cram_codec_decoder2encoder(fd, t->e_byte_array_len.len_codec) == -1 || cram_codec_decoder2encoder(fd, t->e_byte_array_len.val_codec) == -1) { t->free(t); return -1; } // {len,val}_{encoding,dat} are undefined, but unused. // Leaving them unset here means we can test that assertion. *c = *t; free(t); break; } case E_BYTE_ARRAY_STOP: // shares struct with decode c->free = cram_byte_array_stop_encode_free; c->store = cram_byte_array_stop_encode_store; c->encode = cram_byte_array_stop_encode; break; default: return -1; } return 0; } stacks-1.44/htslib/cram/._cram_codecs.h000644 000765 000024 00000000265 12750743246 020556 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/cram_codecs.h000644 000765 000024 00000013644 12750743246 020346 0ustar00catchenstaff000000 000000 /* Copyright (c) 2012-2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #ifndef _CRAM_ENCODINGS_H_ #define _CRAM_ENCODINGS_H_ #include #ifdef __cplusplus extern "C" { #endif struct cram_codec; /* * Slow but simple huffman decoder to start with. * Read a bit at a time, keeping track of {length, value} * eg. 1 1 0 1 => {1,1}, {2,3}, {3,6}, {4,13} * * Keep track of this through the huffman code table. * For fast scanning we have an index of where the first code of length X * appears. */ typedef struct { int32_t symbol; int32_t p; // next code start value, minus index to codes[] int32_t code; int32_t len; } cram_huffman_code; typedef struct { int ncodes; cram_huffman_code *codes; } cram_huffman_decoder; #define MAX_HUFF 128 typedef struct { cram_huffman_code *codes; int nvals; int val2code[MAX_HUFF+1]; // value to code lookup for small values } cram_huffman_encoder; typedef struct { int32_t offset; int32_t nbits; } cram_beta_decoder; typedef struct { int32_t offset; } cram_gamma_decoder; typedef struct { int32_t offset; int32_t k; } cram_subexp_decoder; typedef struct { int32_t content_id; enum cram_external_type type; } cram_external_decoder; typedef struct { struct cram_codec *len_codec; struct cram_codec *val_codec; } cram_byte_array_len_decoder; typedef struct { unsigned char stop; int32_t content_id; } cram_byte_array_stop_decoder; typedef struct { enum cram_encoding len_encoding; enum cram_encoding val_encoding; void *len_dat; void *val_dat; struct cram_codec *len_codec; struct cram_codec *val_codec; } cram_byte_array_len_encoder; /* * A generic codec structure. */ typedef struct cram_codec { enum cram_encoding codec; cram_block *out; void (*free)(struct cram_codec *codec); int (*decode)(cram_slice *slice, struct cram_codec *codec, cram_block *in, char *out, int *out_size); int (*encode)(cram_slice *slice, struct cram_codec *codec, char *in, int in_size); int (*store)(struct cram_codec *codec, cram_block *b, char *prefix, int version); union { cram_huffman_decoder huffman; cram_external_decoder external; cram_beta_decoder beta; cram_gamma_decoder gamma; cram_subexp_decoder subexp; cram_byte_array_len_decoder byte_array_len; cram_byte_array_stop_decoder byte_array_stop; cram_huffman_encoder e_huffman; cram_external_decoder e_external; cram_byte_array_stop_decoder e_byte_array_stop; cram_byte_array_len_encoder e_byte_array_len; cram_beta_decoder e_beta; }; } cram_codec; const char *cram_encoding2str(enum cram_encoding t); cram_codec *cram_decoder_init(enum cram_encoding codec, char *data, int size, enum cram_external_type option, int version); cram_codec *cram_encoder_init(enum cram_encoding codec, cram_stats *st, enum cram_external_type option, void *dat, int version); //int cram_decode(void *codes, char *in, int in_size, char *out, int *out_size); //void cram_decoder_free(void *codes); //#define GET_BIT_MSB(b,v) (void)(v<<=1, v|=(b->data[b->byte] >> b->bit)&1, (--b->bit == -1) && (b->bit = 7, b->byte++)) #define GET_BIT_MSB(b,v) (void)(v<<=1, v|=(b->data[b->byte] >> b->bit)&1, b->byte += (--b->bit<0), b->bit&=7) /* * Check that enough bits are left in a block to satisy a bit-based decoder. * Return 0 if there are enough * 1 if not. */ static inline int cram_not_enough_bits(cram_block *blk, int nbits) { if (nbits < 0 || (blk->byte >= blk->uncomp_size && nbits > 0) || (blk->uncomp_size - blk->byte <= INT32_MAX / 8 + 1 && (blk->uncomp_size - blk->byte) * 8 + blk->bit - 7 < nbits)) { return 1; } return 0; } /* * Returns the content_id used by this codec, also in id2 if byte_array_len. * Returns -1 for the CORE block and -2 for unneeded. * id2 is only filled out for BYTE_ARRAY_LEN which uses 2 codecs. */ int cram_codec_to_id(cram_codec *c, int *id2); /* * cram_codec structures are specialised for decoding or encoding. * Unfortunately this makes turning a decoder into an encoder (such as * when transcoding files) problematic. * * This function converts a cram decoder codec into an encoder version * in-place (ie it modifiers the codec itself). * * Returns 0 on success; * -1 on failure. */ int cram_codec_decoder2encoder(cram_fd *fd, cram_codec *c); #ifdef __cplusplus } #endif #endif /* _CRAM_ENCODINGS_H_ */ stacks-1.44/htslib/cram/._cram_decode.c000644 000765 000024 00000000265 12750743246 020534 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/cram_decode.c000644 000765 000024 00000242326 12750743246 020325 0ustar00catchenstaff000000 000000 /* Copyright (c) 2012-2014 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /* * - In-memory decoding of CRAM data structures. * - Iterator for reading CRAM record by record. */ #include #include #include #include #include #include #include #include #include #include #include #include "cram/cram.h" #include "cram/os.h" #include "htslib/hts.h" //Whether CIGAR has just M or uses = and X to indicate match and mismatch //#define USE_X /* ---------------------------------------------------------------------- * CRAM compression headers */ /* * Decodes the Tag Dictionary record in the preservation map * Updates the cram compression header. * * Returns number of bytes decoded on success * -1 on failure */ int cram_decode_TD(char *cp, const char *endp, cram_block_compression_hdr *h) { char *op = cp; unsigned char *dat; cram_block *b; int32_t blk_size = 0; int nTL, i, sz; if (!(b = cram_new_block(0, 0))) return -1; /* Decode */ cp += safe_itf8_get(cp, endp, &blk_size); if (!blk_size) { h->nTL = 0; h->TL = NULL; cram_free_block(b); return cp - op; } if (blk_size < 0 || endp - cp < blk_size) { cram_free_block(b); return -1; } BLOCK_APPEND(b, cp, blk_size); cp += blk_size; sz = cp - op; // Force nul termination if missing if (BLOCK_DATA(b)[BLOCK_SIZE(b)-1]) BLOCK_APPEND_CHAR(b, '\0'); /* Set up TL lookup table */ dat = BLOCK_DATA(b); // Count for (nTL = i = 0; i < BLOCK_SIZE(b); i++) { nTL++; while (dat[i]) i++; } // Copy h->nTL = nTL; if (!(h->TL = calloc(h->nTL, sizeof(unsigned char *)))) { cram_free_block(b); return -1; } for (nTL = i = 0; i < BLOCK_SIZE(b); i++) { h->TL[nTL++] = &dat[i]; while (dat[i]) i++; } h->TD_blk = b; return sz; } /* * Decodes a CRAM block compression header. * Returns header ptr on success * NULL on failure */ cram_block_compression_hdr *cram_decode_compression_header(cram_fd *fd, cram_block *b) { char *cp, *endp, *cp_copy; cram_block_compression_hdr *hdr = calloc(1, sizeof(*hdr)); int i; int32_t map_size = 0, map_count = 0; if (!hdr) return NULL; if (b->method != RAW) { if (cram_uncompress_block(b)) { free(hdr); return NULL; } } cp = (char *)b->data; endp = cp + b->uncomp_size; if (CRAM_MAJOR_VERS(fd->version) == 1) { cp += safe_itf8_get(cp, endp, &hdr->ref_seq_id); cp += safe_itf8_get(cp, endp, &hdr->ref_seq_start); cp += safe_itf8_get(cp, endp, &hdr->ref_seq_span); cp += safe_itf8_get(cp, endp, &hdr->num_records); cp += safe_itf8_get(cp, endp, &hdr->num_landmarks); if (!(hdr->landmark = malloc(hdr->num_landmarks * sizeof(int32_t)))) { free(hdr); return NULL; } for (i = 0; i < hdr->num_landmarks; i++) { cp += safe_itf8_get(cp, endp, &hdr->landmark[i]); } } hdr->preservation_map = kh_init(map); memset(hdr->rec_encoding_map, 0, CRAM_MAP_HASH * sizeof(hdr->rec_encoding_map[0])); memset(hdr->tag_encoding_map, 0, CRAM_MAP_HASH * sizeof(hdr->tag_encoding_map[0])); if (!hdr->preservation_map) { cram_free_compression_header(hdr); return NULL; } /* Initialise defaults for preservation map */ hdr->mapped_qs_included = 0; hdr->unmapped_qs_included = 0; hdr->unmapped_placed = 0; hdr->qs_included = 0; hdr->read_names_included = 0; hdr->AP_delta = 1; memcpy(hdr->substitution_matrix, "CGTNAGTNACTNACGNACGT", 20); /* Preservation map */ cp += safe_itf8_get(cp, endp, &map_size); cp_copy = cp; cp += safe_itf8_get(cp, endp, &map_count); for (i = 0; i < map_count; i++) { pmap_t hd; khint_t k; int r; if (endp - cp < 2) { cram_free_compression_header(hdr); return NULL; } cp += 2; switch(CRAM_KEY(cp[-2],cp[-1])) { case CRAM_KEY('M','I'): hd.i = *cp++; k = kh_put(map, hdr->preservation_map, "MI", &r); if (-1 == r) { cram_free_compression_header(hdr); return NULL; } kh_val(hdr->preservation_map, k) = hd; hdr->mapped_qs_included = hd.i; break; case CRAM_KEY('U','I'): hd.i = *cp++; k = kh_put(map, hdr->preservation_map, "UI", &r); if (-1 == r) { cram_free_compression_header(hdr); return NULL; } kh_val(hdr->preservation_map, k) = hd; hdr->unmapped_qs_included = hd.i; break; case CRAM_KEY('P','I'): hd.i = *cp++; k = kh_put(map, hdr->preservation_map, "PI", &r); if (-1 == r) { cram_free_compression_header(hdr); return NULL; } kh_val(hdr->preservation_map, k) = hd; hdr->unmapped_placed = hd.i; break; case CRAM_KEY('R','N'): hd.i = *cp++; k = kh_put(map, hdr->preservation_map, "RN", &r); if (-1 == r) { cram_free_compression_header(hdr); return NULL; } kh_val(hdr->preservation_map, k) = hd; hdr->read_names_included = hd.i; break; case CRAM_KEY('A','P'): hd.i = *cp++; k = kh_put(map, hdr->preservation_map, "AP", &r); if (-1 == r) { cram_free_compression_header(hdr); return NULL; } kh_val(hdr->preservation_map, k) = hd; hdr->AP_delta = hd.i; break; case CRAM_KEY('R','R'): hd.i = *cp++; k = kh_put(map, hdr->preservation_map, "RR", &r); if (-1 == r) { cram_free_compression_header(hdr); return NULL; } kh_val(hdr->preservation_map, k) = hd; fd->no_ref = !hd.i; break; case CRAM_KEY('S','M'): if (endp - cp < 5) { cram_free_compression_header(hdr); return NULL; } hdr->substitution_matrix[0][(cp[0]>>6)&3] = 'C'; hdr->substitution_matrix[0][(cp[0]>>4)&3] = 'G'; hdr->substitution_matrix[0][(cp[0]>>2)&3] = 'T'; hdr->substitution_matrix[0][(cp[0]>>0)&3] = 'N'; hdr->substitution_matrix[1][(cp[1]>>6)&3] = 'A'; hdr->substitution_matrix[1][(cp[1]>>4)&3] = 'G'; hdr->substitution_matrix[1][(cp[1]>>2)&3] = 'T'; hdr->substitution_matrix[1][(cp[1]>>0)&3] = 'N'; hdr->substitution_matrix[2][(cp[2]>>6)&3] = 'A'; hdr->substitution_matrix[2][(cp[2]>>4)&3] = 'C'; hdr->substitution_matrix[2][(cp[2]>>2)&3] = 'T'; hdr->substitution_matrix[2][(cp[2]>>0)&3] = 'N'; hdr->substitution_matrix[3][(cp[3]>>6)&3] = 'A'; hdr->substitution_matrix[3][(cp[3]>>4)&3] = 'C'; hdr->substitution_matrix[3][(cp[3]>>2)&3] = 'G'; hdr->substitution_matrix[3][(cp[3]>>0)&3] = 'N'; hdr->substitution_matrix[4][(cp[4]>>6)&3] = 'A'; hdr->substitution_matrix[4][(cp[4]>>4)&3] = 'C'; hdr->substitution_matrix[4][(cp[4]>>2)&3] = 'G'; hdr->substitution_matrix[4][(cp[4]>>0)&3] = 'T'; hd.p = cp; cp += 5; k = kh_put(map, hdr->preservation_map, "SM", &r); if (-1 == r) { cram_free_compression_header(hdr); return NULL; } kh_val(hdr->preservation_map, k) = hd; break; case CRAM_KEY('T','D'): { int sz = cram_decode_TD(cp, endp, hdr); // tag dictionary if (sz < 0) { cram_free_compression_header(hdr); return NULL; } hd.p = cp; cp += sz; k = kh_put(map, hdr->preservation_map, "TD", &r); if (-1 == r) { cram_free_compression_header(hdr); return NULL; } kh_val(hdr->preservation_map, k) = hd; break; } default: fprintf(stderr, "Unrecognised preservation map key %c%c\n", cp[-2], cp[-1]); // guess byte; cp++; break; } } if (cp - cp_copy != map_size) { cram_free_compression_header(hdr); return NULL; } /* Record encoding map */ cp += safe_itf8_get(cp, endp, &map_size); cp_copy = cp; cp += safe_itf8_get(cp, endp, &map_count); for (i = 0; i < map_count; i++) { char *key = cp; int32_t encoding = E_NULL; int32_t size = 0; cram_map *m = malloc(sizeof(*m)); // FIXME: use pooled_alloc if (!m || endp - cp < 4) { free(m); cram_free_compression_header(hdr); return NULL; } cp += 2; cp += safe_itf8_get(cp, endp, &encoding); cp += safe_itf8_get(cp, endp, &size); // Fill out cram_map purely for cram_dump to dump out. m->key = (key[0]<<8)|key[1]; m->encoding = encoding; m->size = size; m->offset = cp - (char *)b->data; m->codec = NULL; if (m->encoding == E_NULL) continue; if (size < 0 || endp - cp < size) { free(m); cram_free_compression_header(hdr); return NULL; } //printf("%s codes for %.2s\n", cram_encoding2str(encoding), key); /* * For CRAM1.0 CF and BF are Byte and not Int. * Practically speaking it makes no difference unless we have a * 1.0 format file that stores these in EXTERNAL as only then * does Byte vs Int matter. * * Neither this C code nor Java reference implementations did this, * so we gloss over it and treat them as int. */ if (key[0] == 'B' && key[1] == 'F') { if (!(hdr->codecs[DS_BF] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'C' && key[1] == 'F') { if (!(hdr->codecs[DS_CF] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'R' && key[1] == 'I') { if (!(hdr->codecs[DS_RI] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'R' && key[1] == 'L') { if (!(hdr->codecs[DS_RL] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'A' && key[1] == 'P') { if (!(hdr->codecs[DS_AP] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'R' && key[1] == 'G') { if (!(hdr->codecs[DS_RG] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'M' && key[1] == 'F') { if (!(hdr->codecs[DS_MF] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'N' && key[1] == 'S') { if (!(hdr->codecs[DS_NS] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'N' && key[1] == 'P') { if (!(hdr->codecs[DS_NP] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'T' && key[1] == 'S') { if (!(hdr->codecs[DS_TS] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'N' && key[1] == 'F') { if (!(hdr->codecs[DS_NF] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'T' && key[1] == 'C') { if (!(hdr->codecs[DS_TC] = cram_decoder_init(encoding, cp, size, E_BYTE, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'T' && key[1] == 'N') { if (!(hdr->codecs[DS_TN] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'F' && key[1] == 'N') { if (!(hdr->codecs[DS_FN] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'F' && key[1] == 'C') { if (!(hdr->codecs[DS_FC] = cram_decoder_init(encoding, cp, size, E_BYTE, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'F' && key[1] == 'P') { if (!(hdr->codecs[DS_FP] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'B' && key[1] == 'S') { if (!(hdr->codecs[DS_BS] = cram_decoder_init(encoding, cp, size, E_BYTE, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'I' && key[1] == 'N') { if (!(hdr->codecs[DS_IN] = cram_decoder_init(encoding, cp, size, E_BYTE_ARRAY, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'S' && key[1] == 'C') { if (!(hdr->codecs[DS_SC] = cram_decoder_init(encoding, cp, size, E_BYTE_ARRAY, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'D' && key[1] == 'L') { if (!(hdr->codecs[DS_DL] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'B' && key[1] == 'A') { if (!(hdr->codecs[DS_BA] = cram_decoder_init(encoding, cp, size, E_BYTE, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'B' && key[1] == 'B') { if (!(hdr->codecs[DS_BB] = cram_decoder_init(encoding, cp, size, E_BYTE_ARRAY, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'R' && key[1] == 'S') { if (!(hdr->codecs[DS_RS] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'P' && key[1] == 'D') { if (!(hdr->codecs[DS_PD] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'H' && key[1] == 'C') { if (!(hdr->codecs[DS_HC] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'M' && key[1] == 'Q') { if (!(hdr->codecs[DS_MQ] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'R' && key[1] == 'N') { if (!(hdr->codecs[DS_RN] = cram_decoder_init(encoding, cp, size, E_BYTE_ARRAY_BLOCK, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'Q' && key[1] == 'S') { if (!(hdr->codecs[DS_QS] = cram_decoder_init(encoding, cp, size, E_BYTE, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'Q' && key[1] == 'Q') { if (!(hdr->codecs[DS_QQ] = cram_decoder_init(encoding, cp, size, E_BYTE_ARRAY, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'T' && key[1] == 'L') { if (!(hdr->codecs[DS_TL] = cram_decoder_init(encoding, cp, size, E_INT, fd->version))) { cram_free_compression_header(hdr); return NULL; } } else if (key[0] == 'T' && key[1] == 'M') { } else if (key[0] == 'T' && key[1] == 'V') { } else fprintf(stderr, "Unrecognised key: %.2s\n", key); cp += size; m->next = hdr->rec_encoding_map[CRAM_MAP(key[0], key[1])]; hdr->rec_encoding_map[CRAM_MAP(key[0], key[1])] = m; } if (cp - cp_copy != map_size) { cram_free_compression_header(hdr); return NULL; } /* Tag encoding map */ cp += safe_itf8_get(cp, endp, &map_size); cp_copy = cp; cp += safe_itf8_get(cp, endp, &map_count); for (i = 0; i < map_count; i++) { int32_t encoding = E_NULL; int32_t size = 0; cram_map *m = malloc(sizeof(*m)); // FIXME: use pooled_alloc char *key; if (!m || endp - cp < 6) { free(m); cram_free_compression_header(hdr); return NULL; } key = cp + 1; m->key = (key[0]<<16)|(key[1]<<8)|key[2]; cp += 4; // Strictly ITF8, but this suffices cp += safe_itf8_get(cp, endp, &encoding); cp += safe_itf8_get(cp, endp, &size); m->encoding = encoding; m->size = size; m->offset = cp - (char *)b->data; if (size < 0 || endp - cp < size || !(m->codec = cram_decoder_init(encoding, cp, size, E_BYTE_ARRAY_BLOCK, fd->version))) { cram_free_compression_header(hdr); free(m); return NULL; } cp += size; m->next = hdr->tag_encoding_map[CRAM_MAP(key[0],key[1])]; hdr->tag_encoding_map[CRAM_MAP(key[0],key[1])] = m; } if (cp - cp_copy != map_size) { cram_free_compression_header(hdr); return NULL; } return hdr; } /* * Note we also need to scan through the record encoding map to * see which data series share the same block, either external or * CORE. For example if we need the BF data series but MQ and CF * are also encoded in the same block then we need to add those in * as a dependency in order to correctly decode BF. * * Returns 0 on success * -1 on failure */ int cram_dependent_data_series(cram_fd *fd, cram_block_compression_hdr *hdr, cram_slice *s) { int *block_used; int core_used = 0; int i; static int i_to_id[] = { DS_BF, DS_AP, DS_FP, DS_RL, DS_DL, DS_NF, DS_BA, DS_QS, DS_FC, DS_FN, DS_BS, DS_IN, DS_RG, DS_MQ, DS_TL, DS_RN, DS_NS, DS_NP, DS_TS, DS_MF, DS_CF, DS_RI, DS_RS, DS_PD, DS_HC, DS_SC, DS_BB, DS_QQ, }; uint32_t orig_ds; /* * Set the data_series bit field based on fd->required_fields * contents. */ if (fd->required_fields && fd->required_fields != INT_MAX) { hdr->data_series = 0; if (fd->required_fields & SAM_QNAME) hdr->data_series |= CRAM_RN; if (fd->required_fields & SAM_FLAG) hdr->data_series |= CRAM_BF; if (fd->required_fields & SAM_RNAME) hdr->data_series |= CRAM_RI | CRAM_BF; if (fd->required_fields & SAM_POS) hdr->data_series |= CRAM_AP | CRAM_BF; if (fd->required_fields & SAM_MAPQ) hdr->data_series |= CRAM_MQ; if (fd->required_fields & SAM_CIGAR) hdr->data_series |= CRAM_CIGAR; if (fd->required_fields & SAM_RNEXT) hdr->data_series |= CRAM_CF | CRAM_NF | CRAM_RI | CRAM_NS |CRAM_BF; if (fd->required_fields & SAM_PNEXT) hdr->data_series |= CRAM_CF | CRAM_NF | CRAM_AP | CRAM_NP | CRAM_BF; if (fd->required_fields & SAM_TLEN) hdr->data_series |= CRAM_CF | CRAM_NF | CRAM_AP | CRAM_TS | CRAM_BF | CRAM_MF | CRAM_RI | CRAM_CIGAR; if (fd->required_fields & SAM_SEQ) hdr->data_series |= CRAM_SEQ; if (!(fd->required_fields & SAM_AUX)) // No easy way to get MD/NM without other tags at present fd->decode_md = 0; if (fd->required_fields & SAM_QUAL) hdr->data_series |= CRAM_QUAL; if (fd->required_fields & SAM_AUX) hdr->data_series |= CRAM_RG | CRAM_TL | CRAM_aux; if (fd->required_fields & SAM_RGAUX) hdr->data_series |= CRAM_RG | CRAM_BF; // Always uncompress CORE block if (cram_uncompress_block(s->block[0])) return -1; } else { hdr->data_series = CRAM_ALL; for (i = 0; i < s->hdr->num_blocks; i++) { if (cram_uncompress_block(s->block[i])) return -1; } return 0; } block_used = calloc(s->hdr->num_blocks+1, sizeof(int)); if (!block_used) return -1; do { /* * Also set data_series based on code prerequisites. Eg if we need * CRAM_QS then we also need to know CRAM_RL so we know how long it * is, or if we need FC/FP then we also need FN (number of features). * * It's not reciprocal though. We may be needing to decode FN * but have no need to decode FC, FP and cigar ops. */ if (hdr->data_series & CRAM_RS) hdr->data_series |= CRAM_FC|CRAM_FP; if (hdr->data_series & CRAM_PD) hdr->data_series |= CRAM_FC|CRAM_FP; if (hdr->data_series & CRAM_HC) hdr->data_series |= CRAM_FC|CRAM_FP; if (hdr->data_series & CRAM_QS) hdr->data_series |= CRAM_FC|CRAM_FP; if (hdr->data_series & CRAM_IN) hdr->data_series |= CRAM_FC|CRAM_FP; if (hdr->data_series & CRAM_SC) hdr->data_series |= CRAM_FC|CRAM_FP; if (hdr->data_series & CRAM_BS) hdr->data_series |= CRAM_FC|CRAM_FP; if (hdr->data_series & CRAM_DL) hdr->data_series |= CRAM_FC|CRAM_FP; if (hdr->data_series & CRAM_BA) hdr->data_series |= CRAM_FC|CRAM_FP; if (hdr->data_series & CRAM_BB) hdr->data_series |= CRAM_FC|CRAM_FP; if (hdr->data_series & CRAM_QQ) hdr->data_series |= CRAM_FC|CRAM_FP; // cram_decode_seq() needs seq[] array if (hdr->data_series & (CRAM_SEQ|CRAM_CIGAR)) hdr->data_series |= CRAM_RL; if (hdr->data_series & CRAM_FP) hdr->data_series |= CRAM_FC; if (hdr->data_series & CRAM_FC) hdr->data_series |= CRAM_FN; if (hdr->data_series & CRAM_aux) hdr->data_series |= CRAM_TL; if (hdr->data_series & CRAM_MF) hdr->data_series |= CRAM_CF; if (hdr->data_series & CRAM_MQ) hdr->data_series |= CRAM_BF; if (hdr->data_series & CRAM_BS) hdr->data_series |= CRAM_RI; if (hdr->data_series & (CRAM_MF |CRAM_NS |CRAM_NP |CRAM_TS |CRAM_NF)) hdr->data_series |= CRAM_CF; if (!hdr->read_names_included && hdr->data_series & CRAM_RN) hdr->data_series |= CRAM_CF | CRAM_NF; if (hdr->data_series & (CRAM_BA | CRAM_QS | CRAM_BB | CRAM_QQ)) hdr->data_series |= CRAM_BF | CRAM_CF | CRAM_RL; orig_ds = hdr->data_series; // Find which blocks are in use. for (i = 0; i < sizeof(i_to_id)/sizeof(*i_to_id); i++) { int bnum1, bnum2, j; cram_codec *c = hdr->codecs[i_to_id[i]]; if (!(hdr->data_series & (1<hdr->num_blocks; j++) { if (s->block[j]->content_type == EXTERNAL && s->block[j]->content_id == bnum1) { block_used[j] = 1; if (cram_uncompress_block(s->block[j])) { free(block_used); return -1; } } } break; } if (bnum2 == -2 || bnum1 == bnum2) break; bnum1 = bnum2; // 2nd pass } } // Tags too if ((fd->required_fields & SAM_AUX) || (hdr->data_series & CRAM_aux)) { for (i = 0; i < CRAM_MAP_HASH; i++) { int bnum1, bnum2, j; cram_map *m = hdr->tag_encoding_map[i]; while (m) { cram_codec *c = m->codec; if (!c) continue; bnum1 = cram_codec_to_id(c, &bnum2); for (;;) { switch (bnum1) { case -2: break; case -1: core_used = 1; break; default: for (j = 0; j < s->hdr->num_blocks; j++) { if (s->block[j]->content_type == EXTERNAL && s->block[j]->content_id == bnum1) { block_used[j] = 1; if (cram_uncompress_block(s->block[j])) { free(block_used); return -1; } } } break; } if (bnum2 == -2 || bnum1 == bnum2) break; bnum1 = bnum2; // 2nd pass } m = m->next; } } } // We now know which blocks are in used, so repeat and find // which other data series need to be added. for (i = 0; i < sizeof(i_to_id)/sizeof(*i_to_id); i++) { int bnum1, bnum2, j; cram_codec *c = hdr->codecs[i_to_id[i]]; if (!c) continue; bnum1 = cram_codec_to_id(c, &bnum2); for (;;) { switch (bnum1) { case -2: break; case -1: if (core_used) { //printf(" + data series %08x:\n", 1<data_series |= 1<hdr->num_blocks; j++) { if (s->block[j]->content_type == EXTERNAL && s->block[j]->content_id == bnum1) { if (block_used[j]) { //printf(" + data series %08x:\n", 1<data_series |= 1<tag_encoding_map[i]; while (m) { cram_codec *c = m->codec; if (!c) continue; bnum1 = cram_codec_to_id(c, &bnum2); for (;;) { switch (bnum1) { case -2: break; case -1: //printf(" + data series %08x:\n", CRAM_aux); hdr->data_series |= CRAM_aux; break; default: for (j = 0; j < s->hdr->num_blocks; j++) { if (s->block[j]->content_type && s->block[j]->content_id == bnum1) { if (block_used[j]) { //printf(" + data series %08x:\n", // CRAM_aux); hdr->data_series |= CRAM_aux; } } } break; } if (bnum2 == -2 || bnum1 == bnum2) break; bnum1 = bnum2; // 2nd pass } m = m->next; } } } while (orig_ds != hdr->data_series); free(block_used); return 0; } /* * Checks whether an external block is used solely by a single data series. * Returns the codec type if so (EXTERNAL, BYTE_ARRAY_LEN, BYTE_ARRAY_STOP) * or 0 if not (E_NULL). */ static int cram_ds_unique(cram_block_compression_hdr *hdr, cram_codec *c, int id) { int i, n_id = 0; enum cram_encoding e_type = 0; for (i = 0; i < DS_END; i++) { cram_codec *c; int bnum1, bnum2, old_n_id; if (!(c = hdr->codecs[i])) continue; bnum1 = cram_codec_to_id(c, &bnum2); old_n_id = n_id; if (bnum1 == id) { n_id++; e_type = c->codec; } if (bnum2 == id) { n_id++; e_type = c->codec; } if (n_id == old_n_id+2) n_id--; // len/val in same place counts once only. } return n_id == 1 ? e_type : 0; } /* * Attempts to estimate the size of some blocks so we can preallocate them * before decoding. Although decoding will automatically grow the blocks, * it is typically more efficient to preallocate. */ void cram_decode_estimate_sizes(cram_block_compression_hdr *hdr, cram_slice *s, int *qual_size, int *name_size, int *q_id) { int bnum1, bnum2; cram_codec *cd; *qual_size = 0; *name_size = 0; /* Qual */ cd = hdr->codecs[DS_QS]; bnum1 = cram_codec_to_id(cd, &bnum2); if (bnum1 < 0 && bnum2 >= 0) bnum1 = bnum2; if (cram_ds_unique(hdr, cd, bnum1)) { cram_block *b = cram_get_block_by_id(s, bnum1); if (b) *qual_size = b->uncomp_size; if (q_id && cd->codec == E_EXTERNAL) *q_id = bnum1; } /* Name */ cd = hdr->codecs[DS_RN]; bnum1 = cram_codec_to_id(cd, &bnum2); if (bnum1 < 0 && bnum2 >= 0) bnum1 = bnum2; if (cram_ds_unique(hdr, cd, bnum1)) { cram_block *b = cram_get_block_by_id(s, bnum1); if (b) *name_size = b->uncomp_size; } } /* ---------------------------------------------------------------------- * CRAM slices */ /* * Decodes a CRAM (un)mapped slice header block. * Returns slice header ptr on success * NULL on failure */ cram_block_slice_hdr *cram_decode_slice_header(cram_fd *fd, cram_block *b) { cram_block_slice_hdr *hdr; char *cp; char *cp_end; int i; if (b->method != RAW) { /* Spec. says slice header should be RAW, but we can future-proof by trying to decode it if it isn't. */ if (cram_uncompress_block(b) < 0) return NULL; } cp = (char *)b->data; cp_end = cp + b->uncomp_size; if (b->content_type != MAPPED_SLICE && b->content_type != UNMAPPED_SLICE) return NULL; if (!(hdr = calloc(1, sizeof(*hdr)))) return NULL; hdr->content_type = b->content_type; if (b->content_type == MAPPED_SLICE) { cp += safe_itf8_get(cp, cp_end, &hdr->ref_seq_id); cp += safe_itf8_get(cp, cp_end, &hdr->ref_seq_start); cp += safe_itf8_get(cp, cp_end, &hdr->ref_seq_span); } cp += safe_itf8_get(cp, cp_end, &hdr->num_records); hdr->record_counter = 0; if (CRAM_MAJOR_VERS(fd->version) == 2) { int32_t i32 = 0; cp += safe_itf8_get(cp, cp_end, &i32); hdr->record_counter = i32; } else if (CRAM_MAJOR_VERS(fd->version) >= 3) { cp += ltf8_get(cp, &hdr->record_counter); } cp += safe_itf8_get(cp, cp_end, &hdr->num_blocks); cp += safe_itf8_get(cp, cp_end, &hdr->num_content_ids); if (hdr->num_content_ids < 1 || hdr->num_content_ids >= SIZE_MAX / sizeof(int32_t)) { /* Slice must have at least one data block, and malloc'd size shouldn't wrap. */ free(hdr); return NULL; } hdr->block_content_ids = malloc(hdr->num_content_ids * sizeof(int32_t)); if (!hdr->block_content_ids) { free(hdr); return NULL; } for (i = 0; i < hdr->num_content_ids; i++) { int l = safe_itf8_get(cp, cp_end, &hdr->block_content_ids[i]); if (l <= 0) { free(hdr->block_content_ids); free(hdr); return NULL; } cp += l; } if (b->content_type == MAPPED_SLICE) { cp += safe_itf8_get(cp, cp_end, &hdr->ref_base_id); } if (CRAM_MAJOR_VERS(fd->version) != 1) { if (cp_end - cp < 16) { free(hdr->block_content_ids); free(hdr); return NULL; } memcpy(hdr->md5, cp, 16); } else { memset(hdr->md5, 0, 16); } return hdr; } #if 0 /* Returns the number of bits set in val; it the highest bit used */ static int nbits(int v) { static const int MultiplyDeBruijnBitPosition[32] = { 1, 10, 2, 11, 14, 22, 3, 30, 12, 15, 17, 19, 23, 26, 4, 31, 9, 13, 21, 29, 16, 18, 25, 8, 20, 28, 24, 7, 27, 6, 5, 32 }; v |= v >> 1; // first up to set all bits 1 after the first 1 */ v |= v >> 2; v |= v >> 4; v |= v >> 8; v |= v >> 16; // DeBruijn magic to find top bit return MultiplyDeBruijnBitPosition[(uint32_t)(v * 0x07C4ACDDU) >> 27]; } #endif #if 0 static int sort_freqs(const void *vp1, const void *vp2) { const int i1 = *(const int *)vp1; const int i2 = *(const int *)vp2; return i1-i2; } #endif /* ---------------------------------------------------------------------- * Primary CRAM sequence decoder */ /* * Internal part of cram_decode_slice(). * Generates the sequence, quality and cigar components. */ static int cram_decode_seq(cram_fd *fd, cram_container *c, cram_slice *s, cram_block *blk, cram_record *cr, SAM_hdr *bfd, int cf, char *seq, char *qual, int has_MD, int has_NM) { int prev_pos = 0, f, r = 0, out_sz = 1; int seq_pos = 1; int cig_len = 0, ref_pos = cr->apos; int32_t fn, i32; enum cigar_op cig_op = BAM_CMATCH; uint32_t *cigar = s->cigar; uint32_t ncigar = s->ncigar; uint32_t cigar_alloc = s->cigar_alloc; uint32_t nm = 0; int32_t md_dist = 0; int orig_aux = 0; int decode_md = fd->decode_md && s->ref && !has_MD; int decode_nm = fd->decode_md && s->ref && !has_NM; uint32_t ds = c->comp_hdr->data_series; if ((ds & CRAM_QS) && !(cf & CRAM_FLAG_PRESERVE_QUAL_SCORES)) { memset(qual, 255, cr->len); } if (cr->cram_flags & CRAM_FLAG_NO_SEQ) decode_md = decode_nm = 0; if (decode_md) { orig_aux = BLOCK_SIZE(s->aux_blk); BLOCK_APPEND(s->aux_blk, "MDZ", 3); } if (ds & CRAM_FN) { if (!c->comp_hdr->codecs[DS_FN]) return -1; r |= c->comp_hdr->codecs[DS_FN]->decode(s,c->comp_hdr->codecs[DS_FN], blk, (char *)&fn, &out_sz); if (r) return r; } else { fn = 0; } ref_pos--; // count from 0 cr->cigar = ncigar; if (!(ds & (CRAM_FC | CRAM_FP))) goto skip_cigar; for (f = 0; f < fn; f++) { int32_t pos = 0; char op; if (ncigar+2 >= cigar_alloc) { cigar_alloc = cigar_alloc ? cigar_alloc*2 : 1024; s->cigar = cigar; if (!(cigar = realloc(cigar, cigar_alloc * sizeof(*cigar)))) return -1; } if (ds & CRAM_FC) { if (!c->comp_hdr->codecs[DS_FC]) return -1; r |= c->comp_hdr->codecs[DS_FC]->decode(s, c->comp_hdr->codecs[DS_FC], blk, &op, &out_sz); if (r) return r; } if (!(ds & CRAM_FP)) continue; if (!c->comp_hdr->codecs[DS_FP]) return -1; r |= c->comp_hdr->codecs[DS_FP]->decode(s, c->comp_hdr->codecs[DS_FP], blk, (char *)&pos, &out_sz); if (r) return r; pos += prev_pos; if (pos <= 0) { fprintf(stderr, "Error: feature position %d before start of read.\n", pos); return -1; } if (pos > seq_pos) { if (pos > cr->len+1) return -1; if (s->ref && cr->ref_id >= 0) { if (ref_pos + pos - seq_pos > bfd->ref[cr->ref_id].len) { static int whinged = 0; int rlen; if (!whinged) fprintf(stderr, "Ref pos outside of ref " "sequence boundary\n"); whinged = 1; rlen = bfd->ref[cr->ref_id].len - ref_pos; if (rlen > 0) { memcpy(&seq[seq_pos-1], &s->ref[ref_pos - s->ref_start +1], rlen); if ((pos - seq_pos) - rlen > 0) memset(&seq[seq_pos-1+rlen], 'N', (pos - seq_pos) - rlen); } else { memset(&seq[seq_pos-1], 'N', cr->len - seq_pos + 1); } } else { memcpy(&seq[seq_pos-1], &s->ref[ref_pos - s->ref_start +1], pos - seq_pos); } } #ifdef USE_X if (cig_len && cig_op != BAM_CBASE_MATCH) { cigar[ncigar++] = (cig_len<<4) + cig_op; cig_len = 0; } cig_op = BAM_CBASE_MATCH; #else if (cig_len && cig_op != BAM_CMATCH) { cigar[ncigar++] = (cig_len<<4) + cig_op; cig_len = 0; } cig_op = BAM_CMATCH; #endif cig_len += pos - seq_pos; ref_pos += pos - seq_pos; if (md_dist >= 0) md_dist += pos - seq_pos; seq_pos = pos; } prev_pos = pos; if (!(ds & CRAM_FC)) goto skip_cigar; if (!(ds & CRAM_FC)) continue; switch(op) { case 'S': { // soft clip: IN int32_t out_sz2 = 1; int have_sc = 0; if (cig_len) { cigar[ncigar++] = (cig_len<<4) + cig_op; cig_len = 0; } switch (CRAM_MAJOR_VERS(fd->version)) { case 1: if (ds & CRAM_IN) { r |= c->comp_hdr->codecs[DS_IN] ? c->comp_hdr->codecs[DS_IN] ->decode(s, c->comp_hdr->codecs[DS_IN], blk, cr->len ? &seq[pos-1] : NULL, &out_sz2) : (seq[pos-1] = 'N', out_sz2 = 1, 0); have_sc = 1; } break; case 2: default: if (ds & CRAM_SC) { r |= c->comp_hdr->codecs[DS_SC] ? c->comp_hdr->codecs[DS_SC] ->decode(s, c->comp_hdr->codecs[DS_SC], blk, cr->len ? &seq[pos-1] : NULL, &out_sz2) : (seq[pos-1] = 'N', out_sz2 = 1, 0); have_sc = 1; } break; // default: // r |= c->comp_hdr->codecs[DS_BB] // ? c->comp_hdr->codecs[DS_BB] // ->decode(s, c->comp_hdr->codecs[DS_BB], // blk, &seq[pos-1], &out_sz2) // : (seq[pos-1] = 'N', out_sz2 = 1, 0); } if (have_sc) { if (r) return r; cigar[ncigar++] = (out_sz2<<4) + BAM_CSOFT_CLIP; cig_op = BAM_CSOFT_CLIP; seq_pos += out_sz2; } break; } case 'X': { // Substitution; BS unsigned char base; #ifdef USE_X if (cig_len && cig_op != BAM_CBASE_MISMATCH) { cigar[ncigar++] = (cig_len<<4) + cig_op; cig_len = 0; } if (ds & CRAM_BS) { if (!c->comp_hdr->codecs[DS_BS]) return -1; r |= c->comp_hdr->codecs[DS_BS] ->decode(s, c->comp_hdr->codecs[DS_BS], blk, (char *)&base, &out_sz); if (pos-1 < cr->len) seq[pos-1] = 'N'; // FIXME look up BS=base value } cig_op = BAM_CBASE_MISMATCH; #else int ref_base; if (cig_len && cig_op != BAM_CMATCH) { cigar[ncigar++] = (cig_len<<4) + cig_op; cig_len = 0; } if (ds & CRAM_BS) { if (!c->comp_hdr->codecs[DS_BS]) return -1; r |= c->comp_hdr->codecs[DS_BS] ->decode(s, c->comp_hdr->codecs[DS_BS], blk, (char *)&base, &out_sz); if (r) return -1; if (ref_pos >= bfd->ref[cr->ref_id].len || !s->ref) { if (pos-1 < cr->len) seq[pos-1] = c->comp_hdr-> substitution_matrix[fd->L1['N']][base]; if (decode_md || decode_nm) { if (md_dist >= 0 && decode_md) BLOCK_APPEND_UINT(s->aux_blk, md_dist); md_dist = -1; nm--; } } else { unsigned char ref_call = ref_pos <= s->ref_end ? (uc)s->ref[ref_pos - s->ref_start +1] : 'N'; ref_base = fd->L1[ref_call]; if (pos-1 < cr->len) seq[pos-1] = c->comp_hdr-> substitution_matrix[ref_base][base]; if (decode_md) { BLOCK_APPEND_UINT(s->aux_blk, md_dist); BLOCK_APPEND_CHAR(s->aux_blk, ref_call); md_dist = 0; } } } cig_op = BAM_CMATCH; #endif nm++; cig_len++; seq_pos++; ref_pos++; break; } case 'D': { // Deletion; DL if (cig_len && cig_op != BAM_CDEL) { cigar[ncigar++] = (cig_len<<4) + cig_op; cig_len = 0; } if (ds & CRAM_DL) { if (!c->comp_hdr->codecs[DS_DL]) return -1; r |= c->comp_hdr->codecs[DS_DL] ->decode(s, c->comp_hdr->codecs[DS_DL], blk, (char *)&i32, &out_sz); if (r) return r; if (decode_md || decode_nm) { if (md_dist >= 0 && decode_md) BLOCK_APPEND_UINT(s->aux_blk, md_dist); if (ref_pos + i32 <= bfd->ref[cr->ref_id].len) { if (decode_md) { BLOCK_APPEND_CHAR(s->aux_blk, '^'); BLOCK_APPEND(s->aux_blk, &s->ref[ref_pos - s->ref_start +1], i32); md_dist = 0; } nm += i32; } else { uint32_t dlen; if (bfd->ref[cr->ref_id].len >= ref_pos) { if (decode_md) { BLOCK_APPEND_CHAR(s->aux_blk, '^'); BLOCK_APPEND(s->aux_blk, &s->ref[ref_pos - s->ref_start+1], bfd->ref[cr->ref_id].len-ref_pos); BLOCK_APPEND_UINT(s->aux_blk, 0); } dlen = i32 - (bfd->ref[cr->ref_id].len - ref_pos); nm += i32 - dlen; } else { dlen = i32; } md_dist = -1; } } cig_op = BAM_CDEL; cig_len += i32; ref_pos += i32; //printf(" %d: DL = %d (ret %d)\n", f, i32, r); } break; } case 'I': { // Insertion (several bases); IN int32_t out_sz2 = 1; if (cig_len && cig_op != BAM_CINS) { cigar[ncigar++] = (cig_len<<4) + cig_op; cig_len = 0; } if (ds & CRAM_IN) { if (!c->comp_hdr->codecs[DS_IN]) return -1; r |= c->comp_hdr->codecs[DS_IN] ->decode(s, c->comp_hdr->codecs[DS_IN], blk, cr->len ? &seq[pos-1] : NULL, &out_sz2); if (r) return r; cig_op = BAM_CINS; cig_len += out_sz2; seq_pos += out_sz2; nm += out_sz2; //printf(" %d: IN(I) = %.*s (ret %d, out_sz %d)\n", f, out_sz2, dat, r, out_sz2); } break; } case 'i': { // Insertion (single base); BA if (cig_len && cig_op != BAM_CINS) { cigar[ncigar++] = (cig_len<<4) + cig_op; cig_len = 0; } if (ds & CRAM_BA) { if (!c->comp_hdr->codecs[DS_BA]) return -1; r |= c->comp_hdr->codecs[DS_BA] ->decode(s, c->comp_hdr->codecs[DS_BA], blk, cr->len ? &seq[pos-1] : NULL, &out_sz); if (r) return r; } cig_op = BAM_CINS; cig_len++; seq_pos++; nm++; break; } case 'b': { // Several bases int32_t len = 1; if (cig_len && cig_op != BAM_CMATCH) { cigar[ncigar++] = (cig_len<<4) + cig_op; cig_len = 0; } if (ds & CRAM_BB) { if (!c->comp_hdr->codecs[DS_BB]) return -1; r |= c->comp_hdr->codecs[DS_BB] ->decode(s, c->comp_hdr->codecs[DS_BB], blk, cr->len ? &seq[pos-1] : NULL, &len); if (r) return r; if (decode_md || decode_nm) { int x; if (md_dist >= 0 && decode_md) BLOCK_APPEND_UINT(s->aux_blk, md_dist); for (x = 0; x < len; x++) { if (x && decode_md) BLOCK_APPEND_UINT(s->aux_blk, 0); if (ref_pos+x >= bfd->ref[cr->ref_id].len || !s->ref) { md_dist = -1; break; } else { if (decode_md) { char r = s->ref[ref_pos+x-s->ref_start +1]; BLOCK_APPEND_CHAR(s->aux_blk, r); } } } nm += x; } } cig_op = BAM_CMATCH; cig_len+=len; seq_pos+=len; ref_pos+=len; //prev_pos+=len; break; } case 'q': { // Several quality values int32_t len = 1; if (cig_len && cig_op != BAM_CMATCH) { cigar[ncigar++] = (cig_len<<4) + cig_op; cig_len = 0; } if (ds & CRAM_QQ) { if (!c->comp_hdr->codecs[DS_QQ]) return -1; r |= c->comp_hdr->codecs[DS_QQ] ->decode(s, c->comp_hdr->codecs[DS_QQ], blk, (char *)&qual[pos-1], &len); if (r) return r; } cig_op = BAM_CMATCH; cig_len+=len; seq_pos+=len; ref_pos+=len; //prev_pos+=len; break; } case 'B': { // Read base; BA, QS #ifdef USE_X if (cig_len && cig_op != BAM_CBASE_MISMATCH) { cigar[ncigar++] = (cig_len<<4) + cig_op; cig_len = 0; } #else if (cig_len && cig_op != BAM_CMATCH) { cigar[ncigar++] = (cig_len<<4) + cig_op; cig_len = 0; } #endif if (ds & CRAM_BA) { if (!c->comp_hdr->codecs[DS_BA]) return -1; r |= c->comp_hdr->codecs[DS_BA] ->decode(s, c->comp_hdr->codecs[DS_BA], blk, cr->len ? &seq[pos-1] : NULL, &out_sz); if (decode_md || decode_nm) { if (md_dist >= 0 && decode_md) BLOCK_APPEND_UINT(s->aux_blk, md_dist); if (ref_pos >= bfd->ref[cr->ref_id].len || !s->ref) { md_dist = -1; } else { if (decode_md) BLOCK_APPEND_CHAR(s->aux_blk, s->ref[ref_pos-s->ref_start +1]); nm++; md_dist = 0; } } } if (ds & CRAM_QS) { if (!c->comp_hdr->codecs[DS_QS]) return -1; r |= c->comp_hdr->codecs[DS_QS] ->decode(s, c->comp_hdr->codecs[DS_QS], blk, (char *)&qual[pos-1], &out_sz); } #ifdef USE_X cig_op = BAM_CBASE_MISMATCH; #else cig_op = BAM_CMATCH; #endif cig_len++; seq_pos++; ref_pos++; //printf(" %d: BA/QS(B) = %c/%d (ret %d)\n", f, i32, qc, r); break; } case 'Q': { // Quality score; QS if (ds & CRAM_QS) { if (!c->comp_hdr->codecs[DS_QS]) return -1; r |= c->comp_hdr->codecs[DS_QS] ->decode(s, c->comp_hdr->codecs[DS_QS], blk, (char *)&qual[pos-1], &out_sz); //printf(" %d: QS = %d (ret %d)\n", f, qc, r); } break; } case 'H': { // hard clip; HC if (cig_len && cig_op != BAM_CHARD_CLIP) { cigar[ncigar++] = (cig_len<<4) + cig_op; cig_len = 0; } if (ds & CRAM_HC) { if (!c->comp_hdr->codecs[DS_HC]) return -1; r |= c->comp_hdr->codecs[DS_HC] ->decode(s, c->comp_hdr->codecs[DS_HC], blk, (char *)&i32, &out_sz); if (r) return r; cig_op = BAM_CHARD_CLIP; cig_len += i32; } break; } case 'P': { // padding; PD if (cig_len && cig_op != BAM_CPAD) { cigar[ncigar++] = (cig_len<<4) + cig_op; cig_len = 0; } if (ds & CRAM_PD) { if (!c->comp_hdr->codecs[DS_PD]) return -1; r |= c->comp_hdr->codecs[DS_PD] ->decode(s, c->comp_hdr->codecs[DS_PD], blk, (char *)&i32, &out_sz); if (r) return r; cig_op = BAM_CPAD; cig_len += i32; nm += i32; } break; } case 'N': { // Ref skip; RS if (cig_len && cig_op != BAM_CREF_SKIP) { cigar[ncigar++] = (cig_len<<4) + cig_op; cig_len = 0; } if (ds & CRAM_RS) { if (!c->comp_hdr->codecs[DS_RS]) return -1; r |= c->comp_hdr->codecs[DS_RS] ->decode(s, c->comp_hdr->codecs[DS_RS], blk, (char *)&i32, &out_sz); if (r) return r; cig_op = BAM_CREF_SKIP; cig_len += i32; ref_pos += i32; nm += i32; } break; } default: fprintf(stderr, "Error: Unknown feature code '%c'\n", op); return -1; } } if (!(ds & CRAM_FC)) goto skip_cigar; /* An implicit match op for any unaccounted for bases */ if ((ds & CRAM_FN) && cr->len >= seq_pos) { if (s->ref) { if (ref_pos + cr->len - seq_pos + 1 > bfd->ref[cr->ref_id].len) { static int whinged = 0; int rlen; if (!whinged) fprintf(stderr, "Ref pos outside of ref sequence boundary\n"); whinged = 1; rlen = bfd->ref[cr->ref_id].len - ref_pos; if (rlen > 0) { if (seq_pos-1 + rlen < cr->len) memcpy(&seq[seq_pos-1], &s->ref[ref_pos - s->ref_start +1], rlen); if ((cr->len - seq_pos + 1) - rlen > 0) memset(&seq[seq_pos-1+rlen], 'N', (cr->len - seq_pos + 1) - rlen); } else { if (cr->len - seq_pos + 1 > 0) memset(&seq[seq_pos-1], 'N', cr->len - seq_pos + 1); } } else { if (cr->len - seq_pos + 1 > 0) memcpy(&seq[seq_pos-1], &s->ref[ref_pos - s->ref_start +1], cr->len - seq_pos + 1); ref_pos += cr->len - seq_pos + 1; if (md_dist >= 0) md_dist += cr->len - seq_pos + 1; } } if (ncigar+1 >= cigar_alloc) { cigar_alloc = cigar_alloc ? cigar_alloc*2 : 1024; s->cigar = cigar; if (!(cigar = realloc(cigar, cigar_alloc * sizeof(*cigar)))) return -1; } #ifdef USE_X if (cig_len && cig_op != BAM_CBASE_MATCH) { cigar[ncigar++] = (cig_len<<4) + cig_op; cig_len = 0; } cig_op = BAM_CBASE_MATCH; #else if (cig_len && cig_op != BAM_CMATCH) { cigar[ncigar++] = (cig_len<<4) + cig_op; cig_len = 0; } cig_op = BAM_CMATCH; #endif cig_len += cr->len - seq_pos+1; } skip_cigar: if ((ds & CRAM_FN) && decode_md) { if (md_dist >= 0) BLOCK_APPEND_UINT(s->aux_blk, md_dist); } if (cig_len) { if (ncigar >= cigar_alloc) { cigar_alloc = cigar_alloc ? cigar_alloc*2 : 1024; s->cigar = cigar; if (!(cigar = realloc(cigar, cigar_alloc * sizeof(*cigar)))) return -1; } cigar[ncigar++] = (cig_len<<4) + cig_op; } cr->ncigar = ncigar - cr->cigar; cr->aend = ref_pos; //printf("2: %.*s %d .. %d\n", cr->name_len, DSTRING_STR(name_ds) + cr->name, cr->apos, ref_pos); if (ds & CRAM_MQ) { if (!c->comp_hdr->codecs[DS_MQ]) return -1; r |= c->comp_hdr->codecs[DS_MQ] ->decode(s, c->comp_hdr->codecs[DS_MQ], blk, (char *)&cr->mqual, &out_sz); } else { cr->mqual = 40; } if ((ds & CRAM_QS) && (cf & CRAM_FLAG_PRESERVE_QUAL_SCORES)) { int32_t out_sz2 = cr->len; if (ds & CRAM_QS) { if (!c->comp_hdr->codecs[DS_QS]) return -1; r |= c->comp_hdr->codecs[DS_QS] ->decode(s, c->comp_hdr->codecs[DS_QS], blk, qual, &out_sz2); } } s->cigar = cigar; s->cigar_alloc = cigar_alloc; s->ncigar = ncigar; if (cr->cram_flags & CRAM_FLAG_NO_SEQ) cr->len = 0; if (decode_md) { BLOCK_APPEND_CHAR(s->aux_blk, '\0'); // null terminate MD:Z: cr->aux_size += BLOCK_SIZE(s->aux_blk) - orig_aux; } if (decode_nm) { char buf[7]; buf[0] = 'N'; buf[1] = 'M'; buf[2] = 'I'; buf[3] = (nm>> 0) & 0xff; buf[4] = (nm>> 8) & 0xff; buf[5] = (nm>>16) & 0xff; buf[6] = (nm>>24) & 0xff; BLOCK_APPEND(s->aux_blk, buf, 7); cr->aux_size += 7; } return r; } /* * Quick and simple hash lookup for cram_map arrays */ static cram_map *map_find(cram_map **map, unsigned char *key, int id) { cram_map *m; m = map[CRAM_MAP(key[0],key[1])]; while (m && m->key != id) m= m->next; return m; } //#define map_find(M,K,I) M[CRAM_MAP(K[0],K[1])];while (m && m->key != I);m= m->next static int cram_decode_aux_1_0(cram_container *c, cram_slice *s, cram_block *blk, cram_record *cr) { int i, r = 0, out_sz = 1; unsigned char ntags; if (!c->comp_hdr->codecs[DS_TC]) return -1; r |= c->comp_hdr->codecs[DS_TC]->decode(s, c->comp_hdr->codecs[DS_TC], blk, (char *)&ntags, &out_sz); cr->ntags = ntags; //printf("TC=%d\n", cr->ntags); cr->aux_size = 0; cr->aux = BLOCK_SIZE(s->aux_blk); for (i = 0; i < cr->ntags; i++) { int32_t id, out_sz = 1; unsigned char tag_data[3]; cram_map *m; //printf("Tag %d/%d\n", i+1, cr->ntags); if (!c->comp_hdr->codecs[DS_TN]) return -1; r |= c->comp_hdr->codecs[DS_TN]->decode(s, c->comp_hdr->codecs[DS_TN], blk, (char *)&id, &out_sz); if (out_sz == 3) { tag_data[0] = ((char *)&id)[0]; tag_data[1] = ((char *)&id)[1]; tag_data[2] = ((char *)&id)[2]; } else { tag_data[0] = (id>>16) & 0xff; tag_data[1] = (id>>8) & 0xff; tag_data[2] = id & 0xff; } m = map_find(c->comp_hdr->tag_encoding_map, tag_data, id); if (!m) return -1; BLOCK_APPEND(s->aux_blk, (char *)tag_data, 3); if (!m->codec) return -1; r |= m->codec->decode(s, m->codec, blk, (char *)s->aux_blk, &out_sz); cr->aux_size += out_sz + 3; } return r; } static int cram_decode_aux(cram_container *c, cram_slice *s, cram_block *blk, cram_record *cr, int *has_MD, int *has_NM) { int i, r = 0, out_sz = 1; int32_t TL = 0; unsigned char *TN; uint32_t ds = c->comp_hdr->data_series; if (!(ds & (CRAM_TL|CRAM_aux))) { cr->aux = 0; cr->aux_size = 0; return 0; } if (!c->comp_hdr->codecs[DS_TL]) return -1; r |= c->comp_hdr->codecs[DS_TL]->decode(s, c->comp_hdr->codecs[DS_TL], blk, (char *)&TL, &out_sz); if (r || TL < 0 || TL >= c->comp_hdr->nTL) return -1; TN = c->comp_hdr->TL[TL]; cr->ntags = strlen((char *)TN)/3; // optimise to remove strlen //printf("TC=%d\n", cr->ntags); cr->aux_size = 0; cr->aux = BLOCK_SIZE(s->aux_blk); if (!(ds & CRAM_aux)) return 0; for (i = 0; i < cr->ntags; i++) { int32_t id, out_sz = 1; unsigned char tag_data[3]; cram_map *m; if (TN[0] == 'M' && TN[1] == 'D' && has_MD) *has_MD = 1; if (TN[0] == 'N' && TN[1] == 'M' && has_NM) *has_NM = 1; //printf("Tag %d/%d\n", i+1, cr->ntags); tag_data[0] = *TN++; tag_data[1] = *TN++; tag_data[2] = *TN++; id = (tag_data[0]<<16) | (tag_data[1]<<8) | tag_data[2]; m = map_find(c->comp_hdr->tag_encoding_map, tag_data, id); if (!m) return -1; BLOCK_APPEND(s->aux_blk, (char *)tag_data, 3); if (!m->codec) return -1; r |= m->codec->decode(s, m->codec, blk, (char *)s->aux_blk, &out_sz); if (r) break; cr->aux_size += out_sz + 3; } return r; } /* Resolve mate pair cross-references between recs within this slice */ static int cram_decode_slice_xref(cram_slice *s, int required_fields) { int rec; if (!(required_fields & (SAM_RNEXT | SAM_PNEXT | SAM_TLEN))) { for (rec = 0; rec < s->hdr->num_records; rec++) { cram_record *cr = &s->crecs[rec]; cr->tlen = 0; cr->mate_pos = 0; cr->mate_ref_id = -1; } return 0; } for (rec = 0; rec < s->hdr->num_records; rec++) { cram_record *cr = &s->crecs[rec]; if (cr->mate_line >= 0) { if (cr->mate_line < s->hdr->num_records) { /* * On the first read, loop through computing lengths. * It's not perfect as we have one slice per reference so we * cannot detect when TLEN should be zero due to seqs that * map to multiple references. * * We also cannot set tlen correct when it spans a slice for * other reasons. This may make tlen too small. Should we * fix this by forcing TLEN to be stored verbatim in such cases? * * Or do we just admit defeat and output 0 for tlen? It's the * safe option... */ if (cr->tlen == INT_MIN) { int id1 = rec, id2 = rec; int aleft = cr->apos, aright = cr->aend; int tlen; int ref = cr->ref_id; // number of segments starting at the same point. int left_cnt = 0; do { if (aleft > s->crecs[id2].apos) aleft = s->crecs[id2].apos, left_cnt = 1; else if (aleft == s->crecs[id2].apos) left_cnt++; if (aright < s->crecs[id2].aend) aright = s->crecs[id2].aend; if (s->crecs[id2].mate_line == -1) { s->crecs[id2].mate_line = rec; break; } if (s->crecs[id2].mate_line <= id2 || s->crecs[id2].mate_line >= s->hdr->num_records) return -1; id2 = s->crecs[id2].mate_line; if (s->crecs[id2].ref_id != ref) ref = -1; } while (id2 != id1); if (ref != -1) { tlen = aright - aleft + 1; id1 = id2 = rec; /* * When we have two seqs with identical start and * end coordinates, set +/- tlen based on 1st/last * bit flags instead, as a tie breaker. */ if (s->crecs[id2].apos == aleft) { if (left_cnt == 1 || (s->crecs[id2].flags & BAM_FREAD1)) s->crecs[id2].tlen = tlen; else s->crecs[id2].tlen = -tlen; } else { s->crecs[id2].tlen = -tlen; } id2 = s->crecs[id2].mate_line; while (id2 != id1) { if (s->crecs[id2].apos == aleft) { if (left_cnt == 1 || (s->crecs[id2].flags & BAM_FREAD1)) s->crecs[id2].tlen = tlen; else s->crecs[id2].tlen = -tlen; } else { s->crecs[id2].tlen = -tlen; } id2 = s->crecs[id2].mate_line; } } else { id1 = id2 = rec; s->crecs[id2].tlen = 0; id2 = s->crecs[id2].mate_line; while (id2 != id1) { s->crecs[id2].tlen = 0; id2 = s->crecs[id2].mate_line; } } } cr->mate_pos = s->crecs[cr->mate_line].apos; cr->mate_ref_id = s->crecs[cr->mate_line].ref_id; // paired cr->flags |= BAM_FPAIRED; // set mate unmapped if needed if (s->crecs[cr->mate_line].flags & BAM_FUNMAP) { cr->flags |= BAM_FMUNMAP; cr->tlen = 0; } if (cr->flags & BAM_FUNMAP) { cr->tlen = 0; } // set mate reversed if needed if (s->crecs[cr->mate_line].flags & BAM_FREVERSE) cr->flags |= BAM_FMREVERSE; } else { fprintf(stderr, "Mate line out of bounds: %d vs [0, %d]\n", cr->mate_line, s->hdr->num_records-1); } /* FIXME: construct read names here too if needed */ } else { if (cr->mate_flags & CRAM_M_REVERSE) { cr->flags |= BAM_FPAIRED | BAM_FMREVERSE; } if (cr->mate_flags & CRAM_M_UNMAP) { cr->flags |= BAM_FMUNMAP; //cr->mate_ref_id = -1; } if (!(cr->flags & BAM_FPAIRED)) cr->mate_ref_id = -1; } if (cr->tlen == INT_MIN) cr->tlen = 0; // Just incase } return 0; } static char *md5_print(unsigned char *md5, char *out) { int i; for (i = 0; i < 16; i++) { out[i*2+0] = "0123456789abcdef"[md5[i]>>4]; out[i*2+1] = "0123456789abcdef"[md5[i]&15]; } out[32] = 0; return out; } /* * Decode an entire slice from container blocks. Fills out s->crecs[] array. * Returns 0 on success * -1 on failure */ int cram_decode_slice(cram_fd *fd, cram_container *c, cram_slice *s, SAM_hdr *bfd) { cram_block *blk = s->block[0]; int32_t bf, ref_id; unsigned char cf; int out_sz, r = 0; int rec; char *seq = NULL, *qual = NULL; int unknown_rg = -1; int embed_ref; char **refs = NULL; uint32_t ds; if (cram_dependent_data_series(fd, c->comp_hdr, s) != 0) return -1; ds = c->comp_hdr->data_series; blk->bit = 7; // MSB first // Study the blocks and estimate approx sizes to preallocate. // This looks to speed up decoding by around 8-9%. // We can always shrink back down at the end if we overestimated. // However it's likely that this also saves memory as own growth // factor (*=1.5) is never applied. { int qsize, nsize, q_id; cram_decode_estimate_sizes(c->comp_hdr, s, &qsize, &nsize, &q_id); //fprintf(stderr, "qsize=%d nsize=%d\n", qsize, nsize); if (qsize && (ds & CRAM_RL)) BLOCK_RESIZE_EXACT(s->seqs_blk, qsize+1); if (qsize && (ds & CRAM_RL)) BLOCK_RESIZE_EXACT(s->qual_blk, qsize+1); if (nsize && (ds & CRAM_NS)) BLOCK_RESIZE_EXACT(s->name_blk, nsize+1); // To do - consider using q_id here to usurp the quality block and // avoid a memcpy during decode. // Specifically when quality is an external block uniquely used by // DS_QS only, then we can set s->qual_blk directly to this // block and save the codec->decode() calls. (Approx 3% cpu saving) } /* Look for unknown RG, added as last by Java CRAM? */ if (bfd->nrg > 0 && bfd->rg[bfd->nrg-1].name != NULL && !strcmp(bfd->rg[bfd->nrg-1].name, "UNKNOWN")) unknown_rg = bfd->nrg-1; if (blk->content_type != CORE) return -1; if (s->crecs) free(s->crecs); if (!(s->crecs = malloc(s->hdr->num_records * sizeof(*s->crecs)))) return -1; ref_id = s->hdr->ref_seq_id; embed_ref = s->hdr->ref_base_id >= 0 ? 1 : 0; if (ref_id >= 0) { if (embed_ref) { cram_block *b; if (s->hdr->ref_base_id < 0) { fprintf(stderr, "No reference specified and " "no embedded reference is available.\n"); return -1; } b = cram_get_block_by_id(s, s->hdr->ref_base_id); if (!b) return -1; if (cram_uncompress_block(b) != 0) return -1; s->ref = (char *)BLOCK_DATA(b); s->ref_start = s->hdr->ref_seq_start; s->ref_end = s->hdr->ref_seq_start + s->hdr->ref_seq_span-1; } else if (!fd->no_ref) { //// Avoid Java cramtools bug by loading entire reference seq //s->ref = cram_get_ref(fd, s->hdr->ref_seq_id, 1, 0); //s->ref_start = 1; if (fd->required_fields & SAM_SEQ) s->ref = cram_get_ref(fd, s->hdr->ref_seq_id, s->hdr->ref_seq_start, s->hdr->ref_seq_start + s->hdr->ref_seq_span -1); s->ref_start = s->hdr->ref_seq_start; s->ref_end = s->hdr->ref_seq_start + s->hdr->ref_seq_span-1; /* Sanity check */ if (s->ref_start < 0) { fprintf(stderr, "Slice starts before base 1.\n"); s->ref_start = 0; } pthread_mutex_lock(&fd->ref_lock); pthread_mutex_lock(&fd->refs->lock); if ((fd->required_fields & SAM_SEQ) && s->ref_end > fd->refs->ref_id[ref_id]->length) { s->ref_end = fd->refs->ref_id[ref_id]->length; } pthread_mutex_unlock(&fd->refs->lock); pthread_mutex_unlock(&fd->ref_lock); } } if ((fd->required_fields & SAM_SEQ) && s->ref == NULL && s->hdr->ref_seq_id >= 0 && !fd->no_ref) { fprintf(stderr, "Unable to fetch reference #%d %d..%d\n", s->hdr->ref_seq_id, s->hdr->ref_seq_start, s->hdr->ref_seq_start + s->hdr->ref_seq_span-1); return -1; } if (CRAM_MAJOR_VERS(fd->version) != 1 && (fd->required_fields & SAM_SEQ) && s->hdr->ref_seq_id >= 0 && !fd->ignore_md5 && memcmp(s->hdr->md5, "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0", 16)) { hts_md5_context *md5; unsigned char digest[16]; if (s->ref && s->hdr->ref_seq_id >= 0) { int start, len; if (s->hdr->ref_seq_start >= s->ref_start) { start = s->hdr->ref_seq_start - s->ref_start; } else { fprintf(stderr, "Slice starts before base 1.\n"); start = 0; } if (s->hdr->ref_seq_span <= s->ref_end - s->ref_start + 1) { len = s->hdr->ref_seq_span; } else { fprintf(stderr, "Slice ends beyond reference end.\n"); len = s->ref_end - s->ref_start + 1; } if (!(md5 = hts_md5_init())) return -1; if (start + len > s->ref_end - s->ref_start + 1) len = s->ref_end - s->ref_start + 1 - start; if (len >= 0) hts_md5_update(md5, s->ref + start, len); hts_md5_final(digest, md5); hts_md5_destroy(md5); } else if (!s->ref && s->hdr->ref_base_id >= 0) { cram_block *b = cram_get_block_by_id(s, s->hdr->ref_base_id); if (b) { if (!(md5 = hts_md5_init())) return -1; hts_md5_update(md5, b->data, b->uncomp_size); hts_md5_final(digest, md5); hts_md5_destroy(md5); } } if ((!s->ref && s->hdr->ref_base_id < 0) || memcmp(digest, s->hdr->md5, 16) != 0) { char M[33]; fprintf(stderr, "ERROR: md5sum reference mismatch for ref " "%d pos %d..%d\n", ref_id, s->ref_start, s->ref_end); fprintf(stderr, "CRAM: %s\n", md5_print(s->hdr->md5, M)); fprintf(stderr, "Ref : %s\n", md5_print(digest, M)); return -1; } } if (ref_id == -2) { pthread_mutex_lock(&fd->ref_lock); pthread_mutex_lock(&fd->refs->lock); refs = calloc(fd->refs->nref, sizeof(char *)); pthread_mutex_unlock(&fd->refs->lock); pthread_mutex_unlock(&fd->ref_lock); if (!refs) return -1; } int last_ref_id = -9; // Arbitrary -ve marker for not-yet-set for (rec = 0; rec < s->hdr->num_records; rec++) { cram_record *cr = &s->crecs[rec]; int has_MD, has_NM; //fprintf(stderr, "Decode seq %d, %d/%d\n", rec, blk->byte, blk->bit); cr->s = s; out_sz = 1; /* decode 1 item */ if (ds & CRAM_BF) { if (!c->comp_hdr->codecs[DS_BF]) return -1; r |= c->comp_hdr->codecs[DS_BF] ->decode(s, c->comp_hdr->codecs[DS_BF], blk, (char *)&bf, &out_sz); if (r || bf < 0 || bf >= sizeof(fd->bam_flag_swap)/sizeof(*fd->bam_flag_swap)) return -1; bf = fd->bam_flag_swap[bf]; cr->flags = bf; } else { cr->flags = bf = 0x4; // unmapped } if (ds & CRAM_CF) { if (CRAM_MAJOR_VERS(fd->version) == 1) { /* CF is byte in 1.0, int32 in 2.0 */ if (!c->comp_hdr->codecs[DS_CF]) return -1; r |= c->comp_hdr->codecs[DS_CF] ->decode(s, c->comp_hdr->codecs[DS_CF], blk, (char *)&cf, &out_sz); if (r) return -1; cr->cram_flags = cf; } else { if (!c->comp_hdr->codecs[DS_CF]) return -1; r |= c->comp_hdr->codecs[DS_CF] ->decode(s, c->comp_hdr->codecs[DS_CF], blk, (char *)&cr->cram_flags, &out_sz); if (r) return -1; cf = cr->cram_flags; } } else { cf = cr->cram_flags = 0; } if (CRAM_MAJOR_VERS(fd->version) != 1 && ref_id == -2) { if (ds & CRAM_RI) { if (!c->comp_hdr->codecs[DS_RI]) return -1; r |= c->comp_hdr->codecs[DS_RI] ->decode(s, c->comp_hdr->codecs[DS_RI], blk, (char *)&cr->ref_id, &out_sz); if (r) return -1; if ((fd->required_fields & (SAM_SEQ|SAM_TLEN)) && cr->ref_id >= 0 && cr->ref_id != last_ref_id) { if (!fd->no_ref) { if (!refs[cr->ref_id]) refs[cr->ref_id] = cram_get_ref(fd, cr->ref_id, 1, 0); s->ref = refs[cr->ref_id]; if (!fd->unsorted && last_ref_id >= 0 && refs[last_ref_id]) { cram_ref_decr(fd->refs, last_ref_id); refs[last_ref_id] = NULL; } } s->ref_start = 1; pthread_mutex_lock(&fd->ref_lock); pthread_mutex_lock(&fd->refs->lock); s->ref_end = fd->refs->ref_id[cr->ref_id]->length; pthread_mutex_unlock(&fd->refs->lock); pthread_mutex_unlock(&fd->ref_lock); last_ref_id = cr->ref_id; } } else { cr->ref_id = -1; } } else { cr->ref_id = ref_id; // Forced constant in CRAM 1.0 } if (cr->ref_id >= bfd->nref) { fprintf(stderr, "Requested unknown reference ID %d\n", cr->ref_id); return -1; } if (ds & CRAM_RL) { if (!c->comp_hdr->codecs[DS_RL]) return -1; r |= c->comp_hdr->codecs[DS_RL] ->decode(s, c->comp_hdr->codecs[DS_RL], blk, (char *)&cr->len, &out_sz); if (r) return r; if (cr->len < 0) { fprintf(stderr, "Read has negative length\n"); return -1; } } if (ds & CRAM_AP) { if (!c->comp_hdr->codecs[DS_AP]) return -1; r |= c->comp_hdr->codecs[DS_AP] ->decode(s, c->comp_hdr->codecs[DS_AP], blk, (char *)&cr->apos, &out_sz); if (r) return r; if (c->comp_hdr->AP_delta) cr->apos += s->last_apos; s->last_apos= cr->apos; } else { cr->apos = c->ref_seq_start; } if (ds & CRAM_RG) { if (!c->comp_hdr->codecs[DS_RG]) return -1; r |= c->comp_hdr->codecs[DS_RG] ->decode(s, c->comp_hdr->codecs[DS_RG], blk, (char *)&cr->rg, &out_sz); if (r) return r; if (cr->rg == unknown_rg) cr->rg = -1; } else { cr->rg = -1; } cr->name_len = 0; if (c->comp_hdr->read_names_included) { int32_t out_sz2 = 1; // Read directly into name cram_block cr->name = BLOCK_SIZE(s->name_blk); if (ds & CRAM_RN) { if (!c->comp_hdr->codecs[DS_RN]) return -1; r |= c->comp_hdr->codecs[DS_RN] ->decode(s, c->comp_hdr->codecs[DS_RN], blk, (char *)s->name_blk, &out_sz2); if (r) return r; cr->name_len = out_sz2; } } cr->mate_pos = 0; cr->mate_line = -1; cr->mate_ref_id = -1; if ((ds & CRAM_CF) && (cf & CRAM_FLAG_DETACHED)) { if (ds & CRAM_MF) { if (CRAM_MAJOR_VERS(fd->version) == 1) { /* MF is byte in 1.0, int32 in 2.0 */ unsigned char mf; if (!c->comp_hdr->codecs[DS_MF]) return -1; r |= c->comp_hdr->codecs[DS_MF] ->decode(s, c->comp_hdr->codecs[DS_MF], blk, (char *)&mf, &out_sz); if (r) return r; cr->mate_flags = mf; } else { if (!c->comp_hdr->codecs[DS_MF]) return -1; r |= c->comp_hdr->codecs[DS_MF] ->decode(s, c->comp_hdr->codecs[DS_MF], blk, (char *)&cr->mate_flags, &out_sz); if (r) return r; } } else { cr->mate_flags = 0; } if (!c->comp_hdr->read_names_included) { int32_t out_sz2 = 1; // Read directly into name cram_block cr->name = BLOCK_SIZE(s->name_blk); if (ds & CRAM_RN) { if (!c->comp_hdr->codecs[DS_RN]) return -1; r |= c->comp_hdr->codecs[DS_RN] ->decode(s, c->comp_hdr->codecs[DS_RN], blk, (char *)s->name_blk, &out_sz2); if (r) return r; cr->name_len = out_sz2; } } if (ds & CRAM_NS) { if (!c->comp_hdr->codecs[DS_NS]) return -1; r |= c->comp_hdr->codecs[DS_NS] ->decode(s, c->comp_hdr->codecs[DS_NS], blk, (char *)&cr->mate_ref_id, &out_sz); if (r) return r; } // Skip as mate_ref of "*" is legit. It doesn't mean unmapped, just unknown. // if (cr->mate_ref_id == -1 && cr->flags & 0x01) { // /* Paired, but unmapped */ // cr->flags |= BAM_FMUNMAP; // } if (ds & CRAM_NP) { if (!c->comp_hdr->codecs[DS_NP]) return -1; r |= c->comp_hdr->codecs[DS_NP] ->decode(s, c->comp_hdr->codecs[DS_NP], blk, (char *)&cr->mate_pos, &out_sz); if (r) return r; } if (ds & CRAM_TS) { if (!c->comp_hdr->codecs[DS_TS]) return -1; r |= c->comp_hdr->codecs[DS_TS] ->decode(s, c->comp_hdr->codecs[DS_TS], blk, (char *)&cr->tlen, &out_sz); if (r) return r; } else { cr->tlen = INT_MIN; } } else if ((ds & CRAM_CF) && (cf & CRAM_FLAG_MATE_DOWNSTREAM)) { if (ds & CRAM_NF) { if (!c->comp_hdr->codecs[DS_NF]) return -1; r |= c->comp_hdr->codecs[DS_NF] ->decode(s, c->comp_hdr->codecs[DS_NF], blk, (char *)&cr->mate_line, &out_sz); if (r) return r; cr->mate_line += rec + 1; //cr->name_len = sprintf(name, "%d", name_id++); //cr->name = DSTRING_LEN(name_ds); //dstring_nappend(name_ds, name, cr->name_len); cr->mate_ref_id = -1; cr->tlen = INT_MIN; cr->mate_pos = 0; } else { cr->mate_flags = 0; cr->tlen = INT_MIN; } } else { cr->mate_flags = 0; cr->tlen = INT_MIN; } /* else if (!name[0]) { //name[0] = '?'; name[1] = 0; //cr->name_len = 1; //cr->name= DSTRING_LEN(s->name_ds); //dstring_nappend(s->name_ds, "?", 1); cr->mate_ref_id = -1; cr->tlen = 0; cr->mate_pos = 0; } */ /* Auxiliary tags */ has_MD = has_NM = 0; if (CRAM_MAJOR_VERS(fd->version) == 1) r |= cram_decode_aux_1_0(c, s, blk, cr); else r |= cram_decode_aux(c, s, blk, cr, &has_MD, &has_NM); if (r) return r; /* Fake up dynamic string growth and appending */ if (ds & CRAM_RL) { cr->seq = BLOCK_SIZE(s->seqs_blk); BLOCK_GROW(s->seqs_blk, cr->len); seq = (char *)BLOCK_END(s->seqs_blk); BLOCK_SIZE(s->seqs_blk) += cr->len; if (!seq) return -1; cr->qual = BLOCK_SIZE(s->qual_blk); BLOCK_GROW(s->qual_blk, cr->len); qual = (char *)BLOCK_END(s->qual_blk); BLOCK_SIZE(s->qual_blk) += cr->len; if (!s->ref) memset(seq, '=', cr->len); } if (!(bf & BAM_FUNMAP)) { if ((ds & CRAM_AP) && cr->apos <= 0) { fprintf(stderr, "Read has alignment position %d but no unmapped flag\n", cr->apos); return -1; } /* Decode sequence and generate CIGAR */ if (ds & (CRAM_SEQ | CRAM_MQ)) { r |= cram_decode_seq(fd, c, s, blk, cr, bfd, cf, seq, qual, has_MD, has_NM); if (r) return r; } else { cr->cigar = 0; cr->ncigar = 0; cr->aend = cr->apos; cr->mqual = 0; } } else { int out_sz2 = cr->len; //puts("Unmapped"); cr->cigar = 0; cr->ncigar = 0; cr->aend = cr->apos; cr->mqual = 0; if (ds & CRAM_BA && cr->len) { if (!c->comp_hdr->codecs[DS_BA]) return -1; r |= c->comp_hdr->codecs[DS_BA] ->decode(s, c->comp_hdr->codecs[DS_BA], blk, (char *)seq, &out_sz2); if (r) return r; } if ((ds & CRAM_CF) && (cf & CRAM_FLAG_PRESERVE_QUAL_SCORES)) { out_sz2 = cr->len; if (ds & CRAM_QS && cr->len >= 0) { if (!c->comp_hdr->codecs[DS_QS]) return -1; r |= c->comp_hdr->codecs[DS_QS] ->decode(s, c->comp_hdr->codecs[DS_QS], blk, qual, &out_sz2); if (r) return r; } } else { if (ds & CRAM_RL) memset(qual, 255, cr->len); } } } pthread_mutex_lock(&fd->ref_lock); if (refs) { int i; for (i = 0; i < fd->refs->nref; i++) { if (refs[i]) cram_ref_decr(fd->refs, i); } free(refs); } else if (ref_id >= 0 && s->ref != fd->ref_free && !embed_ref) { cram_ref_decr(fd->refs, ref_id); } pthread_mutex_unlock(&fd->ref_lock); /* Resolve mate pair cross-references between recs within this slice */ r |= cram_decode_slice_xref(s, fd->required_fields); // Free the original blocks as we no longer need these. { int i; for (i = 0; i < s->hdr->num_blocks; i++) { cram_block *b = s->block[i]; cram_free_block(b); s->block[i] = NULL; } } // Also see initial BLOCK_RESIZE_EXACT at top of function. // As we grow blocks we overallocate by up to 50%. So shrink // back to their final sizes here. // // fprintf(stderr, "%d %d // %d %d // %d %d // %d %d\n", // (int)s->seqs_blk->byte, (int)s->seqs_blk->alloc, // (int)s->qual_blk->byte, (int)s->qual_blk->alloc, // (int)s->name_blk->byte, (int)s->name_blk->alloc, // (int)s->aux_blk->byte, (int)s->aux_blk->alloc); BLOCK_RESIZE_EXACT(s->seqs_blk, BLOCK_SIZE(s->seqs_blk)+1); BLOCK_RESIZE_EXACT(s->qual_blk, BLOCK_SIZE(s->qual_blk)+1); BLOCK_RESIZE_EXACT(s->name_blk, BLOCK_SIZE(s->name_blk)+1); BLOCK_RESIZE_EXACT(s->aux_blk, BLOCK_SIZE(s->aux_blk)+1); return r; } typedef struct { cram_fd *fd; cram_container *c; cram_slice *s; SAM_hdr *h; int exit_code; } cram_decode_job; void *cram_decode_slice_thread(void *arg) { cram_decode_job *j = (cram_decode_job *)arg; j->exit_code = cram_decode_slice(j->fd, j->c, j->s, j->h); return j; } /* * Spawn a multi-threaded version of cram_decode_slice(). */ int cram_decode_slice_mt(cram_fd *fd, cram_container *c, cram_slice *s, SAM_hdr *bfd) { cram_decode_job *j; int nonblock; if (!fd->pool) return cram_decode_slice(fd, c, s, bfd); if (!(j = malloc(sizeof(*j)))) return -1; j->fd = fd; j->c = c; j->s = s; j->h = bfd; nonblock = t_pool_results_queue_sz(fd->rqueue) ? 1 : 0; if (-1 == t_pool_dispatch2(fd->pool, fd->rqueue, cram_decode_slice_thread, j, nonblock)) { /* Would block */ fd->job_pending = j; } else { fd->job_pending = NULL; } // flush too return 0; } /* ---------------------------------------------------------------------- * CRAM sequence iterators. */ /* * Converts a cram in-memory record into a bam in-memory record. We * pass a pointer to a bam_seq_t pointer along with the a pointer to * the allocated size. These can initially be pointers to NULL and zero. * * This function will reallocate the bam buffer as required and update * (*bam)->alloc accordingly, allowing it to be used within a loop * efficiently without needing to allocate new bam objects over and * over again. * * Returns the used size of the bam record on success * -1 on failure. */ static int cram_to_bam(SAM_hdr *bfd, cram_fd *fd, cram_slice *s, cram_record *cr, int rec, bam_seq_t **bam) { int bam_idx, rg_len; char name_a[1024], *name; int name_len; char *aux, *aux_orig; char *seq, *qual; /* Assign names if not explicitly set */ if (fd->required_fields & SAM_QNAME) { if (cr->name_len) { name = (char *)BLOCK_DATA(s->name_blk) + cr->name; name_len = cr->name_len; } else { name = name_a; name_len = strlen(fd->prefix); memcpy(name, fd->prefix, name_len); name += name_len; *name++ = ':'; if (cr->mate_line >= 0 && cr->mate_line < rec) name = (char *)append_uint64((unsigned char *)name, s->hdr->record_counter + cr->mate_line + 1); else name = (char *)append_uint64((unsigned char *)name, s->hdr->record_counter + rec + 1); name_len = name - name_a; name = name_a; } } else { name = "?"; name_len = 1; } /* Generate BAM record */ if (cr->rg < -1 || cr->rg >= bfd->nrg) return -1; rg_len = (cr->rg != -1) ? bfd->rg[cr->rg].name_len + 4 : 0; if (fd->required_fields & (SAM_SEQ | SAM_QUAL)) { if (!BLOCK_DATA(s->seqs_blk)) return -1; seq = (char *)BLOCK_DATA(s->seqs_blk) + cr->seq; } else { seq = "*"; cr->len = 0; } if (fd->required_fields & SAM_QUAL) { if (!BLOCK_DATA(s->qual_blk)) return -1; qual = (char *)BLOCK_DATA(s->qual_blk) + cr->qual; } else { qual = NULL; } bam_idx = bam_construct_seq(bam, cr->aux_size + rg_len, name, name_len, cr->flags, cr->ref_id, cr->apos, cr->aend, cr->mqual, cr->ncigar, &s->cigar[cr->cigar], cr->mate_ref_id, cr->mate_pos, cr->tlen, cr->len, seq, qual); if (bam_idx == -1) return -1; aux = aux_orig = (char *)bam_aux(*bam); /* Auxiliary strings */ if (cr->aux_size != 0) { memcpy(aux, BLOCK_DATA(s->aux_blk) + cr->aux, cr->aux_size); aux += cr->aux_size; } /* RG:Z: */ if (cr->rg != -1) { int len = bfd->rg[cr->rg].name_len; *aux++ = 'R'; *aux++ = 'G'; *aux++ = 'Z'; memcpy(aux, bfd->rg[cr->rg].name, len); aux += len; *aux++ = 0; } return bam_idx + (aux - aux_orig); } /* * Here be dragons! The multi-threading code in this is crufty beyond belief. */ static cram_slice *cram_next_slice(cram_fd *fd, cram_container **cp) { cram_container *c; cram_slice *s = NULL; if (!(c = fd->ctr)) { // Load first container. do { if (!(c = fd->ctr = cram_read_container(fd))) return NULL; } while (c->length == 0); /* * The first container may be a result of a sub-range query. * In which case it may still not be the optimal starting point * due to skipped containers/slices in the index. */ if (fd->range.refid != -2) { while (c->ref_seq_id != -2 && (c->ref_seq_id < fd->range.refid || c->ref_seq_start + c->ref_seq_span-1 < fd->range.start)) { if (0 != cram_seek(fd, c->length, SEEK_CUR)) return NULL; cram_free_container(fd->ctr); do { if (!(c = fd->ctr = cram_read_container(fd))) return NULL; } while (c->length == 0); } if (c->ref_seq_id != -2 && c->ref_seq_id != fd->range.refid) return NULL; } if (!(c->comp_hdr_block = cram_read_block(fd))) return NULL; if (c->comp_hdr_block->content_type != COMPRESSION_HEADER) return NULL; c->comp_hdr = cram_decode_compression_header(fd, c->comp_hdr_block); if (!c->comp_hdr) return NULL; if (!c->comp_hdr->AP_delta && sam_hdr_sort_order(fd->header) != ORDER_COORD) { pthread_mutex_lock(&fd->ref_lock); fd->unsorted = 1; pthread_mutex_unlock(&fd->ref_lock); } } if ((s = c->slice)) { c->slice = NULL; cram_free_slice(s); s = NULL; } if (c->curr_slice == c->max_slice) { cram_free_container(c); c = NULL; } /* Sorry this is so contorted! */ for (;;) { if (fd->job_pending) { cram_decode_job *j = (cram_decode_job *)fd->job_pending; c = j->c; s = j->s; free(fd->job_pending); fd->job_pending = NULL; } else if (!fd->ooc) { empty_container: if (!c || c->curr_slice == c->max_slice) { // new container do { if (!(c = fd->ctr = cram_read_container(fd))) { if (fd->pool) { fd->ooc = 1; break; } return NULL; } } while (c->length == 0); if (fd->ooc) break; /* Skip containers not yet spanning our range */ if (fd->range.refid != -2 && c->ref_seq_id != -2) { fd->required_fields |= SAM_POS; if (c->ref_seq_id != fd->range.refid) { cram_free_container(c); fd->ctr = NULL; fd->ooc = 1; fd->eof = 1; break; } if (c->ref_seq_start > fd->range.end) { cram_free_container(c); fd->ctr = NULL; fd->ooc = 1; fd->eof = 1; break; } if (c->ref_seq_start + c->ref_seq_span-1 < fd->range.start) { c->curr_rec = c->max_rec; c->curr_slice = c->max_slice; cram_seek(fd, c->length, SEEK_CUR); cram_free_container(c); c = NULL; continue; } } if (!(c->comp_hdr_block = cram_read_block(fd))) return NULL; if (c->comp_hdr_block->content_type != COMPRESSION_HEADER) return NULL; c->comp_hdr = cram_decode_compression_header(fd, c->comp_hdr_block); if (!c->comp_hdr) return NULL; if (!c->comp_hdr->AP_delta && sam_hdr_sort_order(fd->header) != ORDER_COORD) { pthread_mutex_lock(&fd->ref_lock); fd->unsorted = 1; pthread_mutex_unlock(&fd->ref_lock); } } if (c->num_records == 0) { cram_free_container(c); c = NULL; goto empty_container; } if (!(s = c->slice = cram_read_slice(fd))) return NULL; c->curr_slice++; c->curr_rec = 0; c->max_rec = s->hdr->num_records; s->last_apos = s->hdr->ref_seq_start; /* Skip slices not yet spanning our range */ if (fd->range.refid != -2 && s->hdr->ref_seq_id != -2) { if (s->hdr->ref_seq_id != fd->range.refid) { fd->eof = 1; cram_free_slice(s); c->slice = NULL; return NULL; } if (s->hdr->ref_seq_start > fd->range.end) { fd->eof = 1; cram_free_slice(s); c->slice = NULL; return NULL; } if (s->hdr->ref_seq_start + s->hdr->ref_seq_span-1 < fd->range.start) { cram_free_slice(s); c->slice = NULL; cram_free_container(c); c = NULL; continue; } } } /* Test decoding of 1st seq */ if (!c || !s) break; if (cram_decode_slice_mt(fd, c, s, fd->header) != 0) { // if (cram_decode_slice(fd, c, s, fd->header) != 0) { fprintf(stderr, "Failure to decode slice\n"); cram_free_slice(s); c->slice = NULL; return NULL; } if (!fd->pool || fd->job_pending) break; // Push it a bit far, to qsize in queue rather than pending arrival, // as cram tends to be a bit bursty in decode timings. if (t_pool_results_queue_len(fd->rqueue) > fd->pool->qsize) break; } if (fd->pool) { t_pool_result *res; cram_decode_job *j; // fprintf(stderr, "Thread pool len = %d, %d\n", // t_pool_results_queue_len(fd->rqueue), // t_pool_results_queue_sz(fd->rqueue)); if (fd->ooc && t_pool_results_queue_empty(fd->rqueue)) return NULL; res = t_pool_next_result_wait(fd->rqueue); if (!res || !res->data) { fprintf(stderr, "t_pool_next_result failure\n"); return NULL; } j = (cram_decode_job *)res->data; c = j->c; s = j->s; fd->ctr = c; t_pool_delete_result(res, 1); } *cp = c; return s; } /* * Read the next cram record and return it. * Note that to decode cram_record the caller will need to look up some data * in the current slice, pointed to by fd->ctr->slice. This is valid until * the next call to cram_get_seq (which may invalidate it). * * Returns record pointer on success (do not free) * NULL on failure */ cram_record *cram_get_seq(cram_fd *fd) { cram_container *c; cram_slice *s; for (;;) { c = fd->ctr; if (c && c->slice && c->curr_rec < c->max_rec) { s = c->slice; } else { if (!(s = cram_next_slice(fd, &c))) return NULL; continue; /* In case slice contains no records */ } if (fd->range.refid != -2) { if (fd->range.refid == -1 && s->crecs[c->curr_rec].ref_id != -1) { // Special case when looking for unmapped blocks at end. // If these are mixed in with mapped data (c->ref_id == -2) // then we need skip until we find the unmapped data, if at all c->curr_rec++; continue; } if (s->crecs[c->curr_rec].ref_id < fd->range.refid && s->crecs[c->curr_rec].ref_id != -1) { // Looking for a mapped read, but not there yet. Special case // as -1 (unmapped) shouldn't be considered < refid. c->curr_rec++; continue; } if (s->crecs[c->curr_rec].ref_id != fd->range.refid) { fd->eof = 1; cram_free_slice(s); c->slice = NULL; return NULL; } if (fd->range.refid != -1 && s->crecs[c->curr_rec].apos > fd->range.end) { fd->eof = 1; cram_free_slice(s); c->slice = NULL; return NULL; } if (fd->range.refid != -1 && s->crecs[c->curr_rec].aend < fd->range.start) { c->curr_rec++; continue; } } break; } fd->ctr = c; c->slice = s; return &s->crecs[c->curr_rec++]; } /* * Read the next cram record and convert it to a bam_seq_t struct. * * Returns 0 on success * -1 on EOF or failure (check fd->err) */ int cram_get_bam_seq(cram_fd *fd, bam_seq_t **bam) { cram_record *cr; cram_container *c; cram_slice *s; if (!(cr = cram_get_seq(fd))) return -1; c = fd->ctr; s = c->slice; return cram_to_bam(fd->header, fd, s, cr, c->curr_rec-1, bam); } stacks-1.44/htslib/cram/._cram_decode.h000644 000765 000024 00000000265 12750743246 020541 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/cram_decode.h000644 000765 000024 00000006673 12750743246 020335 0ustar00catchenstaff000000 000000 /* Copyright (c) 2012-2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /*! \file * Include cram.h instead. * * This is an internal part of the CRAM system and is automatically included * when you #include cram.h. * * Implements the decoding portion of CRAM I/O. Also see * cram_codecs.[ch] for the actual encoding functions themselves. */ #ifndef _CRAM_READ_H_ #define _CRAM_READ_H_ #ifdef __cplusplus extern "C" { #endif /* ---------------------------------------------------------------------- * CRAM sequence iterators. */ /*! Read the next cram record and return it as a cram_record. * * Note that to decode cram_record the caller will need to look up some data * in the current slice, pointed to by fd->ctr->slice. This is valid until * the next call to cram_get_seq (which may invalidate it). * * @return * Returns record pointer on success (do not free); * NULL on failure */ cram_record *cram_get_seq(cram_fd *fd); /*! Read the next cram record and convert it to a bam_seq_t struct. * * @return * Returns 0 on success; * -1 on EOF or failure (check fd->err) */ int cram_get_bam_seq(cram_fd *fd, bam_seq_t **bam); /* ---------------------------------------------------------------------- * Internal functions */ /*! INTERNAL: * Decodes a CRAM block compression header. * * @return * Returns header ptr on success; * NULL on failure */ cram_block_compression_hdr *cram_decode_compression_header(cram_fd *fd, cram_block *b); /*! INTERNAL: * Decodes a CRAM (un)mapped slice header block. * * @return * Returns slice header ptr on success; * NULL on failure */ cram_block_slice_hdr *cram_decode_slice_header(cram_fd *fd, cram_block *b); /*! INTERNAL: * Decode an entire slice from container blocks. Fills out s->crecs[] array. * * @return * Returns 0 on success; * -1 on failure */ int cram_decode_slice(cram_fd *fd, cram_container *c, cram_slice *s, SAM_hdr *hdr); #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/cram/._cram_encode.c000644 000765 000024 00000000265 12750743246 020546 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/cram_encode.c000644 000765 000024 00000252523 12750743246 020337 0ustar00catchenstaff000000 000000 /* Copyright (c) 2012-2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #include #include #include #include #include #include #include #include #include #include #include #include "cram/cram.h" #include "cram/os.h" #include "htslib/hts.h" #define Z_CRAM_STRAT Z_FILTERED //#define Z_CRAM_STRAT Z_RLE //#define Z_CRAM_STRAT Z_HUFFMAN_ONLY //#define Z_CRAM_STRAT Z_DEFAULT_STRATEGY static int process_one_read(cram_fd *fd, cram_container *c, cram_slice *s, cram_record *cr, bam_seq_t *b, int rnum); /* * Returns index of val into key. * Basically strchr(key, val)-key; */ static int sub_idx(char *key, char val) { int i; for (i = 0; *key && *key++ != val; i++); return i; } /* * Encodes a compression header block into a generic cram_block structure. * * Returns cram_block ptr on success * NULL on failure */ cram_block *cram_encode_compression_header(cram_fd *fd, cram_container *c, cram_block_compression_hdr *h) { cram_block *cb = cram_new_block(COMPRESSION_HEADER, 0); cram_block *map = cram_new_block(COMPRESSION_HEADER, 0); int i, mc; if (!cb || !map) return NULL; /* * This is a concatenation of several blocks of data: * header + landmarks, preservation map, read encoding map, and the tag * encoding map. * All 4 are variable sized and we need to know how large these are * before creating the compression header itself as this starts with * the total size (stored as a variable length string). */ // Duplicated from container itself, and removed in 1.1 if (CRAM_MAJOR_VERS(fd->version) == 1) { itf8_put_blk(cb, h->ref_seq_id); itf8_put_blk(cb, h->ref_seq_start); itf8_put_blk(cb, h->ref_seq_span); itf8_put_blk(cb, h->num_records); itf8_put_blk(cb, h->num_landmarks); for (i = 0; i < h->num_landmarks; i++) { itf8_put_blk(cb, h->landmark[i]); } } if (h->preservation_map) kh_destroy(map, h->preservation_map); /* Create in-memory preservation map */ /* FIXME: should create this when we create the container */ { khint_t k; int r; if (!(h->preservation_map = kh_init(map))) return NULL; k = kh_put(map, h->preservation_map, "RN", &r); if (-1 == r) return NULL; kh_val(h->preservation_map, k).i = 1; if (CRAM_MAJOR_VERS(fd->version) == 1) { k = kh_put(map, h->preservation_map, "PI", &r); if (-1 == r) return NULL; kh_val(h->preservation_map, k).i = 0; k = kh_put(map, h->preservation_map, "UI", &r); if (-1 == r) return NULL; kh_val(h->preservation_map, k).i = 1; k = kh_put(map, h->preservation_map, "MI", &r); if (-1 == r) return NULL; kh_val(h->preservation_map, k).i = 1; } else { // Technically SM was in 1.0, but wasn't in Java impl. k = kh_put(map, h->preservation_map, "SM", &r); if (-1 == r) return NULL; kh_val(h->preservation_map, k).i = 0; k = kh_put(map, h->preservation_map, "TD", &r); if (-1 == r) return NULL; kh_val(h->preservation_map, k).i = 0; k = kh_put(map, h->preservation_map, "AP", &r); if (-1 == r) return NULL; kh_val(h->preservation_map, k).i = h->AP_delta; if (fd->no_ref || fd->embed_ref) { // Reference Required == No k = kh_put(map, h->preservation_map, "RR", &r); if (-1 == r) return NULL; kh_val(h->preservation_map, k).i = 0; } } } /* Encode preservation map; could collapse this and above into one */ mc = 0; BLOCK_SIZE(map) = 0; if (h->preservation_map) { khint_t k; for (k = kh_begin(h->preservation_map); k != kh_end(h->preservation_map); k++) { const char *key; khash_t(map) *pmap = h->preservation_map; if (!kh_exist(pmap, k)) continue; key = kh_key(pmap, k); BLOCK_APPEND(map, key, 2); switch(CRAM_KEY(key[0], key[1])) { case CRAM_KEY('M','I'): BLOCK_APPEND_CHAR(map, kh_val(pmap, k).i); break; case CRAM_KEY('U','I'): BLOCK_APPEND_CHAR(map, kh_val(pmap, k).i); break; case CRAM_KEY('P','I'): BLOCK_APPEND_CHAR(map, kh_val(pmap, k).i); break; case CRAM_KEY('A','P'): BLOCK_APPEND_CHAR(map, kh_val(pmap, k).i); break; case CRAM_KEY('R','N'): BLOCK_APPEND_CHAR(map, kh_val(pmap, k).i); break; case CRAM_KEY('R','R'): BLOCK_APPEND_CHAR(map, kh_val(pmap, k).i); break; case CRAM_KEY('S','M'): { char smat[5], *mp = smat; *mp++ = (sub_idx("CGTN", h->substitution_matrix[0][0]) << 6) | (sub_idx("CGTN", h->substitution_matrix[0][1]) << 4) | (sub_idx("CGTN", h->substitution_matrix[0][2]) << 2) | (sub_idx("CGTN", h->substitution_matrix[0][3]) << 0); *mp++ = (sub_idx("AGTN", h->substitution_matrix[1][0]) << 6) | (sub_idx("AGTN", h->substitution_matrix[1][1]) << 4) | (sub_idx("AGTN", h->substitution_matrix[1][2]) << 2) | (sub_idx("AGTN", h->substitution_matrix[1][3]) << 0); *mp++ = (sub_idx("ACTN", h->substitution_matrix[2][0]) << 6) | (sub_idx("ACTN", h->substitution_matrix[2][1]) << 4) | (sub_idx("ACTN", h->substitution_matrix[2][2]) << 2) | (sub_idx("ACTN", h->substitution_matrix[2][3]) << 0); *mp++ = (sub_idx("ACGN", h->substitution_matrix[3][0]) << 6) | (sub_idx("ACGN", h->substitution_matrix[3][1]) << 4) | (sub_idx("ACGN", h->substitution_matrix[3][2]) << 2) | (sub_idx("ACGN", h->substitution_matrix[3][3]) << 0); *mp++ = (sub_idx("ACGT", h->substitution_matrix[4][0]) << 6) | (sub_idx("ACGT", h->substitution_matrix[4][1]) << 4) | (sub_idx("ACGT", h->substitution_matrix[4][2]) << 2) | (sub_idx("ACGT", h->substitution_matrix[4][3]) << 0); BLOCK_APPEND(map, smat, 5); break; } case CRAM_KEY('T','D'): { itf8_put_blk(map, BLOCK_SIZE(h->TD_blk)); BLOCK_APPEND(map, BLOCK_DATA(h->TD_blk), BLOCK_SIZE(h->TD_blk)); break; } default: fprintf(stderr, "Unknown preservation key '%.2s'\n", key); break; } mc++; } } itf8_put_blk(cb, BLOCK_SIZE(map) + itf8_size(mc)); itf8_put_blk(cb, mc); BLOCK_APPEND(cb, BLOCK_DATA(map), BLOCK_SIZE(map)); /* rec encoding map */ mc = 0; BLOCK_SIZE(map) = 0; if (h->codecs[DS_BF]) { if (-1 == h->codecs[DS_BF]->store(h->codecs[DS_BF], map, "BF", fd->version)) return NULL; mc++; } if (h->codecs[DS_CF]) { if (-1 == h->codecs[DS_CF]->store(h->codecs[DS_CF], map, "CF", fd->version)) return NULL; mc++; } if (h->codecs[DS_RL]) { if (-1 == h->codecs[DS_RL]->store(h->codecs[DS_RL], map, "RL", fd->version)) return NULL; mc++; } if (h->codecs[DS_AP]) { if (-1 == h->codecs[DS_AP]->store(h->codecs[DS_AP], map, "AP", fd->version)) return NULL; mc++; } if (h->codecs[DS_RG]) { if (-1 == h->codecs[DS_RG]->store(h->codecs[DS_RG], map, "RG", fd->version)) return NULL; mc++; } if (h->codecs[DS_MF]) { if (-1 == h->codecs[DS_MF]->store(h->codecs[DS_MF], map, "MF", fd->version)) return NULL; mc++; } if (h->codecs[DS_NS]) { if (-1 == h->codecs[DS_NS]->store(h->codecs[DS_NS], map, "NS", fd->version)) return NULL; mc++; } if (h->codecs[DS_NP]) { if (-1 == h->codecs[DS_NP]->store(h->codecs[DS_NP], map, "NP", fd->version)) return NULL; mc++; } if (h->codecs[DS_TS]) { if (-1 == h->codecs[DS_TS]->store(h->codecs[DS_TS], map, "TS", fd->version)) return NULL; mc++; } if (h->codecs[DS_NF]) { if (-1 == h->codecs[DS_NF]->store(h->codecs[DS_NF], map, "NF", fd->version)) return NULL; mc++; } if (h->codecs[DS_TC]) { if (-1 == h->codecs[DS_TC]->store(h->codecs[DS_TC], map, "TC", fd->version)) return NULL; mc++; } if (h->codecs[DS_TN]) { if (-1 == h->codecs[DS_TN]->store(h->codecs[DS_TN], map, "TN", fd->version)) return NULL; mc++; } if (h->codecs[DS_TL]) { if (-1 == h->codecs[DS_TL]->store(h->codecs[DS_TL], map, "TL", fd->version)) return NULL; mc++; } if (h->codecs[DS_FN]) { if (-1 == h->codecs[DS_FN]->store(h->codecs[DS_FN], map, "FN", fd->version)) return NULL; mc++; } if (h->codecs[DS_FC]) { if (-1 == h->codecs[DS_FC]->store(h->codecs[DS_FC], map, "FC", fd->version)) return NULL; mc++; } if (h->codecs[DS_FP]) { if (-1 == h->codecs[DS_FP]->store(h->codecs[DS_FP], map, "FP", fd->version)) return NULL; mc++; } if (h->codecs[DS_BS]) { if (-1 == h->codecs[DS_BS]->store(h->codecs[DS_BS], map, "BS", fd->version)) return NULL; mc++; } if (h->codecs[DS_IN]) { if (-1 == h->codecs[DS_IN]->store(h->codecs[DS_IN], map, "IN", fd->version)) return NULL; mc++; } if (h->codecs[DS_DL]) { if (-1 == h->codecs[DS_DL]->store(h->codecs[DS_DL], map, "DL", fd->version)) return NULL; mc++; } if (h->codecs[DS_BA]) { if (-1 == h->codecs[DS_BA]->store(h->codecs[DS_BA], map, "BA", fd->version)) return NULL; mc++; } if (h->codecs[DS_BB]) { if (-1 == h->codecs[DS_BB]->store(h->codecs[DS_BB], map, "BB", fd->version)) return NULL; mc++; } if (h->codecs[DS_MQ]) { if (-1 == h->codecs[DS_MQ]->store(h->codecs[DS_MQ], map, "MQ", fd->version)) return NULL; mc++; } if (h->codecs[DS_RN]) { if (-1 == h->codecs[DS_RN]->store(h->codecs[DS_RN], map, "RN", fd->version)) return NULL; mc++; } if (h->codecs[DS_QS]) { if (-1 == h->codecs[DS_QS]->store(h->codecs[DS_QS], map, "QS", fd->version)) return NULL; mc++; } if (h->codecs[DS_QQ]) { if (-1 == h->codecs[DS_QQ]->store(h->codecs[DS_QQ], map, "QQ", fd->version)) return NULL; mc++; } if (h->codecs[DS_RI]) { if (-1 == h->codecs[DS_RI]->store(h->codecs[DS_RI], map, "RI", fd->version)) return NULL; mc++; } if (CRAM_MAJOR_VERS(fd->version) != 1) { if (h->codecs[DS_SC]) { if (-1 == h->codecs[DS_SC]->store(h->codecs[DS_SC], map, "SC", fd->version)) return NULL; mc++; } if (h->codecs[DS_RS]) { if (-1 == h->codecs[DS_RS]->store(h->codecs[DS_RS], map, "RS", fd->version)) return NULL; mc++; } if (h->codecs[DS_PD]) { if (-1 == h->codecs[DS_PD]->store(h->codecs[DS_PD], map, "PD", fd->version)) return NULL; mc++; } if (h->codecs[DS_HC]) { if (-1 == h->codecs[DS_HC]->store(h->codecs[DS_HC], map, "HC", fd->version)) return NULL; mc++; } } if (h->codecs[DS_TM]) { if (-1 == h->codecs[DS_TM]->store(h->codecs[DS_TM], map, "TM", fd->version)) return NULL; mc++; } if (h->codecs[DS_TV]) { if (-1 == h->codecs[DS_TV]->store(h->codecs[DS_TV], map, "TV", fd->version)) return NULL; mc++; } itf8_put_blk(cb, BLOCK_SIZE(map) + itf8_size(mc)); itf8_put_blk(cb, mc); BLOCK_APPEND(cb, BLOCK_DATA(map), BLOCK_SIZE(map)); /* tag encoding map */ #if 0 mp = map; mc = 0; if (h->tag_encoding_map) { HashItem *hi; HashIter *iter = HashTableIterCreate(); if (!iter) return NULL; while ((hi = HashTableIterNext(h->tag_encoding_map, iter))) { cram_map *m = hi->data.p; int sz; mp += itf8_put(mp, (hi->key[0]<<16)|(hi->key[1]<<8)|hi->key[2]); if (-1 == (sz = m->codec->store(m->codec, mp, NULL, fd->version))) return NULL; mp += sz; mc++; } HashTableIterDestroy(iter); } #else mc = 0; BLOCK_SIZE(map) = 0; if (c->tags_used) { khint_t k; #define TAG_ID(a) ((#a[0]<<8)+#a[1]) for (k = kh_begin(c->tags_used); k != kh_end(c->tags_used); k++) { int key; if (!kh_exist(c->tags_used, k)) continue; mc++; itf8_put_blk(map, kh_key(c->tags_used, k)); // use block content id 4 switch((key = kh_key(c->tags_used, k)) & 0xff) { case 'Z': case 'H': // string as byte_array_stop if (CRAM_MAJOR_VERS(fd->version) == 1) { BLOCK_APPEND(map, "\005" // BYTE_ARRAY_STOP "\005" // len "\t" // stop-byte is also SAM separator DS_aux_S "\000\000\000", 7); } else { if (key>>8 == TAG_ID(OQ)) BLOCK_APPEND(map, "\005" // BYTE_ARRAY_STOP "\002" // len "\t" // stop-byte is also SAM separator DS_aux_OQ_S, 4); else if (key>>8 == TAG_ID(BQ)) BLOCK_APPEND(map, "\005" // BYTE_ARRAY_STOP "\002" // len "\t" // stop-byte is also SAM separator DS_aux_BQ_S, 4); else if (key>>8 == TAG_ID(BD)) BLOCK_APPEND(map, "\005" // BYTE_ARRAY_STOP "\002" // len "\t" // stop-byte is also SAM separator DS_aux_BD_S, 4); else if (key>>8 == TAG_ID(BI)) BLOCK_APPEND(map, "\005" // BYTE_ARRAY_STOP "\002" // len "\t" // stop-byte is also SAM separator DS_aux_BI_S, 4); else if ((key>>8 == TAG_ID(Q2)) || (key>>8 == TAG_ID(U2)) || (key>>8 == TAG_ID(QT)) || (key>>8 == TAG_ID(CQ))) BLOCK_APPEND(map, "\005" // BYTE_ARRAY_STOP "\002" // len "\t" // stop-byte is also SAM separator DS_aux_oq_S, 4); else if ((key>>8 == TAG_ID(R2)) || (key>>8 == TAG_ID(E2)) || (key>>8 == TAG_ID(CS)) || (key>>8 == TAG_ID(BC)) || (key>>8 == TAG_ID(RT))) BLOCK_APPEND(map, "\005" // BYTE_ARRAY_STOP "\002" // len "\t" // stop-byte is also SAM separator DS_aux_os_S, 4); else BLOCK_APPEND(map, "\005" // BYTE_ARRAY_STOP "\002" // len "\t" // stop-byte is also SAM separator DS_aux_oz_S, 4); } break; case 'A': case 'c': case 'C': // byte array len, 1 byte BLOCK_APPEND(map, "\004" // BYTE_ARRAY_LEN "\011" // length "\003" // HUFFMAN (len) "\004" // huffman-len "\001" // 1 symbol "\001" // symbol=1 byte value "\001" // 1 length "\000" // length=0 "\001" // EXTERNAL (val) "\001" // external-len DS_aux_S,// content-id 11); break; case 's': case 'S': // byte array len, 2 byte BLOCK_APPEND(map, "\004" // BYTE_ARRAY_LEN "\011" // length "\003" // HUFFMAN (len) "\004" // huffman-len "\001" // 1 symbol "\002" // symbol=2 byte value "\001" // 1 length "\000" // length=0 "\001" // EXTERNAL (val) "\001" // external-len DS_aux_S,// content-id 11); break; case 'i': case 'I': case 'f': // byte array len, 4 byte BLOCK_APPEND(map, "\004" // BYTE_ARRAY_LEN "\011" // length "\003" // HUFFMAN (len) "\004" // huffman-len "\001" // 1 symbol "\004" // symbol=4 byte value "\001" // 1 length "\000" // length=0 "\001" // EXTERNAL (val) "\001" // external-len DS_aux_S,// content-id 11); break; case 'B': // Byte array of variable size, but we generate our tag // byte stream at the wrong stage (during reading and not // after slice header construction). So we use // BYTE_ARRAY_LEN with the length codec being external // too. if ((key>>8 == TAG_ID(FZ)) || (key>>8 == TAG_ID(ZM))) BLOCK_APPEND(map, "\004" // BYTE_ARRAY_LEN "\006" // length "\001" // EXTERNAL (len) "\001" // external-len DS_aux_FZ_S // content-id "\001" // EXTERNAL (val) "\001" // external-len DS_aux_FZ_S,// content-id 8); else BLOCK_APPEND(map, "\004" // BYTE_ARRAY_LEN "\006" // length "\001" // EXTERNAL (len) "\001" // external-len DS_aux_S // content-id "\001" // EXTERNAL (val) "\001" // external-len DS_aux_S,// content-id 8); break; default: fprintf(stderr, "Unsupported SAM aux type '%c'\n", kh_key(c->tags_used, k) & 0xff); } //mp += m->codec->store(m->codec, mp, NULL, fd->version); } } #endif itf8_put_blk(cb, BLOCK_SIZE(map) + itf8_size(mc)); itf8_put_blk(cb, mc); BLOCK_APPEND(cb, BLOCK_DATA(map), BLOCK_SIZE(map)); if (fd->verbose) fprintf(stderr, "Wrote compression block header in %d bytes\n", (int)BLOCK_SIZE(cb)); BLOCK_UPLEN(cb); cram_free_block(map); return cb; } /* * Encodes a slice compression header. * * Returns cram_block on success * NULL on failure */ cram_block *cram_encode_slice_header(cram_fd *fd, cram_slice *s) { char *buf; char *cp; cram_block *b = cram_new_block(MAPPED_SLICE, 0); int j; if (!b) return NULL; if (NULL == (cp = buf = malloc(16+5*(8+s->hdr->num_blocks)))) { cram_free_block(b); return NULL; } cp += itf8_put(cp, s->hdr->ref_seq_id); cp += itf8_put(cp, s->hdr->ref_seq_start); cp += itf8_put(cp, s->hdr->ref_seq_span); cp += itf8_put(cp, s->hdr->num_records); if (CRAM_MAJOR_VERS(fd->version) == 2) cp += itf8_put(cp, s->hdr->record_counter); else if (CRAM_MAJOR_VERS(fd->version) >= 3) cp += ltf8_put(cp, s->hdr->record_counter); cp += itf8_put(cp, s->hdr->num_blocks); cp += itf8_put(cp, s->hdr->num_content_ids); for (j = 0; j < s->hdr->num_content_ids; j++) { cp += itf8_put(cp, s->hdr->block_content_ids[j]); } if (s->hdr->content_type == MAPPED_SLICE) cp += itf8_put(cp, s->hdr->ref_base_id); if (CRAM_MAJOR_VERS(fd->version) != 1) { memcpy(cp, s->hdr->md5, 16); cp += 16; } assert(cp-buf <= 16+5*(8+s->hdr->num_blocks)); b->data = (unsigned char *)buf; b->comp_size = b->uncomp_size = cp-buf; return b; } /* * Encodes a single read. * * Returns 0 on success * -1 on failure */ static int cram_encode_slice_read(cram_fd *fd, cram_container *c, cram_block_compression_hdr *h, cram_slice *s, cram_record *cr, int *last_pos) { int r = 0; int32_t i32; unsigned char uc; //fprintf(stderr, "Encode seq %d, %d/%d FN=%d, %s\n", rec, core->byte, core->bit, cr->nfeature, s->name_ds->str + cr->name); //printf("BF=0x%x\n", cr->flags); // bf = cram_flag_swap[cr->flags]; i32 = fd->cram_flag_swap[cr->flags & 0xfff]; r |= h->codecs[DS_BF]->encode(s, h->codecs[DS_BF], (char *)&i32, 1); i32 = cr->cram_flags & CRAM_FLAG_MASK; r |= h->codecs[DS_CF]->encode(s, h->codecs[DS_CF], (char *)&i32, 1); if (CRAM_MAJOR_VERS(fd->version) != 1 && s->hdr->ref_seq_id == -2) r |= h->codecs[DS_RI]->encode(s, h->codecs[DS_RI], (char *)&cr->ref_id, 1); r |= h->codecs[DS_RL]->encode(s, h->codecs[DS_RL], (char *)&cr->len, 1); if (c->pos_sorted) { i32 = cr->apos - *last_pos; r |= h->codecs[DS_AP]->encode(s, h->codecs[DS_AP], (char *)&i32, 1); *last_pos = cr->apos; } else { i32 = cr->apos; r |= h->codecs[DS_AP]->encode(s, h->codecs[DS_AP], (char *)&i32, 1); } r |= h->codecs[DS_RG]->encode(s, h->codecs[DS_RG], (char *)&cr->rg, 1); if (c->comp_hdr->read_names_included) { // RN codec: Already stored in block[3]. } if (cr->cram_flags & CRAM_FLAG_DETACHED) { i32 = cr->mate_flags; r |= h->codecs[DS_MF]->encode(s, h->codecs[DS_MF], (char *)&i32, 1); if (!c->comp_hdr->read_names_included) { // RN codec: Already stored in block[3]. } r |= h->codecs[DS_NS]->encode(s, h->codecs[DS_NS], (char *)&cr->mate_ref_id, 1); r |= h->codecs[DS_NP]->encode(s, h->codecs[DS_NP], (char *)&cr->mate_pos, 1); r |= h->codecs[DS_TS]->encode(s, h->codecs[DS_TS], (char *)&cr->tlen, 1); } else if (cr->cram_flags & CRAM_FLAG_MATE_DOWNSTREAM) { r |= h->codecs[DS_NF]->encode(s, h->codecs[DS_NF], (char *)&cr->mate_line, 1); } /* Aux tags */ if (CRAM_MAJOR_VERS(fd->version) == 1) { int j; uc = cr->ntags; r |= h->codecs[DS_TC]->encode(s, h->codecs[DS_TC], (char *)&uc, 1); for (j = 0; j < cr->ntags; j++) { uint32_t i32 = s->TN[cr->TN_idx + j]; // id r |= h->codecs[DS_TN]->encode(s, h->codecs[DS_TN], (char *)&i32, 1); } } else { r |= h->codecs[DS_TL]->encode(s, h->codecs[DS_TL], (char *)&cr->TL, 1); } // qual // QS codec : Already stored in block[2]. // features (diffs) if (!(cr->flags & BAM_FUNMAP)) { int prev_pos = 0, j; r |= h->codecs[DS_FN]->encode(s, h->codecs[DS_FN], (char *)&cr->nfeature, 1); for (j = 0; j < cr->nfeature; j++) { cram_feature *f = &s->features[cr->feature + j]; uc = f->X.code; r |= h->codecs[DS_FC]->encode(s, h->codecs[DS_FC], (char *)&uc, 1); i32 = f->X.pos - prev_pos; r |= h->codecs[DS_FP]->encode(s, h->codecs[DS_FP], (char *)&i32, 1); prev_pos = f->X.pos; switch(f->X.code) { //char *seq; case 'X': //fprintf(stderr, " FC=%c FP=%d base=%d\n", f->X.code, i32, f->X.base); uc = f->X.base; r |= h->codecs[DS_BS]->encode(s, h->codecs[DS_BS], (char *)&uc, 1); break; case 'S': // Already done // r |= h->codecs[DS_SC]->encode(s, h->codecs[DS_SC], // BLOCK_DATA(s->soft_blk) + f->S.seq_idx, // f->S.len); // if (IS_CRAM_3_VERS(fd)) { // r |= h->codecs[DS_BB]->encode(s, h->codecs[DS_BB], // BLOCK_DATA(s->seqs_blk) + f->S.seq_idx, // f->S.len); // } break; case 'I': //seq = DSTRING_STR(s->seqs_ds) + f->S.seq_idx; //r |= h->codecs[DS_IN]->encode(s, h->codecs[DS_IN], // seq, f->S.len); // if (IS_CRAM_3_VERS(fd)) { // r |= h->codecs[DS_BB]->encode(s, h->codecs[DS_BB], // BLOCK_DATA(s->seqs_blk) + f->I.seq_idx, // f->I.len); // } break; case 'i': uc = f->i.base; r |= h->codecs[DS_BA]->encode(s, h->codecs[DS_BA], (char *)&uc, 1); //seq = DSTRING_STR(s->seqs_ds) + f->S.seq_idx; //r |= h->codecs[DS_IN]->encode(s, h->codecs[DS_IN], // seq, 1); break; case 'D': i32 = f->D.len; r |= h->codecs[DS_DL]->encode(s, h->codecs[DS_DL], (char *)&i32, 1); break; case 'B': // // Used when we try to store a non ACGTN base or an N // // that aligns against a non ACGTN reference uc = f->B.base; r |= h->codecs[DS_BA]->encode(s, h->codecs[DS_BA], (char *)&uc, 1); // Already added // uc = f->B.qual; // r |= h->codecs[DS_QS]->encode(s, h->codecs[DS_QS], // (char *)&uc, 1); break; case 'b': // string of bases r |= h->codecs[DS_BB]->encode(s, h->codecs[DS_BB], (char *)BLOCK_DATA(s->seqs_blk) + f->b.seq_idx, f->b.len); break; case 'Q': // Already added // uc = f->B.qual; // r |= h->codecs[DS_QS]->encode(s, h->codecs[DS_QS], // (char *)&uc, 1); break; case 'N': i32 = f->N.len; r |= h->codecs[DS_RS]->encode(s, h->codecs[DS_RS], (char *)&i32, 1); break; case 'P': i32 = f->P.len; r |= h->codecs[DS_PD]->encode(s, h->codecs[DS_PD], (char *)&i32, 1); break; case 'H': i32 = f->H.len; r |= h->codecs[DS_HC]->encode(s, h->codecs[DS_HC], (char *)&i32, 1); break; default: fprintf(stderr, "unhandled feature code %c\n", f->X.code); return -1; } } r |= h->codecs[DS_MQ]->encode(s, h->codecs[DS_MQ], (char *)&cr->mqual, 1); } else { char *seq = (char *)BLOCK_DATA(s->seqs_blk) + cr->seq; if (cr->len) r |= h->codecs[DS_BA]->encode(s, h->codecs[DS_BA], seq, cr->len); } return r ? -1 : 0; } /* * Applies various compression methods to specific blocks, depending on * known observations of how data series compress. * * Returns 0 on success * -1 on failure */ static int cram_compress_slice(cram_fd *fd, cram_slice *s) { int level = fd->level, i; int method = 1< 5 && s->block[0]->uncomp_size > 500) cram_compress_block(fd, s->block[0], NULL, GZIP, 1); if (fd->use_bz2) method |= 1<use_rans) method |= (1<use_lzma) method |= (1<= 6) methodF = method; /* Specific compression methods for certain block types */ if (cram_compress_block(fd, s->block[DS_IN], fd->m[DS_IN], //IN (seq) method, level)) return -1; if (fd->level == 0) { /* Do nothing */ } else if (fd->level == 1) { if (cram_compress_block(fd, s->block[DS_QS], fd->m[DS_QS], methodF, 1)) return -1; for (i = DS_aux; i <= DS_aux_oz; i++) { if (s->block[i]) if (cram_compress_block(fd, s->block[i], fd->m[i], method, 1)) return -1; } } else if (fd->level < 3) { if (cram_compress_block(fd, s->block[DS_QS], fd->m[DS_QS], method, 1)) return -1; if (cram_compress_block(fd, s->block[DS_BA], fd->m[DS_BA], method, 1)) return -1; if (s->block[DS_BB]) if (cram_compress_block(fd, s->block[DS_BB], fd->m[DS_BB], method, 1)) return -1; for (i = DS_aux; i <= DS_aux_oz; i++) { if (s->block[i]) if (cram_compress_block(fd, s->block[i], fd->m[i], method, level)) return -1; } } else { if (cram_compress_block(fd, s->block[DS_QS], fd->m[DS_QS], method, level)) return -1; if (cram_compress_block(fd, s->block[DS_BA], fd->m[DS_BA], method, level)) return -1; if (s->block[DS_BB]) if (cram_compress_block(fd, s->block[DS_BB], fd->m[DS_BB], method, level)) return -1; for (i = DS_aux; i <= DS_aux_oz; i++) { if (s->block[i]) if (cram_compress_block(fd, s->block[i], fd->m[i], method, level)) return -1; } } // NAME: best is generally xz, bzip2, zlib then rans1 // It benefits well from a little bit extra compression level. if (cram_compress_block(fd, s->block[DS_RN], fd->m[DS_RN], method & ~(1<block[DS_NS] != s->block[0]) if (cram_compress_block(fd, s->block[DS_NS], fd->m[DS_NS], method, level)) return -1; /* * Minimal compression of any block still uncompressed, bar CORE */ { int i; for (i = 1; i < DS_END; i++) { if (!s->block[i] || s->block[i] == s->block[0]) continue; // fast methods only if (s->block[i]->method == RAW) { cram_compress_block(fd, s->block[i], fd->m[i], methodF, level); } } } return 0; } /* * Encodes a single slice from a container * * Returns 0 on success * -1 on failure */ static int cram_encode_slice(cram_fd *fd, cram_container *c, cram_block_compression_hdr *h, cram_slice *s) { int rec, r = 0, last_pos; int embed_ref; enum cram_DS_ID id; embed_ref = fd->embed_ref && s->hdr->ref_seq_id != -1 ? 1 : 0; /* * Slice external blocks: * ID 0 => base calls (insertions, soft-clip) * ID 1 => qualities * ID 2 => names * ID 3 => TS (insert size), NP (next frag) * ID 4 => tag values * ID 6 => tag IDs (TN), if CRAM_V1.0 * ID 7 => TD tag dictionary, if !CRAM_V1.0 */ /* Create cram slice header */ s->hdr->ref_base_id = embed_ref ? DS_ref : -1; s->hdr->record_counter = c->num_records + c->record_counter; c->num_records += s->hdr->num_records; s->block = calloc(DS_END, sizeof(s->block[0])); s->hdr->block_content_ids = malloc(DS_END * sizeof(int32_t)); if (!s->block || !s->hdr->block_content_ids) return -1; // Create first fixed blocks, always external. // CORE if (!(s->block[0] = cram_new_block(CORE, 0))) return -1; // TN block for CRAM v1 if (CRAM_MAJOR_VERS(fd->version) == 1) { if (h->codecs[DS_TN]->codec == E_EXTERNAL) { if (!(s->block[DS_TN] = cram_new_block(EXTERNAL,DS_TN))) return -1; h->codecs[DS_TN]->external.content_id = DS_TN; } else { s->block[DS_TN] = s->block[0]; } s->block[DS_TN] = s->block[DS_TN]; } // Embedded reference if (embed_ref) { if (!(s->block[DS_ref] = cram_new_block(EXTERNAL, DS_ref))) return -1; s->ref_id = DS_ref; // needed? BLOCK_APPEND(s->block[DS_ref], c->ref + c->first_base - c->ref_start, c->last_base - c->first_base + 1); } /* * All the data-series blocks if appropriate. */ for (id = DS_BF; id < DS_TN; id++) { if (h->codecs[id] && (h->codecs[id]->codec == E_EXTERNAL || h->codecs[id]->codec == E_BYTE_ARRAY_STOP || h->codecs[id]->codec == E_BYTE_ARRAY_LEN)) { switch (h->codecs[id]->codec) { case E_EXTERNAL: if (!(s->block[id] = cram_new_block(EXTERNAL, id))) return -1; h->codecs[id]->external.content_id = id; break; case E_BYTE_ARRAY_STOP: if (!(s->block[id] = cram_new_block(EXTERNAL, id))) return -1; h->codecs[id]->byte_array_stop.content_id = id; break; case E_BYTE_ARRAY_LEN: { cram_codec *cc; cc = h->codecs[id]->e_byte_array_len.len_codec; if (cc->codec == E_EXTERNAL) { int eid = cc->external.content_id; if (!(s->block[eid] = cram_new_block(EXTERNAL, eid))) return -1; cc->external.content_id = eid; cc->out = s->block[eid]; } cc = h->codecs[id]->e_byte_array_len.val_codec; if (cc->codec == E_EXTERNAL) { int eid = cc->external.content_id; if (!s->block[eid]) if (!(s->block[eid] = cram_new_block(EXTERNAL, eid))) return -1; cc->external.content_id = eid; cc->out = s->block[eid]; } break; } default: break; } } else { if (!(id == DS_BB && !h->codecs[DS_BB])) s->block[id] = s->block[0]; } if (h->codecs[id]) h->codecs[id]->out = s->block[id]; } /* Encode reads */ last_pos = s->hdr->ref_seq_start; for (rec = 0; rec < s->hdr->num_records; rec++) { cram_record *cr = &s->crecs[rec]; if (cram_encode_slice_read(fd, c, h, s, cr, &last_pos) == -1) return -1; } s->block[0]->uncomp_size = s->block[0]->byte + (s->block[0]->bit < 7); s->block[0]->comp_size = s->block[0]->uncomp_size; // Make sure the fixed blocks point to the correct sources s->block[DS_IN] = s->base_blk; s->base_blk = NULL; s->block[DS_QS] = s->qual_blk; s->qual_blk = NULL; s->block[DS_RN] = s->name_blk; s->name_blk = NULL; s->block[DS_SC] = s->soft_blk; s->soft_blk = NULL; s->block[DS_aux]= s->aux_blk; s->aux_blk = NULL; s->block[DS_aux_OQ]= s->aux_OQ_blk; s->aux_OQ_blk = NULL; s->block[DS_aux_BQ]= s->aux_BQ_blk; s->aux_BQ_blk = NULL; s->block[DS_aux_BD]= s->aux_BD_blk; s->aux_BD_blk = NULL; s->block[DS_aux_BI]= s->aux_BI_blk; s->aux_BI_blk = NULL; s->block[DS_aux_FZ]= s->aux_FZ_blk; s->aux_FZ_blk = NULL; s->block[DS_aux_oq]= s->aux_oq_blk; s->aux_oq_blk = NULL; s->block[DS_aux_os]= s->aux_os_blk; s->aux_os_blk = NULL; s->block[DS_aux_oz]= s->aux_oz_blk; s->aux_oz_blk = NULL; // Ensure block sizes are up to date. for (id = 1; id < DS_END; id++) { if (!s->block[id] || s->block[id] == s->block[0]) continue; if (s->block[id]->uncomp_size == 0) BLOCK_UPLEN(s->block[id]); } // Compress it all if (cram_compress_slice(fd, s) == -1) return -1; // Collapse empty blocks and create hdr_block { int i, j; for (i = j = 1; i < DS_END; i++) { if (!s->block[i] || s->block[i] == s->block[0]) continue; if (s->block[i]->uncomp_size == 0) { cram_free_block(s->block[i]); s->block[i] = NULL; continue; } s->block[j] = s->block[i]; s->hdr->block_content_ids[j-1] = s->block[i]->content_id; j++; } s->hdr->num_content_ids = j-1; s->hdr->num_blocks = j; if (!(s->hdr_block = cram_encode_slice_header(fd, s))) return -1; } return r ? -1 : 0; } /* * Encodes all slices in a container into blocks. * Returns 0 on success * -1 on failure */ int cram_encode_container(cram_fd *fd, cram_container *c) { int i, j, slice_offset; cram_block_compression_hdr *h = c->comp_hdr; cram_block *c_hdr; int multi_ref = 0; int r1, r2, sn, nref; spare_bams *spares; /* Cache references up-front if we have unsorted access patterns */ pthread_mutex_lock(&fd->ref_lock); nref = fd->refs->nref; pthread_mutex_unlock(&fd->ref_lock); if (!fd->no_ref && c->refs_used) { for (i = 0; i < nref; i++) { if (c->refs_used[i]) cram_get_ref(fd, i, 1, 0); } } /* To create M5 strings */ /* Fetch reference sequence */ if (!fd->no_ref) { bam_seq_t *b = c->bams[0]; char *ref; ref = cram_get_ref(fd, bam_ref(b), 1, 0); if (!ref && bam_ref(b) >= 0) { fprintf(stderr, "Failed to load reference #%d\n", bam_ref(b)); return -1; } if ((c->ref_id = bam_ref(b)) >= 0) { c->ref_seq_id = c->ref_id; c->ref = fd->refs->ref_id[c->ref_seq_id]->seq; c->ref_start = 1; c->ref_end = fd->refs->ref_id[c->ref_seq_id]->length; } else { c->ref_seq_id = c->ref_id; // FIXME remove one var! } } else { c->ref_id = bam_ref(c->bams[0]); cram_ref_incr(fd->refs, c->ref_id); c->ref_seq_id = c->ref_id; } /* Turn bams into cram_records and gather basic stats */ for (r1 = sn = 0; r1 < c->curr_c_rec; sn++) { cram_slice *s = c->slices[sn]; int first_base = INT_MAX, last_base = INT_MIN; assert(sn < c->curr_slice); /* FIXME: we could create our slice objects here too instead of * in cram_put_bam_seq. It's more natural here and also this is * bit is threaded so it's less work in the main thread. */ for (r2 = 0; r1 < c->curr_c_rec && r2 < c->max_rec; r1++, r2++) { cram_record *cr = &s->crecs[r2]; bam_seq_t *b = c->bams[r1]; /* If multi-ref we need to cope with changing reference per seq */ if (c->multi_seq && !fd->no_ref) { if (bam_ref(b) != c->ref_seq_id && bam_ref(b) >= 0) { if (c->ref_seq_id >= 0) cram_ref_decr(fd->refs, c->ref_seq_id); if (!cram_get_ref(fd, bam_ref(b), 1, 0)) { fprintf(stderr, "Failed to load reference #%d\n", bam_ref(b)); return -1; } c->ref_seq_id = bam_ref(b); // overwritten later by -2 assert(fd->refs->ref_id[c->ref_seq_id]->seq); c->ref = fd->refs->ref_id[c->ref_seq_id]->seq; c->ref_start = 1; c->ref_end = fd->refs->ref_id[c->ref_seq_id]->length; } } if (process_one_read(fd, c, s, cr, b, r2) != 0) return -1; if (first_base > cr->apos) first_base = cr->apos; if (last_base < cr->aend) last_base = cr->aend; } if (c->multi_seq) { s->hdr->ref_seq_id = -2; s->hdr->ref_seq_start = 0; s->hdr->ref_seq_span = 0; } else { s->hdr->ref_seq_id = c->ref_id; s->hdr->ref_seq_start = first_base; s->hdr->ref_seq_span = last_base - first_base + 1; } s->hdr->num_records = r2; } if (c->multi_seq && !fd->no_ref) { if (c->ref_seq_id >= 0) cram_ref_decr(fd->refs, c->ref_seq_id); } /* Link our bams[] array onto the spare bam list for reuse */ spares = malloc(sizeof(*spares)); pthread_mutex_lock(&fd->bam_list_lock); spares->bams = c->bams; spares->next = fd->bl; fd->bl = spares; pthread_mutex_unlock(&fd->bam_list_lock); c->bams = NULL; /* Detect if a multi-seq container */ cram_stats_encoding(fd, c->stats[DS_RI]); multi_ref = c->stats[DS_RI]->nvals > 1; if (multi_ref) { if (fd->verbose) fprintf(stderr, "Multi-ref container\n"); c->ref_seq_id = -2; c->ref_seq_start = 0; c->ref_seq_span = 0; } /* Compute MD5s */ for (i = 0; i < c->curr_slice; i++) { cram_slice *s = c->slices[i]; if (CRAM_MAJOR_VERS(fd->version) != 1) { if (s->hdr->ref_seq_id >= 0 && c->multi_seq == 0 && !fd->no_ref) { hts_md5_context *md5 = hts_md5_init(); if (!md5) return -1; hts_md5_update(md5, c->ref + s->hdr->ref_seq_start - c->ref_start, s->hdr->ref_seq_span); hts_md5_final(s->hdr->md5, md5); hts_md5_destroy(md5); } else { memset(s->hdr->md5, 0, 16); } } } c->num_records = 0; c->num_blocks = 1; // cram_block_compression_hdr c->length = 0; //fprintf(stderr, "=== BF ===\n"); h->codecs[DS_BF] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_BF]), c->stats[DS_BF], E_INT, NULL, fd->version); //fprintf(stderr, "=== CF ===\n"); h->codecs[DS_CF] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_CF]), c->stats[DS_CF], E_INT, NULL, fd->version); // fprintf(stderr, "=== RN ===\n"); // h->codecs[DS_RN] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_RN]), // c->stats[DS_RN], E_BYTE_ARRAY, NULL, // fd->version); //fprintf(stderr, "=== AP ===\n"); if (c->pos_sorted) { h->codecs[DS_AP] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_AP]), c->stats[DS_AP], E_INT, NULL, fd->version); } else { int p[2] = {0, c->max_apos}; h->codecs[DS_AP] = cram_encoder_init(E_BETA, NULL, E_INT, p, fd->version); } //fprintf(stderr, "=== RG ===\n"); h->codecs[DS_RG] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_RG]), c->stats[DS_RG], E_INT, NULL, fd->version); //fprintf(stderr, "=== MQ ===\n"); h->codecs[DS_MQ] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_MQ]), c->stats[DS_MQ], E_INT, NULL, fd->version); //fprintf(stderr, "=== NS ===\n"); h->codecs[DS_NS] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_NS]), c->stats[DS_NS], E_INT, NULL, fd->version); //fprintf(stderr, "=== MF ===\n"); h->codecs[DS_MF] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_MF]), c->stats[DS_MF], E_INT, NULL, fd->version); //fprintf(stderr, "=== TS ===\n"); h->codecs[DS_TS] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_TS]), c->stats[DS_TS], E_INT, NULL, fd->version); //fprintf(stderr, "=== NP ===\n"); h->codecs[DS_NP] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_NP]), c->stats[DS_NP], E_INT, NULL, fd->version); //fprintf(stderr, "=== NF ===\n"); h->codecs[DS_NF] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_NF]), c->stats[DS_NF], E_INT, NULL, fd->version); //fprintf(stderr, "=== RL ===\n"); h->codecs[DS_RL] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_RL]), c->stats[DS_RL], E_INT, NULL, fd->version); //fprintf(stderr, "=== FN ===\n"); h->codecs[DS_FN] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_FN]), c->stats[DS_FN], E_INT, NULL, fd->version); //fprintf(stderr, "=== FC ===\n"); h->codecs[DS_FC] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_FC]), c->stats[DS_FC], E_BYTE, NULL, fd->version); //fprintf(stderr, "=== FP ===\n"); h->codecs[DS_FP] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_FP]), c->stats[DS_FP], E_INT, NULL, fd->version); //fprintf(stderr, "=== DL ===\n"); h->codecs[DS_DL] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_DL]), c->stats[DS_DL], E_INT, NULL, fd->version); //fprintf(stderr, "=== BA ===\n"); h->codecs[DS_BA] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_BA]), c->stats[DS_BA], E_BYTE, NULL, fd->version); if (CRAM_MAJOR_VERS(fd->version) >= 3) { cram_byte_array_len_encoder e; e.len_encoding = E_EXTERNAL; e.len_dat = (void *)DS_BB_len; //e.len_dat = (void *)DS_BB; e.val_encoding = E_EXTERNAL; e.val_dat = (void *)DS_BB; h->codecs[DS_BB] = cram_encoder_init(E_BYTE_ARRAY_LEN, NULL, E_BYTE_ARRAY, (void *)&e, fd->version); } else { h->codecs[DS_BB] = NULL; } //fprintf(stderr, "=== BS ===\n"); h->codecs[DS_BS] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_BS]), c->stats[DS_BS], E_BYTE, NULL, fd->version); if (CRAM_MAJOR_VERS(fd->version) == 1) { h->codecs[DS_TL] = NULL; h->codecs[DS_RI] = NULL; h->codecs[DS_RS] = NULL; h->codecs[DS_PD] = NULL; h->codecs[DS_HC] = NULL; h->codecs[DS_SC] = NULL; //fprintf(stderr, "=== TC ===\n"); h->codecs[DS_TC] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_TC]), c->stats[DS_TC], E_BYTE, NULL, fd->version); //fprintf(stderr, "=== TN ===\n"); h->codecs[DS_TN] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_TN]), c->stats[DS_TN], E_INT, NULL, fd->version); } else { h->codecs[DS_TC] = NULL; h->codecs[DS_TN] = NULL; //fprintf(stderr, "=== TL ===\n"); h->codecs[DS_TL] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_TL]), c->stats[DS_TL], E_INT, NULL, fd->version); //fprintf(stderr, "=== RI ===\n"); h->codecs[DS_RI] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_RI]), c->stats[DS_RI], E_INT, NULL, fd->version); //fprintf(stderr, "=== RS ===\n"); h->codecs[DS_RS] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_RS]), c->stats[DS_RS], E_INT, NULL, fd->version); //fprintf(stderr, "=== PD ===\n"); h->codecs[DS_PD] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_PD]), c->stats[DS_PD], E_INT, NULL, fd->version); //fprintf(stderr, "=== HC ===\n"); h->codecs[DS_HC] = cram_encoder_init(cram_stats_encoding(fd, c->stats[DS_HC]), c->stats[DS_HC], E_INT, NULL, fd->version); //fprintf(stderr, "=== SC ===\n"); if (1) { int i2[2] = {0, DS_SC}; h->codecs[DS_SC] = cram_encoder_init(E_BYTE_ARRAY_STOP, NULL, E_BYTE_ARRAY, (void *)i2, fd->version); } else { // Appears to be no practical benefit to using this method, // but it may work better if we start mixing SC, IN and BB // elements into the same external block. cram_byte_array_len_encoder e; e.len_encoding = E_EXTERNAL; e.len_dat = (void *)DS_SC_len; e.val_encoding = E_EXTERNAL; e.val_dat = (void *)DS_SC; h->codecs[DS_SC] = cram_encoder_init(E_BYTE_ARRAY_LEN, NULL, E_BYTE_ARRAY, (void *)&e, fd->version); } } //fprintf(stderr, "=== IN ===\n"); { int i2[2] = {0, DS_IN}; h->codecs[DS_IN] = cram_encoder_init(E_BYTE_ARRAY_STOP, NULL, E_BYTE_ARRAY, (void *)i2, fd->version); } h->codecs[DS_QS] = cram_encoder_init(E_EXTERNAL, NULL, E_BYTE, (void *)DS_QS, fd->version); { int i2[2] = {0, DS_RN}; h->codecs[DS_RN] = cram_encoder_init(E_BYTE_ARRAY_STOP, NULL, E_BYTE_ARRAY, (void *)i2, fd->version); } /* Encode slices */ for (i = 0; i < c->curr_slice; i++) { if (fd->verbose) fprintf(stderr, "Encode slice %d\n", i); if (cram_encode_slice(fd, c, h, c->slices[i]) != 0) return -1; } /* Create compression header */ { h->ref_seq_id = c->ref_seq_id; h->ref_seq_start = c->ref_seq_start; h->ref_seq_span = c->ref_seq_span; h->num_records = c->num_records; h->mapped_qs_included = 0; // fixme h->unmapped_qs_included = 0; // fixme h->AP_delta = c->pos_sorted; // h->... fixme memcpy(h->substitution_matrix, CRAM_SUBST_MATRIX, 20); if (!(c_hdr = cram_encode_compression_header(fd, c, h))) return -1; } /* Compute landmarks */ /* Fill out slice landmarks */ c->num_landmarks = c->curr_slice; c->landmark = malloc(c->num_landmarks * sizeof(*c->landmark)); if (!c->landmark) return -1; /* * Slice offset starts after the first block, so we need to simulate * writing it to work out the correct offset */ { slice_offset = c_hdr->method == RAW ? c_hdr->uncomp_size : c_hdr->comp_size; slice_offset += 2 + 4*(CRAM_MAJOR_VERS(fd->version) >= 3) + itf8_size(c_hdr->content_id) + itf8_size(c_hdr->comp_size) + itf8_size(c_hdr->uncomp_size); } c->ref_seq_id = c->slices[0]->hdr->ref_seq_id; c->ref_seq_start = c->slices[0]->hdr->ref_seq_start; c->ref_seq_span = c->slices[0]->hdr->ref_seq_span; for (i = 0; i < c->curr_slice; i++) { cram_slice *s = c->slices[i]; c->num_blocks += s->hdr->num_blocks + 1; // slice header c->landmark[i] = slice_offset; if (s->hdr->ref_seq_start + s->hdr->ref_seq_span > c->ref_seq_start + c->ref_seq_span) { c->ref_seq_span = s->hdr->ref_seq_start + s->hdr->ref_seq_span - c->ref_seq_start; } slice_offset += s->hdr_block->method == RAW ? s->hdr_block->uncomp_size : s->hdr_block->comp_size; slice_offset += 2 + 4*(CRAM_MAJOR_VERS(fd->version) >= 3) + itf8_size(s->hdr_block->content_id) + itf8_size(s->hdr_block->comp_size) + itf8_size(s->hdr_block->uncomp_size); for (j = 0; j < s->hdr->num_blocks; j++) { slice_offset += 2 + 4*(CRAM_MAJOR_VERS(fd->version) >= 3) + itf8_size(s->block[j]->content_id) + itf8_size(s->block[j]->comp_size) + itf8_size(s->block[j]->uncomp_size); slice_offset += s->block[j]->method == RAW ? s->block[j]->uncomp_size : s->block[j]->comp_size; } } c->length += slice_offset; // just past the final slice c->comp_hdr_block = c_hdr; if (c->ref_seq_id >= 0) { cram_ref_decr(fd->refs, c->ref_seq_id); } /* Cache references up-front if we have unsorted access patterns */ if (!fd->no_ref && c->refs_used) { for (i = 0; i < fd->refs->nref; i++) { if (c->refs_used[i]) cram_ref_decr(fd->refs, i); } } return 0; } /* * Adds a feature code to a read within a slice. For purposes of minimising * memory allocations and fragmentation we have one array of features for all * reads within the slice. We return the index into this array for this new * feature. * * Returns feature index on success * -1 on failure. */ static int cram_add_feature(cram_container *c, cram_slice *s, cram_record *r, cram_feature *f) { if (s->nfeatures >= s->afeatures) { s->afeatures = s->afeatures ? s->afeatures*2 : 1024; s->features = realloc(s->features, s->afeatures*sizeof(*s->features)); if (!s->features) return -1; } if (!r->nfeature++) { r->feature = s->nfeatures; cram_stats_add(c->stats[DS_FP], f->X.pos); } else { cram_stats_add(c->stats[DS_FP], f->X.pos - s->features[r->feature + r->nfeature-2].X.pos); } cram_stats_add(c->stats[DS_FC], f->X.code); s->features[s->nfeatures++] = *f; return 0; } static int cram_add_substitution(cram_fd *fd, cram_container *c, cram_slice *s, cram_record *r, int pos, char base, char qual, char ref) { cram_feature f; // seq=ACGTN vs ref=ACGT or seq=ACGT vs ref=ACGTN if (fd->L2[(uc)base]<4 || (fd->L2[(uc)base]<5 && fd->L2[(uc)ref]<4)) { f.X.pos = pos+1; f.X.code = 'X'; f.X.base = fd->cram_sub_matrix[ref&0x1f][base&0x1f]; cram_stats_add(c->stats[DS_BS], f.X.base); } else { f.B.pos = pos+1; f.B.code = 'B'; f.B.base = base; f.B.qual = qual; cram_stats_add(c->stats[DS_BA], f.B.base); cram_stats_add(c->stats[DS_QS], f.B.qual); BLOCK_APPEND_CHAR(s->qual_blk, qual); } return cram_add_feature(c, s, r, &f); } static int cram_add_bases(cram_fd *fd, cram_container *c, cram_slice *s, cram_record *r, int pos, int len, char *base) { cram_feature f; f.b.pos = pos+1; f.b.code = 'b'; f.b.seq_idx = base - (char *)BLOCK_DATA(s->seqs_blk); f.b.len = len; return cram_add_feature(c, s, r, &f); } static int cram_add_base(cram_fd *fd, cram_container *c, cram_slice *s, cram_record *r, int pos, char base, char qual) { cram_feature f; f.B.pos = pos+1; f.B.code = 'B'; f.B.base = base; f.B.qual = qual; cram_stats_add(c->stats[DS_BA], base); cram_stats_add(c->stats[DS_QS], qual); BLOCK_APPEND_CHAR(s->qual_blk, qual); return cram_add_feature(c, s, r, &f); } static int cram_add_quality(cram_fd *fd, cram_container *c, cram_slice *s, cram_record *r, int pos, char qual) { cram_feature f; f.Q.pos = pos+1; f.Q.code = 'Q'; f.Q.qual = qual; cram_stats_add(c->stats[DS_QS], qual); BLOCK_APPEND_CHAR(s->qual_blk, qual); return cram_add_feature(c, s, r, &f); } static int cram_add_deletion(cram_container *c, cram_slice *s, cram_record *r, int pos, int len, char *base) { cram_feature f; f.D.pos = pos+1; f.D.code = 'D'; f.D.len = len; cram_stats_add(c->stats[DS_DL], len); return cram_add_feature(c, s, r, &f); } static int cram_add_softclip(cram_container *c, cram_slice *s, cram_record *r, int pos, int len, char *base, int version) { cram_feature f; f.S.pos = pos+1; f.S.code = 'S'; f.S.len = len; switch (CRAM_MAJOR_VERS(version)) { case 1: f.S.seq_idx = BLOCK_SIZE(s->base_blk); BLOCK_APPEND(s->base_blk, base, len); BLOCK_APPEND_CHAR(s->base_blk, '\0'); break; case 2: default: f.S.seq_idx = BLOCK_SIZE(s->soft_blk); if (base) { BLOCK_APPEND(s->soft_blk, base, len); } else { int i; for (i = 0; i < len; i++) BLOCK_APPEND_CHAR(s->soft_blk, 'N'); } BLOCK_APPEND_CHAR(s->soft_blk, '\0'); break; // default: // // v3.0 onwards uses BB data-series // f.S.seq_idx = BLOCK_SIZE(s->soft_blk); } return cram_add_feature(c, s, r, &f); } static int cram_add_hardclip(cram_container *c, cram_slice *s, cram_record *r, int pos, int len, char *base) { cram_feature f; f.S.pos = pos+1; f.S.code = 'H'; f.S.len = len; cram_stats_add(c->stats[DS_HC], len); return cram_add_feature(c, s, r, &f); } static int cram_add_skip(cram_container *c, cram_slice *s, cram_record *r, int pos, int len, char *base) { cram_feature f; f.S.pos = pos+1; f.S.code = 'N'; f.S.len = len; cram_stats_add(c->stats[DS_RS], len); return cram_add_feature(c, s, r, &f); } static int cram_add_pad(cram_container *c, cram_slice *s, cram_record *r, int pos, int len, char *base) { cram_feature f; f.S.pos = pos+1; f.S.code = 'P'; f.S.len = len; cram_stats_add(c->stats[DS_PD], len); return cram_add_feature(c, s, r, &f); } static int cram_add_insertion(cram_container *c, cram_slice *s, cram_record *r, int pos, int len, char *base) { cram_feature f; f.I.pos = pos+1; if (len == 1) { char b = base ? *base : 'N'; f.i.code = 'i'; f.i.base = b; cram_stats_add(c->stats[DS_BA], b); } else { f.I.code = 'I'; f.I.len = len; f.S.seq_idx = BLOCK_SIZE(s->base_blk); if (base) { BLOCK_APPEND(s->base_blk, base, len); } else { int i; for (i = 0; i < len; i++) BLOCK_APPEND_CHAR(s->base_blk, 'N'); } BLOCK_APPEND_CHAR(s->base_blk, '\0'); } return cram_add_feature(c, s, r, &f); } /* * Encodes auxiliary data. * Returns the read-group parsed out of the BAM aux fields on success * NULL on failure or no rg present (FIXME) */ static char *cram_encode_aux_1_0(cram_fd *fd, bam_seq_t *b, cram_container *c, cram_slice *s, cram_record *cr) { char *aux, *tmp, *rg = NULL; int aux_size = bam_blk_size(b) - ((char *)bam_aux(b) - (char *)&bam_ref(b)); /* Worst case is 1 nul char on every ??:Z: string, so +33% */ BLOCK_GROW(s->aux_blk, aux_size*1.34+1); tmp = (char *)BLOCK_END(s->aux_blk); aux = (char *)bam_aux(b); cr->TN_idx = s->nTN; while (aux[0] != 0) { int32_t i32; int r; if (aux[0] == 'R' && aux[1] == 'G' && aux[2] == 'Z') { rg = &aux[3]; while (*aux++); continue; } if (aux[0] == 'M' && aux[1] == 'D' && aux[2] == 'Z') { while (*aux++); continue; } if (aux[0] == 'N' && aux[1] == 'M') { switch(aux[2]) { case 'A': case 'C': case 'c': aux+=4; break; case 'I': case 'i': case 'f': aux+=7; break; default: fprintf(stderr, "Unhandled type code for NM tag\n"); return NULL; } continue; } cr->ntags++; i32 = (aux[0]<<16) | (aux[1]<<8) | aux[2]; kh_put(s_i2i, c->tags_used, i32, &r); if (-1 == r) return NULL; if (s->nTN >= s->aTN) { s->aTN = s->aTN ? s->aTN*2 : 1024; if (!(s->TN = realloc(s->TN, s->aTN * sizeof(*s->TN)))) return NULL; } s->TN[s->nTN++] = i32; cram_stats_add(c->stats[DS_TN], i32); switch(aux[2]) { case 'A': case 'C': case 'c': aux+=3; //*tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; break; case 'S': case 's': aux+=3; //*tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; break; case 'I': case 'i': case 'f': aux+=3; //*tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; break; case 'd': aux+=3; //*tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; break; case 'Z': case 'H': aux+=3; //*tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; while ((*tmp++=*aux++)); *tmp++ = '\t'; // stop byte break; case 'B': { int type = aux[3], blen; uint32_t count = (uint32_t)((((unsigned char *)aux)[4]<< 0) + (((unsigned char *)aux)[5]<< 8) + (((unsigned char *)aux)[6]<<16) + (((unsigned char *)aux)[7]<<24)); // skip TN field aux+=3; //*tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; // We use BYTE_ARRAY_LEN with external length, so store that first switch (type) { case 'c': case 'C': blen = count; break; case 's': case 'S': blen = 2*count; break; case 'i': case 'I': case 'f': blen = 4*count; break; default: fprintf(stderr, "Unknown sub-type '%c' for aux type 'B'\n", type); return NULL; } tmp += itf8_put(tmp, blen+5); *tmp++=*aux++; // sub-type & length *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; // The tag data itself memcpy(tmp, aux, blen); tmp += blen; aux += blen; //cram_stats_add(c->aux_B_stats, blen); break; } default: fprintf(stderr, "Unknown aux type '%c'\n", aux[2]); return NULL; } } cram_stats_add(c->stats[DS_TC], cr->ntags); cr->aux = BLOCK_SIZE(s->aux_blk); cr->aux_size = (uc *)tmp - (BLOCK_DATA(s->aux_blk) + cr->aux); BLOCK_SIZE(s->aux_blk) = (uc *)tmp - BLOCK_DATA(s->aux_blk); assert(s->aux_blk->byte <= s->aux_blk->alloc); return rg; } /* * Encodes auxiliary data. Largely duplicated from above, but done so to * keep it simple and avoid a myriad of version ifs. * * Returns the read-group parsed out of the BAM aux fields on success * NULL on failure or no rg present (FIXME) */ static char *cram_encode_aux(cram_fd *fd, bam_seq_t *b, cram_container *c, cram_slice *s, cram_record *cr) { char *aux, *orig, *tmp, *rg = NULL; int aux_size = bam_get_l_aux(b); cram_block *td_b = c->comp_hdr->TD_blk; int TD_blk_size = BLOCK_SIZE(td_b), new; char *key; khint_t k; /* Worst case is 1 nul char on every ??:Z: string, so +33% */ BLOCK_GROW(s->aux_blk, aux_size*1.34+1); tmp = (char *)BLOCK_END(s->aux_blk); orig = aux = (char *)bam_aux(b); // Copy aux keys to td_b and aux values to s->aux_blk while (aux - orig < aux_size && aux[0] != 0) { uint32_t i32; int r; // RG:Z if (aux[0] == 'R' && aux[1] == 'G' && aux[2] == 'Z') { rg = &aux[3]; while (*aux++); continue; } // MD:Z if (aux[0] == 'M' && aux[1] == 'D' && aux[2] == 'Z') { if (cr->len && !fd->no_ref && !(cr->flags & BAM_FUNMAP)) { while (*aux++); continue; } } // NM:i if (aux[0] == 'N' && aux[1] == 'M') { if (cr->len && !fd->no_ref && !(cr->flags & BAM_FUNMAP)) { switch(aux[2]) { case 'A': case 'C': case 'c': aux+=4; break; case 'S': case 's': aux+=5; break; case 'I': case 'i': case 'f': aux+=7; break; default: fprintf(stderr, "Unhandled type code for NM tag\n"); return NULL; } continue; } } BLOCK_APPEND(td_b, aux, 3); i32 = (aux[0]<<16) | (aux[1]<<8) | aux[2]; kh_put(s_i2i, c->tags_used, i32, &r); if (-1 == r) return NULL; // BQ:Z if (aux[0] == 'B' && aux[1] == 'Q' && aux[2] == 'Z') { char *tmp; if (!s->aux_BQ_blk) if (!(s->aux_BQ_blk = cram_new_block(EXTERNAL, DS_aux_BQ))) return NULL; BLOCK_GROW(s->aux_BQ_blk, aux_size*1.34+1); tmp = (char *)BLOCK_END(s->aux_BQ_blk); aux += 3; while ((*tmp++=*aux++)); *tmp++ = '\t'; BLOCK_SIZE(s->aux_BQ_blk) = (uc *)tmp - BLOCK_DATA(s->aux_BQ_blk); continue; } // BD:Z if (aux[0] == 'B' && aux[1]=='D' && aux[2] == 'Z') { char *tmp; if (!s->aux_BD_blk) if (!(s->aux_BD_blk = cram_new_block(EXTERNAL, DS_aux_BD))) return NULL; BLOCK_GROW(s->aux_BD_blk, aux_size*1.34+1); tmp = (char *)BLOCK_END(s->aux_BD_blk); aux += 3; while ((*tmp++=*aux++)); *tmp++ = '\t'; BLOCK_SIZE(s->aux_BD_blk) = (uc *)tmp - BLOCK_DATA(s->aux_BD_blk); continue; } // BI:Z if (aux[0] == 'B' && aux[1]=='I' && aux[2] == 'Z') { char *tmp; if (!s->aux_BI_blk) if (!(s->aux_BI_blk = cram_new_block(EXTERNAL, DS_aux_BI))) return NULL; BLOCK_GROW(s->aux_BI_blk, aux_size*1.34+1); tmp = (char *)BLOCK_END(s->aux_BI_blk); aux += 3; while ((*tmp++=*aux++)); *tmp++ = '\t'; BLOCK_SIZE(s->aux_BI_blk) = (uc *)tmp - BLOCK_DATA(s->aux_BI_blk); continue; } // OQ:Z: if (aux[0] == 'O' && aux[1] == 'Q' && aux[2] == 'Z') { char *tmp; if (!s->aux_OQ_blk) if (!(s->aux_OQ_blk = cram_new_block(EXTERNAL, DS_aux_OQ))) return NULL; BLOCK_GROW(s->aux_OQ_blk, aux_size*1.34+1); tmp = (char *)BLOCK_END(s->aux_OQ_blk); aux += 3; while ((*tmp++=*aux++)); *tmp++ = '\t'; BLOCK_SIZE(s->aux_OQ_blk) = (uc *)tmp - BLOCK_DATA(s->aux_OQ_blk); continue; } // FZ:B or ZM:B if ((aux[0] == 'F' && aux[1] == 'Z' && aux[2] == 'B') || (aux[0] == 'Z' && aux[1] == 'M' && aux[2] == 'B')) { int type = aux[3], blen; uint32_t count = (uint32_t)((((unsigned char *)aux)[4]<< 0) + (((unsigned char *)aux)[5]<< 8) + (((unsigned char *)aux)[6]<<16) + (((unsigned char *)aux)[7]<<24)); char *tmp; if (!s->aux_FZ_blk) if (!(s->aux_FZ_blk = cram_new_block(EXTERNAL, DS_aux_FZ))) return NULL; BLOCK_GROW(s->aux_FZ_blk, aux_size*1.34+1); tmp = (char *)BLOCK_END(s->aux_FZ_blk); // skip TN field aux+=3; // We use BYTE_ARRAY_LEN with external length, so store that first switch (type) { case 'c': case 'C': blen = count; break; case 's': case 'S': blen = 2*count; break; case 'i': case 'I': case 'f': blen = 4*count; break; default: fprintf(stderr, "Unknown sub-type '%c' for aux type 'B'\n", type); return NULL; } blen += 5; // sub-type & length tmp += itf8_put(tmp, blen); // The tag data itself memcpy(tmp, aux, blen); tmp += blen; aux += blen; BLOCK_SIZE(s->aux_FZ_blk) = (uc *)tmp - BLOCK_DATA(s->aux_FZ_blk); continue; } // Other quality data - {Q2,E2,U2,CQ}:Z and similar if (((aux[0] == 'Q' && aux[1] == '2') || (aux[0] == 'U' && aux[1] == '2') || (aux[0] == 'Q' && aux[1] == 'T') || (aux[0] == 'C' && aux[1] == 'Q')) && aux[2] == 'Z') { char *tmp; if (!s->aux_oq_blk) if (!(s->aux_oq_blk = cram_new_block(EXTERNAL, DS_aux_oq))) return NULL; BLOCK_GROW(s->aux_oq_blk, aux_size*1.34+1); tmp = (char *)BLOCK_END(s->aux_oq_blk); aux += 3; while ((*tmp++=*aux++)); *tmp++ = '\t'; BLOCK_SIZE(s->aux_oq_blk) = (uc *)tmp - BLOCK_DATA(s->aux_oq_blk); continue; } // Other sequence data - {R2,E2,CS,BC,RT}:Z and similar if (((aux[0] == 'R' && aux[1] == '2') || (aux[0] == 'E' && aux[1] == '2') || (aux[0] == 'C' && aux[1] == 'S') || (aux[0] == 'B' && aux[1] == 'C') || (aux[0] == 'R' && aux[1] == 'T')) && aux[2] == 'Z') { char *tmp; if (!s->aux_os_blk) if (!(s->aux_os_blk = cram_new_block(EXTERNAL, DS_aux_os))) return NULL; BLOCK_GROW(s->aux_os_blk, aux_size*1.34+1); tmp = (char *)BLOCK_END(s->aux_os_blk); aux += 3; while ((*tmp++=*aux++)); *tmp++ = '\t'; BLOCK_SIZE(s->aux_os_blk) = (uc *)tmp - BLOCK_DATA(s->aux_os_blk); continue; } switch(aux[2]) { case 'A': case 'C': case 'c': aux+=3; *tmp++=*aux++; break; case 'S': case 's': aux+=3; *tmp++=*aux++; *tmp++=*aux++; break; case 'I': case 'i': case 'f': aux+=3; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; break; case 'd': aux+=3; //*tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; *tmp++=*aux++; break; case 'Z': case 'H': { char *tmp; if (!s->aux_oz_blk) if (!(s->aux_oz_blk = cram_new_block(EXTERNAL, DS_aux_oz))) return NULL; BLOCK_GROW(s->aux_oz_blk, aux_size*1.34+1); tmp = (char *)BLOCK_END(s->aux_oz_blk); aux += 3; while ((*tmp++=*aux++)); *tmp++ = '\t'; BLOCK_SIZE(s->aux_oz_blk) = (uc *)tmp - BLOCK_DATA(s->aux_oz_blk); } break; case 'B': { int type = aux[3], blen; uint32_t count = (uint32_t)((((unsigned char *)aux)[4]<< 0) + (((unsigned char *)aux)[5]<< 8) + (((unsigned char *)aux)[6]<<16) + (((unsigned char *)aux)[7]<<24)); // skip TN field aux+=3; // We use BYTE_ARRAY_LEN with external length, so store that first switch (type) { case 'c': case 'C': blen = count; break; case 's': case 'S': blen = 2*count; break; case 'i': case 'I': case 'f': blen = 4*count; break; default: fprintf(stderr, "Unknown sub-type '%c' for aux type 'B'\n", type); return NULL; } blen += 5; // sub-type & length tmp += itf8_put(tmp, blen); // The tag data itself memcpy(tmp, aux, blen); tmp += blen; aux += blen; //cram_stats_add(c->aux_B_stats, blen); break; } default: fprintf(stderr, "Unknown aux type '%c'\n", aux[2]); return NULL; } } // FIXME: sort BLOCK_DATA(td_b) by char[3] triples // And and increment TD hash entry BLOCK_APPEND_CHAR(td_b, 0); // Duplicate key as BLOCK_DATA() can be realloced to a new pointer. key = string_ndup(c->comp_hdr->TD_keys, (char *)BLOCK_DATA(td_b) + TD_blk_size, BLOCK_SIZE(td_b) - TD_blk_size); k = kh_put(m_s2i, c->comp_hdr->TD_hash, key, &new); if (new < 0) { return NULL; } else if (new == 0) { BLOCK_SIZE(td_b) = TD_blk_size; } else { kh_val(c->comp_hdr->TD_hash, k) = c->comp_hdr->nTL; c->comp_hdr->nTL++; } cr->TL = kh_val(c->comp_hdr->TD_hash, k); cram_stats_add(c->stats[DS_TL], cr->TL); cr->aux = BLOCK_SIZE(s->aux_blk); cr->aux_size = (uc *)tmp - (BLOCK_DATA(s->aux_blk) + cr->aux); BLOCK_SIZE(s->aux_blk) = (uc *)tmp - BLOCK_DATA(s->aux_blk); assert(s->aux_blk->byte <= s->aux_blk->alloc); return rg; } /* * Handles creation of a new container or new slice, flushing any * existing containers when appropriate. * * Really this is next slice, which may or may not lead to a new container. * * Returns cram_container pointer on success * NULL on failure. */ static cram_container *cram_next_container(cram_fd *fd, bam_seq_t *b) { cram_container *c = fd->ctr; cram_slice *s; int i; /* First occurence */ if (c->curr_ref == -2) c->curr_ref = bam_ref(b); if (c->slice) { s = c->slice; if (c->multi_seq) { s->hdr->ref_seq_id = -2; s->hdr->ref_seq_start = 0; s->hdr->ref_seq_span = 0; } else { s->hdr->ref_seq_id = c->curr_ref; s->hdr->ref_seq_start = c->first_base; s->hdr->ref_seq_span = c->last_base - c->first_base + 1; } s->hdr->num_records = c->curr_rec; if (c->curr_slice == 0) { if (c->ref_seq_id != s->hdr->ref_seq_id) c->ref_seq_id = s->hdr->ref_seq_id; c->ref_seq_start = c->first_base; } c->curr_slice++; } /* Flush container */ if (c->curr_slice == c->max_slice || (bam_ref(b) != c->curr_ref && !c->multi_seq)) { c->ref_seq_span = fd->last_base - c->ref_seq_start + 1; if (fd->verbose) fprintf(stderr, "Flush container %d/%d..%d\n", c->ref_seq_id, c->ref_seq_start, c->ref_seq_start + c->ref_seq_span -1); /* Encode slices */ if (fd->pool) { if (-1 == cram_flush_container_mt(fd, c)) return NULL; } else { if (-1 == cram_flush_container(fd, c)) return NULL; // Move to sep func, as we need cram_flush_container for // the closing phase to flush the partial container. for (i = 0; i < c->max_slice; i++) { cram_free_slice(c->slices[i]); c->slices[i] = NULL; } c->slice = NULL; c->curr_slice = 0; /* Easy approach for purposes of freeing stats */ cram_free_container(c); } c = fd->ctr = cram_new_container(fd->seqs_per_slice, fd->slices_per_container); if (!c) return NULL; c->record_counter = fd->record_counter; c->curr_ref = bam_ref(b); } c->last_pos = c->first_base = c->last_base = bam_pos(b)+1; /* New slice */ c->slice = c->slices[c->curr_slice] = cram_new_slice(MAPPED_SLICE, c->max_rec); if (!c->slice) return NULL; if (c->multi_seq) { c->slice->hdr->ref_seq_id = -2; c->slice->hdr->ref_seq_start = 0; c->slice->last_apos = 1; } else { c->slice->hdr->ref_seq_id = bam_ref(b); // wrong for unsorted data, will fix during encoding. c->slice->hdr->ref_seq_start = bam_pos(b)+1; c->slice->last_apos = bam_pos(b)+1; } c->curr_rec = 0; return c; } /* * Converts a single bam record into a cram record. * Possibly used within a thread. * * Returns 0 on success; * -1 on failure */ static int process_one_read(cram_fd *fd, cram_container *c, cram_slice *s, cram_record *cr, bam_seq_t *b, int rnum) { int i, fake_qual = -1; char *cp, *rg; char *ref, *seq, *qual; // FIXME: multi-ref containers ref = c->ref; cr->flags = bam_flag(b); cr->len = bam_seq_len(b); //fprintf(stderr, "%s => %d\n", rg ? rg : "\"\"", cr->rg); // Fields to resolve later //cr->mate_line; // index to another cram_record //cr->mate_flags; // MF //cr->ntags; // TC cr->ntags = 0; //cram_stats_add(c->stats[DS_TC], cr->ntags); if (CRAM_MAJOR_VERS(fd->version) == 1) rg = cram_encode_aux_1_0(fd, b, c, s, cr); else rg = cram_encode_aux(fd, b, c, s, cr); //cr->aux_size = b->blk_size - ((char *)bam_aux(b) - (char *)&bam_ref(b)); //cr->aux = DSTRING_LEN(s->aux_ds); //dstring_nappend(s->aux_ds, bam_aux(b), cr->aux_size); /* Read group, identified earlier */ if (rg) { SAM_RG *brg = sam_hdr_find_rg(fd->header, rg); cr->rg = brg ? brg->id : -1; } else if (CRAM_MAJOR_VERS(fd->version) == 1) { SAM_RG *brg = sam_hdr_find_rg(fd->header, "UNKNOWN"); assert(brg); } else { cr->rg = -1; } cram_stats_add(c->stats[DS_RG], cr->rg); cr->ref_id = bam_ref(b); cram_stats_add(c->stats[DS_RI], cr->ref_id); cram_stats_add(c->stats[DS_BF], fd->cram_flag_swap[cr->flags & 0xfff]); // Non reference based encoding means storing the bases verbatim as features, which in // turn means every base also has a quality already stored. if (!fd->no_ref || CRAM_MAJOR_VERS(fd->version) >= 3) cr->cram_flags = CRAM_FLAG_PRESERVE_QUAL_SCORES; else cr->cram_flags = 0; if (cr->len <= 0 && CRAM_MAJOR_VERS(fd->version) >= 3) cr->cram_flags |= CRAM_FLAG_NO_SEQ; //cram_stats_add(c->stats[DS_CF], cr->cram_flags & CRAM_FLAG_MASK); c->num_bases += cr->len; cr->apos = bam_pos(b)+1; if (c->pos_sorted) { if (cr->apos < s->last_apos) { c->pos_sorted = 0; } else { cram_stats_add(c->stats[DS_AP], cr->apos - s->last_apos); s->last_apos = cr->apos; } } else { //cram_stats_add(c->stats[DS_AP], cr->apos); } c->max_apos += (cr->apos > c->max_apos) * (cr->apos - c->max_apos); cr->name = BLOCK_SIZE(s->name_blk); cr->name_len = bam_name_len(b); cram_stats_add(c->stats[DS_RN], cr->name_len); BLOCK_APPEND(s->name_blk, bam_name(b), bam_name_len(b)); /* * This seqs_ds is largely pointless and it could reuse the same memory * over and over. * s->base_blk is what we need for encoding. */ cr->seq = BLOCK_SIZE(s->seqs_blk); cr->qual = BLOCK_SIZE(s->qual_blk); BLOCK_GROW(s->seqs_blk, cr->len+1); BLOCK_GROW(s->qual_blk, cr->len); seq = cp = (char *)BLOCK_END(s->seqs_blk); *seq = 0; #ifdef ALLOW_UAC { // Convert seq 2 bases at a time for speed. static const uint16_t code2base[256] = { 15677, 16701, 17213, 19773, 18237, 21053, 21309, 22077, 21565, 22333, 22845, 18493, 19261, 17469, 16957, 20029, 15681, 16705, 17217, 19777, 18241, 21057, 21313, 22081, 21569, 22337, 22849, 18497, 19265, 17473, 16961, 20033, 15683, 16707, 17219, 19779, 18243, 21059, 21315, 22083, 21571, 22339, 22851, 18499, 19267, 17475, 16963, 20035, 15693, 16717, 17229, 19789, 18253, 21069, 21325, 22093, 21581, 22349, 22861, 18509, 19277, 17485, 16973, 20045, 15687, 16711, 17223, 19783, 18247, 21063, 21319, 22087, 21575, 22343, 22855, 18503, 19271, 17479, 16967, 20039, 15698, 16722, 17234, 19794, 18258, 21074, 21330, 22098, 21586, 22354, 22866, 18514, 19282, 17490, 16978, 20050, 15699, 16723, 17235, 19795, 18259, 21075, 21331, 22099, 21587, 22355, 22867, 18515, 19283, 17491, 16979, 20051, 15702, 16726, 17238, 19798, 18262, 21078, 21334, 22102, 21590, 22358, 22870, 18518, 19286, 17494, 16982, 20054, 15700, 16724, 17236, 19796, 18260, 21076, 21332, 22100, 21588, 22356, 22868, 18516, 19284, 17492, 16980, 20052, 15703, 16727, 17239, 19799, 18263, 21079, 21335, 22103, 21591, 22359, 22871, 18519, 19287, 17495, 16983, 20055, 15705, 16729, 17241, 19801, 18265, 21081, 21337, 22105, 21593, 22361, 22873, 18521, 19289, 17497, 16985, 20057, 15688, 16712, 17224, 19784, 18248, 21064, 21320, 22088, 21576, 22344, 22856, 18504, 19272, 17480, 16968, 20040, 15691, 16715, 17227, 19787, 18251, 21067, 21323, 22091, 21579, 22347, 22859, 18507, 19275, 17483, 16971, 20043, 15684, 16708, 17220, 19780, 18244, 21060, 21316, 22084, 21572, 22340, 22852, 18500, 19268, 17476, 16964, 20036, 15682, 16706, 17218, 19778, 18242, 21058, 21314, 22082, 21570, 22338, 22850, 18498, 19266, 17474, 16962, 20034, 15694, 16718, 17230, 19790, 18254, 21070, 21326, 22094, 21582, 22350, 22862, 18510, 19278, 17486, 16974, 20046 }; int l2 = cr->len / 2; unsigned char *from = (unsigned char *)bam_seq(b); uint16_t *cpi = (uint16_t *)cp; cp[0] = 0; for (i = 0; i < l2; i++) cpi[i] = le_int2(code2base[from[i]]); if ((i *= 2) < cr->len) cp[i] = seq_nt16_str[bam_seqi(bam_seq(b), i)]; } #else for (i = 0; i < cr->len; i++) cp[i] = seq_nt16_str[bam_seqi(bam_seq(b), i)]; #endif BLOCK_SIZE(s->seqs_blk) += cr->len; qual = cp = (char *)bam_qual(b); /* Copy and parse */ if (!(cr->flags & BAM_FUNMAP)) { uint32_t *cig_to, *cig_from; int apos = cr->apos-1, spos = 0; cr->cigar = s->ncigar; cr->ncigar = bam_cigar_len(b); while (cr->cigar + cr->ncigar >= s->cigar_alloc) { s->cigar_alloc = s->cigar_alloc ? s->cigar_alloc*2 : 1024; s->cigar = realloc(s->cigar, s->cigar_alloc * sizeof(*s->cigar)); if (!s->cigar) return -1; } cig_to = (uint32_t *)s->cigar; cig_from = (uint32_t *)bam_cigar(b); cr->feature = 0; cr->nfeature = 0; for (i = 0; i < cr->ncigar; i++) { enum cigar_op cig_op = cig_from[i] & BAM_CIGAR_MASK; uint32_t cig_len = cig_from[i] >> BAM_CIGAR_SHIFT; cig_to[i] = cig_from[i]; /* Can also generate events from here for CRAM diffs */ switch (cig_op) { int l; // Don't trust = and X ops to be correct. case BAM_CMATCH: case BAM_CBASE_MATCH: case BAM_CBASE_MISMATCH: //fprintf(stderr, "\nBAM_CMATCH\nR: %.*s\nS: %.*s\n", // cig_len, &ref[apos], cig_len, &seq[spos]); l = 0; if (!fd->no_ref && cr->len) { int end = cig_len+apos < c->ref_end ? cig_len : c->ref_end - apos; char *sp = &seq[spos]; char *rp = &ref[apos]; char *qp = &qual[spos]; if (end > cr->len) { fprintf(stderr, "CIGAR and query sequence are of " "different length\n"); return -1; } for (l = 0; l < end; l++) { if (rp[l] != sp[l]) { if (!sp[l]) break; if (0 && CRAM_MAJOR_VERS(fd->version) >= 3) { // Disabled for the time being as it doesn't // seem to gain us much. int ol=l; while (l 1) { if (cram_add_bases(fd, c, s, cr, spos+ol, l-ol, &seq[spos+ol])) return -1; l--; } else { l = ol; if (cram_add_substitution(fd, c, s, cr, spos+l, sp[l], qp[l], rp[l])) return -1; } } else { if (cram_add_substitution(fd, c, s, cr, spos+l, sp[l], qp[l], rp[l])) return -1; } } } spos += l; apos += l; } if (l < cig_len && cr->len) { if (fd->no_ref) { if (CRAM_MAJOR_VERS(fd->version) == 3) { if (cram_add_bases(fd, c, s, cr, spos, cig_len-l, &seq[spos])) return -1; spos += cig_len-l; } else { for (; l < cig_len && seq[spos]; l++, spos++) { if (cram_add_base(fd, c, s, cr, spos, seq[spos], qual[spos])) return -1; } } } else { /* off end of sequence or non-ref based output */ for (; l < cig_len && seq[spos]; l++, spos++) { if (cram_add_base(fd, c, s, cr, spos, seq[spos], qual[spos])) return -1; } } apos += cig_len; } else if (!cr->len) { /* Seq "*" */ apos += cig_len; spos += cig_len; } break; case BAM_CDEL: if (cram_add_deletion(c, s, cr, spos, cig_len, &seq[spos])) return -1; apos += cig_len; break; case BAM_CREF_SKIP: if (cram_add_skip(c, s, cr, spos, cig_len, &seq[spos])) return -1; apos += cig_len; break; case BAM_CINS: if (cram_add_insertion(c, s, cr, spos, cig_len, cr->len ? &seq[spos] : NULL)) return -1; if (fd->no_ref && cr->len) { for (l = 0; l < cig_len; l++, spos++) { cram_add_quality(fd, c, s, cr, spos, qual[spos]); } } else { spos += cig_len; } break; case BAM_CSOFT_CLIP: if (cram_add_softclip(c, s, cr, spos, cig_len, cr->len ? &seq[spos] : NULL, fd->version)) return -1; if (fd->no_ref && !(cr->cram_flags & CRAM_FLAG_PRESERVE_QUAL_SCORES)) { if (cr->len) { for (l = 0; l < cig_len; l++, spos++) { cram_add_quality(fd, c, s, cr, spos, qual[spos]); } } else { for (l = 0; l < cig_len; l++, spos++) { cram_add_quality(fd, c, s, cr, spos, -1); } } } else { spos += cig_len; } break; case BAM_CHARD_CLIP: if (cram_add_hardclip(c, s, cr, spos, cig_len, &seq[spos])) return -1; break; case BAM_CPAD: if (cram_add_pad(c, s, cr, spos, cig_len, &seq[spos])) return -1; break; } } if (cr->len && spos != cr->len) { fprintf(stderr, "CIGAR and query sequence are of different " "length\n"); return -1; } fake_qual = spos; cr->aend = fd->no_ref ? apos : MIN(apos, c->ref_end); cram_stats_add(c->stats[DS_FN], cr->nfeature); } else { // Unmapped cr->cram_flags |= CRAM_FLAG_PRESERVE_QUAL_SCORES; cr->cigar = 0; cr->ncigar = 0; cr->nfeature = 0; cr->aend = cr->apos; for (i = 0; i < cr->len; i++) cram_stats_add(c->stats[DS_BA], seq[i]); fake_qual = 0; } /* * Append to the qual block now. We do this here as * cram_add_substitution() can generate BA/QS events which need to * be in the qual block before we append the rest of the data. */ if (cr->cram_flags & CRAM_FLAG_PRESERVE_QUAL_SCORES) { /* Special case of seq "*" */ if (cr->len == 0) { cr->len = fake_qual; BLOCK_GROW(s->qual_blk, cr->len); cp = (char *)BLOCK_END(s->qual_blk); memset(cp, 255, cr->len); } else { BLOCK_GROW(s->qual_blk, cr->len); cp = (char *)BLOCK_END(s->qual_blk); char *from = (char *)&bam_qual(b)[0]; char *to = &cp[0]; memcpy(to, from, cr->len); //for (i = 0; i < cr->len; i++) cp[i] = from[i]; } BLOCK_SIZE(s->qual_blk) += cr->len; } else { if (cr->len == 0) cr->len = fake_qual >= 0 ? fake_qual : cr->aend - cr->apos + 1; } cram_stats_add(c->stats[DS_RL], cr->len); /* Now we know apos and aend both, update mate-pair information */ { int new; khint_t k; int sec = (cr->flags & BAM_FSECONDARY) ? 1 : 0; //fprintf(stderr, "Checking %"PRId64"/%.*s\t", rnum, // cr->name_len, DSTRING_STR(s->name_ds)+cr->name); if (cr->flags & BAM_FPAIRED) { char *key = string_ndup(s->pair_keys, (char *)BLOCK_DATA(s->name_blk)+cr->name, cr->name_len); if (!key) return -1; k = kh_put(m_s2i, s->pair[sec], key, &new); if (-1 == new) return -1; else if (new > 0) kh_val(s->pair[sec], k) = rnum; } else { new = 1; } if (new == 0) { cram_record *p = &s->crecs[kh_val(s->pair[sec], k)]; int aleft, aright, sign; aleft = MIN(cr->apos, p->apos); aright = MAX(cr->aend, p->aend); if (cr->apos < p->apos) { sign = 1; } else if (cr->apos > p->apos) { sign = -1; } else if (cr->flags & BAM_FREAD1) { sign = 1; } else { sign = -1; } //fprintf(stderr, "paired %"PRId64"\n", kh_val(s->pair[sec], k)); // This vs p: tlen, matepos, flags if (bam_ins_size(b) != sign*(aright-aleft+1)) goto detached; if (MAX(bam_mate_pos(b)+1, 0) != p->apos) goto detached; if (((bam_flag(b) & BAM_FMUNMAP) != 0) != ((p->flags & BAM_FUNMAP) != 0)) goto detached; if (((bam_flag(b) & BAM_FMREVERSE) != 0) != ((p->flags & BAM_FREVERSE) != 0)) goto detached; // p vs this: tlen, matepos, flags if (p->ref_id != cr->ref_id) goto detached; if (p->tlen != -sign*(aright-aleft+1)) goto detached; if (p->mate_pos != cr->apos) goto detached; if (((p->flags & BAM_FMUNMAP) != 0) != ((p->mate_flags & CRAM_M_UNMAP) != 0)) goto detached; if (((p->flags & BAM_FMREVERSE) != 0) != ((p->mate_flags & CRAM_M_REVERSE) != 0)) goto detached; // Supplementary reads are just too ill defined if ((cr->flags & BAM_FSUPPLEMENTARY) || (p->flags & BAM_FSUPPLEMENTARY)) goto detached; /* * The fields below are unused when encoding this read as it is * no longer detached. In theory they may get referred to when * processing a 3rd or 4th read in this template?, so we set them * here just to be sure. * * They do not need cram_stats_add() calls those as they are * not emitted. */ cr->mate_pos = p->apos; cr->tlen = sign*(aright-aleft+1); cr->mate_flags = ((p->flags & BAM_FMUNMAP) == BAM_FMUNMAP) * CRAM_M_UNMAP + ((p->flags & BAM_FMREVERSE) == BAM_FMREVERSE) * CRAM_M_REVERSE; // Decrement statistics aggregated earlier if (p->cram_flags & CRAM_FLAG_STATS_ADDED) { cram_stats_del(c->stats[DS_NP], p->mate_pos); cram_stats_del(c->stats[DS_MF], p->mate_flags); cram_stats_del(c->stats[DS_TS], p->tlen); cram_stats_del(c->stats[DS_NS], p->mate_ref_id); } /* Similarly we could correct the p-> values too, but these will no * longer have any code that refers back to them as the new 'p' * for this template is our current 'cr'. */ //p->mate_pos = cr->apos; //p->mate_flags = // ((cr->flags & BAM_FMUNMAP) == BAM_FMUNMAP) * CRAM_M_UNMAP + // ((cr->flags & BAM_FMREVERSE) == BAM_FMREVERSE)* CRAM_M_REVERSE; //p->tlen = p->apos - cr->aend; // Clear detached from cr flags cr->cram_flags &= ~CRAM_FLAG_DETACHED; cram_stats_add(c->stats[DS_CF], cr->cram_flags & CRAM_FLAG_MASK); // Clear detached from p flags and set downstream if (p->cram_flags & CRAM_FLAG_STATS_ADDED) { cram_stats_del(c->stats[DS_CF], p->cram_flags & CRAM_FLAG_MASK); p->cram_flags &= ~CRAM_FLAG_STATS_ADDED; } p->cram_flags &= ~CRAM_FLAG_DETACHED; p->cram_flags |= CRAM_FLAG_MATE_DOWNSTREAM; cram_stats_add(c->stats[DS_CF], p->cram_flags & CRAM_FLAG_MASK); p->mate_line = rnum - (kh_val(s->pair[sec], k) + 1); cram_stats_add(c->stats[DS_NF], p->mate_line); kh_val(s->pair[sec], k) = rnum; } else { detached: //fprintf(stderr, "unpaired\n"); /* Derive mate flags from this flag */ cr->mate_flags = 0; if (bam_flag(b) & BAM_FMUNMAP) cr->mate_flags |= CRAM_M_UNMAP; if (bam_flag(b) & BAM_FMREVERSE) cr->mate_flags |= CRAM_M_REVERSE; cram_stats_add(c->stats[DS_MF], cr->mate_flags); cr->mate_pos = MAX(bam_mate_pos(b)+1, 0); cram_stats_add(c->stats[DS_NP], cr->mate_pos); cr->tlen = bam_ins_size(b); cram_stats_add(c->stats[DS_TS], cr->tlen); cr->cram_flags |= CRAM_FLAG_DETACHED; cram_stats_add(c->stats[DS_CF], cr->cram_flags & CRAM_FLAG_MASK); cram_stats_add(c->stats[DS_NS], bam_mate_ref(b)); cr->cram_flags |= CRAM_FLAG_STATS_ADDED; } } cr->mqual = bam_map_qual(b); cram_stats_add(c->stats[DS_MQ], cr->mqual); cr->mate_ref_id = bam_mate_ref(b); if (!(bam_flag(b) & BAM_FUNMAP)) { if (c->first_base > cr->apos) c->first_base = cr->apos; if (c->last_base < cr->aend) c->last_base = cr->aend; } return 0; } /* * Write iterator: put BAM format sequences into a CRAM file. * We buffer up a containers worth of data at a time. * * Returns 0 on success * -1 on failure */ int cram_put_bam_seq(cram_fd *fd, bam_seq_t *b) { cram_container *c; if (!fd->ctr) { fd->ctr = cram_new_container(fd->seqs_per_slice, fd->slices_per_container); if (!fd->ctr) return -1; fd->ctr->record_counter = fd->record_counter; } c = fd->ctr; if (!c->slice || c->curr_rec == c->max_rec || (bam_ref(b) != c->curr_ref && c->curr_ref >= -1)) { int slice_rec, curr_rec, multi_seq = fd->multi_seq == 1; int curr_ref = c->slice ? c->curr_ref : bam_ref(b); /* * Start packing slices when we routinely have under 1/4tr full. * * This option isn't available if we choose to embed references * since we can only have one per slice. */ if (fd->multi_seq == -1 && c->curr_rec < c->max_rec/4+10 && fd->last_slice && fd->last_slice < c->max_rec/4+10 && !fd->embed_ref) { if (fd->verbose && !c->multi_seq) fprintf(stderr, "Multi-ref enabled for this container\n"); multi_seq = 1; } slice_rec = c->slice_rec; curr_rec = c->curr_rec; if (CRAM_MAJOR_VERS(fd->version) == 1 || c->curr_rec == c->max_rec || fd->multi_seq != 1 || !c->slice) { if (NULL == (c = cram_next_container(fd, b))) { if (fd->ctr) { // prevent cram_close attempting to flush cram_free_container(fd->ctr); fd->ctr = NULL; } return -1; } } /* * Due to our processing order, some things we've already done we * cannot easily undo. So when we first notice we should be packing * multiple sequences per container we emit the small partial * container as-is and then start a fresh one in a different mode. */ if (multi_seq) { fd->multi_seq = 1; c->multi_seq = 1; c->pos_sorted = 0; // required atm for multi_seq slices if (!c->refs_used) { pthread_mutex_lock(&fd->ref_lock); c->refs_used = calloc(fd->refs->nref, sizeof(int)); pthread_mutex_unlock(&fd->ref_lock); if (!c->refs_used) return -1; } } fd->last_slice = curr_rec - slice_rec; c->slice_rec = c->curr_rec; // Have we seen this reference before? if (bam_ref(b) >= 0 && bam_ref(b) != curr_ref && !fd->embed_ref && !fd->unsorted && multi_seq) { if (!c->refs_used) { pthread_mutex_lock(&fd->ref_lock); c->refs_used = calloc(fd->refs->nref, sizeof(int)); pthread_mutex_unlock(&fd->ref_lock); if (!c->refs_used) return -1; } else if (c->refs_used && c->refs_used[bam_ref(b)]) { pthread_mutex_lock(&fd->ref_lock); fd->unsorted = 1; pthread_mutex_unlock(&fd->ref_lock); fd->multi_seq = 1; } } c->curr_ref = bam_ref(b); if (c->refs_used && c->curr_ref >= 0) c->refs_used[c->curr_ref]++; } if (!c->bams) { /* First time through, allocate a set of bam pointers */ pthread_mutex_lock(&fd->bam_list_lock); if (fd->bl) { spare_bams *spare = fd->bl; c->bams = spare->bams; fd->bl = spare->next; free(spare); } else { c->bams = calloc(c->max_c_rec, sizeof(bam_seq_t *)); if (!c->bams) return -1; } pthread_mutex_unlock(&fd->bam_list_lock); } /* Copy or alloc+copy the bam record, for later encoding */ if (c->bams[c->curr_c_rec]) bam_copy1(c->bams[c->curr_c_rec], b); else c->bams[c->curr_c_rec] = bam_dup(b); c->curr_rec++; c->curr_c_rec++; fd->record_counter++; return 0; } stacks-1.44/htslib/cram/._cram_encode.h000644 000765 000024 00000000265 12750743246 020553 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/cram_encode.h000644 000765 000024 00000006363 12750743246 020343 0ustar00catchenstaff000000 000000 /* Copyright (c) 2012-2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /*! \file * Include cram.h instead. * * This is an internal part of the CRAM system and is automatically included * when you #include cram.h. * * Implements the encoding portion of CRAM I/O. Also see * cram_codecs.[ch] for the actual encoding functions themselves. */ #ifndef _CRAM_WRITE_H_ #define _CRAM_WRITE_H_ #ifdef __cplusplus extern "C" { #endif /* ---------------------------------------------------------------------- * CRAM sequence iterators. */ /*! Write iterator: put BAM format sequences into a CRAM file. * * We buffer up a containers worth of data at a time. * * FIXME: break this into smaller pieces. * * @return * Returns 0 on success; * -1 on failure */ int cram_put_bam_seq(cram_fd *fd, bam_seq_t *b); /* ---------------------------------------------------------------------- * Internal functions */ /*! INTERNAL: * Encodes a compression header block into a generic cram_block structure. * * @return * Returns cram_block ptr on success; * NULL on failure */ cram_block *cram_encode_compression_header(cram_fd *fd, cram_container *c, cram_block_compression_hdr *h); /*! INTERNAL: * Encodes a slice compression header. * * @return * Returns cram_block on success; * NULL on failure */ cram_block *cram_encode_slice_header(cram_fd *fd, cram_slice *s); /*! INTERNAL: * Encodes all slices in a container into blocks. * * @return * Returns 0 on success; * -1 on failure * * FIXME: separate into encode_container and write_container. Ideally * we should be able to do read_container / write_container or * decode_container / encode_container. */ int cram_encode_container(cram_fd *fd, cram_container *c); #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/cram/._cram_external.c000644 000765 000024 00000000265 12750743246 021133 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/cram_external.c000644 000765 000024 00000030306 12750743246 020715 0ustar00catchenstaff000000 000000 /* Copyright (c) 2015 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /*! \file * External CRAM interface. * * Internally we're happy to use macros and to grub around in the cram * structures. This isn't very sustainable for an externally usable * ABI though, so we have anonymous structs and accessor functions too * to permit software such as samtools reheader to manipulate cram * containers and blocks in a robust manner. */ #include #include "htslib/hfile.h" #include "cram/cram.h" /* *----------------------------------------------------------------------------- * cram_fd */ SAM_hdr *cram_fd_get_header(cram_fd *fd) { return fd->header; } void cram_fd_set_header(cram_fd *fd, SAM_hdr *hdr) { fd->header = hdr; } int cram_fd_get_version(cram_fd *fd) { return fd->version; } void cram_fd_set_version(cram_fd *fd, int vers) { fd->version = vers; } int cram_major_vers(cram_fd *fd) { return CRAM_MAJOR_VERS(fd->version); } int cram_minor_vers(cram_fd *fd) { return CRAM_MINOR_VERS(fd->version); } hFILE *cram_fd_get_fp(cram_fd *fd) { return fd->fp; } void cram_fd_set_fp(cram_fd *fd, hFILE *fp) { fd->fp = fp; } /* *----------------------------------------------------------------------------- * cram_container */ int32_t cram_container_get_length(cram_container *c) { return c->length; } void cram_container_set_length(cram_container *c, int32_t length) { c->length = length; } int32_t cram_container_get_num_blocks(cram_container *c) { return c->num_blocks; } void cram_container_set_num_blocks(cram_container *c, int32_t num_blocks) { c->num_blocks = num_blocks; } /* Returns the landmarks[] array and the number of elements * in num_landmarks. */ int32_t *cram_container_get_landmarks(cram_container *c, int32_t *num_landmarks) { *num_landmarks = c->num_landmarks; return c->landmark; } /* Sets the landmarks[] array (pointer copy, not a memory dup) and * num_landmarks value. */ void cram_container_set_landmarks(cram_container *c, int32_t num_landmarks, int32_t *landmarks) { c->num_landmarks = num_landmarks; c->landmark = landmarks; } /* Returns true if the container is empty (EOF marker) */ int cram_container_is_empty(cram_fd *fd) { return fd->empty_container; } /* *----------------------------------------------------------------------------- * cram_block_compression_hdr */ /* * Utility function to edit an RG id. * This is only possible if there is one single RG value used and it * is in the container compression header using HUFFMAN or BETA * codec. In this case it is essentially hard coded and needs no * editing of external (or worse, CORE) blocks. * * Returns 0 on success * -1 on failure */ // Or arbitrary set compression header constant? static int cram_block_compression_hdr_set_DS(cram_block_compression_hdr *ch, int ds, int new_rg) { if (!ch || !ch->codecs[ds]) return -1; switch (ch->codecs[ds]->codec) { case E_HUFFMAN: if (ch->codecs[ds]->huffman.ncodes != 1) return -1; ch->codecs[ds]->huffman.codes[0].symbol = new_rg; return 0; case E_BETA: if (ch->codecs[ds]->beta.nbits != 0) return -1; ch->codecs[ds]->beta.offset = -new_rg; return 0; default: return -1; } return 0; } int cram_block_compression_hdr_set_rg(cram_block_compression_hdr *ch, int new_rg) { return cram_block_compression_hdr_set_DS(ch, DS_RG, new_rg); } /* * Converts a cram_block_compression_hdr struct used for decoding to * one used for encoding. Maybe this should be a transparent * operation applied on-demand. * * Returns 0 on success * -1 on failure */ int cram_block_compression_hdr_decoder2encoder(cram_fd *fd, cram_block_compression_hdr *ch) { int i; if (!ch) return -1; for (i = 0; i < DS_END; i++) { cram_codec *co = ch->codecs[i]; if (!co) continue; if (-1 == cram_codec_decoder2encoder(fd, co)) return -1; } return 0; } /* *----------------------------------------------------------------------------- * cram_slice */ int32_t cram_slice_hdr_get_num_blocks(cram_block_slice_hdr *hdr) { return hdr->num_blocks; } /* *----------------------------------------------------------------------------- * cram_block */ int32_t cram_block_get_content_id(cram_block *b) { return b->content_id; } int32_t cram_block_get_comp_size(cram_block *b) { return b->comp_size; } int32_t cram_block_get_uncomp_size(cram_block *b) { return b->uncomp_size; } int32_t cram_block_get_crc32(cram_block *b) { return b->crc32; } void * cram_block_get_data(cram_block *b) { return BLOCK_DATA(b); } int32_t cram_block_get_size(cram_block *b) { return BLOCK_SIZE(b); } enum cram_content_type cram_block_get_content_type(cram_block *b) { return b->content_type; } void cram_block_set_content_id(cram_block *b, int32_t id) { b->content_id = id; } void cram_block_set_comp_size(cram_block *b, int32_t size) { b->comp_size = size; } void cram_block_set_uncomp_size(cram_block *b, int32_t size) { b->uncomp_size = size; } void cram_block_set_crc32(cram_block *b, int32_t crc) { b->crc32 = crc; } void cram_block_set_data(cram_block *b, void *data) { BLOCK_DATA(b) = data; } void cram_block_set_size(cram_block *b, int32_t size) { BLOCK_SIZE(b) = size; } int cram_block_append(cram_block *b, void *data, int size) { BLOCK_APPEND(b, data, size); return BLOCK_DATA(b) ? 0 : -1; // It'll do for now... } void cram_block_update_size(cram_block *b) { BLOCK_UPLEN(b); } // Offset is known as "size" internally, but it can be confusing. size_t cram_block_get_offset(cram_block *b) { return BLOCK_SIZE(b); } void cram_block_set_offset(cram_block *b, size_t offset) { BLOCK_SIZE(b) = offset; } /* * Copies the blocks representing the next num_slice slices from a * container from 'in' to 'out'. It is expected that the file pointer * is just after the read of the cram_container and cram compression * header. * * Returns 0 on success * -1 on failure */ int cram_copy_slice(cram_fd *in, cram_fd *out, int32_t num_slice) { int32_t i, j; for (i = 0; i < num_slice; i++) { cram_block *blk; cram_block_slice_hdr *hdr; if (!(blk = cram_read_block(in))) return -1; if (!(hdr = cram_decode_slice_header(in, blk))) { cram_free_block(blk); return -1; } if (cram_write_block(out, blk) != 0) { cram_free_block(blk); return -1; } cram_free_block(blk); int num_blocks = cram_slice_hdr_get_num_blocks(hdr); for (j = 0; j < num_blocks; j++) { blk = cram_read_block(in); if (!blk || cram_write_block(out, blk) != 0) { if (blk) cram_free_block(blk); return -1; } cram_free_block(blk); } cram_free_slice_header(hdr); } return 0; } /* * Renumbers RG numbers in a cram compression header. * * CRAM stores RG as the Nth number in the header, rather than a * string holding the ID: tag. This is smaller in space, but means * "samtools cat" to join files together that contain single but * different RG lines needs a way of renumbering them. * * The file descriptor is expected to be immediately after the * cram_container structure (ie before the cram compression header). * Due to the nature of the CRAM format, this needs to read and write * the blocks itself. Note that there may be multiple slices within * the container, meaning multiple compression headers to manipulate. * Changing RG may change the size of the compression header and * therefore the length field in the container. Hence we rewrite all * blocks just incase and also emit the adjusted container. * * The current implementation can only cope with renumbering a single * RG (and only then if it is using HUFFMAN or BETA codecs). In * theory it *may* be possible to renumber multiple RGs if they use * HUFFMAN to the CORE block or use an external block unshared by any * other data series. So we have an API that can be upgraded to * support this, but do not implement it for now. An example * implementation of RG as an EXTERNAL block would be to find that * block and rewrite it, returning the number of blocks consumed. * * Returns 0 on success; * -1 if unable to edit; * -2 on other errors (eg I/O). */ int cram_transcode_rg(cram_fd *in, cram_fd *out, cram_container *c, int nrg, int *in_rg, int *out_rg) { int new_rg = *out_rg, old_size, new_size; cram_block *o_blk, *n_blk; cram_block_compression_hdr *ch; if (nrg != 1) { fprintf(stderr, "[%s] ERROR: not implemented for nrg != 1\n", __func__); return -2; } // Produce a new block holding the updated compression header, // with RG transcoded to a new value. (Single only supported.) o_blk = cram_read_block(in); old_size = cram_block_size(o_blk); ch = cram_decode_compression_header(in, o_blk); if (cram_block_compression_hdr_set_rg(ch, new_rg) != 0) return -1; cram_block_compression_hdr_decoder2encoder(in, ch); n_blk = cram_encode_compression_header(in, c, ch); cram_free_compression_header(ch); /* * Warning: this has internal knowledge of the cram compression * header format. * * The decoder doesn't set c->tags_used, so the encoder puts a two * byte blank segment. This means n_blk is too short. We skip * through the decoded old block (o_blk) and copy from there. */ char *cp = cram_block_get_data(o_blk); char *op = cp; char *endp = cp + cram_block_get_uncomp_size(o_blk); //fprintf(stderr, "sz = %d\n", (int)(endp-cp)); int32_t i32; cp += safe_itf8_get(cp, endp, &i32); cp += i32; cp += safe_itf8_get(cp, endp, &i32); cp += i32; op = cp; cp += safe_itf8_get(cp, endp, &i32); i32 += (cp-op); //fprintf(stderr, "remaining %d bytes\n", i32); cram_block_set_size(n_blk, cram_block_get_size(n_blk)-2); cram_block_append(n_blk, op, i32); cram_block_update_size(n_blk); new_size = cram_block_size(n_blk); //fprintf(stderr, "size %d -> %d\n", old_size, new_size); // Now we've constructedthe updated compression header, // amend the container too (it may have changed size). int32_t *landmarks, num_landmarks; landmarks = cram_container_get_landmarks(c, &num_landmarks); if (old_size != new_size) { int diff = new_size - old_size, j; for (j = 0; j < num_landmarks; j++) landmarks[j] += diff; //cram_container_set_landmarks(c, num_landmarks, landmarks); cram_container_set_length(c, cram_container_get_length(c) + diff); } // Finally write it all out; container, compression header, // and then all the remaining slice blocks. if (cram_write_container(out, c) != 0) return -2; cram_write_block(out, n_blk); cram_free_block(o_blk); cram_free_block(n_blk); // Container num_blocks can be invalid, due to a bug. // Instead we iterate in slice context instead. return cram_copy_slice(in, out, num_landmarks); } stacks-1.44/htslib/cram/._cram_index.c000644 000765 000024 00000000265 12750743246 020420 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/cram_index.c000644 000765 000024 00000034755 12750743246 020216 0ustar00catchenstaff000000 000000 /* Copyright (c) 2013-2014 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /* * The index is a gzipped tab-delimited text file with one line per slice. * The columns are: * 1: reference number (0 to N-1, as per BAM ref_id) * 2: reference position of 1st read in slice (1..?) * 3: number of reads in slice * 4: offset of container start (relative to end of SAM header, so 1st * container is offset 0). * 5: slice number within container (ie which landmark). * * In memory, we hold this in a nested containment list. Each list element is * a cram_index struct. Each element in turn can contain its own list of * cram_index structs. * * Any start..end range which is entirely contained within another (and * earlier as it is sorted) range will be held within it. This ensures that * the outer list will never have containments and we can safely do a * binary search to find the first range which overlaps any given coordinate. */ #include #include #include #include #include #include #include #include #include #include #include #include "htslib/hfile.h" #include "hts_internal.h" #include "cram/cram.h" #include "cram/os.h" #include "cram/zfio.h" #if 0 static void dump_index_(cram_index *e, int level) { int i, n; n = printf("%*s%d / %d .. %d, ", level*4, "", e->refid, e->start, e->end); printf("%*soffset %"PRId64"\n", MAX(0,50-n), "", e->offset); for (i = 0; i < e->nslice; i++) { dump_index_(&e->e[i], level+1); } } static void dump_index(cram_fd *fd) { int i; for (i = 0; i < fd->index_sz; i++) { dump_index_(&fd->index[i], 0); } } #endif static int kget_int32(kstring_t *k, size_t *pos, int32_t *val_p) { int sign = 1; int32_t val = 0; size_t p = *pos; while (p < k->l && (k->s[p] == ' ' || k->s[p] == '\t')) p++; if (p < k->l && k->s[p] == '-') sign = -1, p++; if (p >= k->l || !(k->s[p] >= '0' && k->s[p] <= '9')) return -1; while (p < k->l && k->s[p] >= '0' && k->s[p] <= '9') val = val*10 + k->s[p++]-'0'; *pos = p; *val_p = sign*val; return 0; } static int kget_int64(kstring_t *k, size_t *pos, int64_t *val_p) { int sign = 1; int64_t val = 0; size_t p = *pos; while (p < k->l && (k->s[p] == ' ' || k->s[p] == '\t')) p++; if (p < k->l && k->s[p] == '-') sign = -1, p++; if (p >= k->l || !(k->s[p] >= '0' && k->s[p] <= '9')) return -1; while (p < k->l && k->s[p] >= '0' && k->s[p] <= '9') val = val*10 + k->s[p++]-'0'; *pos = p; *val_p = sign*val; return 0; } /* * Loads a CRAM .crai index into memory. * * Returns 0 for success * -1 for failure */ int cram_index_load(cram_fd *fd, const char *fn, const char *fn_idx) { char *fn2 = NULL; char buf[65536]; ssize_t len; kstring_t kstr = {0}; FILE *fp; cram_index *idx; cram_index **idx_stack = NULL, *ep, e; int idx_stack_alloc = 0, idx_stack_ptr = 0; size_t pos = 0; /* Check if already loaded */ if (fd->index) return 0; fd->index = calloc((fd->index_sz = 1), sizeof(*fd->index)); if (!fd->index) return -1; idx = &fd->index[0]; idx->refid = -1; idx->start = INT_MIN; idx->end = INT_MAX; idx_stack = calloc(++idx_stack_alloc, sizeof(*idx_stack)); idx_stack[idx_stack_ptr] = idx; if (!fn_idx) { fn2 = hts_idx_getfn(fn, ".crai"); if (!fn2) { free(idx_stack); return -1; } fn_idx = fn2; } if (!(fp = fopen(fn_idx, "r"))) { perror(fn_idx); free(idx_stack); free(fn2); return -1; } // Load the file into memory while ((len = fread(buf, 1, 65536, fp)) > 0) kputsn(buf, len, &kstr); if (len < 0 || kstr.l < 2) { if (kstr.s) free(kstr.s); free(idx_stack); free(fn2); return -1; } if (fclose(fp)) { if (kstr.s) free(kstr.s); free(idx_stack); free(fn2); return -1; } // Uncompress if required if (kstr.s[0] == 31 && (uc)kstr.s[1] == 139) { size_t l; char *s = zlib_mem_inflate(kstr.s, kstr.l, &l); free(kstr.s); if (!s) { free(idx_stack); free(fn2); return -1; } kstr.s = s; kstr.l = l; kstr.m = l; // conservative estimate of the size allocated kputsn("", 0, &kstr); // ensure kstr.s is NUL-terminated } // Parse it line at a time do { /* 1.1 layout */ if (kget_int32(&kstr, &pos, &e.refid) == -1) { free(kstr.s); free(idx_stack); free(fn2); return -1; } if (kget_int32(&kstr, &pos, &e.start) == -1) { free(kstr.s); free(idx_stack); free(fn2); return -1; } if (kget_int32(&kstr, &pos, &e.end) == -1) { free(kstr.s); free(idx_stack); free(fn2); return -1; } if (kget_int64(&kstr, &pos, &e.offset) == -1) { free(kstr.s); free(idx_stack); free(fn2); return -1; } if (kget_int32(&kstr, &pos, &e.slice) == -1) { free(kstr.s); free(idx_stack); free(fn2); return -1; } if (kget_int32(&kstr, &pos, &e.len) == -1) { free(kstr.s); free(idx_stack); free(fn2); return -1; } e.end += e.start-1; //printf("%d/%d..%d\n", e.refid, e.start, e.end); if (e.refid < -1) { free(kstr.s); free(idx_stack); free(fn2); fprintf(stderr, "Malformed index file, refid %d\n", e.refid); return -1; } if (e.refid != idx->refid) { if (fd->index_sz < e.refid+2) { size_t index_end = fd->index_sz * sizeof(*fd->index); fd->index_sz = e.refid+2; fd->index = realloc(fd->index, fd->index_sz * sizeof(*fd->index)); memset(((char *)fd->index) + index_end, 0, fd->index_sz * sizeof(*fd->index) - index_end); } idx = &fd->index[e.refid+1]; idx->refid = e.refid; idx->start = INT_MIN; idx->end = INT_MAX; idx->nslice = idx->nalloc = 0; idx->e = NULL; idx_stack[(idx_stack_ptr = 0)] = idx; } while (!(e.start >= idx->start && e.end <= idx->end) || idx->end == 0) { idx = idx_stack[--idx_stack_ptr]; } // Now contains, so append if (idx->nslice+1 >= idx->nalloc) { idx->nalloc = idx->nalloc ? idx->nalloc*2 : 16; idx->e = realloc(idx->e, idx->nalloc * sizeof(*idx->e)); } e.nalloc = e.nslice = 0; e.e = NULL; *(ep = &idx->e[idx->nslice++]) = e; idx = ep; if (++idx_stack_ptr >= idx_stack_alloc) { idx_stack_alloc *= 2; idx_stack = realloc(idx_stack, idx_stack_alloc*sizeof(*idx_stack)); } idx_stack[idx_stack_ptr] = idx; while (pos < kstr.l && kstr.s[pos] != '\n') pos++; pos++; } while (pos < kstr.l); free(idx_stack); free(kstr.s); free(fn2); // dump_index(fd); return 0; } static void cram_index_free_recurse(cram_index *e) { if (e->e) { int i; for (i = 0; i < e->nslice; i++) { cram_index_free_recurse(&e->e[i]); } free(e->e); } } void cram_index_free(cram_fd *fd) { int i; if (!fd->index) return; for (i = 0; i < fd->index_sz; i++) { cram_index_free_recurse(&fd->index[i]); } free(fd->index); fd->index = NULL; } /* * Searches the index for the first slice overlapping a reference ID * and position, or one immediately preceding it if none is found in * the index to overlap this position. (Our index may have missing * entries, but we require at least one per reference.) * * If the index finds multiple slices overlapping this position we * return the first one only. Subsequent calls should specifying * "from" as the last slice we checked to find the next one. Otherwise * set "from" to be NULL to find the first one. * * Returns the cram_index pointer on sucess * NULL on failure */ cram_index *cram_index_query(cram_fd *fd, int refid, int pos, cram_index *from) { int i, j, k; cram_index *e; if (refid+1 < 0 || refid+1 >= fd->index_sz) return NULL; if (!from) from = &fd->index[refid+1]; // Ref with nothing aligned against it. if (!from->e) return NULL; // This sequence is covered by the index, so binary search to find // the optimal starting block. i = 0, j = fd->index[refid+1].nslice-1; for (k = j/2; k != i; k = (j-i)/2 + i) { if (from->e[k].refid > refid) { j = k; continue; } if (from->e[k].refid < refid) { i = k; continue; } if (from->e[k].start >= pos) { j = k; continue; } if (from->e[k].start < pos) { i = k; continue; } } // i==j or i==j-1. Check if j is better. if (j >= 0 && from->e[j].start < pos && from->e[j].refid == refid) i = j; /* The above found *a* bin overlapping, but not necessarily the first */ while (i > 0 && from->e[i-1].end >= pos) i--; /* We may be one bin before the optimum, so check */ while (i+1 < from->nslice && (from->e[i].refid < refid || from->e[i].end < pos)) i++; e = &from->e[i]; return e; } /* * Skips to a container overlapping the start coordinate listed in * cram_range. * * In theory we call cram_index_query multiple times, once per slice * overlapping the range. However slices may be absent from the index * which makes this problematic. Instead we find the left-most slice * and then read from then on, skipping decoding of slices and/or * whole containers when they don't overlap the specified cram_range. * * Returns 0 on success * -1 on general failure * -2 on no-data (empty chromosome) */ int cram_seek_to_refpos(cram_fd *fd, cram_range *r) { cram_index *e; // Ideally use an index, so see if we have one. if ((e = cram_index_query(fd, r->refid, r->start, NULL))) { if (0 != cram_seek(fd, e->offset, SEEK_SET)) if (0 != cram_seek(fd, e->offset - fd->first_container, SEEK_CUR)) return -1; } else { // Absent from index, but this most likely means it simply has no data. return -2; } if (fd->ctr) { cram_free_container(fd->ctr); fd->ctr = NULL; fd->ooc = 0; } return 0; } /* * A specialised form of cram_index_build (below) that deals with slices * having multiple references in this (ref_id -2). In this scenario we * decode the slice to look at the RI data series instead. * * Returns 0 on success * -1 on failure */ static int cram_index_build_multiref(cram_fd *fd, cram_container *c, cram_slice *s, zfp *fp, off_t cpos, int32_t landmark, int sz) { int i, ref = -2, ref_start = 0, ref_end; char buf[1024]; if (0 != cram_decode_slice(fd, c, s, fd->header)) return -1; ref_end = INT_MIN; for (i = 0; i < s->hdr->num_records; i++) { if (s->crecs[i].ref_id == ref) { if (ref_end < s->crecs[i].aend) ref_end = s->crecs[i].aend; continue; } if (ref != -2) { sprintf(buf, "%d\t%d\t%d\t%"PRId64"\t%d\t%d\n", ref, ref_start, ref_end - ref_start + 1, (int64_t)cpos, landmark, sz); zfputs(buf, fp); } ref = s->crecs[i].ref_id; ref_start = s->crecs[i].apos; ref_end = INT_MIN; } if (ref != -2) { sprintf(buf, "%d\t%d\t%d\t%"PRId64"\t%d\t%d\n", ref, ref_start, ref_end - ref_start + 1, (int64_t)cpos, landmark, sz); zfputs(buf, fp); } return 0; } /* * Builds an index file. * * fd is a newly opened cram file that we wish to index. * fn_base is the filename of the associated CRAM file. * fn_idx is the filename of the index file to be written; * if NULL, we add ".crai" to fn_base to get the index filename. * * Returns 0 on success * -1 on failure */ int cram_index_build(cram_fd *fd, const char *fn_base, const char *fn_idx) { cram_container *c; off_t cpos, spos, hpos; zfp *fp; kstring_t fn_idx_str = {0}; if (! fn_idx) { kputs(fn_base, &fn_idx_str); kputs(".crai", &fn_idx_str); fn_idx = fn_idx_str.s; } if (!(fp = zfopen(fn_idx, "wz"))) { perror(fn_idx); free(fn_idx_str.s); return -1; } free(fn_idx_str.s); cpos = htell(fd->fp); while ((c = cram_read_container(fd))) { int j; if (fd->err) { perror("Cram container read"); return -1; } hpos = htell(fd->fp); if (!(c->comp_hdr_block = cram_read_block(fd))) return -1; assert(c->comp_hdr_block->content_type == COMPRESSION_HEADER); c->comp_hdr = cram_decode_compression_header(fd, c->comp_hdr_block); if (!c->comp_hdr) return -1; // 2.0 format for (j = 0; j < c->num_landmarks; j++) { char buf[1024]; cram_slice *s; int sz; spos = htell(fd->fp); assert(spos - cpos - c->offset == c->landmark[j]); if (!(s = cram_read_slice(fd))) { zfclose(fp); return -1; } sz = (int)(htell(fd->fp) - spos); if (s->hdr->ref_seq_id == -2) { cram_index_build_multiref(fd, c, s, fp, cpos, c->landmark[j], sz); } else { sprintf(buf, "%d\t%d\t%d\t%"PRId64"\t%d\t%d\n", s->hdr->ref_seq_id, s->hdr->ref_seq_start, s->hdr->ref_seq_span, (int64_t)cpos, c->landmark[j], sz); zfputs(buf, fp); } cram_free_slice(s); } cpos = htell(fd->fp); assert(cpos == hpos + c->length); cram_free_container(c); } if (fd->err) { zfclose(fp); return -1; } return (zfclose(fp) >= 0)? 0 : -1; } stacks-1.44/htslib/cram/._cram_index.h000644 000765 000024 00000000265 12750743246 020425 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/cram_index.h000644 000765 000024 00000006357 12750743246 020220 0ustar00catchenstaff000000 000000 /* Copyright (c) 2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #ifndef _CRAM_INDEX_H_ #define _CRAM_INDEX_H_ #ifdef __cplusplus extern "C" { #endif /* * Loads a CRAM .crai index into memory. * Returns 0 for success * -1 for failure */ int cram_index_load(cram_fd *fd, const char *fn, const char *fn_idx); void cram_index_free(cram_fd *fd); /* * Searches the index for the first slice overlapping a reference ID * and position. * * Returns the cram_index pointer on sucess * NULL on failure */ cram_index *cram_index_query(cram_fd *fd, int refid, int pos, cram_index *frm); /* * Skips to a container overlapping the start coordinate listed in * cram_range. * * Returns 0 on success * -1 on failure */ int cram_seek_to_refpos(cram_fd *fd, cram_range *r); void cram_index_free(cram_fd *fd); /* * Skips to a container overlapping the start coordinate listed in * cram_range. * * In theory we call cram_index_query multiple times, once per slice * overlapping the range. However slices may be absent from the index * which makes this problematic. Instead we find the left-most slice * and then read from then on, skipping decoding of slices and/or * whole containers when they don't overlap the specified cram_range. * * Returns 0 on success * -1 on failure */ int cram_seek_to_refpos(cram_fd *fd, cram_range *r); /* * Builds an index file. * * fd is a newly opened cram file that we wish to index. * fn_base is the filename of the associated CRAM file. * fn_idx is the filename of the index file to be written; * if NULL, we add ".crai" to fn_base to get the index filename. * * Returns 0 on success * -1 on failure */ int cram_index_build(cram_fd *fd, const char *fn_base, const char *fn_idx); #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/cram/._cram_io.c000644 000765 000024 00000000265 12750743246 017720 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/cram_io.c000644 000765 000024 00000334331 12750743246 017507 0ustar00catchenstaff000000 000000 /* Copyright (c) 2012-2014 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /* * CRAM I/O primitives. * * - ITF8 encoding and decoding. * - Block based I/O * - Zlib inflating and deflating (memory) * - CRAM basic data structure reading and writing * - File opening / closing * - Reference sequence handling */ /* * TODO: BLOCK_GROW, BLOCK_RESIZE, BLOCK_APPEND and itf8_put_blk all need * a way to return errors for when malloc fails. */ #include #include #include #include #include #include #include #include #ifdef HAVE_LIBBZ2 #include #endif #ifdef HAVE_LIBLZMA #include #endif #include #include #include #include #include #include "cram/cram.h" #include "cram/os.h" #include "htslib/hts.h" #include "cram/open_trace_file.h" #include "cram/rANS_static.h" //#define REF_DEBUG #ifdef REF_DEBUG #include #define gettid() (int)syscall(SYS_gettid) #define RP(...) fprintf (stderr, __VA_ARGS__) #else #define RP(...) #endif #include "htslib/hfile.h" #include "htslib/bgzf.h" #include "htslib/faidx.h" #define TRIAL_SPAN 50 #define NTRIALS 3 /* ---------------------------------------------------------------------- * ITF8 encoding and decoding. * * Also see the itf8_get and itf8_put macros in cram_io.h */ /* * LEGACY: consider using itf8_decode_crc. * * Reads an integer in ITF-8 encoding from 'cp' and stores it in * *val. * * Returns the number of bytes read on success * -1 on failure */ int itf8_decode(cram_fd *fd, int32_t *val_p) { static int nbytes[16] = { 0,0,0,0, 0,0,0,0, // 0000xxxx - 0111xxxx 1,1,1,1, // 1000xxxx - 1011xxxx 2,2, // 1100xxxx - 1101xxxx 3, // 1110xxxx 4, // 1111xxxx }; static int nbits[16] = { 0x7f, 0x7f, 0x7f, 0x7f, 0x7f, 0x7f, 0x7f, 0x7f, // 0000xxxx - 0111xxxx 0x3f, 0x3f, 0x3f, 0x3f, // 1000xxxx - 1011xxxx 0x1f, 0x1f, // 1100xxxx - 1101xxxx 0x0f, // 1110xxxx 0x0f, // 1111xxxx }; int32_t val = hgetc(fd->fp); if (val == -1) return -1; int i = nbytes[val>>4]; val &= nbits[val>>4]; switch(i) { case 0: *val_p = val; return 1; case 1: val = (val<<8) | (unsigned char)hgetc(fd->fp); *val_p = val; return 2; case 2: val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); *val_p = val; return 3; case 3: val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); *val_p = val; return 4; case 4: // really 3.5 more, why make it different? val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<4) | (((unsigned char)hgetc(fd->fp)) & 0x0f); *val_p = val; } return 5; } int itf8_decode_crc(cram_fd *fd, int32_t *val_p, uint32_t *crc) { static int nbytes[16] = { 0,0,0,0, 0,0,0,0, // 0000xxxx - 0111xxxx 1,1,1,1, // 1000xxxx - 1011xxxx 2,2, // 1100xxxx - 1101xxxx 3, // 1110xxxx 4, // 1111xxxx }; static int nbits[16] = { 0x7f, 0x7f, 0x7f, 0x7f, 0x7f, 0x7f, 0x7f, 0x7f, // 0000xxxx - 0111xxxx 0x3f, 0x3f, 0x3f, 0x3f, // 1000xxxx - 1011xxxx 0x1f, 0x1f, // 1100xxxx - 1101xxxx 0x0f, // 1110xxxx 0x0f, // 1111xxxx }; unsigned char c[5]; int32_t val = hgetc(fd->fp); if (val == -1) return -1; c[0]=val; int i = nbytes[val>>4]; val &= nbits[val>>4]; switch(i) { case 0: *val_p = val; *crc = crc32(*crc, c, 1); return 1; case 1: val = (val<<8) | (c[1]=hgetc(fd->fp)); *val_p = val; *crc = crc32(*crc, c, 2); return 2; case 2: val = (val<<8) | (c[1]=hgetc(fd->fp)); val = (val<<8) | (c[2]=hgetc(fd->fp)); *val_p = val; *crc = crc32(*crc, c, 3); return 3; case 3: val = (val<<8) | (c[1]=hgetc(fd->fp)); val = (val<<8) | (c[2]=hgetc(fd->fp)); val = (val<<8) | (c[3]=hgetc(fd->fp)); *val_p = val; *crc = crc32(*crc, c, 4); return 4; case 4: // really 3.5 more, why make it different? val = (val<<8) | (c[1]=hgetc(fd->fp)); val = (val<<8) | (c[2]=hgetc(fd->fp)); val = (val<<8) | (c[3]=hgetc(fd->fp)); val = (val<<4) | (((c[4]=hgetc(fd->fp))) & 0x0f); *val_p = val; *crc = crc32(*crc, c, 5); } return 5; } /* * Encodes and writes a single integer in ITF-8 format. * Returns 0 on success * -1 on failure */ int itf8_encode(cram_fd *fd, int32_t val) { char buf[5]; int len = itf8_put(buf, val); return hwrite(fd->fp, buf, len) == len ? 0 : -1; } const int itf8_bytes[16] = { 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 4, 5 }; #ifndef ITF8_MACROS /* * As above, but decoding from memory */ int itf8_get(char *cp, int32_t *val_p) { unsigned char *up = (unsigned char *)cp; if (up[0] < 0x80) { *val_p = up[0]; return 1; } else if (up[0] < 0xc0) { *val_p = ((up[0] <<8) | up[1]) & 0x3fff; return 2; } else if (up[0] < 0xe0) { *val_p = ((up[0]<<16) | (up[1]<< 8) | up[2]) & 0x1fffff; return 3; } else if (up[0] < 0xf0) { *val_p = ((up[0]<<24) | (up[1]<<16) | (up[2]<<8) | up[3]) & 0x0fffffff; return 4; } else { *val_p = ((up[0] & 0x0f)<<28) | (up[1]<<20) | (up[2]<<12) | (up[3]<<4) | (up[4] & 0x0f); return 5; } } /* * Stores a value to memory in ITF-8 format. * * Returns the number of bytes required to store the number. * This is a maximum of 5 bytes. */ int itf8_put(char *cp, int32_t val) { if (!(val & ~0x00000007f)) { // 1 byte *cp = val; return 1; } else if (!(val & ~0x00003fff)) { // 2 byte *cp++ = (val >> 8 ) | 0x80; *cp = val & 0xff; return 2; } else if (!(val & ~0x01fffff)) { // 3 byte *cp++ = (val >> 16) | 0xc0; *cp++ = (val >> 8 ) & 0xff; *cp = val & 0xff; return 3; } else if (!(val & ~0x0fffffff)) { // 4 byte *cp++ = (val >> 24) | 0xe0; *cp++ = (val >> 16) & 0xff; *cp++ = (val >> 8 ) & 0xff; *cp = val & 0xff; return 4; } else { // 5 byte *cp++ = 0xf0 | ((val>>28) & 0xff); *cp++ = (val >> 20) & 0xff; *cp++ = (val >> 12) & 0xff; *cp++ = (val >> 4 ) & 0xff; *cp = val & 0x0f; return 5; } } #endif /* 64-bit itf8 variant */ int ltf8_put(char *cp, int64_t val) { if (!(val & ~((1LL<<7)-1))) { *cp = val; return 1; } else if (!(val & ~((1LL<<(6+8))-1))) { *cp++ = (val >> 8 ) | 0x80; *cp = val & 0xff; return 2; } else if (!(val & ~((1LL<<(5+2*8))-1))) { *cp++ = (val >> 16) | 0xc0; *cp++ = (val >> 8 ) & 0xff; *cp = val & 0xff; return 3; } else if (!(val & ~((1LL<<(4+3*8))-1))) { *cp++ = (val >> 24) | 0xe0; *cp++ = (val >> 16) & 0xff; *cp++ = (val >> 8 ) & 0xff; *cp = val & 0xff; return 4; } else if (!(val & ~((1LL<<(3+4*8))-1))) { *cp++ = (val >> 32) | 0xf0; *cp++ = (val >> 24) & 0xff; *cp++ = (val >> 16) & 0xff; *cp++ = (val >> 8 ) & 0xff; *cp = val & 0xff; return 5; } else if (!(val & ~((1LL<<(2+5*8))-1))) { *cp++ = (val >> 40) | 0xf8; *cp++ = (val >> 32) & 0xff; *cp++ = (val >> 24) & 0xff; *cp++ = (val >> 16) & 0xff; *cp++ = (val >> 8 ) & 0xff; *cp = val & 0xff; return 6; } else if (!(val & ~((1LL<<(1+6*8))-1))) { *cp++ = (val >> 48) | 0xfc; *cp++ = (val >> 40) & 0xff; *cp++ = (val >> 32) & 0xff; *cp++ = (val >> 24) & 0xff; *cp++ = (val >> 16) & 0xff; *cp++ = (val >> 8 ) & 0xff; *cp = val & 0xff; return 7; } else if (!(val & ~((1LL<<(7*8))-1))) { *cp++ = (val >> 56) | 0xfe; *cp++ = (val >> 48) & 0xff; *cp++ = (val >> 40) & 0xff; *cp++ = (val >> 32) & 0xff; *cp++ = (val >> 24) & 0xff; *cp++ = (val >> 16) & 0xff; *cp++ = (val >> 8 ) & 0xff; *cp = val & 0xff; return 8; } else { *cp++ = 0xff; *cp++ = (val >> 56) & 0xff; *cp++ = (val >> 48) & 0xff; *cp++ = (val >> 40) & 0xff; *cp++ = (val >> 32) & 0xff; *cp++ = (val >> 24) & 0xff; *cp++ = (val >> 16) & 0xff; *cp++ = (val >> 8 ) & 0xff; *cp = val & 0xff; return 9; } } int ltf8_get(char *cp, int64_t *val_p) { unsigned char *up = (unsigned char *)cp; if (up[0] < 0x80) { *val_p = up[0]; return 1; } else if (up[0] < 0xc0) { *val_p = (((uint64_t)up[0]<< 8) | (uint64_t)up[1]) & (((1LL<<(6+8)))-1); return 2; } else if (up[0] < 0xe0) { *val_p = (((uint64_t)up[0]<<16) | ((uint64_t)up[1]<< 8) | (uint64_t)up[2]) & ((1LL<<(5+2*8))-1); return 3; } else if (up[0] < 0xf0) { *val_p = (((uint64_t)up[0]<<24) | ((uint64_t)up[1]<<16) | ((uint64_t)up[2]<< 8) | (uint64_t)up[3]) & ((1LL<<(4+3*8))-1); return 4; } else if (up[0] < 0xf8) { *val_p = (((uint64_t)up[0]<<32) | ((uint64_t)up[1]<<24) | ((uint64_t)up[2]<<16) | ((uint64_t)up[3]<< 8) | (uint64_t)up[4]) & ((1LL<<(3+4*8))-1); return 5; } else if (up[0] < 0xfc) { *val_p = (((uint64_t)up[0]<<40) | ((uint64_t)up[1]<<32) | ((uint64_t)up[2]<<24) | ((uint64_t)up[3]<<16) | ((uint64_t)up[4]<< 8) | (uint64_t)up[5]) & ((1LL<<(2+5*8))-1); return 6; } else if (up[0] < 0xfe) { *val_p = (((uint64_t)up[0]<<48) | ((uint64_t)up[1]<<40) | ((uint64_t)up[2]<<32) | ((uint64_t)up[3]<<24) | ((uint64_t)up[4]<<16) | ((uint64_t)up[5]<< 8) | (uint64_t)up[6]) & ((1LL<<(1+6*8))-1); return 7; } else if (up[0] < 0xff) { *val_p = (((uint64_t)up[1]<<48) | ((uint64_t)up[2]<<40) | ((uint64_t)up[3]<<32) | ((uint64_t)up[4]<<24) | ((uint64_t)up[5]<<16) | ((uint64_t)up[6]<< 8) | (uint64_t)up[7]) & ((1LL<<(7*8))-1); return 8; } else { *val_p = (((uint64_t)up[1]<<56) | ((uint64_t)up[2]<<48) | ((uint64_t)up[3]<<40) | ((uint64_t)up[4]<<32) | ((uint64_t)up[5]<<24) | ((uint64_t)up[6]<<16) | ((uint64_t)up[7]<< 8) | (uint64_t)up[8]); return 9; } } /* * LEGACY: consider using ltf8_decode_crc. */ int ltf8_decode(cram_fd *fd, int64_t *val_p) { int c = hgetc(fd->fp); int64_t val = (unsigned char)c; if (c == -1) return -1; if (val < 0x80) { *val_p = val; return 1; } else if (val < 0xc0) { val = (val<<8) | (unsigned char)hgetc(fd->fp); *val_p = val & (((1LL<<(6+8)))-1); return 2; } else if (val < 0xe0) { val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); *val_p = val & ((1LL<<(5+2*8))-1); return 3; } else if (val < 0xf0) { val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); *val_p = val & ((1LL<<(4+3*8))-1); return 4; } else if (val < 0xf8) { val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); *val_p = val & ((1LL<<(3+4*8))-1); return 5; } else if (val < 0xfc) { val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); *val_p = val & ((1LL<<(2+5*8))-1); return 6; } else if (val < 0xfe) { val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); *val_p = val & ((1LL<<(1+6*8))-1); return 7; } else if (val < 0xff) { val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); *val_p = val & ((1LL<<(7*8))-1); return 8; } else { val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); val = (val<<8) | (unsigned char)hgetc(fd->fp); *val_p = val; } return 9; } int ltf8_decode_crc(cram_fd *fd, int64_t *val_p, uint32_t *crc) { unsigned char c[9]; int64_t val = (unsigned char)hgetc(fd->fp); if (val == -1) return -1; c[0] = val; if (val < 0x80) { *val_p = val; *crc = crc32(*crc, c, 1); return 1; } else if (val < 0xc0) { val = (val<<8) | (c[1]=hgetc(fd->fp));; *val_p = val & (((1LL<<(6+8)))-1); *crc = crc32(*crc, c, 2); return 2; } else if (val < 0xe0) { val = (val<<8) | (c[1]=hgetc(fd->fp));; val = (val<<8) | (c[2]=hgetc(fd->fp));; *val_p = val & ((1LL<<(5+2*8))-1); *crc = crc32(*crc, c, 3); return 3; } else if (val < 0xf0) { val = (val<<8) | (c[1]=hgetc(fd->fp));; val = (val<<8) | (c[2]=hgetc(fd->fp));; val = (val<<8) | (c[3]=hgetc(fd->fp));; *val_p = val & ((1LL<<(4+3*8))-1); *crc = crc32(*crc, c, 4); return 4; } else if (val < 0xf8) { val = (val<<8) | (c[1]=hgetc(fd->fp));; val = (val<<8) | (c[2]=hgetc(fd->fp));; val = (val<<8) | (c[3]=hgetc(fd->fp));; val = (val<<8) | (c[4]=hgetc(fd->fp));; *val_p = val & ((1LL<<(3+4*8))-1); *crc = crc32(*crc, c, 5); return 5; } else if (val < 0xfc) { val = (val<<8) | (c[1]=hgetc(fd->fp));; val = (val<<8) | (c[2]=hgetc(fd->fp));; val = (val<<8) | (c[3]=hgetc(fd->fp));; val = (val<<8) | (c[4]=hgetc(fd->fp));; val = (val<<8) | (c[5]=hgetc(fd->fp));; *val_p = val & ((1LL<<(2+5*8))-1); *crc = crc32(*crc, c, 6); return 6; } else if (val < 0xfe) { val = (val<<8) | (c[1]=hgetc(fd->fp));; val = (val<<8) | (c[2]=hgetc(fd->fp));; val = (val<<8) | (c[3]=hgetc(fd->fp));; val = (val<<8) | (c[4]=hgetc(fd->fp));; val = (val<<8) | (c[5]=hgetc(fd->fp));; val = (val<<8) | (c[6]=hgetc(fd->fp));; *val_p = val & ((1LL<<(1+6*8))-1); *crc = crc32(*crc, c, 7); return 7; } else if (val < 0xff) { val = (val<<8) | (c[1]=hgetc(fd->fp));; val = (val<<8) | (c[2]=hgetc(fd->fp));; val = (val<<8) | (c[3]=hgetc(fd->fp));; val = (val<<8) | (c[4]=hgetc(fd->fp));; val = (val<<8) | (c[5]=hgetc(fd->fp));; val = (val<<8) | (c[6]=hgetc(fd->fp));; val = (val<<8) | (c[7]=hgetc(fd->fp));; *val_p = val & ((1LL<<(7*8))-1); *crc = crc32(*crc, c, 8); return 8; } else { val = (val<<8) | (c[1]=hgetc(fd->fp));; val = (val<<8) | (c[2]=hgetc(fd->fp));; val = (val<<8) | (c[3]=hgetc(fd->fp));; val = (val<<8) | (c[4]=hgetc(fd->fp));; val = (val<<8) | (c[5]=hgetc(fd->fp));; val = (val<<8) | (c[6]=hgetc(fd->fp));; val = (val<<8) | (c[7]=hgetc(fd->fp));; val = (val<<8) | (c[8]=hgetc(fd->fp));; *crc = crc32(*crc, c, 9); *val_p = val; } return 9; } /* * Pushes a value in ITF8 format onto the end of a block. * This shouldn't be used for high-volume data as it is not the fastest * method. * * Returns the number of bytes written */ int itf8_put_blk(cram_block *blk, int val) { char buf[5]; int sz; sz = itf8_put(buf, val); BLOCK_APPEND(blk, buf, sz); return sz; } /* * Decodes a 32-bit little endian value from fd and stores in val. * * Returns the number of bytes read on success * -1 on failure */ int int32_decode(cram_fd *fd, int32_t *val) { int32_t i; if (4 != hread(fd->fp, &i, 4)) return -1; *val = le_int4(i); return 4; } /* * Encodes a 32-bit little endian value 'val' and writes to fd. * * Returns the number of bytes written on success * -1 on failure */ int int32_encode(cram_fd *fd, int32_t val) { val = le_int4(val); if (4 != hwrite(fd->fp, &val, 4)) return -1; return 4; } /* As int32_decoded/encode, but from/to blocks instead of cram_fd */ int int32_get_blk(cram_block *b, int32_t *val) { if (b->uncomp_size - BLOCK_SIZE(b) < 4) return -1; *val = b->data[b->byte ] | (b->data[b->byte+1] << 8) | (b->data[b->byte+2] << 16) | (b->data[b->byte+3] << 24); BLOCK_SIZE(b) += 4; return 4; } /* As int32_decoded/encode, but from/to blocks instead of cram_fd */ int int32_put_blk(cram_block *b, int32_t val) { unsigned char cp[4]; cp[0] = ( val & 0xff); cp[1] = ((val>>8) & 0xff); cp[2] = ((val>>16) & 0xff); cp[3] = ((val>>24) & 0xff); BLOCK_APPEND(b, cp, 4); return b->data ? 0 : -1; } /* ---------------------------------------------------------------------- * zlib compression code - from Gap5's tg_iface_g.c * They're static here as they're only used within the cram_compress_block * and cram_uncompress_block functions, which are the external interface. */ char *zlib_mem_inflate(char *cdata, size_t csize, size_t *size) { z_stream s; unsigned char *data = NULL; /* Uncompressed output */ int data_alloc = 0; int err; /* Starting point at uncompressed size, and scale after that */ data = malloc(data_alloc = csize*1.2+100); if (!data) return NULL; /* Initialise zlib stream */ s.zalloc = Z_NULL; /* use default allocation functions */ s.zfree = Z_NULL; s.opaque = Z_NULL; s.next_in = (unsigned char *)cdata; s.avail_in = csize; s.total_in = 0; s.next_out = data; s.avail_out = data_alloc; s.total_out = 0; //err = inflateInit(&s); err = inflateInit2(&s, 15 + 32); if (err != Z_OK) { fprintf(stderr, "zlib inflateInit error: %s\n", s.msg); free(data); return NULL; } /* Decode to 'data' array */ for (;s.avail_in;) { unsigned char *data_tmp; int alloc_inc; s.next_out = &data[s.total_out]; err = inflate(&s, Z_NO_FLUSH); if (err == Z_STREAM_END) break; if (err != Z_OK) { fprintf(stderr, "zlib inflate error: %s\n", s.msg); if (data) free(data); return NULL; } /* More to come, so realloc based on growth so far */ alloc_inc = (double)s.avail_in/s.total_in * s.total_out + 100; data = realloc((data_tmp = data), data_alloc += alloc_inc); if (!data) { free(data_tmp); return NULL; } s.avail_out += alloc_inc; } inflateEnd(&s); *size = s.total_out; return (char *)data; } static char *zlib_mem_deflate(char *data, size_t size, size_t *cdata_size, int level, int strat) { z_stream s; unsigned char *cdata = NULL; /* Compressed output */ int cdata_alloc = 0; int cdata_pos = 0; int err; cdata = malloc(cdata_alloc = size*1.05+100); if (!cdata) return NULL; cdata_pos = 0; /* Initialise zlib stream */ s.zalloc = Z_NULL; /* use default allocation functions */ s.zfree = Z_NULL; s.opaque = Z_NULL; s.next_in = (unsigned char *)data; s.avail_in = size; s.total_in = 0; s.next_out = cdata; s.avail_out = cdata_alloc; s.total_out = 0; s.data_type = Z_BINARY; err = deflateInit2(&s, level, Z_DEFLATED, 15|16, 9, strat); if (err != Z_OK) { fprintf(stderr, "zlib deflateInit2 error: %s\n", s.msg); return NULL; } /* Encode to 'cdata' array */ for (;s.avail_in;) { s.next_out = &cdata[cdata_pos]; s.avail_out = cdata_alloc - cdata_pos; if (cdata_alloc - cdata_pos <= 0) { fprintf(stderr, "Deflate produced larger output than expected. Abort\n"); return NULL; } err = deflate(&s, Z_NO_FLUSH); cdata_pos = cdata_alloc - s.avail_out; if (err != Z_OK) { fprintf(stderr, "zlib deflate error: %s\n", s.msg); break; } } if (deflate(&s, Z_FINISH) != Z_STREAM_END) { fprintf(stderr, "zlib deflate error: %s\n", s.msg); } *cdata_size = s.total_out; if (deflateEnd(&s) != Z_OK) { fprintf(stderr, "zlib deflate error: %s\n", s.msg); } return (char *)cdata; } #ifdef HAVE_LIBLZMA /* ------------------------------------------------------------------------ */ /* * Data compression routines using liblzma (xz) * * On a test set this shrunk the main db from 136157104 bytes to 114796168, but * caused tg_index to grow from 2m43.707s to 15m3.961s. Exporting as bfastq * went from 18.3s to 36.3s. So decompression suffers too, but not as bad * as compression times. * * For now we disable this functionality. If it's to be reenabled make sure you * improve the mem_inflate implementation as it's just a test hack at the * moment. */ static char *lzma_mem_deflate(char *data, size_t size, size_t *cdata_size, int level) { char *out; size_t out_size = lzma_stream_buffer_bound(size); *cdata_size = 0; out = malloc(out_size); /* Single call compression */ if (LZMA_OK != lzma_easy_buffer_encode(level, LZMA_CHECK_CRC32, NULL, (uint8_t *)data, size, (uint8_t *)out, cdata_size, out_size)) return NULL; return out; } static char *lzma_mem_inflate(char *cdata, size_t csize, size_t *size) { lzma_stream strm = LZMA_STREAM_INIT; size_t out_size = 0, out_pos = 0; char *out = NULL; int r; /* Initiate the decoder */ if (LZMA_OK != lzma_stream_decoder(&strm, 50000000, 0)) return NULL; /* Decode loop */ strm.avail_in = csize; strm.next_in = (uint8_t *)cdata; for (;strm.avail_in;) { if (strm.avail_in > out_size - out_pos) { out_size += strm.avail_in * 4 + 32768; out = realloc(out, out_size); } strm.avail_out = out_size - out_pos; strm.next_out = (uint8_t *)&out[out_pos]; r = lzma_code(&strm, LZMA_RUN); if (LZMA_OK != r && LZMA_STREAM_END != r) { fprintf(stderr, "r=%d\n", r); fprintf(stderr, "mem=%"PRId64"d\n", (int64_t)lzma_memusage(&strm)); return NULL; } out_pos = strm.total_out; if (r == LZMA_STREAM_END) break; } /* finish up any unflushed data; necessary? */ r = lzma_code(&strm, LZMA_FINISH); if (r != LZMA_OK && r != LZMA_STREAM_END) { fprintf(stderr, "r=%d\n", r); return NULL; } out = realloc(out, strm.total_out); *size = strm.total_out; lzma_end(&strm); return out; } #endif /* ---------------------------------------------------------------------- * CRAM blocks - the dynamically growable data block. We have code to * create, update, (un)compress and read/write. * * These are derived from the deflate_interlaced.c blocks, but with the * CRAM extension of content types and IDs. */ /* * Allocates a new cram_block structure with a specified content_type and * id. * * Returns block pointer on success * NULL on failure */ cram_block *cram_new_block(enum cram_content_type content_type, int content_id) { cram_block *b = malloc(sizeof(*b)); if (!b) return NULL; b->method = b->orig_method = RAW; b->content_type = content_type; b->content_id = content_id; b->comp_size = 0; b->uncomp_size = 0; b->data = NULL; b->alloc = 0; b->byte = 0; b->bit = 7; // MSB return b; } /* * Reads a block from a cram file. * Returns cram_block pointer on success. * NULL on failure */ cram_block *cram_read_block(cram_fd *fd) { cram_block *b = malloc(sizeof(*b)); unsigned char c; uint32_t crc = 0; if (!b) return NULL; //fprintf(stderr, "Block at %d\n", (int)ftell(fd->fp)); if (-1 == (b->method = hgetc(fd->fp))) { free(b); return NULL; } c = b->method; crc = crc32(crc, &c, 1); if (-1 == (b->content_type= hgetc(fd->fp))) { free(b); return NULL; } c = b->content_type; crc = crc32(crc, &c, 1); if (-1 == itf8_decode_crc(fd, &b->content_id, &crc)) { free(b); return NULL; } if (-1 == itf8_decode_crc(fd, &b->comp_size, &crc)) { free(b); return NULL; } if (-1 == itf8_decode_crc(fd, &b->uncomp_size, &crc)) { free(b); return NULL; } // fprintf(stderr, " method %d, ctype %d, cid %d, csize %d, ucsize %d\n", // b->method, b->content_type, b->content_id, b->comp_size, b->uncomp_size); if (b->method == RAW) { b->alloc = b->uncomp_size; if (!(b->data = malloc(b->uncomp_size))){ free(b); return NULL; } if (b->uncomp_size != hread(fd->fp, b->data, b->uncomp_size)) { free(b->data); free(b); return NULL; } } else { b->alloc = b->comp_size; if (!(b->data = malloc(b->comp_size))) { free(b); return NULL; } if (b->comp_size != hread(fd->fp, b->data, b->comp_size)) { free(b->data); free(b); return NULL; } } if (CRAM_MAJOR_VERS(fd->version) >= 3) { if (-1 == int32_decode(fd, (int32_t *)&b->crc32)) { free(b); return NULL; } crc = crc32(crc, b->data ? b->data : (uc *)"", b->alloc); if (crc != b->crc32) { fprintf(stderr, "Block CRC32 failure\n"); free(b->data); free(b); return NULL; } } b->orig_method = b->method; b->idx = 0; b->byte = 0; b->bit = 7; // MSB return b; } /* * Computes the size of a cram block, including the block * header itself. */ uint32_t cram_block_size(cram_block *b) { unsigned char dat[100], *cp = dat;; uint32_t sz; *cp++ = b->method; *cp++ = b->content_type; cp += itf8_put(cp, b->content_id); cp += itf8_put(cp, b->comp_size); cp += itf8_put(cp, b->uncomp_size); sz = cp-dat + 4; sz += b->method == RAW ? b->uncomp_size : b->comp_size; return sz; } /* * Writes a CRAM block. * Returns 0 on success * -1 on failure */ int cram_write_block(cram_fd *fd, cram_block *b) { assert(b->method != RAW || (b->comp_size == b->uncomp_size)); if (hputc(b->method, fd->fp) == EOF) return -1; if (hputc(b->content_type, fd->fp) == EOF) return -1; if (itf8_encode(fd, b->content_id) == -1) return -1; if (itf8_encode(fd, b->comp_size) == -1) return -1; if (itf8_encode(fd, b->uncomp_size) == -1) return -1; if (b->method == RAW) { if (b->uncomp_size != hwrite(fd->fp, b->data, b->uncomp_size)) return -1; } else { if (b->comp_size != hwrite(fd->fp, b->data, b->comp_size)) return -1; } if (CRAM_MAJOR_VERS(fd->version) >= 3) { unsigned char dat[100], *cp = dat;; uint32_t crc; *cp++ = b->method; *cp++ = b->content_type; cp += itf8_put(cp, b->content_id); cp += itf8_put(cp, b->comp_size); cp += itf8_put(cp, b->uncomp_size); crc = crc32(0L, dat, cp-dat); if (b->method == RAW) { b->crc32 = crc32(crc, b->data ? b->data : (uc*)"", b->uncomp_size); } else { b->crc32 = crc32(crc, b->data ? b->data : (uc*)"", b->comp_size); } if (-1 == int32_encode(fd, b->crc32)) return -1; } return 0; } /* * Frees a CRAM block, deallocating internal data too. */ void cram_free_block(cram_block *b) { if (!b) return; if (b->data) free(b->data); free(b); } /* * Uncompresses a CRAM block, if compressed. */ int cram_uncompress_block(cram_block *b) { char *uncomp; size_t uncomp_size = 0; if (b->uncomp_size == 0) { // blank block b->method = RAW; return 0; } switch (b->method) { case RAW: return 0; case GZIP: uncomp = zlib_mem_inflate((char *)b->data, b->comp_size, &uncomp_size); if (!uncomp) return -1; if ((int)uncomp_size != b->uncomp_size) { free(uncomp); return -1; } free(b->data); b->data = (unsigned char *)uncomp; b->alloc = uncomp_size; b->method = RAW; break; #ifdef HAVE_LIBBZ2 case BZIP2: { unsigned int usize = b->uncomp_size; if (!(uncomp = malloc(usize))) return -1; if (BZ_OK != BZ2_bzBuffToBuffDecompress(uncomp, &usize, (char *)b->data, b->comp_size, 0, 0)) { free(uncomp); return -1; } free(b->data); b->data = (unsigned char *)uncomp; b->alloc = usize; b->method = RAW; b->uncomp_size = usize; // Just incase it differs break; } #else case BZIP2: fprintf(stderr, "Bzip2 compression is not compiled into this " "version.\nPlease rebuild and try again.\n"); return -1; #endif #ifdef HAVE_LIBLZMA case LZMA: uncomp = lzma_mem_inflate((char *)b->data, b->comp_size, &uncomp_size); if (!uncomp) return -1; if ((int)uncomp_size != b->uncomp_size) return -1; free(b->data); b->data = (unsigned char *)uncomp; b->alloc = uncomp_size; b->method = RAW; break; #else case LZMA: fprintf(stderr, "Lzma compression is not compiled into this " "version.\nPlease rebuild and try again.\n"); return -1; break; #endif case RANS: { unsigned int usize = b->uncomp_size, usize2; uncomp = (char *)rans_uncompress(b->data, b->comp_size, &usize2); if (!uncomp || usize != usize2) return -1; free(b->data); b->data = (unsigned char *)uncomp; b->alloc = usize2; b->method = RAW; b->uncomp_size = usize2; // Just incase it differs //fprintf(stderr, "Expanded %d to %d\n", b->comp_size, b->uncomp_size); break; } default: return -1; } return 0; } static char *cram_compress_by_method(char *in, size_t in_size, size_t *out_size, enum cram_block_method method, int level, int strat) { switch (method) { case GZIP: return zlib_mem_deflate(in, in_size, out_size, level, strat); case BZIP2: { #ifdef HAVE_LIBBZ2 unsigned int comp_size = in_size*1.01 + 600; char *comp = malloc(comp_size); if (!comp) return NULL; if (BZ_OK != BZ2_bzBuffToBuffCompress(comp, &comp_size, in, in_size, level, 0, 30)) { free(comp); return NULL; } *out_size = comp_size; return comp; #else return NULL; #endif } case LZMA: #ifdef HAVE_LIBLZMA return lzma_mem_deflate(in, in_size, out_size, level); #else return NULL; #endif case RANS0: { unsigned int out_size_i; unsigned char *cp; cp = rans_compress((unsigned char *)in, in_size, &out_size_i, 0); *out_size = out_size_i; return (char *)cp; } case RANS1: { unsigned int out_size_i; unsigned char *cp; cp = rans_compress((unsigned char *)in, in_size, &out_size_i, 1); *out_size = out_size_i; return (char *)cp; } case RAW: break; default: return NULL; } return NULL; } /* * Compresses a block using one of two different zlib strategies. If we only * want one choice set strat2 to be -1. * * The logic here is that sometimes Z_RLE does a better job than Z_FILTERED * or Z_DEFAULT_STRATEGY on quality data. If so, we'd rather use it as it is * significantly faster. * * Method and level -1 implies defaults, as specified in cram_fd. */ int cram_compress_block(cram_fd *fd, cram_block *b, cram_metrics *metrics, int method, int level) { char *comp = NULL; size_t comp_size = 0; int strat; if (b->method != RAW) { // Maybe already compressed if s->block[0] was compressed and // we have e.g. s->block[DS_BA] set to s->block[0] due to only // one base type present and hence using E_HUFFMAN on block 0. // A second explicit attempt to compress the same block then // occurs. return 0; } if (method == -1) { method = 1<use_bz2) method |= 1<use_lzma) method |= 1<level; //fprintf(stderr, "IN: block %d, sz %d\n", b->content_id, b->uncomp_size); if (method == RAW || level == 0 || b->uncomp_size == 0) { b->method = RAW; b->comp_size = b->uncomp_size; //fprintf(stderr, "Skip block id %d\n", b->content_id); return 0; } if (metrics) { pthread_mutex_lock(&fd->metrics_lock); if (metrics->trial > 0 || --metrics->next_trial <= 0) { size_t sz_best = INT_MAX; size_t sz_gz_rle = 0; size_t sz_gz_def = 0; size_t sz_rans0 = 0; size_t sz_rans1 = 0; size_t sz_bzip2 = 0; size_t sz_lzma = 0; int method_best = 0; char *c_best = NULL, *c = NULL; if (metrics->revised_method) method = metrics->revised_method; else metrics->revised_method = method; if (metrics->next_trial == 0) { metrics->next_trial = TRIAL_SPAN; metrics->trial = NTRIALS; metrics->sz_gz_rle /= 2; metrics->sz_gz_def /= 2; metrics->sz_rans0 /= 2; metrics->sz_rans1 /= 2; metrics->sz_bzip2 /= 2; metrics->sz_lzma /= 2; } pthread_mutex_unlock(&fd->metrics_lock); if (method & (1<data, b->uncomp_size, &sz_gz_rle, GZIP, 1, Z_RLE); if (c && sz_best > sz_gz_rle) { sz_best = sz_gz_rle; method_best = GZIP_RLE; if (c_best) free(c_best); c_best = c; } else if (c) { free(c); } else { sz_gz_rle = b->uncomp_size*2+1000; } //fprintf(stderr, "Block %d; %d->%d\n", b->content_id, b->uncomp_size, sz_gz_rle); } if (method & (1<data, b->uncomp_size, &sz_gz_def, GZIP, level, Z_FILTERED); if (c && sz_best > sz_gz_def) { sz_best = sz_gz_def; method_best = GZIP; if (c_best) free(c_best); c_best = c; } else if (c) { free(c); } else { sz_gz_def = b->uncomp_size*2+1000; } //fprintf(stderr, "Block %d; %d->%d\n", b->content_id, b->uncomp_size, sz_gz_def); } if (method & (1<data, b->uncomp_size, &sz_rans0, RANS0, 0, 0); if (c && sz_best > sz_rans0) { sz_best = sz_rans0; method_best = RANS0; if (c_best) free(c_best); c_best = c; } else if (c) { free(c); } else { sz_rans0 = b->uncomp_size*2+1000; } } if (method & (1<data, b->uncomp_size, &sz_rans1, RANS1, 0, 0); if (c && sz_best > sz_rans1) { sz_best = sz_rans1; method_best = RANS1; if (c_best) free(c_best); c_best = c; } else if (c) { free(c); } else { sz_rans1 = b->uncomp_size*2+1000; } } if (method & (1<data, b->uncomp_size, &sz_bzip2, BZIP2, level, 0); if (c && sz_best > sz_bzip2) { sz_best = sz_bzip2; method_best = BZIP2; if (c_best) free(c_best); c_best = c; } else if (c) { free(c); } else { sz_bzip2 = b->uncomp_size*2+1000; } } if (method & (1<data, b->uncomp_size, &sz_lzma, LZMA, level, 0); if (c && sz_best > sz_lzma) { sz_best = sz_lzma; method_best = LZMA; if (c_best) free(c_best); c_best = c; } else if (c) { free(c); } else { sz_lzma = b->uncomp_size*2+1000; } } //fprintf(stderr, "sz_best = %d\n", sz_best); free(b->data); b->data = (unsigned char *)c_best; //printf("method_best = %s\n", cram_block_method2str(method_best)); b->method = method_best == GZIP_RLE ? GZIP : method_best; b->comp_size = sz_best; pthread_mutex_lock(&fd->metrics_lock); metrics->sz_gz_rle += sz_gz_rle; metrics->sz_gz_def += sz_gz_def; metrics->sz_rans0 += sz_rans0; metrics->sz_rans1 += sz_rans1; metrics->sz_bzip2 += sz_bzip2; metrics->sz_lzma += sz_lzma; if (--metrics->trial == 0) { int best_method = RAW; int best_sz = INT_MAX; // Scale methods by cost if (fd->level <= 3) { metrics->sz_rans1 *= 1.02; metrics->sz_gz_def *= 1.04; metrics->sz_bzip2 *= 1.08; metrics->sz_lzma *= 1.10; } else if (fd->level <= 6) { metrics->sz_rans1 *= 1.01; metrics->sz_gz_def *= 1.02; metrics->sz_bzip2 *= 1.03; metrics->sz_lzma *= 1.05; } if (method & (1< metrics->sz_gz_rle) best_sz = metrics->sz_gz_rle, best_method = GZIP_RLE; if (method & (1< metrics->sz_gz_def) best_sz = metrics->sz_gz_def, best_method = GZIP; if (method & (1< metrics->sz_rans0) best_sz = metrics->sz_rans0, best_method = RANS0; if (method & (1< metrics->sz_rans1) best_sz = metrics->sz_rans1, best_method = RANS1; if (method & (1< metrics->sz_bzip2) best_sz = metrics->sz_bzip2, best_method = BZIP2; if (method & (1< metrics->sz_lzma) best_sz = metrics->sz_lzma, best_method = LZMA; if (best_method == GZIP_RLE) { metrics->method = GZIP; metrics->strat = Z_RLE; } else { metrics->method = best_method; metrics->strat = Z_FILTERED; } // If we see at least MAXFAIL trials in a row for a specific // compression method with more than MAXDELTA aggregate // size then we drop this from the list of methods used // for this block type. #define MAXDELTA 0.20 #define MAXFAILS 4 if (best_method == GZIP_RLE) { metrics->gz_rle_cnt = 0; metrics->gz_rle_extra = 0; } else if (best_sz < metrics->sz_gz_rle) { double r = (double)metrics->sz_gz_rle / best_sz - 1; if (++metrics->gz_rle_cnt >= MAXFAILS && (metrics->gz_rle_extra += r) >= MAXDELTA) method &= ~(1<gz_def_cnt = 0; metrics->gz_def_extra = 0; } else if (best_sz < metrics->sz_gz_def) { double r = (double)metrics->sz_gz_def / best_sz - 1; if (++metrics->gz_def_cnt >= MAXFAILS && (metrics->gz_def_extra += r) >= MAXDELTA) method &= ~(1<rans0_cnt = 0; metrics->rans0_extra = 0; } else if (best_sz < metrics->sz_rans0) { double r = (double)metrics->sz_rans0 / best_sz - 1; if (++metrics->rans0_cnt >= MAXFAILS && (metrics->rans0_extra += r) >= MAXDELTA) method &= ~(1<rans1_cnt = 0; metrics->rans1_extra = 0; } else if (best_sz < metrics->sz_rans1) { double r = (double)metrics->sz_rans1 / best_sz - 1; if (++metrics->rans1_cnt >= MAXFAILS && (metrics->rans1_extra += r) >= MAXDELTA) method &= ~(1<bzip2_cnt = 0; metrics->bzip2_extra = 0; } else if (best_sz < metrics->sz_bzip2) { double r = (double)metrics->sz_bzip2 / best_sz - 1; if (++metrics->bzip2_cnt >= MAXFAILS && (metrics->bzip2_extra += r) >= MAXDELTA) method &= ~(1<lzma_cnt = 0; metrics->lzma_extra = 0; } else if (best_sz < metrics->sz_lzma) { double r = (double)metrics->sz_lzma / best_sz - 1; if (++metrics->lzma_cnt >= MAXFAILS && (metrics->lzma_extra += r) >= MAXDELTA) method &= ~(1<revised_method) // fprintf(stderr, "%d: method from %x to %x\n", // b->content_id, metrics->revised_method, method); metrics->revised_method = method; } pthread_mutex_unlock(&fd->metrics_lock); } else { strat = metrics->strat; method = metrics->method; pthread_mutex_unlock(&fd->metrics_lock); comp = cram_compress_by_method((char *)b->data, b->uncomp_size, &comp_size, method, level, strat); if (!comp) return -1; free(b->data); b->data = (unsigned char *)comp; b->comp_size = comp_size; b->method = method; } } else { // no cached metrics, so just do zlib? comp = cram_compress_by_method((char *)b->data, b->uncomp_size, &comp_size, GZIP, level, Z_FILTERED); if (!comp) { fprintf(stderr, "Compression failed!\n"); return -1; } free(b->data); b->data = (unsigned char *)comp; b->comp_size = comp_size; b->method = GZIP; } if (fd->verbose) fprintf(stderr, "Compressed block ID %d from %d to %d by method %s\n", b->content_id, b->uncomp_size, b->comp_size, cram_block_method2str(b->method)); if (b->method == RANS1) b->method = RANS0; // Spec just has RANS (not 0/1) with auto-sensing return 0; } cram_metrics *cram_new_metrics(void) { cram_metrics *m = calloc(1, sizeof(*m)); if (!m) return NULL; m->trial = NTRIALS-1; m->next_trial = TRIAL_SPAN; m->method = RAW; m->strat = 0; m->revised_method = 0; return m; } char *cram_block_method2str(enum cram_block_method m) { switch(m) { case RAW: return "RAW"; case GZIP: return "GZIP"; case BZIP2: return "BZIP2"; case LZMA: return "LZMA"; case RANS0: return "RANS0"; case RANS1: return "RANS1"; case GZIP_RLE: return "GZIP_RLE"; case ERROR: break; } return "?"; } char *cram_content_type2str(enum cram_content_type t) { switch (t) { case FILE_HEADER: return "FILE_HEADER"; case COMPRESSION_HEADER: return "COMPRESSION_HEADER"; case MAPPED_SLICE: return "MAPPED_SLICE"; case UNMAPPED_SLICE: return "UNMAPPED_SLICE"; case EXTERNAL: return "EXTERNAL"; case CORE: return "CORE"; case CT_ERROR: break; } return "?"; } /* * Extra error checking on fclose to really ensure data is written. * Care needs to be taken to handle pipes vs real files. * * Returns 0 on success * -1 on failure. */ int paranoid_fclose(FILE *fp) { if (-1 == fflush(fp) && errno != EBADF) { fclose(fp); return -1; } errno = 0; if (-1 == fsync(fileno(fp))) { if (errno != EINVAL) { // eg pipe fclose(fp); return -1; } } return fclose(fp); } /* ---------------------------------------------------------------------- * Reference sequence handling * * These revolve around the refs_t structure, which may potentially be * shared between multiple cram_fd. * * We start with refs_create() to allocate an empty refs_t and then * populate it with @SQ line data using refs_from_header(). This is done on * cram_open(). Also at start up we can call cram_load_reference() which * is used with "scramble -r foo.fa". This replaces the fd->refs with the * new one specified. In either case refs2id() is then called which * maps ref_entry names to @SQ ids (refs_t->ref_id[]). * * Later, possibly within a thread, we will want to know the actual ref * seq itself, obtained by calling cram_get_ref(). This may use the * UR: or M5: fields or the filename specified in the original * cram_load_reference() call. * * Given the potential for multi-threaded reference usage, we have * reference counting (sorry for the confusing double use of "ref") to * track the number of callers interested in any specific reference. */ /* * Frees/unmaps a reference sequence and associated file handles. */ static void ref_entry_free_seq(ref_entry *e) { if (e->mf) mfclose(e->mf); if (e->seq && !e->mf) free(e->seq); e->seq = NULL; e->mf = NULL; } void refs_free(refs_t *r) { RP("refs_free()\n"); if (--r->count > 0) return; if (!r) return; if (r->pool) string_pool_destroy(r->pool); if (r->h_meta) { khint_t k; for (k = kh_begin(r->h_meta); k != kh_end(r->h_meta); k++) { ref_entry *e; if (!kh_exist(r->h_meta, k)) continue; if (!(e = kh_val(r->h_meta, k))) continue; ref_entry_free_seq(e); free(e); } kh_destroy(refs, r->h_meta); } if (r->ref_id) free(r->ref_id); if (r->fp) bgzf_close(r->fp); pthread_mutex_destroy(&r->lock); free(r); } static refs_t *refs_create(void) { refs_t *r = calloc(1, sizeof(*r)); RP("refs_create()\n"); if (!r) return NULL; if (!(r->pool = string_pool_create(8192))) goto err; r->ref_id = NULL; // see refs2id() to populate. r->count = 1; r->last = NULL; r->last_id = -1; if (!(r->h_meta = kh_init(refs))) goto err; pthread_mutex_init(&r->lock, NULL); return r; err: refs_free(r); return NULL; } /* * Opens a reference fasta file as a BGZF stream, allowing for * compressed files. It automatically builds a .fai file if * required and if compressed a .gzi bgzf index too. * * Returns a BGZF handle on success; * NULL on failure. */ static BGZF *bgzf_open_ref(char *fn, char *mode, int is_md5) { BGZF *fp; if (!is_md5) { char fai_file[PATH_MAX]; snprintf(fai_file, PATH_MAX, "%s.fai", fn); if (access(fai_file, R_OK) != 0) if (fai_build(fn) != 0) return NULL; } if (!(fp = bgzf_open(fn, mode))) { perror(fn); return NULL; } if (fp->is_compressed == 1 && bgzf_index_load(fp, fn, ".gzi") < 0) { fprintf(stderr, "Unable to load .gzi index '%s.gzi'\n", fn); bgzf_close(fp); return NULL; } return fp; } /* * Loads a FAI file for a reference.fasta. * "is_err" indicates whether failure to load is worthy of emitting an * error message. In some cases (eg with embedded references) we * speculatively load, just incase, and silently ignore errors. * * Returns the refs_t struct on success (maybe newly allocated); * NULL on failure */ static refs_t *refs_load_fai(refs_t *r_orig, char *fn, int is_err) { struct stat sb; FILE *fp = NULL; char fai_fn[PATH_MAX]; char line[8192]; refs_t *r = r_orig; size_t fn_l = strlen(fn); int id = 0, id_alloc = 0; RP("refs_load_fai %s\n", fn); if (!r) if (!(r = refs_create())) goto err; /* Open reference, for later use */ if (stat(fn, &sb) != 0) { if (is_err) perror(fn); goto err; } if (r->fp) if (bgzf_close(r->fp) != 0) goto err; r->fp = NULL; if (!(r->fn = string_dup(r->pool, fn))) goto err; if (fn_l > 4 && strcmp(&fn[fn_l-4], ".fai") == 0) r->fn[fn_l-4] = 0; if (!(r->fp = bgzf_open_ref(r->fn, "r", 0))) goto err; /* Parse .fai file and load meta-data */ sprintf(fai_fn, "%.*s.fai", PATH_MAX-5, r->fn); if (stat(fai_fn, &sb) != 0) { if (is_err) perror(fai_fn); goto err; } if (!(fp = fopen(fai_fn, "r"))) { if (is_err) perror(fai_fn); goto err; } while (fgets(line, 8192, fp) != NULL) { ref_entry *e = malloc(sizeof(*e)); char *cp; int n; khint_t k; if (!e) return NULL; // id for (cp = line; *cp && !isspace(*cp); cp++) ; *cp++ = 0; e->name = string_dup(r->pool, line); // length while (*cp && isspace(*cp)) cp++; e->length = strtoll(cp, &cp, 10); // offset while (*cp && isspace(*cp)) cp++; e->offset = strtoll(cp, &cp, 10); // bases per line while (*cp && isspace(*cp)) cp++; e->bases_per_line = strtol(cp, &cp, 10); // line length while (*cp && isspace(*cp)) cp++; e->line_length = strtol(cp, &cp, 10); // filename e->fn = r->fn; e->count = 0; e->seq = NULL; e->mf = NULL; e->is_md5 = 0; k = kh_put(refs, r->h_meta, e->name, &n); if (-1 == n) { free(e); return NULL; } if (n) { kh_val(r->h_meta, k) = e; } else { ref_entry *re = kh_val(r->h_meta, k); if (re && (re->count != 0 || re->length != 0)) { /* Keep old */ free(e); } else { /* Replace old */ if (re) free(re); kh_val(r->h_meta, k) = e; } } if (id >= id_alloc) { int x; id_alloc = id_alloc ?id_alloc*2 : 16; r->ref_id = realloc(r->ref_id, id_alloc * sizeof(*r->ref_id)); for (x = id; x < id_alloc; x++) r->ref_id[x] = NULL; } r->ref_id[id] = e; r->nref = ++id; } return r; err: if (fp) fclose(fp); if (!r_orig) refs_free(r); return NULL; } /* * Verifies that the CRAM @SQ lines and .fai files match. */ static void sanitise_SQ_lines(cram_fd *fd) { int i; if (!fd->header) return; if (!fd->refs || !fd->refs->h_meta) return; for (i = 0; i < fd->header->nref; i++) { char *name = fd->header->ref[i].name; khint_t k = kh_get(refs, fd->refs->h_meta, name); ref_entry *r; // We may have @SQ lines which have no known .fai, but do not // in themselves pose a problem because they are unused in the file. if (k == kh_end(fd->refs->h_meta)) continue; if (!(r = (ref_entry *)kh_val(fd->refs->h_meta, k))) continue; if (r->length && r->length != fd->header->ref[i].len) { assert(strcmp(r->name, fd->header->ref[i].name) == 0); // Should we also check MD5sums here to ensure the correct // reference was given? fprintf(stderr, "WARNING: Header @SQ length mismatch for " "ref %s, %d vs %d\n", r->name, fd->header->ref[i].len, (int)r->length); // Fixing the parsed @SQ header will make MD:Z: strings work // and also stop it producing N for the sequence. fd->header->ref[i].len = r->length; } } } /* * Indexes references by the order they appear in a BAM file. This may not * necessarily be the same order they appear in the fasta reference file. * * Returns 0 on success * -1 on failure */ int refs2id(refs_t *r, SAM_hdr *h) { int i; if (r->ref_id) free(r->ref_id); if (r->last) r->last = NULL; r->ref_id = calloc(h->nref, sizeof(*r->ref_id)); if (!r->ref_id) return -1; r->nref = h->nref; for (i = 0; i < h->nref; i++) { khint_t k = kh_get(refs, r->h_meta, h->ref[i].name); if (k != kh_end(r->h_meta)) { r->ref_id[i] = kh_val(r->h_meta, k); } else { fprintf(stderr, "Unable to find ref name '%s'\n", h->ref[i].name); } } return 0; } /* * Generates refs_t entries based on @SQ lines in the header. * Returns 0 on success * -1 on failure */ static int refs_from_header(refs_t *r, cram_fd *fd, SAM_hdr *h) { int i, j; if (!r) return -1; if (!h || h->nref == 0) return 0; //fprintf(stderr, "refs_from_header for %p mode %c\n", fd, fd->mode); /* Existing refs are fine, as long as they're compatible with the hdr. */ if (!(r->ref_id = realloc(r->ref_id, (r->nref + h->nref) * sizeof(*r->ref_id)))) return -1; /* Copy info from h->ref[i] over to r */ for (i = 0, j = r->nref; i < h->nref; i++) { SAM_hdr_type *ty; SAM_hdr_tag *tag; khint_t k; int n; k = kh_get(refs, r->h_meta, h->ref[i].name); if (k != kh_end(r->h_meta)) // Ref already known about continue; if (!(r->ref_id[j] = calloc(1, sizeof(ref_entry)))) return -1; if (!h->ref[i].name) return -1; r->ref_id[j]->name = string_dup(r->pool, h->ref[i].name); r->ref_id[j]->length = 0; // marker for not yet loaded /* Initialise likely filename if known */ if ((ty = sam_hdr_find(h, "SQ", "SN", h->ref[i].name))) { if ((tag = sam_hdr_find_key(h, ty, "M5", NULL))) { r->ref_id[j]->fn = string_dup(r->pool, tag->str+3); //fprintf(stderr, "Tagging @SQ %s / %s\n", r->ref_id[h]->name, r->ref_id[h]->fn); } } k = kh_put(refs, r->h_meta, r->ref_id[j]->name, &n); if (n <= 0) // already exists or error return -1; kh_val(r->h_meta, k) = r->ref_id[j]; j++; } r->nref = j; return 0; } /* * Attaches a header to a cram_fd. * * This should be used when creating a new cram_fd for writing where * we have an SAM_hdr already constructed (eg from a file we've read * in). */ int cram_set_header(cram_fd *fd, SAM_hdr *hdr) { if (fd->header) sam_hdr_free(fd->header); fd->header = hdr; return refs_from_header(fd->refs, fd, hdr); } /* * Converts a directory and a filename into an expanded path, replacing %s * in directory with the filename and %[0-9]+s with portions of the filename * Any remaining parts of filename are added to the end with /%s. */ void expand_cache_path(char *path, char *dir, char *fn) { char *cp; while ((cp = strchr(dir, '%'))) { strncpy(path, dir, cp-dir); path += cp-dir; if (*++cp == 's') { strcpy(path, fn); path += strlen(fn); fn += strlen(fn); cp++; } else if (*cp >= '0' && *cp <= '9') { char *endp; long l; l = strtol(cp, &endp, 10); l = MIN(l, strlen(fn)); if (*endp == 's') { strncpy(path, fn, l); path += l; fn += l; *path = 0; cp = endp+1; } else { *path++ = '%'; *path++ = *cp++; } } else { *path++ = '%'; *path++ = *cp++; } dir = cp; } strcpy(path, dir); path += strlen(dir); if (*fn && path[-1] != '/') *path++ = '/'; strcpy(path, fn); } /* * Make the directory containing path and any prefix directories. */ void mkdir_prefix(char *path, int mode) { char *cp = strrchr(path, '/'); if (!cp) return; *cp = 0; if (is_directory(path)) { *cp = '/'; return; } if (mkdir(path, mode) == 0) { chmod(path, mode); *cp = '/'; return; } mkdir_prefix(path, mode); mkdir(path, mode); chmod(path, mode); *cp = '/'; } /* * Return the cache directory to use, based on the first of these * environment variables to be set to a non-empty value. */ static const char *get_cache_basedir(const char **extra) { char *base; *extra = ""; base = getenv("XDG_CACHE_HOME"); if (base && *base) return base; base = getenv("HOME"); if (base && *base) { *extra = "/.cache"; return base; } base = getenv("TMPDIR"); if (base && *base) return base; base = getenv("TEMP"); if (base && *base) return base; return "/tmp"; } /* * Return an integer representation of pthread_self(). */ static unsigned get_int_threadid() { pthread_t pt = pthread_self(); unsigned char *s = (unsigned char *) &pt; size_t i; unsigned h = 0; for (i = 0; i < sizeof(pthread_t); i++) h = (h << 5) - h + s[i]; return h; } /* * Queries the M5 string from the header and attempts to populate the * reference from this using the REF_PATH environment. * * Returns 0 on sucess * -1 on failure */ static int cram_populate_ref(cram_fd *fd, int id, ref_entry *r) { char *ref_path = getenv("REF_PATH"); SAM_hdr_type *ty; SAM_hdr_tag *tag; char path[PATH_MAX], path_tmp[PATH_MAX]; char cache[PATH_MAX], cache_root[PATH_MAX]; char *local_cache = getenv("REF_CACHE"); mFILE *mf; int local_path = 0; if (fd->verbose) fprintf(stderr, "cram_populate_ref on fd %p, id %d\n", fd, id); cache_root[0] = '\0'; if (!ref_path || *ref_path == '\0') { /* * If we have no ref path, we use the EBI server. * However to avoid spamming it we require a local ref cache too. */ ref_path = "http://www.ebi.ac.uk:80/ena/cram/md5/%s"; if (!local_cache || *local_cache == '\0') { const char *extra; const char *base = get_cache_basedir(&extra); snprintf(cache_root, PATH_MAX, "%s%s/hts-ref", base, extra); snprintf(cache,PATH_MAX, "%s%s/hts-ref/%%2s/%%2s/%%s", base, extra); local_cache = cache; if (fd->verbose) fprintf(stderr, "Populating local cache: %s\n", local_cache); } } if (!r->name) return -1; if (!(ty = sam_hdr_find(fd->header, "SQ", "SN", r->name))) return -1; if (!(tag = sam_hdr_find_key(fd->header, ty, "M5", NULL))) goto no_M5; if (fd->verbose) fprintf(stderr, "Querying ref %s\n", tag->str+3); /* Use cache if available */ if (local_cache && *local_cache) { expand_cache_path(path, local_cache, tag->str+3); local_path = 1; } #ifndef HAVE_MMAP char *path2; /* Search local files in REF_PATH; we can open them and return as above */ if (!local_path && (path2 = find_path(tag->str+3, ref_path))) { strncpy(path, path2, PATH_MAX); free(path2); if (is_file(path)) // incase it's too long local_path = 1; } #endif /* Found via REF_CACHE or local REF_PATH file */ if (local_path) { struct stat sb; BGZF *fp; if (0 == stat(path, &sb) && (fp = bgzf_open(path, "r"))) { r->length = sb.st_size; r->offset = r->line_length = r->bases_per_line = 0; r->fn = string_dup(fd->refs->pool, path); if (fd->refs->fp) if (bgzf_close(fd->refs->fp) != 0) return -1; fd->refs->fp = fp; fd->refs->fn = r->fn; r->is_md5 = 1; // Fall back to cram_get_ref() where it'll do the actual // reading of the file. return 0; } } /* Otherwise search full REF_PATH; slower as loads entire file */ if ((mf = open_path_mfile(tag->str+3, ref_path, NULL))) { size_t sz; r->seq = mfsteal(mf, &sz); if (r->seq) { r->mf = NULL; } else { // keep mf around as we couldn't detach r->seq = mf->data; r->mf = mf; } r->length = sz; r->is_md5 = 1; } else { refs_t *refs; char *fn; no_M5: /* Failed to find in search path or M5 cache, see if @SQ UR: tag? */ if (!(tag = sam_hdr_find_key(fd->header, ty, "UR", NULL))) return -1; fn = (strncmp(tag->str+3, "file:", 5) == 0) ? tag->str+8 : tag->str+3; if (fd->refs->fp) { if (bgzf_close(fd->refs->fp) != 0) return -1; fd->refs->fp = NULL; } if (!(refs = refs_load_fai(fd->refs, fn, 0))) return -1; sanitise_SQ_lines(fd); fd->refs = refs; if (fd->refs->fp) { if (bgzf_close(fd->refs->fp) != 0) return -1; fd->refs->fp = NULL; } if (!fd->refs->fn) return -1; if (-1 == refs2id(fd->refs, fd->header)) return -1; if (!fd->refs->ref_id || !fd->refs->ref_id[id]) return -1; // Local copy already, so fall back to cram_get_ref(). return 0; } /* Populate the local disk cache if required */ if (local_cache && *local_cache) { int pid = (int) getpid(); unsigned thrid = get_int_threadid(); FILE *fp; if (*cache_root && !is_directory(cache_root) && hts_verbose >= 1) fprintf(stderr, "Creating reference cache directory %s\n" "This may become large; see the samtools(1) manual page REF_CACHE discussion\n", cache_root); expand_cache_path(path, local_cache, tag->str+3); if (fd->verbose) fprintf(stderr, "Writing cache file '%s'\n", path); mkdir_prefix(path, 01777); do { // Attempt to further uniquify the temporary filename unsigned t = ((unsigned) time(NULL)) ^ ((unsigned) clock()); thrid++; // Ensure filename changes even if time/clock haven't sprintf(path_tmp, "%s.tmp_%d_%u_%u", path, pid, thrid, t); fp = fopen(path_tmp, "wx"); } while (fp == NULL && errno == EEXIST); if (!fp) { perror(path_tmp); // Not fatal - we have the data already so keep going. return 0; } // Check md5sum hts_md5_context *md5; char unsigned md5_buf1[16]; char md5_buf2[33]; if (!(md5 = hts_md5_init())) { unlink(path_tmp); fclose(fp); return -1; } hts_md5_update(md5, r->seq, r->length); hts_md5_final(md5_buf1, md5); hts_md5_destroy(md5); hts_md5_hex(md5_buf2, md5_buf1); if (strncmp(tag->str+3, md5_buf2, 32) != 0) { fprintf(stderr, "[E::%s] mismatching md5sum for downloaded reference.\n", __func__); unlink(path_tmp); fclose(fp); return -1; } if (r->length != fwrite(r->seq, 1, r->length, fp)) { perror(path); } if (-1 == paranoid_fclose(fp)) { unlink(path_tmp); } else { if (0 == chmod(path_tmp, 0444)) rename(path_tmp, path); else unlink(path_tmp); } } return 0; } static void cram_ref_incr_locked(refs_t *r, int id) { RP("%d INC REF %d, %d %p\n", gettid(), id, (int)(id>=0?r->ref_id[id]->count+1:-999), id>=0?r->ref_id[id]->seq:(char *)1); if (id < 0 || !r->ref_id[id]->seq) return; if (r->last_id == id) r->last_id = -1; ++r->ref_id[id]->count; } void cram_ref_incr(refs_t *r, int id) { pthread_mutex_lock(&r->lock); cram_ref_incr_locked(r, id); pthread_mutex_unlock(&r->lock); } static void cram_ref_decr_locked(refs_t *r, int id) { RP("%d DEC REF %d, %d %p\n", gettid(), id, (int)(id>=0?r->ref_id[id]->count-1:-999), id>=0?r->ref_id[id]->seq:(char *)1); if (id < 0 || !r->ref_id[id]->seq) { assert(r->ref_id[id]->count >= 0); return; } if (--r->ref_id[id]->count <= 0) { assert(r->ref_id[id]->count == 0); if (r->last_id >= 0) { if (r->ref_id[r->last_id]->count <= 0 && r->ref_id[r->last_id]->seq) { RP("%d FREE REF %d (%p)\n", gettid(), r->last_id, r->ref_id[r->last_id]->seq); ref_entry_free_seq(r->ref_id[r->last_id]); r->ref_id[r->last_id]->length = 0; } } r->last_id = id; } } void cram_ref_decr(refs_t *r, int id) { pthread_mutex_lock(&r->lock); cram_ref_decr_locked(r, id); pthread_mutex_unlock(&r->lock); } /* * Used by cram_ref_load and cram_ref_get. The file handle will have * already been opened, so we can catch it. The ref_entry *e informs us * of whether this is a multi-line fasta file or a raw MD5 style file. * Either way we create a single contiguous sequence. * * Returns all or part of a reference sequence on success (malloced); * NULL on failure. */ static char *load_ref_portion(BGZF *fp, ref_entry *e, int start, int end) { off_t offset, len; char *seq; if (end < start) end = start; /* * Compute locations in file. This is trivial for the MD5 files, but * is still necessary for the fasta variants. */ offset = e->line_length ? e->offset + (start-1)/e->bases_per_line * e->line_length + (start-1) % e->bases_per_line : start-1; len = (e->line_length ? e->offset + (end-1)/e->bases_per_line * e->line_length + (end-1) % e->bases_per_line : end-1) - offset + 1; if (bgzf_useek(fp, offset, SEEK_SET) < 0) { perror("bgzf_useek() on reference file"); return NULL; } if (len == 0 || !(seq = malloc(len))) { return NULL; } if (len != bgzf_read(fp, seq, len)) { perror("bgzf_read() on reference file"); free(seq); return NULL; } /* Strip white-space if required. */ if (len != end-start+1) { int i, j; char *cp = seq; char *cp_to; for (i = j = 0; i < len; i++) { if (cp[i] >= '!' && cp[i] <= '~') cp[j++] = toupper(cp[i]); } cp_to = cp+j; if (cp_to - seq != end-start+1) { fprintf(stderr, "Malformed reference file?\n"); free(seq); return NULL; } } else { int i; for (i = 0; i < len; i++) { seq[i] = toupper(seq[i]); } } return seq; } /* * Load the entire reference 'id'. * This also increments the reference count by 1. * * Returns ref_entry on success; * NULL on failure */ ref_entry *cram_ref_load(refs_t *r, int id, int is_md5) { ref_entry *e = r->ref_id[id]; int start = 1, end = e->length; char *seq; if (e->seq) { return e; } assert(e->count == 0); if (r->last) { #ifdef REF_DEBUG int idx = 0; for (idx = 0; idx < r->nref; idx++) if (r->last == r->ref_id[idx]) break; RP("%d cram_ref_load DECR %d\n", gettid(), idx); #endif assert(r->last->count > 0); if (--r->last->count <= 0) { RP("%d FREE REF %d (%p)\n", gettid(), id, r->ref_id[id]->seq); if (r->last->seq) ref_entry_free_seq(r->last); } } /* Open file if it's not already the current open reference */ if (strcmp(r->fn, e->fn) || r->fp == NULL) { if (r->fp) if (bgzf_close(r->fp) != 0) return NULL; r->fn = e->fn; if (!(r->fp = bgzf_open_ref(r->fn, "r", is_md5))) return NULL; } RP("%d Loading ref %d (%d..%d)\n", gettid(), id, start, end); if (!(seq = load_ref_portion(r->fp, e, start, end))) { return NULL; } RP("%d Loaded ref %d (%d..%d) = %p\n", gettid(), id, start, end, seq); RP("%d INC REF %d, %d\n", gettid(), id, (int)(e->count+1)); e->seq = seq; e->mf = NULL; e->count++; /* * Also keep track of last used ref so incr/decr loops on the same * sequence don't cause load/free loops. */ RP("%d cram_ref_load INCR %d => %d\n", gettid(), id, e->count+1); r->last = e; e->count++; return e; } /* * Returns a portion of a reference sequence from start to end inclusive. * The returned pointer is owned by either the cram_file fd or by the * internal refs_t structure and should not be freed by the caller. * * The difference is whether or not this refs_t is in use by just the one * cram_fd or by multiples, or whether we have multiple threads accessing * references. In either case fd->shared will be true and we start using * reference counting to track the number of users of a specific reference * sequence. * * Otherwise the ref seq returned is allocated as part of cram_fd itself * and will be freed up on the next call to cram_get_ref or cram_close. * * To return the entire reference sequence, specify start as 1 and end * as 0. * * To cease using a reference, call cram_ref_decr(). * * Returns reference on success, * NULL on failure */ char *cram_get_ref(cram_fd *fd, int id, int start, int end) { ref_entry *r; char *seq; int ostart = start; if (id == -1) return NULL; /* FIXME: axiomatic query of r->seq being true? * Or shortcut for unsorted data where we load once and never free? */ //fd->shared_ref = 1; // hard code for now to simplify things pthread_mutex_lock(&fd->ref_lock); RP("%d cram_get_ref on fd %p, id %d, range %d..%d\n", gettid(), fd, id, start, end); /* * Unsorted data implies we want to fetch an entire reference at a time. * We just deal with this at the moment by claiming we're sharing * references instead, which has the same requirement. */ if (fd->unsorted) fd->shared_ref = 1; /* Sanity checking: does this ID exist? */ if (id >= fd->refs->nref) { fprintf(stderr, "No reference found for id %d\n", id); pthread_mutex_unlock(&fd->ref_lock); return NULL; } if (!fd->refs || !fd->refs->ref_id[id]) { fprintf(stderr, "No reference found for id %d\n", id); pthread_mutex_unlock(&fd->ref_lock); return NULL; } if (!(r = fd->refs->ref_id[id])) { fprintf(stderr, "No reference found for id %d\n", id); pthread_mutex_unlock(&fd->ref_lock); return NULL; } /* * It has an entry, but may not have been populated yet. * Any manually loaded .fai files have their lengths known. * A ref entry computed from @SQ lines (M5 or UR field) will have * r->length == 0 unless it's been loaded once and verified that we have * an on-disk filename for it. * * 19 Sep 2013: Moved the lock here as the cram_populate_ref code calls * open_path_mfile and libcurl, which isn't multi-thread safe unless I * rewrite my code to have one curl handle per thread. */ pthread_mutex_lock(&fd->refs->lock); if (r->length == 0) { if (cram_populate_ref(fd, id, r) == -1) { fprintf(stderr, "Failed to populate reference for id %d\n", id); pthread_mutex_unlock(&fd->refs->lock); pthread_mutex_unlock(&fd->ref_lock); return NULL; } r = fd->refs->ref_id[id]; if (fd->unsorted) cram_ref_incr_locked(fd->refs, id); } /* * We now know that we the filename containing the reference, so check * for limits. If it's over half the reference we'll load all of it in * memory as this will speed up subsequent calls. */ if (end < 1) end = r->length; if (end >= r->length) end = r->length; assert(start >= 1); if (end - start >= 0.5*r->length || fd->shared_ref) { start = 1; end = r->length; } /* * Maybe we have it cached already? If so use it. * * Alternatively if we don't have the sequence but we're sharing * references and/or are asking for the entire length of it, then * load the full reference into the refs structure and return * a pointer to that one instead. */ if (fd->shared_ref || r->seq || (start == 1 && end == r->length)) { char *cp; if (id >= 0) { if (r->seq) { cram_ref_incr_locked(fd->refs, id); } else { ref_entry *e; if (!(e = cram_ref_load(fd->refs, id, r->is_md5))) { pthread_mutex_unlock(&fd->refs->lock); pthread_mutex_unlock(&fd->ref_lock); return NULL; } /* unsorted data implies cache ref indefinitely, to avoid * continually loading and unloading. */ if (fd->unsorted) cram_ref_incr_locked(fd->refs, id); } fd->ref = NULL; /* We never access it directly */ fd->ref_start = 1; fd->ref_end = r->length; fd->ref_id = id; cp = fd->refs->ref_id[id]->seq + ostart-1; } else { fd->ref = NULL; cp = NULL; } RP("%d cram_get_ref returning for id %d, count %d\n", gettid(), id, (int)r->count); pthread_mutex_unlock(&fd->refs->lock); pthread_mutex_unlock(&fd->ref_lock); return cp; } /* * Otherwise we're not sharing, we don't have a copy of it already and * we're only asking for a small portion of it. * * In this case load up just that segment ourselves, freeing any old * small segments in the process. */ /* Unmapped ref ID */ if (id < 0) { if (fd->ref_free) { free(fd->ref_free); fd->ref_free = NULL; } fd->ref = NULL; fd->ref_id = id; pthread_mutex_unlock(&fd->refs->lock); pthread_mutex_unlock(&fd->ref_lock); return NULL; } /* Open file if it's not already the current open reference */ if (strcmp(fd->refs->fn, r->fn) || fd->refs->fp == NULL) { if (fd->refs->fp) if (bgzf_close(fd->refs->fp) != 0) return NULL; fd->refs->fn = r->fn; if (!(fd->refs->fp = bgzf_open_ref(fd->refs->fn, "r", r->is_md5))) { pthread_mutex_unlock(&fd->refs->lock); pthread_mutex_unlock(&fd->ref_lock); return NULL; } } if (!(fd->ref = load_ref_portion(fd->refs->fp, r, start, end))) { pthread_mutex_unlock(&fd->refs->lock); pthread_mutex_unlock(&fd->ref_lock); return NULL; } if (fd->ref_free) free(fd->ref_free); fd->ref_id = id; fd->ref_start = start; fd->ref_end = end; fd->ref_free = fd->ref; seq = fd->ref; pthread_mutex_unlock(&fd->refs->lock); pthread_mutex_unlock(&fd->ref_lock); return seq + ostart - start; } /* * If fd has been opened for reading, it may be permitted to specify 'fn' * as NULL and let the code auto-detect the reference by parsing the * SAM header @SQ lines. */ int cram_load_reference(cram_fd *fd, char *fn) { int ret = 0; if (fn) { fd->refs = refs_load_fai(fd->refs, fn, !(fd->embed_ref && fd->mode == 'r')); fn = fd->refs ? fd->refs->fn : NULL; if (!fn) ret = -1; sanitise_SQ_lines(fd); } fd->ref_fn = fn; if ((!fd->refs || (fd->refs->nref == 0 && !fn)) && fd->header) { if (fd->refs) refs_free(fd->refs); if (!(fd->refs = refs_create())) return -1; if (-1 == refs_from_header(fd->refs, fd, fd->header)) return -1; } if (fd->header) if (-1 == refs2id(fd->refs, fd->header)) return -1; return ret; } /* ---------------------------------------------------------------------- * Containers */ /* * Creates a new container, specifying the maximum number of slices * and records permitted. * * Returns cram_container ptr on success * NULL on failure */ cram_container *cram_new_container(int nrec, int nslice) { cram_container *c = calloc(1, sizeof(*c)); enum cram_DS_ID id; if (!c) return NULL; c->curr_ref = -2; c->max_c_rec = nrec * nslice; c->curr_c_rec = 0; c->max_rec = nrec; c->record_counter = 0; c->num_bases = 0; c->max_slice = nslice; c->curr_slice = 0; c->pos_sorted = 1; c->max_apos = 0; c->multi_seq = 0; c->bams = NULL; if (!(c->slices = (cram_slice **)calloc(nslice, sizeof(cram_slice *)))) goto err; c->slice = NULL; if (!(c->comp_hdr = cram_new_compression_header())) goto err; c->comp_hdr_block = NULL; for (id = DS_RN; id < DS_TN; id++) if (!(c->stats[id] = cram_stats_create())) goto err; //c->aux_B_stats = cram_stats_create(); if (!(c->tags_used = kh_init(s_i2i))) goto err; c->refs_used = 0; return c; err: if (c) { if (c->slices) free(c->slices); free(c); } return NULL; } void cram_free_container(cram_container *c) { enum cram_DS_ID id; int i; if (!c) return; if (c->refs_used) free(c->refs_used); if (c->landmark) free(c->landmark); if (c->comp_hdr) cram_free_compression_header(c->comp_hdr); if (c->comp_hdr_block) cram_free_block(c->comp_hdr_block); if (c->slices) { for (i = 0; i < c->max_slice; i++) if (c->slices[i]) cram_free_slice(c->slices[i]); free(c->slices); } for (id = DS_RN; id < DS_TN; id++) if (c->stats[id]) cram_stats_free(c->stats[id]); //if (c->aux_B_stats) cram_stats_free(c->aux_B_stats); if (c->tags_used) kh_destroy(s_i2i, c->tags_used); free(c); } /* * Reads a container header. * * Returns cram_container on success * NULL on failure or no container left (fd->err == 0). */ cram_container *cram_read_container(cram_fd *fd) { cram_container c2, *c; int i, s; size_t rd = 0; uint32_t crc = 0; fd->err = 0; fd->eof = 0; memset(&c2, 0, sizeof(c2)); if (CRAM_MAJOR_VERS(fd->version) == 1) { if ((s = itf8_decode_crc(fd, &c2.length, &crc)) == -1) { fd->eof = fd->empty_container ? 1 : 2; return NULL; } else { rd+=s; } } else { uint32_t len; if ((s = int32_decode(fd, &c2.length)) == -1) { if (CRAM_MAJOR_VERS(fd->version) == 2 && CRAM_MINOR_VERS(fd->version) == 0) fd->eof = 1; // EOF blocks arrived in v2.1 else fd->eof = fd->empty_container ? 1 : 2; return NULL; } else { rd+=s; } len = le_int4(c2.length); crc = crc32(0L, (unsigned char *)&len, 4); } if ((s = itf8_decode_crc(fd, &c2.ref_seq_id, &crc)) == -1) return NULL; else rd+=s; if ((s = itf8_decode_crc(fd, &c2.ref_seq_start, &crc))== -1) return NULL; else rd+=s; if ((s = itf8_decode_crc(fd, &c2.ref_seq_span, &crc)) == -1) return NULL; else rd+=s; if ((s = itf8_decode_crc(fd, &c2.num_records, &crc)) == -1) return NULL; else rd+=s; if (CRAM_MAJOR_VERS(fd->version) == 1) { c2.record_counter = 0; c2.num_bases = 0; } else { if (CRAM_MAJOR_VERS(fd->version) >= 3) { if ((s = ltf8_decode_crc(fd, &c2.record_counter, &crc)) == -1) return NULL; else rd += s; } else { int32_t i32; if ((s = itf8_decode_crc(fd, &i32, &crc)) == -1) return NULL; else rd += s; c2.record_counter = i32; } if ((s = ltf8_decode_crc(fd, &c2.num_bases, &crc))== -1) return NULL; else rd += s; } if ((s = itf8_decode_crc(fd, &c2.num_blocks, &crc)) == -1) return NULL; else rd+=s; if ((s = itf8_decode_crc(fd, &c2.num_landmarks, &crc))== -1) return NULL; else rd+=s; if (!(c = calloc(1, sizeof(*c)))) return NULL; *c = c2; if (!(c->landmark = malloc(c->num_landmarks * sizeof(int32_t))) && c->num_landmarks) { fd->err = errno; cram_free_container(c); return NULL; } for (i = 0; i < c->num_landmarks; i++) { if ((s = itf8_decode_crc(fd, &c->landmark[i], &crc)) == -1) { cram_free_container(c); return NULL; } else { rd += s; } } if (CRAM_MAJOR_VERS(fd->version) >= 3) { if (-1 == int32_decode(fd, (int32_t *)&c->crc32)) return NULL; else rd+=4; if (crc != c->crc32) { fprintf(stderr, "Container header CRC32 failure\n"); cram_free_container(c); return NULL; } } c->offset = rd; c->slices = NULL; c->curr_slice = 0; c->max_slice = c->num_landmarks; c->slice_rec = 0; c->curr_rec = 0; c->max_rec = 0; if (c->ref_seq_id == -2) { c->multi_seq = 1; fd->multi_seq = 1; } fd->empty_container = (c->num_records == 0 && c->ref_seq_id == -1 && c->ref_seq_start == 0x454f46 /* EOF */) ? 1 : 0; return c; } /* MAXIMUM storage size needed for the container. */ int cram_container_size(cram_container *c) { return 55 + 5*c->num_landmarks; } /* * Stores the container structure in dat and returns *size as the * number of bytes written to dat[]. The input size of dat is also * held in *size and should be initialised to cram_container_size(c). * * Returns 0 on success; * -1 on failure */ int cram_store_container(cram_fd *fd, cram_container *c, char *dat, int *size) { char *cp = dat; int i; // Check the input buffer is large enough according to our stated // requirements. (NOTE: it may actually take less.) if (cram_container_size(c) > *size) return -1; if (CRAM_MAJOR_VERS(fd->version) == 1) { cp += itf8_put(cp, c->length); } else { *(int32_t *)cp = le_int4(c->length); cp += 4; } if (c->multi_seq) { cp += itf8_put(cp, -2); cp += itf8_put(cp, 0); cp += itf8_put(cp, 0); } else { cp += itf8_put(cp, c->ref_seq_id); cp += itf8_put(cp, c->ref_seq_start); cp += itf8_put(cp, c->ref_seq_span); } cp += itf8_put(cp, c->num_records); if (CRAM_MAJOR_VERS(fd->version) == 2) { cp += itf8_put(cp, c->record_counter); cp += ltf8_put(cp, c->num_bases); } else if (CRAM_MAJOR_VERS(fd->version) >= 3) { cp += ltf8_put(cp, c->record_counter); cp += ltf8_put(cp, c->num_bases); } cp += itf8_put(cp, c->num_blocks); cp += itf8_put(cp, c->num_landmarks); for (i = 0; i < c->num_landmarks; i++) cp += itf8_put(cp, c->landmark[i]); if (CRAM_MAJOR_VERS(fd->version) >= 3) { c->crc32 = crc32(0L, (uc *)dat, cp-dat); cp[0] = c->crc32 & 0xff; cp[1] = (c->crc32 >> 8) & 0xff; cp[2] = (c->crc32 >> 16) & 0xff; cp[3] = (c->crc32 >> 24) & 0xff; cp += 4; } *size = cp-dat; // actual used size return 0; } /* * Writes a container structure. * * Returns 0 on success * -1 on failure */ int cram_write_container(cram_fd *fd, cram_container *c) { char buf_a[1024], *buf = buf_a, *cp; int i; if (55 + c->num_landmarks * 5 >= 1024) buf = malloc(55 + c->num_landmarks * 5); cp = buf; if (CRAM_MAJOR_VERS(fd->version) == 1) { cp += itf8_put(cp, c->length); } else { *(int32_t *)cp = le_int4(c->length); cp += 4; } if (c->multi_seq) { cp += itf8_put(cp, -2); cp += itf8_put(cp, 0); cp += itf8_put(cp, 0); } else { cp += itf8_put(cp, c->ref_seq_id); cp += itf8_put(cp, c->ref_seq_start); cp += itf8_put(cp, c->ref_seq_span); } cp += itf8_put(cp, c->num_records); if (CRAM_MAJOR_VERS(fd->version) == 2) { cp += itf8_put(cp, c->record_counter); cp += ltf8_put(cp, c->num_bases); } else if (CRAM_MAJOR_VERS(fd->version) >= 3) { cp += ltf8_put(cp, c->record_counter); cp += ltf8_put(cp, c->num_bases); } cp += itf8_put(cp, c->num_blocks); cp += itf8_put(cp, c->num_landmarks); for (i = 0; i < c->num_landmarks; i++) cp += itf8_put(cp, c->landmark[i]); if (CRAM_MAJOR_VERS(fd->version) >= 3) { c->crc32 = crc32(0L, (uc *)buf, cp-buf); cp[0] = c->crc32 & 0xff; cp[1] = (c->crc32 >> 8) & 0xff; cp[2] = (c->crc32 >> 16) & 0xff; cp[3] = (c->crc32 >> 24) & 0xff; cp += 4; } if (cp-buf != hwrite(fd->fp, buf, cp-buf)) { if (buf != buf_a) free(buf); return -1; } if (buf != buf_a) free(buf); return 0; } // common component shared by cram_flush_container{,_mt} static int cram_flush_container2(cram_fd *fd, cram_container *c) { int i, j; if (c->curr_slice > 0 && !c->slices) return -1; //fprintf(stderr, "Writing container %d, sum %u\n", c->record_counter, sum); /* Write the container struct itself */ if (0 != cram_write_container(fd, c)) return -1; /* And the compression header */ if (0 != cram_write_block(fd, c->comp_hdr_block)) return -1; /* Followed by the slice blocks */ for (i = 0; i < c->curr_slice; i++) { cram_slice *s = c->slices[i]; if (0 != cram_write_block(fd, s->hdr_block)) return -1; for (j = 0; j < s->hdr->num_blocks; j++) { if (0 != cram_write_block(fd, s->block[j])) return -1; } } return hflush(fd->fp) == 0 ? 0 : -1; } /* * Flushes a completely or partially full container to disk, writing * container structure, header and blocks. This also calls the encoder * functions. * * Returns 0 on success * -1 on failure */ int cram_flush_container(cram_fd *fd, cram_container *c) { /* Encode the container blocks and generate compression header */ if (0 != cram_encode_container(fd, c)) return -1; return cram_flush_container2(fd, c); } typedef struct { cram_fd *fd; cram_container *c; } cram_job; void *cram_flush_thread(void *arg) { cram_job *j = (cram_job *)arg; /* Encode the container blocks and generate compression header */ if (0 != cram_encode_container(j->fd, j->c)) { fprintf(stderr, "cram_encode_container failed\n"); return NULL; } return arg; } static int cram_flush_result(cram_fd *fd) { int i, ret = 0; t_pool_result *r; while ((r = t_pool_next_result(fd->rqueue))) { cram_job *j = (cram_job *)r->data; cram_container *c; if (!j) { t_pool_delete_result(r, 0); return -1; } fd = j->fd; c = j->c; if (0 != cram_flush_container2(fd, c)) return -1; /* Free the container */ for (i = 0; i < c->max_slice; i++) { cram_free_slice(c->slices[i]); c->slices[i] = NULL; } c->slice = NULL; c->curr_slice = 0; cram_free_container(c); ret |= hflush(fd->fp) == 0 ? 0 : -1; t_pool_delete_result(r, 1); } return ret; } int cram_flush_container_mt(cram_fd *fd, cram_container *c) { cram_job *j; if (!fd->pool) return cram_flush_container(fd, c); if (!(j = malloc(sizeof(*j)))) return -1; j->fd = fd; j->c = c; t_pool_dispatch(fd->pool, fd->rqueue, cram_flush_thread, j); return cram_flush_result(fd); } /* ---------------------------------------------------------------------- * Compression headers; the first part of the container */ /* * Creates a new blank container compression header * * Returns header ptr on success * NULL on failure */ cram_block_compression_hdr *cram_new_compression_header(void) { cram_block_compression_hdr *hdr = calloc(1, sizeof(*hdr)); if (!hdr) return NULL; if (!(hdr->TD_blk = cram_new_block(CORE, 0))) { free(hdr); return NULL; } if (!(hdr->TD_hash = kh_init(m_s2i))) { cram_free_block(hdr->TD_blk); free(hdr); return NULL; } if (!(hdr->TD_keys = string_pool_create(8192))) { kh_destroy(m_s2i, hdr->TD_hash); cram_free_block(hdr->TD_blk); free(hdr); return NULL; } return hdr; } void cram_free_compression_header(cram_block_compression_hdr *hdr) { int i; if (hdr->landmark) free(hdr->landmark); if (hdr->preservation_map) kh_destroy(map, hdr->preservation_map); for (i = 0; i < CRAM_MAP_HASH; i++) { cram_map *m, *m2; for (m = hdr->rec_encoding_map[i]; m; m = m2) { m2 = m->next; if (m->codec) m->codec->free(m->codec); free(m); } } for (i = 0; i < CRAM_MAP_HASH; i++) { cram_map *m, *m2; for (m = hdr->tag_encoding_map[i]; m; m = m2) { m2 = m->next; if (m->codec) m->codec->free(m->codec); free(m); } } for (i = 0; i < DS_END; i++) { if (hdr->codecs[i]) hdr->codecs[i]->free(hdr->codecs[i]); } if (hdr->TL) free(hdr->TL); if (hdr->TD_blk) cram_free_block(hdr->TD_blk); if (hdr->TD_hash) kh_destroy(m_s2i, hdr->TD_hash); if (hdr->TD_keys) string_pool_destroy(hdr->TD_keys); free(hdr); } /* ---------------------------------------------------------------------- * Slices and slice headers */ void cram_free_slice_header(cram_block_slice_hdr *hdr) { if (!hdr) return; if (hdr->block_content_ids) free(hdr->block_content_ids); free(hdr); return; } void cram_free_slice(cram_slice *s) { if (!s) return; if (s->hdr_block) cram_free_block(s->hdr_block); if (s->block) { int i; if (s->hdr) { for (i = 0; i < s->hdr->num_blocks; i++) { cram_free_block(s->block[i]); } } free(s->block); } if (s->block_by_id) free(s->block_by_id); if (s->hdr) cram_free_slice_header(s->hdr); if (s->seqs_blk) cram_free_block(s->seqs_blk); if (s->qual_blk) cram_free_block(s->qual_blk); if (s->name_blk) cram_free_block(s->name_blk); if (s->aux_blk) cram_free_block(s->aux_blk); if (s->aux_OQ_blk) cram_free_block(s->aux_OQ_blk); if (s->aux_BQ_blk) cram_free_block(s->aux_BQ_blk); if (s->aux_FZ_blk) cram_free_block(s->aux_FZ_blk); if (s->aux_oq_blk) cram_free_block(s->aux_oq_blk); if (s->aux_os_blk) cram_free_block(s->aux_os_blk); if (s->aux_oz_blk) cram_free_block(s->aux_oz_blk); if (s->base_blk) cram_free_block(s->base_blk); if (s->soft_blk) cram_free_block(s->soft_blk); if (s->cigar) free(s->cigar); if (s->crecs) free(s->crecs); if (s->features) free(s->features); if (s->TN) free(s->TN); if (s->pair_keys) string_pool_destroy(s->pair_keys); if (s->pair[0]) kh_destroy(m_s2i, s->pair[0]); if (s->pair[1]) kh_destroy(m_s2i, s->pair[1]); free(s); } /* * Creates a new empty slice in memory, for subsequent writing to * disk. * * Returns cram_slice ptr on success * NULL on failure */ cram_slice *cram_new_slice(enum cram_content_type type, int nrecs) { cram_slice *s = calloc(1, sizeof(*s)); if (!s) return NULL; if (!(s->hdr = (cram_block_slice_hdr *)calloc(1, sizeof(*s->hdr)))) goto err; s->hdr->content_type = type; s->hdr_block = NULL; s->block = NULL; s->block_by_id = NULL; s->last_apos = 0; if (!(s->crecs = malloc(nrecs * sizeof(cram_record)))) goto err; s->cigar = NULL; s->cigar_alloc = 0; s->ncigar = 0; if (!(s->seqs_blk = cram_new_block(EXTERNAL, 0))) goto err; if (!(s->qual_blk = cram_new_block(EXTERNAL, DS_QS))) goto err; if (!(s->name_blk = cram_new_block(EXTERNAL, DS_RN))) goto err; if (!(s->aux_blk = cram_new_block(EXTERNAL, DS_aux))) goto err; if (!(s->base_blk = cram_new_block(EXTERNAL, DS_IN))) goto err; if (!(s->soft_blk = cram_new_block(EXTERNAL, DS_SC))) goto err; s->features = NULL; s->nfeatures = s->afeatures = 0; #ifndef TN_external s->TN = NULL; s->nTN = s->aTN = 0; #endif // Volatile keys as we do realloc in dstring if (!(s->pair_keys = string_pool_create(8192))) goto err; if (!(s->pair[0] = kh_init(m_s2i))) goto err; if (!(s->pair[1] = kh_init(m_s2i))) goto err; #ifdef BA_external s->BA_len = 0; #endif return s; err: if (s) cram_free_slice(s); return NULL; } /* * Loads an entire slice. * FIXME: In 1.0 the native unit of slices within CRAM is broken * as slices contain references to objects in other slices. * To work around this while keeping the slice oriented outer loop * we read all slices and stitch them together into a fake large * slice instead. * * Returns cram_slice ptr on success * NULL on failure */ cram_slice *cram_read_slice(cram_fd *fd) { cram_block *b = cram_read_block(fd); cram_slice *s = calloc(1, sizeof(*s)); int i, n, max_id, min_id; if (!b || !s) goto err; s->hdr_block = b; switch (b->content_type) { case MAPPED_SLICE: case UNMAPPED_SLICE: if (!(s->hdr = cram_decode_slice_header(fd, b))) goto err; break; default: fprintf(stderr, "Unexpected block of type %s\n", cram_content_type2str(b->content_type)); goto err; } if (s->hdr->num_blocks < 1) { fprintf(stderr, "Slice does not include any data blocks.\n"); goto err; } s->block = calloc(n = s->hdr->num_blocks, sizeof(*s->block)); if (!s->block) goto err; for (max_id = i = 0, min_id = INT_MAX; i < n; i++) { if (!(s->block[i] = cram_read_block(fd))) goto err; if (s->block[i]->content_type == EXTERNAL) { if (max_id < s->block[i]->content_id) max_id = s->block[i]->content_id; if (min_id > s->block[i]->content_id) min_id = s->block[i]->content_id; } } if (min_id >= 0 && max_id < 1024) { if (!(s->block_by_id = calloc(1024, sizeof(s->block[0])))) goto err; for (i = 0; i < n; i++) { if (s->block[i]->content_type != EXTERNAL) continue; s->block_by_id[s->block[i]->content_id] = s->block[i]; } } /* Initialise encoding/decoding tables */ s->cigar = NULL; s->cigar_alloc = 0; s->ncigar = 0; if (!(s->seqs_blk = cram_new_block(EXTERNAL, 0))) goto err; if (!(s->qual_blk = cram_new_block(EXTERNAL, DS_QS))) goto err; if (!(s->name_blk = cram_new_block(EXTERNAL, DS_RN))) goto err; if (!(s->aux_blk = cram_new_block(EXTERNAL, DS_aux))) goto err; if (!(s->base_blk = cram_new_block(EXTERNAL, DS_IN))) goto err; if (!(s->soft_blk = cram_new_block(EXTERNAL, DS_SC))) goto err; s->crecs = NULL; s->last_apos = s->hdr->ref_seq_start; return s; err: if (b) cram_free_block(b); if (s) { s->hdr_block = NULL; cram_free_slice(s); } return NULL; } /* ---------------------------------------------------------------------- * CRAM file definition (header) */ /* * Reads a CRAM file definition structure. * Returns file_def ptr on success * NULL on failure */ cram_file_def *cram_read_file_def(cram_fd *fd) { cram_file_def *def = malloc(sizeof(*def)); if (!def) return NULL; if (26 != hread(fd->fp, &def->magic[0], 26)) { free(def); return NULL; } if (memcmp(def->magic, "CRAM", 4) != 0) { free(def); return NULL; } if (def->major_version > 3) { fprintf(stderr, "CRAM version number mismatch\n" "Expected 1.x, 2.x or 3.x, got %d.%d\n", def->major_version, def->minor_version); free(def); return NULL; } fd->first_container += 26; fd->last_slice = 0; return def; } /* * Writes a cram_file_def structure to cram_fd. * Returns 0 on success * -1 on failure */ int cram_write_file_def(cram_fd *fd, cram_file_def *def) { return (hwrite(fd->fp, &def->magic[0], 26) == 26) ? 0 : -1; } void cram_free_file_def(cram_file_def *def) { if (def) free(def); } /* ---------------------------------------------------------------------- * SAM header I/O */ /* * Reads the SAM header from the first CRAM data block. * Also performs minimal parsing to extract read-group * and sample information. * Returns SAM hdr ptr on success * NULL on failure */ SAM_hdr *cram_read_SAM_hdr(cram_fd *fd) { int32_t header_len; char *header; SAM_hdr *hdr; /* 1.1 onwards stores the header in the first block of a container */ if (CRAM_MAJOR_VERS(fd->version) == 1) { /* Length */ if (-1 == int32_decode(fd, &header_len)) return NULL; /* Alloc and read */ if (header_len < 0 || NULL == (header = malloc((size_t) header_len+1))) return NULL; if (header_len != hread(fd->fp, header, header_len)) return NULL; header[header_len] = '\0'; fd->first_container += 4 + header_len; } else { cram_container *c = cram_read_container(fd); cram_block *b; int i, len; if (!c) return NULL; if (c->num_blocks < 1) { cram_free_container(c); return NULL; } if (!(b = cram_read_block(fd))) { cram_free_container(c); return NULL; } if (cram_uncompress_block(b) != 0) { cram_free_container(c); return NULL; } len = b->comp_size + 2 + 4*(CRAM_MAJOR_VERS(fd->version) >= 3) + itf8_size(b->content_id) + itf8_size(b->uncomp_size) + itf8_size(b->comp_size); /* Extract header from 1st block */ if (-1 == int32_get_blk(b, &header_len) || header_len < 0 || /* Spec. says signed... why? */ b->uncomp_size - 4 < header_len) { cram_free_container(c); cram_free_block(b); return NULL; } if (NULL == (header = malloc((size_t) header_len+1))) { cram_free_container(c); cram_free_block(b); return NULL; } memcpy(header, BLOCK_END(b), header_len); header[header_len]='\0'; cram_free_block(b); /* Consume any remaining blocks */ for (i = 1; i < c->num_blocks; i++) { if (!(b = cram_read_block(fd))) { cram_free_container(c); return NULL; } len += b->comp_size + 2 + 4*(CRAM_MAJOR_VERS(fd->version) >= 3) + itf8_size(b->content_id) + itf8_size(b->uncomp_size) + itf8_size(b->comp_size); cram_free_block(b); } if (c->length > 0 && len > 0 && c->length > len) { // Consume padding char *pads = malloc(c->length - len); if (!pads) { cram_free_container(c); return NULL; } if (c->length - len != hread(fd->fp, pads, c->length - len)) { cram_free_container(c); return NULL; } free(pads); } cram_free_container(c); } /* Parse */ hdr = sam_hdr_parse_(header, header_len); free(header); return hdr; } /* * Converts 'in' to a full pathname to store in out. * Out must be at least PATH_MAX bytes long. */ static void full_path(char *out, char *in) { if (*in == '/') { strncpy(out, in, PATH_MAX); out[PATH_MAX-1] = 0; } else { int len; // unable to get dir or out+in is too long if (!getcwd(out, PATH_MAX) || (len = strlen(out))+1+strlen(in) >= PATH_MAX) { strncpy(out, in, PATH_MAX); out[PATH_MAX-1] = 0; return; } sprintf(out+len, "/%.*s", PATH_MAX - len, in); // FIXME: cope with `pwd`/../../../foo.fa ? } } /* * Writes a CRAM SAM header. * Returns 0 on success * -1 on failure */ int cram_write_SAM_hdr(cram_fd *fd, SAM_hdr *hdr) { int header_len; int blank_block = (CRAM_MAJOR_VERS(fd->version) >= 3); /* Write CRAM MAGIC if not yet written. */ if (fd->file_def->major_version == 0) { fd->file_def->major_version = CRAM_MAJOR_VERS(fd->version); fd->file_def->minor_version = CRAM_MINOR_VERS(fd->version); if (0 != cram_write_file_def(fd, fd->file_def)) return -1; } /* 1.0 requires an UNKNOWN read-group */ if (CRAM_MAJOR_VERS(fd->version) == 1) { if (!sam_hdr_find_rg(hdr, "UNKNOWN")) if (sam_hdr_add(hdr, "RG", "ID", "UNKNOWN", "SM", "UNKNOWN", NULL)) return -1; } /* Fix M5 strings */ if (fd->refs && !fd->no_ref) { int i; for (i = 0; i < hdr->nref; i++) { SAM_hdr_type *ty; char *ref; if (!(ty = sam_hdr_find(hdr, "SQ", "SN", hdr->ref[i].name))) return -1; if (!sam_hdr_find_key(hdr, ty, "M5", NULL)) { char unsigned buf[16]; char buf2[33]; int rlen; hts_md5_context *md5; if (!fd->refs || !fd->refs->ref_id || !fd->refs->ref_id[i]) { return -1; } rlen = fd->refs->ref_id[i]->length; if (!(md5 = hts_md5_init())) return -1; ref = cram_get_ref(fd, i, 1, rlen); if (NULL == ref) return -1; rlen = fd->refs->ref_id[i]->length; /* In case it just loaded */ hts_md5_update(md5, ref, rlen); hts_md5_final(buf, md5); hts_md5_destroy(md5); cram_ref_decr(fd->refs, i); hts_md5_hex(buf2, buf); if (sam_hdr_update(hdr, ty, "M5", buf2, NULL)) return -1; } if (fd->ref_fn) { char ref_fn[PATH_MAX]; full_path(ref_fn, fd->ref_fn); if (sam_hdr_update(hdr, ty, "UR", ref_fn, NULL)) return -1; } } } if (sam_hdr_rebuild(hdr)) return -1; /* Length */ header_len = sam_hdr_length(hdr); if (CRAM_MAJOR_VERS(fd->version) == 1) { if (-1 == int32_encode(fd, header_len)) return -1; /* Text data */ if (header_len != hwrite(fd->fp, sam_hdr_str(hdr), header_len)) return -1; } else { /* Create block(s) inside a container */ cram_block *b = cram_new_block(FILE_HEADER, 0); cram_container *c = cram_new_container(0, 0); int padded_length; char *pads; int is_cram_3 = (CRAM_MAJOR_VERS(fd->version) >= 3); if (!b || !c) { if (b) cram_free_block(b); if (c) cram_free_container(c); return -1; } int32_put_blk(b, header_len); BLOCK_APPEND(b, sam_hdr_str(hdr), header_len); BLOCK_UPLEN(b); // Compress header block if V3.0 and above if (CRAM_MAJOR_VERS(fd->version) >= 3) cram_compress_block(fd, b, NULL, -1, -1); if (blank_block) { c->length = b->comp_size + 2 + 4*is_cram_3 + itf8_size(b->content_id) + itf8_size(b->uncomp_size) + itf8_size(b->comp_size); c->num_blocks = 2; c->num_landmarks = 2; if (!(c->landmark = malloc(2*sizeof(*c->landmark)))) { cram_free_block(b); cram_free_container(c); return -1; } c->landmark[0] = 0; c->landmark[1] = c->length; // Plus extra storage for uncompressed secondary blank block padded_length = MIN(c->length*.5, 10000); c->length += padded_length + 2 + 4*is_cram_3 + itf8_size(b->content_id) + itf8_size(padded_length)*2; } else { // Pad the block instead. c->num_blocks = 1; c->num_landmarks = 1; if (!(c->landmark = malloc(sizeof(*c->landmark)))) return -1; c->landmark[0] = 0; padded_length = MAX(c->length*1.5, 10000) - c->length; c->length = b->comp_size + padded_length + 2 + 4*is_cram_3 + itf8_size(b->content_id) + itf8_size(b->uncomp_size) + itf8_size(b->comp_size); if (NULL == (pads = calloc(1, padded_length))) { cram_free_block(b); cram_free_container(c); return -1; } BLOCK_APPEND(b, pads, padded_length); BLOCK_UPLEN(b); free(pads); } if (-1 == cram_write_container(fd, c)) { cram_free_block(b); cram_free_container(c); return -1; } if (-1 == cram_write_block(fd, b)) { cram_free_block(b); cram_free_container(c); return -1; } if (blank_block) { BLOCK_RESIZE(b, padded_length); memset(BLOCK_DATA(b), 0, padded_length); BLOCK_SIZE(b) = padded_length; BLOCK_UPLEN(b); b->method = RAW; if (-1 == cram_write_block(fd, b)) { cram_free_block(b); cram_free_container(c); return -1; } } cram_free_block(b); cram_free_container(c); } if (-1 == refs_from_header(fd->refs, fd, fd->header)) return -1; if (-1 == refs2id(fd->refs, fd->header)) return -1; if (0 != hflush(fd->fp)) return -1; RP("=== Finishing saving header ===\n"); return 0; } /* ---------------------------------------------------------------------- * The top-level cram opening, closing and option handling */ /* * Initialises the lookup tables. These could be global statics, but they're * clumsy to setup in a multi-threaded environment unless we generate * verbatim code and include that. */ static void cram_init_tables(cram_fd *fd) { int i; memset(fd->L1, 4, 256); fd->L1['A'] = 0; fd->L1['a'] = 0; fd->L1['C'] = 1; fd->L1['c'] = 1; fd->L1['G'] = 2; fd->L1['g'] = 2; fd->L1['T'] = 3; fd->L1['t'] = 3; memset(fd->L2, 5, 256); fd->L2['A'] = 0; fd->L2['a'] = 0; fd->L2['C'] = 1; fd->L2['c'] = 1; fd->L2['G'] = 2; fd->L2['g'] = 2; fd->L2['T'] = 3; fd->L2['t'] = 3; fd->L2['N'] = 4; fd->L2['n'] = 4; if (CRAM_MAJOR_VERS(fd->version) == 1) { for (i = 0; i < 0x200; i++) { int f = 0; if (i & CRAM_FPAIRED) f |= BAM_FPAIRED; if (i & CRAM_FPROPER_PAIR) f |= BAM_FPROPER_PAIR; if (i & CRAM_FUNMAP) f |= BAM_FUNMAP; if (i & CRAM_FREVERSE) f |= BAM_FREVERSE; if (i & CRAM_FREAD1) f |= BAM_FREAD1; if (i & CRAM_FREAD2) f |= BAM_FREAD2; if (i & CRAM_FSECONDARY) f |= BAM_FSECONDARY; if (i & CRAM_FQCFAIL) f |= BAM_FQCFAIL; if (i & CRAM_FDUP) f |= BAM_FDUP; fd->bam_flag_swap[i] = f; } for (i = 0; i < 0x1000; i++) { int g = 0; if (i & BAM_FPAIRED) g |= CRAM_FPAIRED; if (i & BAM_FPROPER_PAIR) g |= CRAM_FPROPER_PAIR; if (i & BAM_FUNMAP) g |= CRAM_FUNMAP; if (i & BAM_FREVERSE) g |= CRAM_FREVERSE; if (i & BAM_FREAD1) g |= CRAM_FREAD1; if (i & BAM_FREAD2) g |= CRAM_FREAD2; if (i & BAM_FSECONDARY) g |= CRAM_FSECONDARY; if (i & BAM_FQCFAIL) g |= CRAM_FQCFAIL; if (i & BAM_FDUP) g |= CRAM_FDUP; fd->cram_flag_swap[i] = g; } } else { /* NOP */ for (i = 0; i < 0x1000; i++) fd->bam_flag_swap[i] = i; for (i = 0; i < 0x1000; i++) fd->cram_flag_swap[i] = i; } memset(fd->cram_sub_matrix, 4, 32*32); for (i = 0; i < 32; i++) { fd->cram_sub_matrix[i]['A'&0x1f]=0; fd->cram_sub_matrix[i]['C'&0x1f]=1; fd->cram_sub_matrix[i]['G'&0x1f]=2; fd->cram_sub_matrix[i]['T'&0x1f]=3; fd->cram_sub_matrix[i]['N'&0x1f]=4; } for (i = 0; i < 20; i+=4) { int j; for (j = 0; j < 20; j++) { fd->cram_sub_matrix["ACGTN"[i>>2]&0x1f][j]=3; fd->cram_sub_matrix["ACGTN"[i>>2]&0x1f][j]=3; fd->cram_sub_matrix["ACGTN"[i>>2]&0x1f][j]=3; fd->cram_sub_matrix["ACGTN"[i>>2]&0x1f][j]=3; } fd->cram_sub_matrix["ACGTN"[i>>2]&0x1f][CRAM_SUBST_MATRIX[i+0]&0x1f]=0; fd->cram_sub_matrix["ACGTN"[i>>2]&0x1f][CRAM_SUBST_MATRIX[i+1]&0x1f]=1; fd->cram_sub_matrix["ACGTN"[i>>2]&0x1f][CRAM_SUBST_MATRIX[i+2]&0x1f]=2; fd->cram_sub_matrix["ACGTN"[i>>2]&0x1f][CRAM_SUBST_MATRIX[i+3]&0x1f]=3; } } // Default version numbers for CRAM static int major_version = 3; static int minor_version = 0; /* * Opens a CRAM file for read (mode "rb") or write ("wb"). * The filename may be "-" to indicate stdin or stdout. * * Returns file handle on success * NULL on failure. */ cram_fd *cram_open(const char *filename, const char *mode) { hFILE *fp; cram_fd *fd; char fmode[3]= { mode[0], '\0', '\0' }; if (strlen(mode) > 1 && (mode[1] == 'b' || mode[1] == 'c')) { fmode[1] = 'b'; } fp = hopen(filename, fmode); if (!fp) return NULL; fd = cram_dopen(fp, filename, mode); if (!fd) hclose_abruptly(fp); return fd; } /* Opens an existing stream for reading or writing. * * Returns file handle on success; * NULL on failure. */ cram_fd *cram_dopen(hFILE *fp, const char *filename, const char *mode) { int i; char *cp; cram_fd *fd = calloc(1, sizeof(*fd)); if (!fd) return NULL; fd->level = 5; for (i = 0; mode[i]; i++) { if (mode[i] >= '0' && mode[i] <= '9') { fd->level = mode[i] - '0'; break; } } fd->fp = fp; fd->mode = *mode; fd->first_container = 0; if (fd->mode == 'r') { /* Reader */ if (!(fd->file_def = cram_read_file_def(fd))) goto err; fd->version = fd->file_def->major_version * 256 + fd->file_def->minor_version; if (!(fd->header = cram_read_SAM_hdr(fd))) goto err; } else { /* Writer */ cram_file_def *def = calloc(1, sizeof(*def)); if (!def) return NULL; fd->file_def = def; def->magic[0] = 'C'; def->magic[1] = 'R'; def->magic[2] = 'A'; def->magic[3] = 'M'; def->major_version = 0; // Indicator to write file def later. def->minor_version = 0; memset(def->file_id, 0, 20); strncpy(def->file_id, filename, 20); fd->version = major_version * 256 + minor_version; /* SAM header written later along with this file_def */ } cram_init_tables(fd); fd->prefix = strdup((cp = strrchr(filename, '/')) ? cp+1 : filename); if (!fd->prefix) goto err; fd->first_base = fd->last_base = -1; fd->record_counter = 0; fd->ctr = NULL; fd->refs = refs_create(); if (!fd->refs) goto err; fd->ref_id = -2; fd->ref = NULL; fd->decode_md = 0; fd->verbose = 0; fd->seqs_per_slice = SEQS_PER_SLICE; fd->slices_per_container = SLICE_PER_CNT; fd->embed_ref = 0; fd->no_ref = 0; fd->ignore_md5 = 0; fd->use_bz2 = 0; fd->use_rans = (CRAM_MAJOR_VERS(fd->version) >= 3); fd->use_lzma = 0; fd->multi_seq = -1; fd->unsorted = 0; fd->shared_ref = 0; fd->index = NULL; fd->own_pool = 0; fd->pool = NULL; fd->rqueue = NULL; fd->job_pending = NULL; fd->ooc = 0; fd->required_fields = INT_MAX; for (i = 0; i < DS_END; i++) fd->m[i] = cram_new_metrics(); fd->range.refid = -2; // no ref. fd->eof = 1; // See samtools issue #150 fd->ref_fn = NULL; fd->bl = NULL; /* Initialise dummy refs from the @SQ headers */ if (-1 == refs_from_header(fd->refs, fd, fd->header)) goto err; return fd; err: if (fd) free(fd); return NULL; } /* * Seek within a CRAM file. * * Returns 0 on success * -1 on failure */ int cram_seek(cram_fd *fd, off_t offset, int whence) { char buf[65536]; fd->ooc = 0; if (hseek(fd->fp, offset, whence) >= 0) return 0; if (!(whence == SEEK_CUR && offset >= 0)) return -1; /* Couldn't fseek, but we're in SEEK_CUR mode so read instead */ while (offset > 0) { int len = MIN(65536, offset); if (len != hread(fd->fp, buf, len)) return -1; offset -= len; } return 0; } /* * Flushes a CRAM file. * Useful for when writing to stdout without wishing to close the stream. * * Returns 0 on success * -1 on failure */ int cram_flush(cram_fd *fd) { if (!fd) return -1; if (fd->mode == 'w' && fd->ctr) { if(fd->ctr->slice) fd->ctr->curr_slice++; if (-1 == cram_flush_container_mt(fd, fd->ctr)) return -1; } return 0; } /* * Closes a CRAM file. * Returns 0 on success * -1 on failure */ int cram_close(cram_fd *fd) { spare_bams *bl, *next; int i; if (!fd) return -1; if (fd->mode == 'w' && fd->ctr) { if(fd->ctr->slice) fd->ctr->curr_slice++; if (-1 == cram_flush_container_mt(fd, fd->ctr)) return -1; } if (fd->pool && fd->eof >= 0) { t_pool_flush(fd->pool); if (0 != cram_flush_result(fd)) return -1; pthread_mutex_destroy(&fd->metrics_lock); pthread_mutex_destroy(&fd->ref_lock); pthread_mutex_destroy(&fd->bam_list_lock); fd->ctr = NULL; // prevent double freeing //fprintf(stderr, "CRAM: destroy queue %p\n", fd->rqueue); t_results_queue_destroy(fd->rqueue); } if (fd->mode == 'w') { /* Write EOF block */ if (CRAM_MAJOR_VERS(fd->version) == 3) { if (38 != hwrite(fd->fp, "\x0f\x00\x00\x00\xff\xff\xff\xff" // Cont HDR "\x0f\xe0\x45\x4f\x46\x00\x00\x00" // Cont HDR "\x00\x01\x00" // Cont HDR "\x05\xbd\xd9\x4f" // CRC32 "\x00\x01\x00\x06\x06" // Comp.HDR blk "\x01\x00\x01\x00\x01\x00" // Comp.HDR blk "\xee\x63\x01\x4b", // CRC32 38)) return -1; } else { if (30 != hwrite(fd->fp, "\x0b\x00\x00\x00\xff\xff\xff\xff" "\x0f\xe0\x45\x4f\x46\x00\x00\x00" "\x00\x01\x00\x00\x01\x00\x06\x06" "\x01\x00\x01\x00\x01\x00", 30)) return -1; } } for (bl = fd->bl; bl; bl = next) { int i, max_rec = fd->seqs_per_slice * fd->slices_per_container; next = bl->next; for (i = 0; i < max_rec; i++) { if (bl->bams[i]) bam_free(bl->bams[i]); } free(bl->bams); free(bl); } if (hclose(fd->fp) != 0) return -1; if (fd->file_def) cram_free_file_def(fd->file_def); if (fd->header) sam_hdr_free(fd->header); free(fd->prefix); if (fd->ctr) cram_free_container(fd->ctr); if (fd->refs) refs_free(fd->refs); if (fd->ref_free) free(fd->ref_free); for (i = 0; i < DS_END; i++) if (fd->m[i]) free(fd->m[i]); if (fd->index) cram_index_free(fd); if (fd->own_pool && fd->pool) t_pool_destroy(fd->pool, 0); free(fd); return 0; } /* * Returns 1 if we hit an EOF while reading. */ int cram_eof(cram_fd *fd) { return fd->eof; } /* * Sets options on the cram_fd. See CRAM_OPT_* definitions in cram_structs.h. * Use this immediately after opening. * * Returns 0 on success * -1 on failure */ int cram_set_option(cram_fd *fd, enum hts_fmt_option opt, ...) { int r; va_list args; va_start(args, opt); r = cram_set_voption(fd, opt, args); va_end(args); return r; } /* * Sets options on the cram_fd. See CRAM_OPT_* definitions in cram_structs.h. * Use this immediately after opening. * * Returns 0 on success * -1 on failure */ int cram_set_voption(cram_fd *fd, enum hts_fmt_option opt, va_list args) { refs_t *refs; if (!fd) { errno = EBADF; return -1; } switch (opt) { case CRAM_OPT_DECODE_MD: fd->decode_md = va_arg(args, int); break; case CRAM_OPT_PREFIX: if (fd->prefix) free(fd->prefix); if (!(fd->prefix = strdup(va_arg(args, char *)))) return -1; break; case CRAM_OPT_VERBOSITY: fd->verbose = va_arg(args, int); break; case CRAM_OPT_SEQS_PER_SLICE: fd->seqs_per_slice = va_arg(args, int); break; case CRAM_OPT_SLICES_PER_CONTAINER: fd->slices_per_container = va_arg(args, int); break; case CRAM_OPT_EMBED_REF: fd->embed_ref = va_arg(args, int); break; case CRAM_OPT_NO_REF: fd->no_ref = va_arg(args, int); break; case CRAM_OPT_IGNORE_MD5: fd->ignore_md5 = va_arg(args, int); break; case CRAM_OPT_USE_BZIP2: fd->use_bz2 = va_arg(args, int); break; case CRAM_OPT_USE_RANS: fd->use_rans = va_arg(args, int); break; case CRAM_OPT_USE_LZMA: fd->use_lzma = va_arg(args, int); break; case CRAM_OPT_SHARED_REF: fd->shared_ref = 1; refs = va_arg(args, refs_t *); if (refs != fd->refs) { if (fd->refs) refs_free(fd->refs); fd->refs = refs; fd->refs->count++; } break; case CRAM_OPT_RANGE: fd->range = *va_arg(args, cram_range *); return cram_seek_to_refpos(fd, &fd->range); case CRAM_OPT_REFERENCE: return cram_load_reference(fd, va_arg(args, char *)); case CRAM_OPT_VERSION: { int major, minor; char *s = va_arg(args, char *); if (2 != sscanf(s, "%d.%d", &major, &minor)) { fprintf(stderr, "Malformed version string %s\n", s); return -1; } if (!((major == 1 && minor == 0) || (major == 2 && (minor == 0 || minor == 1)) || (major == 3 && minor == 0))) { fprintf(stderr, "Unknown version string; " "use 1.0, 2.0, 2.1 or 3.0\n"); errno = EINVAL; return -1; } fd->version = major*256 + minor; if (CRAM_MAJOR_VERS(fd->version) >= 3) fd->use_rans = 1; break; } case CRAM_OPT_MULTI_SEQ_PER_SLICE: fd->multi_seq = va_arg(args, int); break; case CRAM_OPT_NTHREADS: { int nthreads = va_arg(args, int); if (nthreads > 1) { if (!(fd->pool = t_pool_init(nthreads*2, nthreads))) return -1; fd->rqueue = t_results_queue_init(); pthread_mutex_init(&fd->metrics_lock, NULL); pthread_mutex_init(&fd->ref_lock, NULL); pthread_mutex_init(&fd->bam_list_lock, NULL); fd->shared_ref = 1; fd->own_pool = 1; } break; } case CRAM_OPT_THREAD_POOL: fd->pool = va_arg(args, t_pool *); if (fd->pool) { fd->rqueue = t_results_queue_init(); pthread_mutex_init(&fd->metrics_lock, NULL); pthread_mutex_init(&fd->ref_lock, NULL); pthread_mutex_init(&fd->bam_list_lock, NULL); } fd->shared_ref = 1; // Needed to avoid clobbering ref between threads fd->own_pool = 0; //fd->qsize = 1; //fd->decoded = calloc(fd->qsize, sizeof(cram_container *)); //t_pool_dispatch(fd->pool, cram_decoder_thread, fd); break; case CRAM_OPT_REQUIRED_FIELDS: fd->required_fields = va_arg(args, int); break; case HTS_OPT_COMPRESSION_LEVEL: fd->level = va_arg(args, int); break; default: fprintf(stderr, "Unknown CRAM option code %d\n", opt); errno = EINVAL; return -1; } return 0; } stacks-1.44/htslib/cram/._cram_io.h000644 000765 000024 00000000265 12750743246 017725 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/cram_io.h000644 000765 000024 00000045156 12750743246 017520 0ustar00catchenstaff000000 000000 /* Copyright (c) 2012-2014 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /*! \file * Include cram.h instead. * * This is an internal part of the CRAM system and is automatically included * when you #include cram.h. * * Implements the low level CRAM I/O primitives. * This includes basic data types such as byte, int, ITF-8, * maps, bitwise I/O, etc. */ #ifndef _CRAM_IO_H_ #define _CRAM_IO_H_ #define ITF8_MACROS #include #include #ifdef __cplusplus extern "C" { #endif /**@{ ---------------------------------------------------------------------- * ITF8 encoding and decoding. * * Also see the itf8_get and itf8_put macros. */ /*! INTERNAL: Converts two characters into an integer for use in switch{} */ #define CRAM_KEY(a,b) (((a)<<8)|((b))) /*! Reads an integer in ITF-8 encoding from 'fd' and stores it in * *val. * * @return * Returns the number of bytes read on success; * -1 on failure */ int itf8_decode(cram_fd *fd, int32_t *val); #ifndef ITF8_MACROS /*! Reads an integer in ITF-8 encoding from 'cp' and stores it in * *val. * * @return * Returns the number of bytes read on success; * -1 on failure */ int itf8_get(char *cp, int32_t *val_p); /*! Stores a value to memory in ITF-8 format. * * @return * Returns the number of bytes required to store the number. * This is a maximum of 5 bytes. */ int itf8_put(char *cp, int32_t val); #else /* * Macro implementations of the above */ #define itf8_get(c,v) (((uc)(c)[0]<0x80)?(*(v)=(uc)(c)[0],1):(((uc)(c)[0]<0xc0)?(*(v)=(((uc)(c)[0]<<8)|(uc)(c)[1])&0x3fff,2):(((uc)(c)[0]<0xe0)?(*(v)=(((uc)(c)[0]<<16)|((uc)(c)[1]<<8)|(uc)(c)[2])&0x1fffff,3):(((uc)(c)[0]<0xf0)?(*(v)=(((uc)(c)[0]<<24)|((uc)(c)[1]<<16)|((uc)(c)[2]<<8)|(uc)(c)[3])&0x0fffffff,4):(*(v)=(((uc)(c)[0]&0x0f)<<28)|((uc)(c)[1]<<20)|((uc)(c)[2]<<12)|((uc)(c)[3]<<4)|((uc)(c)[4]&0x0f),5))))) #define itf8_put(c,v) ((!((v)&~0x7f))?((c)[0]=(v),1):(!((v)&~0x3fff))?((c)[0]=((v)>>8)|0x80,(c)[1]=(v)&0xff,2):(!((v)&~0x1fffff))?((c)[0]=((v)>>16)|0xc0,(c)[1]=((v)>>8)&0xff,(c)[2]=(v)&0xff,3):(!((v)&~0xfffffff))?((c)[0]=((v)>>24)|0xe0,(c)[1]=((v)>>16)&0xff,(c)[2]=((v)>>8)&0xff,(c)[3]=(v)&0xff,4):((c)[0]=0xf0|(((v)>>28)&0xff),(c)[1]=((v)>>20)&0xff,(c)[2]=((v)>>12)&0xff,(c)[3]=((v)>>4)&0xff,(c)[4]=(v)&0xf,5)) #define itf8_size(v) ((!((v)&~0x7f))?1:(!((v)&~0x3fff))?2:(!((v)&~0x1fffff))?3:(!((v)&~0xfffffff))?4:5) #endif int ltf8_get(char *cp, int64_t *val_p); int ltf8_put(char *cp, int64_t val); /* Version of itf8_get that checks it hasn't run out of input */ extern const int itf8_bytes[16]; static inline int safe_itf8_get(const char *cp, const char *endp, int32_t *val_p) { const unsigned char *up = (unsigned char *)cp; if (endp - cp < 5 && (cp >= endp || endp - cp < itf8_bytes[up[0]>>4])) { *val_p = 0; return 0; } if (up[0] < 0x80) { *val_p = up[0]; return 1; } else if (up[0] < 0xc0) { *val_p = ((up[0] <<8) | up[1]) & 0x3fff; return 2; } else if (up[0] < 0xe0) { *val_p = ((up[0]<<16) | (up[1]<< 8) | up[2]) & 0x1fffff; return 3; } else if (up[0] < 0xf0) { *val_p = ((up[0]<<24) | (up[1]<<16) | (up[2]<<8) | up[3]) & 0x0fffffff; return 4; } else { *val_p = ((up[0] & 0x0f)<<28) | (up[1]<<20) | (up[2]<<12) | (up[3]<<4) | (up[4] & 0x0f); return 5; } } /*! Pushes a value in ITF8 format onto the end of a block. * * This shouldn't be used for high-volume data as it is not the fastest * method. * * @return * Returns the number of bytes written */ int itf8_put_blk(cram_block *blk, int val); /*! Pulls a literal 32-bit value from a block. * * @returns the number of bytes decoded; * -1 on failure. */ int int32_get_blk(cram_block *b, int32_t *val); /*! Pushes a literal 32-bit value onto the end of a block. * * @return * Returns 0 on success; * -1 on failure. */ int int32_put_blk(cram_block *blk, int32_t val); /**@}*/ /**@{ ---------------------------------------------------------------------- * CRAM blocks - the dynamically growable data block. We have code to * create, update, (un)compress and read/write. * * These are derived from the deflate_interlaced.c blocks, but with the * CRAM extension of content types and IDs. */ /*! Allocates a new cram_block structure with a specified content_type and * id. * * @return * Returns block pointer on success; * NULL on failure */ cram_block *cram_new_block(enum cram_content_type content_type, int content_id); /*! Reads a block from a cram file. * * @return * Returns cram_block pointer on success; * NULL on failure */ cram_block *cram_read_block(cram_fd *fd); /*! Writes a CRAM block. * * @return * Returns 0 on success; * -1 on failure */ int cram_write_block(cram_fd *fd, cram_block *b); /*! Frees a CRAM block, deallocating internal data too. */ void cram_free_block(cram_block *b); /*! Uncompress a memory block using Zlib. * * @return * Returns 0 on success; * -1 on failure */ char *zlib_mem_inflate(char *cdata, size_t csize, size_t *size); /*! Uncompresses a CRAM block, if compressed. * * @return * Returns 0 on success; * -1 on failure */ int cram_uncompress_block(cram_block *b); /*! Compresses a block. * * Compresses a block using one of two different zlib strategies. If we only * want one choice set strat2 to be -1. * * The logic here is that sometimes Z_RLE does a better job than Z_FILTERED * or Z_DEFAULT_STRATEGY on quality data. If so, we'd rather use it as it is * significantly faster. * * @return * Returns 0 on success; * -1 on failure */ int cram_compress_block(cram_fd *fd, cram_block *b, cram_metrics *metrics, int method, int level); cram_metrics *cram_new_metrics(void); char *cram_block_method2str(enum cram_block_method m); char *cram_content_type2str(enum cram_content_type t); /* * Find an external block by its content_id */ static inline cram_block *cram_get_block_by_id(cram_slice *slice, int id) { if (slice->block_by_id && id >= 0 && id < 1024) { return slice->block_by_id[id]; } else { int i; for (i = 0; i < slice->hdr->num_blocks; i++) { cram_block *b = slice->block[i]; if (b && b->content_type == EXTERNAL && b->content_id == id) return b; } } return NULL; } /* --- Accessor macros for manipulating blocks on a byte by byte basis --- */ /* Block size and data pointer. */ #define BLOCK_SIZE(b) ((b)->byte) #define BLOCK_DATA(b) ((b)->data) /* Returns the address one past the end of the block */ #define BLOCK_END(b) (&(b)->data[(b)->byte]) /* Request block to be at least 'l' bytes long */ #define BLOCK_RESIZE(b,l) \ do { \ while((b)->alloc <= (l)) { \ (b)->alloc = (b)->alloc ? (b)->alloc*1.5 : 1024; \ (b)->data = realloc((b)->data, (b)->alloc); \ } \ } while(0) /* Make block exactly 'l' bytes long */ #define BLOCK_RESIZE_EXACT(b,l) \ do { \ (b)->alloc = (l); \ (b)->data = realloc((b)->data, (b)->alloc); \ } while(0) /* Ensure the block can hold at least another 'l' bytes */ #define BLOCK_GROW(b,l) BLOCK_RESIZE((b), BLOCK_SIZE((b)) + (l)) /* Append string 's' of length 'l' */ #define BLOCK_APPEND(b,s,l) \ do { \ BLOCK_GROW((b),(l)); \ memcpy(BLOCK_END((b)), (s), (l)); \ BLOCK_SIZE((b)) += (l); \ } while (0) /* Append as single character 'c' */ #define BLOCK_APPEND_CHAR(b,c) \ do { \ BLOCK_GROW((b),1); \ (b)->data[(b)->byte++] = (c); \ } while (0) /* Append a single unsigned integer */ #define BLOCK_APPEND_UINT(b,i) \ do { \ unsigned char *cp; \ BLOCK_GROW((b),11); \ cp = &(b)->data[(b)->byte]; \ (b)->byte += append_uint32(cp, (i)) - cp; \ } while (0) static inline unsigned char *append_uint32(unsigned char *cp, uint32_t i) { uint32_t j; if (i == 0) { *cp++ = '0'; return cp; } if (i < 100) goto b1; if (i < 10000) goto b3; if (i < 1000000) goto b5; if (i < 100000000) goto b7; if ((j = i / 1000000000)) {*cp++ = j + '0'; i -= j*1000000000; goto x8;} if ((j = i / 100000000)) {*cp++ = j + '0'; i -= j*100000000; goto x7;} b7:if ((j = i / 10000000)) {*cp++ = j + '0'; i -= j*10000000; goto x6;} if ((j = i / 1000000)) {*cp++ = j + '0', i -= j*1000000; goto x5;} b5:if ((j = i / 100000)) {*cp++ = j + '0', i -= j*100000; goto x4;} if ((j = i / 10000)) {*cp++ = j + '0', i -= j*10000; goto x3;} b3:if ((j = i / 1000)) {*cp++ = j + '0', i -= j*1000; goto x2;} if ((j = i / 100)) {*cp++ = j + '0', i -= j*100; goto x1;} b1:if ((j = i / 10)) {*cp++ = j + '0', i -= j*10; goto x0;} if (i) *cp++ = i + '0'; return cp; x8: *cp++ = i / 100000000 + '0', i %= 100000000; x7: *cp++ = i / 10000000 + '0', i %= 10000000; x6: *cp++ = i / 1000000 + '0', i %= 1000000; x5: *cp++ = i / 100000 + '0', i %= 100000; x4: *cp++ = i / 10000 + '0', i %= 10000; x3: *cp++ = i / 1000 + '0', i %= 1000; x2: *cp++ = i / 100 + '0', i %= 100; x1: *cp++ = i / 10 + '0', i %= 10; x0: *cp++ = i + '0'; return cp; } static inline unsigned char *append_sub32(unsigned char *cp, uint32_t i) { *cp++ = i / 100000000 + '0', i %= 100000000; *cp++ = i / 10000000 + '0', i %= 10000000; *cp++ = i / 1000000 + '0', i %= 1000000; *cp++ = i / 100000 + '0', i %= 100000; *cp++ = i / 10000 + '0', i %= 10000; *cp++ = i / 1000 + '0', i %= 1000; *cp++ = i / 100 + '0', i %= 100; *cp++ = i / 10 + '0', i %= 10; *cp++ = i + '0'; return cp; } static inline unsigned char *append_uint64(unsigned char *cp, uint64_t i) { uint64_t j; if (i <= 0xffffffff) return append_uint32(cp, i); if ((j = i/1000000000) > 1000000000) { cp = append_uint32(cp, j/1000000000); j %= 1000000000; cp = append_sub32(cp, j); } else { cp = append_uint32(cp, i / 1000000000); } cp = append_sub32(cp, i % 1000000000); return cp; } #define BLOCK_UPLEN(b) \ (b)->comp_size = (b)->uncomp_size = BLOCK_SIZE((b)) /**@}*/ /**@{ ---------------------------------------------------------------------- * Reference sequence handling */ /*! Loads a reference set from fn and stores in the cram_fd. * * @return * Returns 0 on success; * -1 on failure */ int cram_load_reference(cram_fd *fd, char *fn); /*! Generates a lookup table in refs based on the SQ headers in SAM_hdr. * * Indexes references by the order they appear in a BAM file. This may not * necessarily be the same order they appear in the fasta reference file. * * @return * Returns 0 on success; * -1 on failure */ int refs2id(refs_t *r, SAM_hdr *bfd); void refs_free(refs_t *r); /*! Returns a portion of a reference sequence from start to end inclusive. * * The returned pointer is owned by the cram_file fd and should not be freed * by the caller. It is valid only until the next cram_get_ref is called * with the same fd parameter (so is thread-safe if given multiple files). * * To return the entire reference sequence, specify start as 1 and end * as 0. * * @return * Returns reference on success; * NULL on failure */ char *cram_get_ref(cram_fd *fd, int id, int start, int end); void cram_ref_incr(refs_t *r, int id); void cram_ref_decr(refs_t *r, int id); /**@}*/ /**@{ ---------------------------------------------------------------------- * Containers */ /*! Creates a new container, specifying the maximum number of slices * and records permitted. * * @return * Returns cram_container ptr on success; * NULL on failure */ cram_container *cram_new_container(int nrec, int nslice); void cram_free_container(cram_container *c); /*! Reads a container header. * * @return * Returns cram_container on success; * NULL on failure or no container left (fd->err == 0). */ cram_container *cram_read_container(cram_fd *fd); /*! Writes a container structure. * * @return * Returns 0 on success; * -1 on failure */ int cram_write_container(cram_fd *fd, cram_container *h); /*! Flushes a container to disk. * * Flushes a completely or partially full container to disk, writing * container structure, header and blocks. This also calls the encoder * functions. * * @return * Returns 0 on success; * -1 on failure */ int cram_flush_container(cram_fd *fd, cram_container *c); int cram_flush_container_mt(cram_fd *fd, cram_container *c); /**@}*/ /**@{ ---------------------------------------------------------------------- * Compression headers; the first part of the container */ /*! Creates a new blank container compression header * * @return * Returns header ptr on success; * NULL on failure */ cram_block_compression_hdr *cram_new_compression_header(void); /*! Frees a cram_block_compression_hdr */ void cram_free_compression_header(cram_block_compression_hdr *hdr); /**@}*/ /**@{ ---------------------------------------------------------------------- * Slices and slice headers */ /*! Frees a slice header */ void cram_free_slice_header(cram_block_slice_hdr *hdr); /*! Frees a slice */ void cram_free_slice(cram_slice *s); /*! Creates a new empty slice in memory, for subsequent writing to * disk. * * @return * Returns cram_slice ptr on success; * NULL on failure */ cram_slice *cram_new_slice(enum cram_content_type type, int nrecs); /*! Loads an entire slice. * * FIXME: In 1.0 the native unit of slices within CRAM is broken * as slices contain references to objects in other slices. * To work around this while keeping the slice oriented outer loop * we read all slices and stitch them together into a fake large * slice instead. * * @return * Returns cram_slice ptr on success; * NULL on failure */ cram_slice *cram_read_slice(cram_fd *fd); /**@}*/ /**@{ ---------------------------------------------------------------------- * CRAM file definition (header) */ /*! Reads a CRAM file definition structure. * * @return * Returns file_def ptr on success; * NULL on failure */ cram_file_def *cram_read_file_def(cram_fd *fd); /*! Writes a cram_file_def structure to cram_fd. * * @return * Returns 0 on success; * -1 on failure */ int cram_write_file_def(cram_fd *fd, cram_file_def *def); /*! Frees a cram_file_def structure. */ void cram_free_file_def(cram_file_def *def); /**@}*/ /**@{ ---------------------------------------------------------------------- * SAM header I/O */ /*! Reads the SAM header from the first CRAM data block. * * Also performs minimal parsing to extract read-group * and sample information. * * @return * Returns SAM hdr ptr on success; * NULL on failure */ SAM_hdr *cram_read_SAM_hdr(cram_fd *fd); /*! Writes a CRAM SAM header. * * @return * Returns 0 on success; * -1 on failure */ int cram_write_SAM_hdr(cram_fd *fd, SAM_hdr *hdr); /**@}*/ /**@{ ---------------------------------------------------------------------- * The top-level cram opening, closing and option handling */ /*! Opens a CRAM file for read (mode "rb") or write ("wb"). * * The filename may be "-" to indicate stdin or stdout. * * @return * Returns file handle on success; * NULL on failure. */ cram_fd *cram_open(const char *filename, const char *mode); /*! Opens an existing stream for reading or writing. * * @return * Returns file handle on success; * NULL on failure. */ cram_fd *cram_dopen(struct hFILE *fp, const char *filename, const char *mode); /*! Closes a CRAM file. * * @return * Returns 0 on success; * -1 on failure */ int cram_close(cram_fd *fd); /* * Seek within a CRAM file. * * Returns 0 on success * -1 on failure */ int cram_seek(cram_fd *fd, off_t offset, int whence); /* * Flushes a CRAM file. * Useful for when writing to stdout without wishing to close the stream. * * Returns 0 on success * -1 on failure */ int cram_flush(cram_fd *fd); /*! Checks for end of file on a cram_fd stream. * * @return * Returns 0 if not at end of file * 1 if we hit an expected EOF (end of range or EOF block) * 2 for other EOF (end of stream without EOF block) */ int cram_eof(cram_fd *fd); /*! Sets options on the cram_fd. * * See CRAM_OPT_* definitions in cram_structs.h. * Use this immediately after opening. * * @return * Returns 0 on success; * -1 on failure */ int cram_set_option(cram_fd *fd, enum hts_fmt_option opt, ...); /*! Sets options on the cram_fd. * * See CRAM_OPT_* definitions in cram_structs.h. * Use this immediately after opening. * * @return * Returns 0 on success; * -1 on failure */ int cram_set_voption(cram_fd *fd, enum hts_fmt_option opt, va_list args); /*! * Attaches a header to a cram_fd. * * This should be used when creating a new cram_fd for writing where * we have an SAM_hdr already constructed (eg from a file we've read * in). * * @return * Returns 0 on success; * -1 on failure */ int cram_set_header(cram_fd *fd, SAM_hdr *hdr); #ifdef __cplusplus } #endif #endif /* _CRAM_IO_H_ */ stacks-1.44/htslib/cram/._cram_samtools.c000644 000765 000024 00000000265 12750743246 021152 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/cram_samtools.c000644 000765 000024 00000011456 12750743246 020741 0ustar00catchenstaff000000 000000 /* Copyright (c) 2010-2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #include #include #include #include #include "cram/cram.h" #include "htslib/sam.h" /*--------------------------------------------------------------------------- * Samtools compatibility portion */ int bam_construct_seq(bam_seq_t **bp, size_t extra_len, const char *qname, size_t qname_len, int flag, int rname, // Ref ID int pos, int end, // aligned start/end coords int mapq, uint32_t ncigar, const uint32_t *cigar, int mrnm, // Mate Ref ID int mpos, int isize, int len, const char *seq, const char *qual) { static const char L[256] = { 15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15, 15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15, 15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15, 15,15,15,15,15,15,15,15,15,15,15,15,15, 0,15,15, 15, 1,14, 2,13,15,15, 4,11,15,15,12,15, 3,15,15, 15,15, 5, 6, 8,15, 7, 9,15,10,15,15,15,15,15,15, 15, 1,14, 2,13,15,15, 4,11,15,15,12,15, 3,15,15, 15,15, 5, 6, 8,15, 7, 9,15,10,15,15,15,15,15,15, 15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15, 15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15, 15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15, 15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15, 15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15, 15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15, 15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15, 15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15 }; bam1_t *b = (bam1_t *)*bp; uint8_t *cp; int i, bam_len; //b->l_aux = extra_len; // we fill this out later bam_len = qname_len + 1 + ncigar*4 + (len+1)/2 + len + extra_len; if (b->m_data < bam_len) { b->m_data = bam_len; kroundup32(b->m_data); b->data = (uint8_t*)realloc(b->data, b->m_data); if (!b->data) return -1; } b->l_data = bam_len; b->core.tid = rname; b->core.pos = pos-1; b->core.bin = bam_reg2bin(pos, end); b->core.qual = mapq; b->core.l_qname = qname_len+1; b->core.flag = flag; b->core.n_cigar = ncigar; b->core.l_qseq = len; b->core.mtid = mrnm; b->core.mpos = mpos-1; b->core.isize = isize; cp = b->data; strncpy((char *)cp, qname, qname_len); cp[qname_len] = 0; cp += qname_len+1; memcpy(cp, cigar, ncigar*4); cp += ncigar*4; for (i = 0; i+1 < len; i+=2) { *cp++ = (L[(uc)seq[i]]<<4) + L[(uc)seq[i+1]]; } if (i < len) *cp++ = L[(uc)seq[i]]<<4; if (qual) memcpy(cp, qual, len); else memset(cp, '\xff', len); return 0; } bam_hdr_t *cram_header_to_bam(SAM_hdr *h) { int i; bam_hdr_t *header = bam_hdr_init(); header->l_text = ks_len(&h->text); header->text = malloc(header->l_text+1); memcpy(header->text, ks_str(&h->text), header->l_text); header->text[header->l_text] = 0; header->n_targets = h->nref; header->target_name = (char **)calloc(header->n_targets, sizeof(char *)); header->target_len = (uint32_t *)calloc(header->n_targets, 4); for (i = 0; i < h->nref; i++) { header->target_name[i] = strdup(h->ref[i].name); header->target_len[i] = h->ref[i].len; } return header; } SAM_hdr *bam_header_to_cram(bam_hdr_t *h) { return sam_hdr_parse_(h->text, h->l_text); } stacks-1.44/htslib/cram/._cram_samtools.h000644 000765 000024 00000000265 12750743246 021157 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/cram_samtools.h000644 000765 000024 00000006746 12750743246 020754 0ustar00catchenstaff000000 000000 /* Copyright (c) 2010-2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #ifndef _CRAM_SAMTOOLS_H_ #define _CRAM_SAMTOOLS_H_ /* Samtools compatible API */ #define bam_blk_size(b) ((b)->l_data) #define bam_set_blk_size(b,v) ((b)->data_len = (v)) #define bam_ref(b) (b)->core.tid #define bam_pos(b) (b)->core.pos #define bam_mate_pos(b) (b)->core.mpos #define bam_mate_ref(b) (b)->core.mtid #define bam_ins_size(b) (b)->core.isize #define bam_seq_len(b) (b)->core.l_qseq #define bam_cigar_len(b) (b)->core.n_cigar #define bam_flag(b) (b)->core.flag #define bam_bin(b) (b)->core.bin #define bam_map_qual(b) (b)->core.qual #define bam_name_len(b) (b)->core.l_qname #define bam_name(b) bam_get_qname((b)) #define bam_qual(b) bam_get_qual((b)) #define bam_seq(b) bam_get_seq((b)) #define bam_cigar(b) bam_get_cigar((b)) #define bam_aux(b) bam_get_aux((b)) #define bam_dup(b) bam_copy1(bam_init1(), (b)) #define bam_free(b) bam_destroy1((b)) #define bam_reg2bin(beg,end) hts_reg2bin((beg),(end),14,5) #include "htslib/sam.h" enum cigar_op { BAM_CMATCH_=BAM_CMATCH, BAM_CINS_=BAM_CINS, BAM_CDEL_=BAM_CDEL, BAM_CREF_SKIP_=BAM_CREF_SKIP, BAM_CSOFT_CLIP_=BAM_CSOFT_CLIP, BAM_CHARD_CLIP_=BAM_CHARD_CLIP, BAM_CPAD_=BAM_CPAD, BAM_CBASE_MATCH=BAM_CEQUAL, BAM_CBASE_MISMATCH=BAM_CDIFF }; typedef bam1_t bam_seq_t; #include "cram/sam_header.h" #ifdef __cplusplus extern "C" { #endif bam_hdr_t *cram_header_to_bam(SAM_hdr *h); SAM_hdr *bam_header_to_cram(bam_hdr_t *h); int bam_construct_seq(bam_seq_t **bp, size_t extra_len, const char *qname, size_t qname_len, int flag, int rname, // Ref ID int pos, int end, // aligned start/end coords int mapq, uint32_t ncigar, const uint32_t *cigar, int mrnm, // Mate Ref ID int mpos, int isize, int len, const char *seq, const char *qual); #ifdef __cplusplus } #endif #endif /* _CRAM_SAMTOOLS_H_ */ stacks-1.44/htslib/cram/._cram_stats.c000644 000765 000024 00000000265 12750743246 020447 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/cram_stats.c000644 000765 000024 00000027567 12750743246 020250 0ustar00catchenstaff000000 000000 /* Copyright (c) 2012-2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #include #include #include #include #include #include #include #include #include #include #include #include "cram/cram.h" #include "cram/os.h" cram_stats *cram_stats_create(void) { return calloc(1, sizeof(cram_stats)); } void cram_stats_add(cram_stats *st, int32_t val) { st->nsamp++; //assert(val >= 0); if (val < MAX_STAT_VAL && val >= 0) { st->freqs[val]++; } else { khint_t k; int r; if (!st->h) { st->h = kh_init(m_i2i); } k = kh_put(m_i2i, st->h, val, &r); if (r == 0) kh_val(st->h, k)++; else if (r != -1) kh_val(st->h, k) = 1; else ; // FIXME: handle error } } void cram_stats_del(cram_stats *st, int32_t val) { st->nsamp--; //assert(val >= 0); if (val < MAX_STAT_VAL && val >= 0) { st->freqs[val]--; assert(st->freqs[val] >= 0); } else if (st->h) { khint_t k = kh_get(m_i2i, st->h, val); if (k != kh_end(st->h)) { if (--kh_val(st->h, k) == 0) kh_del(m_i2i, st->h, k); } else { fprintf(stderr, "Failed to remove val %d from cram_stats\n", val); st->nsamp++; } } else { fprintf(stderr, "Failed to remove val %d from cram_stats\n", val); st->nsamp++; } } void cram_stats_dump(cram_stats *st) { int i; fprintf(stderr, "cram_stats:\n"); for (i = 0; i < MAX_STAT_VAL; i++) { if (!st->freqs[i]) continue; fprintf(stderr, "\t%d\t%d\n", i, st->freqs[i]); } if (st->h) { khint_t k; for (k = kh_begin(st->h); k != kh_end(st->h); k++) { if (!kh_exist(st->h, k)) continue; fprintf(stderr, "\t%d\t%d\n", kh_key(st->h, k), kh_val(st->h, k)); } } } #if 1 /* Returns the number of bits set in val; it the highest bit used */ static int nbits(int v) { static const int MultiplyDeBruijnBitPosition[32] = { 1, 10, 2, 11, 14, 22, 3, 30, 12, 15, 17, 19, 23, 26, 4, 31, 9, 13, 21, 29, 16, 18, 25, 8, 20, 28, 24, 7, 27, 6, 5, 32 }; v |= v >> 1; // first up to set all bits 1 after the first 1 */ v |= v >> 2; v |= v >> 4; v |= v >> 8; v |= v >> 16; // DeBruijn magic to find top bit return MultiplyDeBruijnBitPosition[(uint32_t)(v * 0x07C4ACDDU) >> 27]; } #endif /* * Computes entropy from integer frequencies for various encoding methods and * picks the best encoding. * * FIXME: we could reuse some of the code here for the actual encoding * parameters too. Eg the best 'k' for SUBEXP or the code lengths for huffman. * * Returns the best codec to use. */ enum cram_encoding cram_stats_encoding(cram_fd *fd, cram_stats *st) { enum cram_encoding best_encoding = E_NULL; int best_size = INT_MAX, bits; int nvals, i, ntot = 0, max_val = 0, min_val = INT_MAX, k; int *vals = NULL, *freqs = NULL, vals_alloc = 0, *codes; //cram_stats_dump(st); /* Count number of unique symbols */ for (nvals = i = 0; i < MAX_STAT_VAL; i++) { if (!st->freqs[i]) continue; if (nvals >= vals_alloc) { vals_alloc = vals_alloc ? vals_alloc*2 : 1024; vals = realloc(vals, vals_alloc * sizeof(int)); freqs = realloc(freqs, vals_alloc * sizeof(int)); if (!vals || !freqs) { if (vals) free(vals); if (freqs) free(freqs); return E_HUFFMAN; // Cannot do much else atm } } vals[nvals] = i; freqs[nvals] = st->freqs[i]; ntot += freqs[nvals]; if (max_val < i) max_val = i; if (min_val > i) min_val = i; nvals++; } if (st->h) { khint_t k; int i; for (k = kh_begin(st->h); k != kh_end(st->h); k++) { if (!kh_exist(st->h, k)) continue; if (nvals >= vals_alloc) { vals_alloc = vals_alloc ? vals_alloc*2 : 1024; vals = realloc(vals, vals_alloc * sizeof(int)); freqs = realloc(freqs, vals_alloc * sizeof(int)); if (!vals || !freqs) return E_HUFFMAN; // Cannot do much else atm } i = kh_key(st->h, k); vals[nvals]=i; freqs[nvals] = kh_val(st->h, k); ntot += freqs[nvals]; if (max_val < i) max_val = i; if (min_val > i) min_val = i; nvals++; } } st->nvals = nvals; assert(ntot == st->nsamp); if (nvals <= 1) { free(vals); free(freqs); return E_HUFFMAN; } if (fd->verbose > 1) fprintf(stderr, "Range = %d..%d, nvals=%d, ntot=%d\n", min_val, max_val, nvals, ntot); /* Theoretical entropy */ // if (fd->verbose > 1) { // double dbits = 0; // for (i = 0; i < nvals; i++) { // dbits += freqs[i] * log((double)freqs[i]/ntot); // } // dbits /= -log(2); // if (fd->verbose > 1) // fprintf(stderr, "Entropy = %f\n", dbits); // } if (nvals > 1 && ntot > 256) { #if 0 /* * CRUDE huffman estimator. Round to closest and round up from 0 * to 1 bit. * * With and without ITF8 incase we have a few discrete values but with * large magnitude. * * Note rans0/arith0 and Z_HUFFMAN_ONLY vs internal huffman can be * compared in this way, but order-1 (eg rans1) or maybe LZ77 modes * may detect the correlation of high bytes to low bytes in multi- * byte values. So this predictor breaks down. */ double dbits = 0; // entropy + ~huffman double dbitsH = 0; double dbitsE = 0; // external entropy + ~huffman double dbitsEH = 0; int F[256] = {0}, n = 0; double e = 0; // accumulated error bits for (i = 0; i < nvals; i++) { double x; int X; unsigned int v = vals[i]; //Better encoding would cope with sign. //v = ABS(vals[i])*2+(vals[i]<0); if (!(v & ~0x7f)) { F[v] += freqs[i], n+=freqs[i]; } else if (!(v & ~0x3fff)) { F[(v>>8) |0x80] += freqs[i]; F[ v &0xff] += freqs[i], n+=2*freqs[i]; } else if (!(v & ~0x1fffff)) { F[(v>>16)|0xc0] += freqs[i]; F[(v>>8 )&0xff] += freqs[i]; F[ v &0xff] += freqs[i], n+=3*freqs[i]; } else if (!(v & ~0x0fffffff)) { F[(v>>24)|0xe0] += freqs[i]; F[(v>>16)&0xff] += freqs[i]; F[(v>>8 )&0xff] += freqs[i]; F[ v &0xff] += freqs[i], n+=4*freqs[i]; } else { F[(v>>28)|0xf0] += freqs[i]; F[(v>>20)&0xff] += freqs[i]; F[(v>>12)&0xff] += freqs[i]; F[(v>>4 )&0xff] += freqs[i]; F[ v &0x0f] += freqs[i], n+=5*freqs[i]; } x = -log((double)freqs[i]/ntot)/.69314718055994530941; X = x+0.5; if ((int)(x+((double)e/freqs[i])+.5)>X) { X++; } else if ((int)(x+((double)e/freqs[i])+.5) 1.1) { //fprintf(stderr, "=> %d < 200 ? E_HUFFMAN : E_BETA\n", nvals); free(vals); free(freqs); return nvals < 200 ? E_HUFFMAN : E_BETA; } #endif free(vals); free(freqs); return E_EXTERNAL; } /* * Avoid complex stats for now, just do heuristic of HUFFMAN for small * alphabets and BETA for anything large. */ free(vals); free(freqs); return nvals < 200 ? E_HUFFMAN : E_BETA; //return E_HUFFMAN; //return E_EXTERNAL; /* We only support huffman now anyway... */ //free(vals); free(freqs); return E_HUFFMAN; /* Beta */ bits = nbits(max_val - min_val) * ntot; if (fd->verbose > 1) fprintf(stderr, "BETA = %d\n", bits); if (best_size > bits) best_size = bits, best_encoding = E_BETA; #if 0 /* Unary */ if (min_val >= 0) { for (bits = i = 0; i < nvals; i++) bits += freqs[i]*(vals[i]+1); if (fd->verbose > 1) fprintf(stderr, "UNARY = %d\n", bits); if (best_size > bits) best_size = bits, best_encoding = E_NULL; //E_UNARY; } /* Gamma */ for (bits = i = 0; i < nvals; i++) bits += ((nbits(vals[i]-min_val+1)-1) + nbits(vals[i]-min_val+1)) * freqs[i]; if (fd->verbose > 1) fprintf(stderr, "GAMMA = %d\n", bits); if (best_size > bits) best_size = bits, best_encoding = E_GAMMA; /* Subexponential */ for (k = 0; k < 10; k++) { for (bits = i = 0; i < nvals; i++) { if (vals[i]-min_val < (1<verbose > 1) fprintf(stderr, "SUBEXP%d = %d\n", k, bits); if (best_size > bits) best_size = bits, best_encoding = E_SUBEXP; } #endif /* byte array len */ /* byte array stop */ /* External? Guesswork! */ /* Huffman */ // qsort(freqs, nvals, sizeof(freqs[0]), sort_freqs); // for (i = 0; i < nvals; i++) { // fprintf(stderr, "%d = %d\n", i, freqs[i]); // vals[i] = 0; // } /* Grow freqs to 2*freqs, to store sums */ /* Vals holds link data */ freqs = realloc(freqs, 2*nvals*sizeof(*freqs)); codes = calloc(2*nvals, sizeof(*codes)); if (!freqs || !codes) return E_HUFFMAN; // Cannot do much else atm /* Inefficient, use pointers to form chain so we can insert and maintain * a sorted list? This is currently O(nvals^2) complexity. */ for (;;) { int low1 = INT_MAX, low2 = INT_MAX; int ind1 = 0, ind2 = 0; for (i = 0; i < nvals; i++) { if (freqs[i] < 0) continue; if (low1 > freqs[i]) low2 = low1, ind2 = ind1, low1 = freqs[i], ind1 = i; else if (low2 > freqs[i]) low2 = freqs[i], ind2 = i; } if (low2 == INT_MAX) break; //fprintf(stderr, "Merge ind %d (%d), %d (%d) = %d+%d, => %d=%d\n", // ind1, vals[ind1], ind2, vals[ind2], low1, low2, // nvals, low1+low2); freqs[nvals] = low1 + low2; codes[ind1] = nvals; codes[ind2] = nvals; freqs[ind1] *= -1; freqs[ind2] *= -1; nvals++; } nvals = nvals/2+1; for (i = 0; i < nvals; i++) { int code_len = 0; for (k = codes[i]; k; k = codes[k]) code_len++; codes[i] = code_len; freqs[i] *= -1; //fprintf(stderr, "%d / %d => %d\n", vals[i], freqs[i], codes[i]); } for (bits = i = 0; i < nvals; i++) { bits += freqs[i] * codes[i]; } if (fd->verbose > 1) fprintf(stderr, "HUFFMAN = %d\n", bits); if (best_size >= bits) best_size = bits, best_encoding = E_HUFFMAN; free(codes); free(vals); free(freqs); return best_encoding; } void cram_stats_free(cram_stats *st) { if (st->h) kh_destroy(m_i2i, st->h); free(st); } stacks-1.44/htslib/cram/._cram_stats.h000644 000765 000024 00000000265 12750743246 020454 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/cram_stats.h000644 000765 000024 00000004347 12750743246 020244 0ustar00catchenstaff000000 000000 /* Copyright (c) 2012-2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #ifndef _CRAM_STATS_H_ #define _CRAM_STATS_H_ #ifdef __cplusplus extern "C" { #endif cram_stats *cram_stats_create(void); void cram_stats_add(cram_stats *st, int32_t val); void cram_stats_del(cram_stats *st, int32_t val); void cram_stats_dump(cram_stats *st); void cram_stats_free(cram_stats *st); /* * Computes entropy from integer frequencies for various encoding methods and * picks the best encoding. * * FIXME: we could reuse some of the code here for the actual encoding * parameters too. Eg the best 'k' for SUBEXP or the code lengths for huffman. * * Returns the best codec to use. */ enum cram_encoding cram_stats_encoding(cram_fd *fd, cram_stats *st); #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/cram/._cram_structs.h000644 000765 000024 00000000265 12750743246 021025 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/cram_structs.h000644 000765 000024 00000054440 12750743246 020614 0ustar00catchenstaff000000 000000 /* Copyright (c) 2012-2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #ifndef _CRAM_STRUCTS_H_ #define _CRAM_STRUCTS_H_ /* * Defines in-memory structs for the basic file-format objects in the * CRAM format. * * The basic file format is: * File-def SAM-hdr Container Container ... * * Container: * Service-block data-block data-block ... * * Multiple blocks in a container are grouped together as slices, * also sometimes referred to as landmarks in the spec. */ #include #include "cram/thread_pool.h" #include "cram/string_alloc.h" #include "cram/mFILE.h" #include "htslib/khash.h" #ifdef __cplusplus extern "C" { #endif // Generic hash-map integer -> integer KHASH_MAP_INIT_INT(m_i2i, int) // Generic hash-set integer -> (existance) KHASH_SET_INIT_INT(s_i2i) // For brevity typedef unsigned char uc; /* * A union for the preservation map. Required for khash. */ typedef union { int i; char *p; } pmap_t; // Generates static functions here which isn't ideal, but we have no way // currently to declare the kh_map_t structure here without also declaring a // duplicate in the .c files due to the nature of the KHASH macros. KHASH_MAP_INIT_STR(map, pmap_t) struct hFILE; #define SEQS_PER_SLICE 10000 #define SLICE_PER_CNT 1 #define CRAM_SUBST_MATRIX "CGTNAGTNACTNACGNACGT" #define MAX_STAT_VAL 1024 //#define MAX_STAT_VAL 16 typedef struct cram_stats { int freqs[MAX_STAT_VAL]; khash_t(m_i2i) *h; int nsamp; // total number of values added int nvals; // total number of unique values added } cram_stats; /* NB: matches java impl, not the spec */ enum cram_encoding { E_NULL = 0, E_EXTERNAL = 1, E_GOLOMB = 2, E_HUFFMAN = 3, E_BYTE_ARRAY_LEN = 4, E_BYTE_ARRAY_STOP = 5, E_BETA = 6, E_SUBEXP = 7, E_GOLOMB_RICE = 8, E_GAMMA = 9, E_NUM_CODECS = 10, /* Number of codecs, not a real one. */ }; enum cram_external_type { E_INT = 1, E_LONG = 2, E_BYTE = 3, E_BYTE_ARRAY = 4, E_BYTE_ARRAY_BLOCK = 5, }; /* External IDs used by this implementation (only assumed during writing) */ enum cram_DS_ID { DS_CORE = 0, DS_aux = 1, // aux_blk DS_aux_OQ = 2, DS_aux_BQ = 3, DS_aux_BD = 4, DS_aux_BI = 5, DS_aux_FZ = 6, // also ZM:B DS_aux_oq = 7, // other qualities DS_aux_os = 8, // other sequences DS_aux_oz = 9, // other strings DS_ref, DS_RN, // name_blk DS_QS, // qual_blk DS_IN, // base_blk DS_SC, // soft_blk DS_BF, // start loop DS_CF, DS_AP, DS_RG, DS_MQ, DS_NS, DS_MF, DS_TS, DS_NP, DS_NF, DS_RL, DS_FN, DS_FC, DS_FP, DS_DL, DS_BA, DS_BS, DS_TL, DS_RI, DS_RS, DS_PD, DS_HC, DS_BB, DS_QQ, DS_TN, // end loop DS_RN_len, DS_SC_len, DS_BB_len, DS_QQ_len, DS_TC, // CRAM v1.0 tags DS_TM, // test DS_TV, // test DS_END, }; /* "File Definition Structure" */ typedef struct cram_file_def { char magic[4]; uint8_t major_version; uint8_t minor_version; char file_id[20]; // Filename or SHA1 checksum } cram_file_def; #define CRAM_MAJOR_VERS(v) ((v) >> 8) #define CRAM_MINOR_VERS(v) ((v) & 0xff) struct cram_slice; enum cram_block_method { ERROR = -1, RAW = 0, GZIP = 1, BZIP2 = 2, LZMA = 3, RANS = 4, // Generic; either order RANS0 = 4, RANS1 = 10, // Not externalised; stored as RANS (generic) GZIP_RLE = 11, // NB: not externalised in CRAM }; enum cram_content_type { CT_ERROR = -1, FILE_HEADER = 0, COMPRESSION_HEADER = 1, MAPPED_SLICE = 2, UNMAPPED_SLICE = 3, // CRAM V1.0 only EXTERNAL = 4, CORE = 5, }; /* Compression metrics */ typedef struct { // number of trials and time to next trial int trial; int next_trial; // aggregate sizes during trials int sz_gz_rle; int sz_gz_def; int sz_rans0; int sz_rans1; int sz_bzip2; int sz_lzma; // resultant method from trials int method; int strat; // Revisions of method, to allow culling of continually failing ones. int gz_rle_cnt; int gz_def_cnt; int rans0_cnt; int rans1_cnt; int bzip2_cnt; int lzma_cnt; int revised_method; double gz_rle_extra; double gz_def_extra; double rans0_extra; double rans1_extra; double bzip2_extra; double lzma_extra; } cram_metrics; /* Block */ typedef struct cram_block { enum cram_block_method method, orig_method; enum cram_content_type content_type; int32_t content_id; int32_t comp_size; int32_t uncomp_size; uint32_t crc32; int32_t idx; /* offset into data */ unsigned char *data; // For bit I/O size_t alloc; size_t byte; int bit; } cram_block; struct cram_codec; /* defined in cram_codecs.h */ struct cram_map; #define CRAM_MAP_HASH 32 #define CRAM_MAP(a,b) (((a)*3+(b))&(CRAM_MAP_HASH-1)) /* Compression header block */ typedef struct cram_block_compression_hdr { int32_t ref_seq_id; int32_t ref_seq_start; int32_t ref_seq_span; int32_t num_records; int32_t num_landmarks; int32_t *landmark; /* Flags from preservation map */ int mapped_qs_included; int unmapped_qs_included; int unmapped_placed; int qs_included; int read_names_included; int AP_delta; // indexed by ref-base and subst. code char substitution_matrix[5][4]; // TD Dictionary as a concatenated block cram_block *TD_blk; // Tag Dictionary int nTL; // number of TL entries in TD unsigned char **TL; // array of size nTL, pointer into TD_blk. khash_t(m_s2i) *TD_hash; // Keyed on TD strings, map to TL[] indices string_alloc_t *TD_keys; // Pooled keys for TD hash. khash_t(map) *preservation_map; struct cram_map *rec_encoding_map[CRAM_MAP_HASH]; struct cram_map *tag_encoding_map[CRAM_MAP_HASH]; struct cram_codec *codecs[DS_END]; char *uncomp; // A single block of uncompressed data size_t uncomp_size, uncomp_alloc; unsigned int data_series; // See cram_fields enum below } cram_block_compression_hdr; typedef struct cram_map { int key; /* 0xe0 + 3 bytes */ enum cram_encoding encoding; int offset; /* Offset into a single block of memory */ int size; /* Size */ struct cram_codec *codec; struct cram_map *next; // for noddy internal hash } cram_map; /* Mapped or unmapped slice header block */ typedef struct cram_block_slice_hdr { enum cram_content_type content_type; int32_t ref_seq_id; /* if content_type == MAPPED_SLICE */ int32_t ref_seq_start; /* if content_type == MAPPED_SLICE */ int32_t ref_seq_span; /* if content_type == MAPPED_SLICE */ int32_t num_records; int64_t record_counter; int32_t num_blocks; int32_t num_content_ids; int32_t *block_content_ids; int32_t ref_base_id; /* if content_type == MAPPED_SLICE */ unsigned char md5[16]; } cram_block_slice_hdr; struct ref_entry; /* * Container. * * Conceptually a container is split into slices, and slices into blocks. * However on disk it's just a list of blocks and we need to query the * block types to identify the start/end points of the slices. * * OR... are landmarks the start/end points of slices? */ typedef struct cram_container { int32_t length; int32_t ref_seq_id; int32_t ref_seq_start; int32_t ref_seq_span; int64_t record_counter; int64_t num_bases; int32_t num_records; int32_t num_blocks; int32_t num_landmarks; int32_t *landmark; /* Size of container header above */ size_t offset; /* Compression header is always the first block? */ cram_block_compression_hdr *comp_hdr; cram_block *comp_hdr_block; /* For construction purposes */ int max_slice, curr_slice; // maximum number of slices int max_rec, curr_rec; // current and max recs per slice int max_c_rec, curr_c_rec; // current and max recs per container int slice_rec; // rec no. for start of this slice int curr_ref; // current ref ID. -2 for no previous int last_pos; // last record position struct cram_slice **slices, *slice; int pos_sorted; // boolean, 1=>position sorted data int max_apos; // maximum position, used if pos_sorted==0 int last_slice; // number of reads in last slice (0 for 1st) int multi_seq; // true if packing multi seqs per cont/slice int unsorted; // true is AP_delta is 0. /* Copied from fd before encoding, to allow multi-threading */ int ref_start, first_base, last_base, ref_id, ref_end; char *ref; //struct ref_entry *ref; /* For multi-threading */ bam_seq_t **bams; /* Statistics for encoding */ cram_stats *stats[DS_END]; khash_t(s_i2i) *tags_used; // set of tag types in use, for tag encoding map int *refs_used; // array of frequency of ref seq IDs uint32_t crc32; // CRC32 } cram_container; /* * A single cram record */ typedef struct cram_record { struct cram_slice *s; // Filled out by cram_decode only int32_t ref_id; // fixed for all recs in slice? int32_t flags; // BF int32_t cram_flags; // CF int32_t len; // RL int32_t apos; // AP int32_t rg; // RG int32_t name; // RN; idx to s->names_blk int32_t name_len; int32_t mate_line; // index to another cram_record int32_t mate_ref_id; int32_t mate_pos; // NP int32_t tlen; // TS // Auxiliary data int32_t ntags; // TC int32_t aux; // idx to s->aux_blk int32_t aux_size; // total size of packed ntags in aux_blk #ifndef TN_external int32_t TN_idx; // TN; idx to s->TN; #else int32_t tn; // idx to s->tn_blk #endif int TL; int32_t seq; // idx to s->seqs_blk int32_t qual; // idx to s->qual_blk int32_t cigar; // idx to s->cigar int32_t ncigar; int32_t aend; // alignment end int32_t mqual; // MQ int32_t feature; // idx to s->feature int32_t nfeature; // number of features int32_t mate_flags; // MF } cram_record; // Accessor macros as an analogue of the bam ones #define cram_qname(c) (&(c)->s->name_blk->data[(c)->name]) #define cram_seq(c) (&(c)->s->seqs_blk->data[(c)->seq]) #define cram_qual(c) (&(c)->s->qual_blk->data[(c)->qual]) #define cram_aux(c) (&(c)->s->aux_blk->data[(c)->aux]) #define cram_seqi(c,i) (cram_seq((c))[(i)]) #define cram_name_len(c) ((c)->name_len) #define cram_strand(c) (((c)->flags & BAM_FREVERSE) != 0) #define cram_mstrand(c) (((c)->flags & BAM_FMREVERSE) != 0) #define cram_cigar(c) (&((cr)->s->cigar)[(c)->cigar]) /* * A feature is a base difference, used for the sequence reference encoding. * (We generate these internally when writing CRAM.) */ typedef struct cram_feature { union { struct { int pos; int code; int base; // substitution code } X; struct { int pos; int code; int base; // actual base & qual int qual; } B; struct { int pos; int code; int seq_idx; // index to s->seqs_blk int len; } b; struct { int pos; int code; int qual; } Q; struct { int pos; int code; int len; int seq_idx; // soft-clip multiple bases } S; struct { int pos; int code; int len; int seq_idx; // insertion multiple bases } I; struct { int pos; int code; int base; // insertion single base } i; struct { int pos; int code; int len; } D; struct { int pos; int code; int len; } N; struct { int pos; int code; int len; } P; struct { int pos; int code; int len; } H; }; } cram_feature; /* * A slice is really just a set of blocks, but it * is the logical unit for decoding a number of * sequences. */ typedef struct cram_slice { cram_block_slice_hdr *hdr; cram_block *hdr_block; cram_block **block; cram_block **block_by_id; /* State used during encoding/decoding */ int last_apos, max_apos; /* Array of decoded cram records */ cram_record *crecs; /* An dynamically growing buffers for data pointed * to by crecs[] array. */ uint32_t *cigar; uint32_t cigar_alloc; uint32_t ncigar; cram_feature *features; int nfeatures; int afeatures; // allocated size of features #ifndef TN_external // TN field (Tag Name) uint32_t *TN; int nTN, aTN; // used and allocated size for TN[] #else cram_block *tn_blk; int tn_id; #endif // For variable sized elements which are always external blocks. cram_block *name_blk; cram_block *seqs_blk; cram_block *qual_blk; cram_block *base_blk; cram_block *soft_blk; cram_block *aux_blk; cram_block *aux_OQ_blk; cram_block *aux_BQ_blk; cram_block *aux_BD_blk; cram_block *aux_BI_blk; cram_block *aux_FZ_blk; cram_block *aux_oq_blk; cram_block *aux_os_blk; cram_block *aux_oz_blk; string_alloc_t *pair_keys; // Pooled keys for pair hash. khash_t(m_s2i) *pair[2]; // for identifying read-pairs in this slice. char *ref; // slice of current reference int ref_start; // start position of current reference; int ref_end; // end position of current reference; int ref_id; } cram_slice; /*----------------------------------------------------------------------------- * Consider moving reference handling to cram_refs.[ch] */ // from fa.fai / samtools faidx files typedef struct ref_entry { char *name; char *fn; int64_t length; int64_t offset; int bases_per_line; int line_length; int64_t count; // for shared references so we know to dealloc seq char *seq; mFILE *mf; int is_md5; // Reference comes from a raw seq found by MD5 } ref_entry; KHASH_MAP_INIT_STR(refs, ref_entry*) // References structure. typedef struct { string_alloc_t *pool; // String pool for holding filenames and SN vals khash_t(refs) *h_meta; // ref_entry*, index by name ref_entry **ref_id; // ref_entry*, index by ID int nref; // number of ref_entry char *fn; // current file opened BGZF *fp; // and the hFILE* to go with it. int count; // how many cram_fd sharing this refs struct pthread_mutex_t lock; // Mutex for multi-threaded updating ref_entry *last; // Last queried sequence int last_id; // Used in cram_ref_decr_locked to delay free } refs_t; /*----------------------------------------------------------------------------- * CRAM index * * Detect format by number of entries per line. * 5 => 1.0 (refid, start, nseq, C offset, slice) * 6 => 1.1 (refid, start, span, C offset, S offset, S size) * * Indices are stored in a nested containment list, which is trivial to set * up as the indices are on sorted data so we're appending to the nclist * in sorted order. Basically if a slice entirely fits within a previous * slice then we append to that slices list. This is done recursively. * * Lists are sorted on two dimensions: ref id + slice coords. */ typedef struct cram_index { int nslice, nalloc; // total number of slices struct cram_index *e; // array of size nslice int refid; // 1.0 1.1 int start; // 1.0 1.1 int end; // 1.1 int nseq; // 1.0 - undocumented int slice; // 1.0 landmark index, 1.1 landmark value int len; // 1.1 - size of slice in bytes int64_t offset; // 1.0 1.1 } cram_index; typedef struct { int refid; int start; int end; } cram_range; /*----------------------------------------------------------------------------- */ /* CRAM File handle */ typedef struct spare_bams { bam_seq_t **bams; struct spare_bams *next; } spare_bams; typedef struct cram_fd { struct hFILE *fp; int mode; // 'r' or 'w' int version; cram_file_def *file_def; SAM_hdr *header; char *prefix; int64_t record_counter; int err; // Most recent compression header decoded //cram_block_compression_hdr *comp_hdr; //cram_block_slice_hdr *slice_hdr; // Current container being processed. cram_container *ctr; // positions for encoding or decoding int first_base, last_base; // cached reference portion refs_t *refs; // ref meta-data structure char *ref, *ref_free; // current portion held in memory int ref_id; int ref_start; int ref_end; char *ref_fn; // reference fasta filename // compression level and metrics int level; cram_metrics *m[DS_END]; // options int decode_md; // Whether to export MD and NM tags int verbose; int seqs_per_slice; int slices_per_container; int embed_ref; int no_ref; int ignore_md5; int use_bz2; int use_rans; int use_lzma; int shared_ref; unsigned int required_fields; cram_range range; // lookup tables, stored here so we can be trivially multi-threaded unsigned int bam_flag_swap[0x1000]; // cram -> bam flags unsigned int cram_flag_swap[0x1000];// bam -> cram flags unsigned char L1[256]; // ACGT{*} ->0123{4} unsigned char L2[256]; // ACGTN{*}->01234{5} char cram_sub_matrix[32][32]; // base substituion codes int index_sz; cram_index *index; // array, sizeof index_sz off_t first_container; int eof; int last_slice; // number of recs encoded in last slice int multi_seq; int unsorted; int empty_container; // Marker for EOF block // thread pool int own_pool; t_pool *pool; t_results_queue *rqueue; pthread_mutex_t metrics_lock; pthread_mutex_t ref_lock; spare_bams *bl; pthread_mutex_t bam_list_lock; void *job_pending; int ooc; // out of containers. } cram_fd; // Translation of required fields to cram data series enum cram_fields { CRAM_BF = 0x00000001, CRAM_AP = 0x00000002, CRAM_FP = 0x00000004, CRAM_RL = 0x00000008, CRAM_DL = 0x00000010, CRAM_NF = 0x00000020, CRAM_BA = 0x00000040, CRAM_QS = 0x00000080, CRAM_FC = 0x00000100, CRAM_FN = 0x00000200, CRAM_BS = 0x00000400, CRAM_IN = 0x00000800, CRAM_RG = 0x00001000, CRAM_MQ = 0x00002000, CRAM_TL = 0x00004000, CRAM_RN = 0x00008000, CRAM_NS = 0x00010000, CRAM_NP = 0x00020000, CRAM_TS = 0x00040000, CRAM_MF = 0x00080000, CRAM_CF = 0x00100000, CRAM_RI = 0x00200000, CRAM_RS = 0x00400000, CRAM_PD = 0x00800000, CRAM_HC = 0x01000000, CRAM_SC = 0x02000000, CRAM_BB = 0x04000000, CRAM_BB_len = 0x08000000, CRAM_QQ = 0x10000000, CRAM_QQ_len = 0x20000000, CRAM_aux= 0x40000000, CRAM_ALL= 0x7fffffff, }; // A CIGAR opcode, but not necessarily the implications of it. Eg FC/FP may // encode a base difference, but we don't need to know what it is for CIGAR. // If we have a soft-clip or insertion, we do need SC/IN though to know how // long that array is. #define CRAM_CIGAR (CRAM_FN | CRAM_FP | CRAM_FC | CRAM_DL | CRAM_IN | \ CRAM_SC | CRAM_HC | CRAM_PD | CRAM_RS | CRAM_RL | CRAM_BF) #define CRAM_SEQ (CRAM_CIGAR | CRAM_BA | CRAM_BS | \ CRAM_RL | CRAM_AP | CRAM_BB) #define CRAM_QUAL (CRAM_CIGAR | CRAM_RL | CRAM_AP | CRAM_QS | CRAM_QQ) /* BF bitfields */ /* Corrected in 1.1. Use bam_flag_swap[bf] and BAM_* macros for 1.0 & 1.1 */ #define CRAM_FPAIRED 256 #define CRAM_FPROPER_PAIR 128 #define CRAM_FUNMAP 64 #define CRAM_FREVERSE 32 #define CRAM_FREAD1 16 #define CRAM_FREAD2 8 #define CRAM_FSECONDARY 4 #define CRAM_FQCFAIL 2 #define CRAM_FDUP 1 #define DS_aux_S "\001" #define DS_aux_OQ_S "\002" #define DS_aux_BQ_S "\003" #define DS_aux_BD_S "\004" #define DS_aux_BI_S "\005" #define DS_aux_FZ_S "\006" #define DS_aux_oq_S "\007" #define DS_aux_os_S "\010" #define DS_aux_oz_S "\011" #define CRAM_M_REVERSE 1 #define CRAM_M_UNMAP 2 /* CF bitfields */ #define CRAM_FLAG_PRESERVE_QUAL_SCORES (1<<0) #define CRAM_FLAG_DETACHED (1<<1) #define CRAM_FLAG_MATE_DOWNSTREAM (1<<2) #define CRAM_FLAG_NO_SEQ (1<<3) #define CRAM_FLAG_MASK ((1<<4)-1) /* Internal only */ #define CRAM_FLAG_STATS_ADDED (1<<30) #ifdef __cplusplus } #endif #endif /* _CRAM_STRUCTS_H_ */ stacks-1.44/htslib/cram/._files.c000644 000765 000024 00000000265 12750743246 017411 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/files.c000644 000765 000024 00000004651 12750743246 017177 0ustar00catchenstaff000000 000000 /* Copyright (c) 1994, 1996-1997, 2000, 2003 MEDICAL RESEARCH COUNCIL All rights reserved Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1 Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2 Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3 Neither the name of the MEDICAL RESEARCH COUNCIL, THE LABORATORY OF MOLECULAR BIOLOGY nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #include #include "cram/misc.h" #include #include /* Alliant's Concentrix is hugely deficient */ /* Define things we require in this program */ /* Methinks S_IFMT and S_IFDIR aren't defined in POSIX */ #ifndef S_ISDIR #define S_ISDIR(m) (((m)&S_IFMT) == S_IFDIR) #endif /*!S_ISDIR*/ #ifndef S_ISREG #define S_ISREG(m) (((m)&S_IFMT) == S_IFREG) #endif /*!S_ISREG*/ int is_directory(char * fn) { struct stat buf; if ( stat(fn,&buf) ) return 0; return S_ISDIR(buf.st_mode); } int is_file(char * fn) { struct stat buf; if ( stat(fn,&buf) ) return 0; return S_ISREG(buf.st_mode); } int file_exists(char * fn) { struct stat buf; return ( stat(fn,&buf) == 0); } int file_size(char * fn) { struct stat buf; if ( stat(fn,&buf) != 0) return 0; return buf.st_size; } stacks-1.44/htslib/cram/._mFILE.c000644 000765 000024 00000000265 12750743246 017203 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/mFILE.c000644 000765 000024 00000040607 12750743246 016772 0ustar00catchenstaff000000 000000 /* Copyright (c) 2005-2006, 2008-2009, 2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #include #include #include #include #include #include #include #include #include #include #include "cram/os.h" #include "cram/mFILE.h" #include "cram/vlen.h" #ifdef HAVE_MMAP #include #endif /* * This file contains memory-based versions of the most commonly used * (by io_lib) stdio functions. * * Actual file IO takes place either on opening or closing an mFILE. * * Coupled to this are a bunch of rather scary macros which can be obtained * by including stdio_hack.h. It is recommended though that you use mFILE.h * instead and replace fopen with mfopen (etc). This is more or less * mandatory if you wish to use both FILE and mFILE structs in a single file. */ static mFILE *m_channel[3]; /* stdin, stdout and stderr fakes */ /* * Reads the entirety of fp into memory. If 'fn' exists it is the filename * associated with fp. This will be used for more optimal reading (via a * stat to identify the size and a single read). Otherwise we use successive * reads until EOF. * * Returns a malloced buffer on success of length *size * NULL on failure */ static char *mfload(FILE *fp, const char *fn, size_t *size, int binary) { struct stat sb; char *data = NULL; size_t allocated = 0, used = 0; int bufsize = 8192; #ifdef _WIN32 if (binary) _setmode(_fileno(fp), _O_BINARY); else _setmode(_fileno(fp), _O_TEXT); #endif if (fn && -1 != stat(fn, &sb)) { data = malloc(allocated = sb.st_size); bufsize = sb.st_size; } else { fn = NULL; } do { size_t len; if (used + bufsize > allocated) { allocated += bufsize; data = realloc(data, allocated); } len = fread(data + used, 1, allocated - used, fp); if (len > 0) used += len; } while (!feof(fp) && (fn == NULL || used < sb.st_size)); *size = used; return data; } #ifdef HAVE_MMAP /* * mmaps in the file, but only for reading currently. * * Returns 0 on success * -1 on failure */ int mfmmap(mFILE *mf, FILE *fp, const char *fn) { struct stat sb; if (stat(fn, &sb) != 0) return -1; mf->size = sb.st_size; mf->data = mmap(NULL, mf->size, PROT_READ, MAP_SHARED, fileno(fp), 0); if (!mf->data) return -1; mf->alloced = 0; return 0; } #endif /* * Creates and returns m_channel[0]. * We initialise this on the first attempted read, which then slurps in * all of stdin until EOF is met. */ mFILE *mstdin(void) { if (m_channel[0]) return m_channel[0]; m_channel[0] = mfcreate(NULL, 0); if (NULL == m_channel[0]) return NULL; m_channel[0]->fp = stdin; return m_channel[0]; } static void init_mstdin(void) { static int done_stdin = 0; if (done_stdin) return; m_channel[0]->data = mfload(stdin, NULL, &m_channel[0]->size, 1); m_channel[0]->mode = MF_READ; done_stdin = 1; } /* * Creates and returns m_channel[1]. This is the fake for stdout. It starts as * an empty buffer which is physically written out only when mfflush or * mfclose are called. */ mFILE *mstdout(void) { if (m_channel[1]) return m_channel[1]; m_channel[1] = mfcreate(NULL, 0); if (NULL == m_channel[1]) return NULL; m_channel[1]->fp = stdout; m_channel[1]->mode = MF_WRITE; return m_channel[1]; } /* * Stderr as an mFILE. * The code handles stderr by returning m_channel[2], but also checking * for stderr in fprintf (the common usage of it) to auto-flush. */ mFILE *mstderr(void) { if (m_channel[2]) return m_channel[2]; m_channel[2] = mfcreate(NULL, 0); if (NULL == m_channel[2]) return NULL; m_channel[2]->fp = stderr; m_channel[2]->mode = MF_WRITE; return m_channel[2]; } /* * For creating existing mFILE pointers directly from memory buffers. */ mFILE *mfcreate(char *data, int size) { mFILE *mf = (mFILE *)malloc(sizeof(*mf)); if (NULL == mf) return NULL; mf->fp = NULL; mf->data = data; mf->alloced = size; mf->size = size; mf->eof = 0; mf->offset = 0; mf->flush_pos = 0; mf->mode = MF_READ | MF_WRITE; return mf; } /* * Recreate an existing mFILE to house new data/size. * It also rewinds the file. */ void mfrecreate(mFILE *mf, char *data, int size) { if (mf->data) free(mf->data); mf->data = data; mf->size = size; mf->alloced = size; mf->eof = 0; mf->offset = 0; mf->flush_pos = 0; } /* * Creates a new mFILE to contain the contents of the FILE pointer. * This mFILE is purely for in-memory operations and has no links to the * original FILE* it came from. It also doesn't close the FILE pointer. * Consider using mfreopen() is you need different behaviour. * * Returns mFILE * on success * NULL on failure. */ mFILE *mfcreate_from(const char *path, const char *mode_str, FILE *fp) { mFILE *mf; /* Open using mfreopen() */ if (NULL == (mf = mfreopen(path, mode_str, fp))) return NULL; /* Disassociate from the input stream */ mf->fp = NULL; return mf; } /* * Converts a FILE * to an mFILE *. * Use this for wrapper functions to turn external prototypes requring * FILE * as an argument into internal code using mFILE *. */ mFILE *mfreopen(const char *path, const char *mode_str, FILE *fp) { mFILE *mf; int r = 0, w = 0, a = 0, b = 0, x = 0, mode = 0; /* Parse mode: * r = read file contents (if truncated => don't read) * w = write on close * a = position at end of buffer * x = position at same location as the original fp, don't seek on flush * + = for update (read and write) * m = mmap (read only) */ if (strchr(mode_str, 'r')) r = 1, mode |= MF_READ; if (strchr(mode_str, 'w')) w = 1, mode |= MF_WRITE | MF_TRUNC; if (strchr(mode_str, 'a')) w = a = 1, mode |= MF_WRITE | MF_APPEND; if (strchr(mode_str, 'b')) b = 1, mode |= MF_BINARY; if (strchr(mode_str, 'x')) x = 1; if (strchr(mode_str, '+')) { w = 1, mode |= MF_READ | MF_WRITE; if (a) r = 1; } #ifdef HAVE_MMAP if (strchr(mode_str, 'm')) if (!w) mode |= MF_MMAP; #endif if (r) { mf = mfcreate(NULL, 0); if (NULL == mf) return NULL; if (!(mode & MF_TRUNC)) { #ifdef HAVE_MMAP if (mode & MF_MMAP) { if (mfmmap(mf, fp, path) == -1) { mf->data = NULL; mode &= ~MF_MMAP; } } #endif if (!mf->data) { mf->data = mfload(fp, path, &mf->size, b); mf->alloced = mf->size; if (!a) fseek(fp, 0, SEEK_SET); } } } else if (w) { /* Write - initialise the data structures */ mf = mfcreate(NULL, 0); if (NULL == mf) return NULL; } else { fprintf(stderr, "Must specify either r, w or a for mode\n"); return NULL; } mf->fp = fp; mf->mode = mode; if (x) { mf->mode |= MF_MODEX; } if (a) { mf->flush_pos = mf->size; fseek(fp, 0, SEEK_END); } return mf; } /* * Opens a file. If we have read access (r or a+) then it loads the entire * file into memory. If We have write access then the pathname is stored. * We do not actually write until an mfclose, which then checks this pathname. */ mFILE *mfopen(const char *path, const char *mode) { FILE *fp; if (NULL == (fp = fopen(path, mode))) return NULL; return mfreopen(path, mode, fp); } /* * Closes an mFILE. If the filename is known (implying write access) then this * also writes the data to disk. * * Stdout is handled by calling mfflush which writes to stdout if appropriate. */ int mfclose(mFILE *mf) { if (!mf) return -1; mfflush(mf); #ifdef HAVE_MMAP if ((mf->mode & MF_MMAP) && mf->data) { /* Mmaped */ munmap(mf->data, mf->size); mf->data = NULL; } #endif if (mf->fp) fclose(mf->fp); mfdestroy(mf); return 0; } /* * Closes the file pointer contained within the mFILE without destroying * the in-memory data. * * Attempting to do this on an mmaped buffer is an error. */ int mfdetach(mFILE *mf) { if (!mf) return -1; mfflush(mf); if (mf->mode & MF_MMAP) return -1; if (mf->fp) { fclose(mf->fp); mf->fp = NULL; } return 0; } /* * Destroys an mFILE structure but does not flush or close it */ int mfdestroy(mFILE *mf) { if (!mf) return -1; if (mf->data) free(mf->data); free(mf); return 0; } /* * Steals that data out of an mFILE. The mFILE itself will be closed. * It is up to the caller to free the stolen buffer. If size_out is * not NULL, mf->size will be stored in it. * This is more-or-less the opposite of mfcreate(). * * Note, we cannot steal the allocated buffer from an mmaped mFILE. */ void *mfsteal(mFILE *mf, size_t *size_out) { void *data; if (!mf) return NULL; data = mf->data; if (NULL != size_out) *size_out = mf->size; if (mfdetach(mf) != 0) return NULL; mf->data = NULL; mfdestroy(mf); return data; } /* * Seek/tell functions. Nothing more than updating and reporting an * in-memory index. NB we can seek on stdin or stdout even provided we * haven't been flushing. */ int mfseek(mFILE *mf, long offset, int whence) { switch (whence) { case SEEK_SET: mf->offset = offset; break; case SEEK_CUR: mf->offset += offset; break; case SEEK_END: mf->offset = mf->size + offset; break; default: errno = EINVAL; return -1; } mf->eof = 0; return 0; } long mftell(mFILE *mf) { return mf->offset; } void mrewind(mFILE *mf) { mf->offset = 0; mf->eof = 0; } /* * mftruncate is not directly a translation of ftruncate as the latter * takes a file descriptor instead of a FILE *. It performs the analogous * role though. * * If offset is -1 then the file is truncated to be the current file * offset. */ void mftruncate(mFILE *mf, long offset) { mf->size = offset != -1 ? offset : mf->offset; if (mf->offset > mf->size) mf->offset = mf->size; } int mfeof(mFILE *mf) { return mf->eof; } /* * mFILE read/write functions. Basically these turn fread/fwrite syntax * into memcpy statements, with appropriate memory handling for writing. */ size_t mfread(void *ptr, size_t size, size_t nmemb, mFILE *mf) { size_t len; char *cptr = (char *)ptr; if (mf == m_channel[0]) init_mstdin(); if (mf->size <= mf->offset) return 0; len = size * nmemb <= mf->size - mf->offset ? size * nmemb : mf->size - mf->offset; if (!size) return 0; memcpy(cptr, &mf->data[mf->offset], len); mf->offset += len; if (len != size * nmemb) { mf->eof = 1; } return len / size; } size_t mfwrite(void *ptr, size_t size, size_t nmemb, mFILE *mf) { if (!(mf->mode & MF_WRITE)) return 0; /* Append mode => forced all writes to end of file */ if (mf->mode & MF_APPEND) mf->offset = mf->size; /* Make sure we have enough room */ while (size * nmemb + mf->offset > mf->alloced) { size_t new_alloced = mf->alloced ? mf->alloced * 2 : 1024; void * new_data = realloc(mf->data, new_alloced); if (NULL == new_data) return 0; mf->alloced = new_alloced; mf->data = new_data; } /* Record where we need to reflush from */ if (mf->offset < mf->flush_pos) mf->flush_pos = mf->offset; /* Copy the data over */ memcpy(&mf->data[mf->offset], ptr, size * nmemb); mf->offset += size * nmemb; if (mf->size < mf->offset) mf->size = mf->offset; return nmemb; } int mfgetc(mFILE *mf) { if (mf == m_channel[0]) init_mstdin(); if (mf->offset < mf->size) { return (unsigned char)mf->data[mf->offset++]; } mf->eof = 1; return -1; } int mungetc(int c, mFILE *mf) { if (mf->offset > 0) { mf->data[--mf->offset] = c; return c; } mf->eof = 1; return -1; } char *mfgets(char *s, int size, mFILE *mf) { int i; if (mf == m_channel[0]) init_mstdin(); *s = 0; for (i = 0; i < size-1;) { if (mf->offset < mf->size) { s[i] = mf->data[mf->offset++]; if (s[i++] == '\n') break; } else { mf->eof = 1; break; } } s[i] = 0; return i ? s : NULL; } /* * Flushes an mFILE. If this is a real open of a file in write mode then * mFILE->fp will be set. We then write out any new data in mFILE since the * last flush. We cannot tell what may have been modified as we don't keep * track of that, so we typically rewrite out the entire file contents between * the last flush_pos and the end of file. * * For stderr/stdout we also reset the offsets so we cannot modify things * we've already output. */ int mfflush(mFILE *mf) { if (!mf->fp) return 0; /* FIXME: only do this when opened in write mode */ if (mf == m_channel[1] || mf == m_channel[2]) { if (mf->flush_pos < mf->size) { size_t bytes = mf->size - mf->flush_pos; if (fwrite(mf->data + mf->flush_pos, 1, bytes, mf->fp) < bytes) return -1; if (0 != fflush(mf->fp)) return -1; } /* Stdout & stderr are non-seekable streams so throw away the data */ mf->offset = mf->size = mf->flush_pos = 0; } /* only flush when opened in write mode */ if (mf->mode & MF_WRITE) { if (mf->flush_pos < mf->size) { size_t bytes = mf->size - mf->flush_pos; if (!(mf->mode & MF_MODEX)) { fseek(mf->fp, mf->flush_pos, SEEK_SET); } if (fwrite(mf->data + mf->flush_pos, 1, bytes, mf->fp) < bytes) return -1; if (0 != fflush(mf->fp)) return -1; } if (ftell(mf->fp) != -1 && ftruncate(fileno(mf->fp), ftell(mf->fp)) == -1) return -1; mf->flush_pos = mf->size; } return 0; } /* * A wrapper around vsprintf() to write to an mFILE. This also uses vflen() to * estimate how many additional bytes of storage will be required for the * vsprintf to work. */ int mfprintf(mFILE *mf, char *fmt, ...) { int ret; size_t est_length; va_list args; va_start(args, fmt); est_length = vflen(fmt, args); va_end(args); while (est_length + mf->offset > mf->alloced) { size_t new_alloced = mf->alloced ? mf->alloced * 2 : 1024; void * new_data = realloc(mf->data, new_alloced); if (NULL == new_data) return -1; mf->alloced = new_alloced; mf->data = new_data; } va_start(args, fmt); ret = vsprintf(&mf->data[mf->offset], fmt, args); va_end(args); if (ret > 0) { mf->offset += ret; if (mf->size < mf->offset) mf->size = mf->offset; } if (mf->fp == stderr) { /* Auto-flush for stderr */ if (0 != mfflush(mf)) return -1; } return ret; } /* * Converts an mFILE from binary to ascii mode by replacing all * cr-nl with nl. * * Primarily used on windows when we've uncompressed a binary file which * happens to be a text file (eg Experiment File). Previously we would have * seeked back to the start and used _setmode(fileno(fp), _O_TEXT). * * Side effect: resets offset and flush_pos back to the start. */ void mfascii(mFILE *mf) { size_t p1, p2; for (p1 = p2 = 1; p1 < mf->size; p1++, p2++) { if (mf->data[p1] == '\n' && mf->data[p1-1] == '\r') { p2--; /* delete the \r */ } mf->data[p2] = mf->data[p1]; } mf->size = p2; mf->offset = mf->flush_pos = 0; } stacks-1.44/htslib/cram/._mFILE.h000644 000765 000024 00000000265 12750743246 017210 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/mFILE.h000644 000765 000024 00000005714 12750743246 016777 0ustar00catchenstaff000000 000000 /* Copyright (c) 2005-2006, 2008-2009 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #ifndef _MFILE_H_ #define _MFILE_H_ #include #ifdef __cplusplus extern "C" { #endif typedef struct { FILE *fp; char *data; size_t alloced; int eof; int mode; /* open mode in MF_?? define bit pattern */ size_t size; size_t offset; size_t flush_pos; } mFILE; #define MF_READ 1 #define MF_WRITE 2 #define MF_APPEND 4 #define MF_BINARY 8 #define MF_TRUNC 16 #define MF_MODEX 32 #define MF_MMAP 64 mFILE *mfreopen(const char *path, const char *mode, FILE *fp); mFILE *mfopen(const char *path, const char *mode); int mfdetach(mFILE *mf); int mfclose(mFILE *mf); int mfdestroy(mFILE *mf); int mfseek(mFILE *mf, long offset, int whence); long mftell(mFILE *mf); void mrewind(mFILE *mf); void mftruncate(mFILE *mf, long offset); int mfeof(mFILE *mf); size_t mfread(void *ptr, size_t size, size_t nmemb, mFILE *mf); size_t mfwrite(void *ptr, size_t size, size_t nmemb, mFILE *mf); int mfgetc(mFILE *mf); int mungetc(int c, mFILE *mf); mFILE *mfcreate(char *data, int size); mFILE *mfcreate_from(const char *path, const char *mode_str, FILE *fp); void mfrecreate(mFILE *mf, char *data, int size); void *mfsteal(mFILE *mf, size_t *size_out); char *mfgets(char *s, int size, mFILE *mf); int mfflush(mFILE *mf); int mfprintf(mFILE *mf, char *fmt, ...); mFILE *mstdin(void); mFILE *mstdout(void); mFILE *mstderr(void); void mfascii(mFILE *mf); #ifdef __cplusplus } #endif #endif /* _MFILE_H_ */ stacks-1.44/htslib/cram/._misc.h000644 000765 000024 00000000265 12750743246 017247 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/misc.h000644 000765 000024 00000010322 12750743246 017025 0ustar00catchenstaff000000 000000 /* Copyright (c) 1994-1997, 2001-2002 MEDICAL RESEARCH COUNCIL All rights reserved Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1 Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2 Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3 Neither the name of the MEDICAL RESEARCH COUNCIL, THE LABORATORY OF MOLECULAR BIOLOGY nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /* Copyright (c) 2003-2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #ifndef _misc_h #define _misc_h #include "cram/os.h" #include #include /* varargs needed for v*printf() prototypes */ #include #ifdef __cplusplus extern "C" { #endif /* * This informs gcc that crash() doesn't return, so it doesn't need to * concern itself that code paths going via crash could mean some variables * being undefined and then issuing uninitialised variable warnings. * This particularly affected convert. */ #ifdef __GNUC__ # define __NORETURN__ __attribute__ ((__noreturn__)) #else # define __NORETURN__ #endif /* * Used for printf style argument checking. We can request a function such * as vTcl_SetResult does argument checking, avoiding bugs with using * %d and passing in a 64-bit record. */ #ifdef __GNUC__ # define __PRINTF_FORMAT__(a,b) __attribute__ ((format (printf, a, b))) #else # define __PRINTF_FORMAT__(a,b) #endif extern int is_directory(char * fn); extern int is_file(char * fn); extern int file_size(char * fn); #define MIN(A,B) ( ( (A) < (B) ) ? (A) : (B) ) #define MAX(A,B) ( ( (A) > (B) ) ? (A) : (B) ) #ifdef __cplusplus } #endif #endif /*_misc_h*/ stacks-1.44/htslib/cram/._open_trace_file.c000644 000765 000024 00000000265 12750743246 021425 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/open_trace_file.c000644 000765 000024 00000026666 12750743246 021225 0ustar00catchenstaff000000 000000 /* Author: James Bonfield Copyright (c) 2000-2001 MEDICAL RESEARCH COUNCIL All rights reserved Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the name of the MEDICAL RESEARCH COUNCIL, THE LABORATORY OF MOLECULAR BIOLOGY nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /* Copyright (c) 2008, 2009, 2013, 2014 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #include #include #include #include #include #include #include #include #include #include "cram/os.h" #ifndef PATH_MAX # define PATH_MAX 1024 #endif #include "cram/open_trace_file.h" #include "cram/misc.h" #include "htslib/hfile.h" /* * Tokenises the search path splitting on colons (unix) or semicolons * (windows). * We also explicitly add a "./" to the end of the search path * * Returns: A new search path with items separated by nul chars. Two nul * chars in a row represent the end of the tokenised path. * Returns NULL for a failure. * * The returned data has been malloced. It is up to the caller to free this * memory. */ char *tokenise_search_path(char *searchpath) { char *newsearch; unsigned int i, j; size_t len; #ifdef _WIN32 char path_sep = ';'; #else char path_sep = ':'; #endif if (!searchpath) searchpath=""; newsearch = (char *)malloc((len = strlen(searchpath))+5); if (!newsearch) return NULL; for (i = 0, j = 0; i < len; i++) { /* "::" => ":". Used for escaping colons in http://foo */ if (i < len-1 && searchpath[i] == ':' && searchpath[i+1] == ':') { newsearch[j++] = ':'; i++; continue; } /* Handle http:// and ftp:// too without :: */ if (path_sep == ':') { if ((i == 0 || (i > 0 && searchpath[i-1] == ':')) && (!strncmp(&searchpath[i], "http:", 5) || !strncmp(&searchpath[i], "ftp:", 4) || !strncmp(&searchpath[i], "|http:", 6) || !strncmp(&searchpath[i], "|ftp:", 5) || !strncmp(&searchpath[i], "URL=http:", 9) || !strncmp(&searchpath[i], "URL=ftp:", 8))) { do { newsearch[j++] = searchpath[i]; } while (i 0) { if (mfwrite(buf, len, 1, mf) <= 0) { hclose_abruptly(hf); mfdestroy(mf); return NULL; } } if (hclose(hf) < 0 || len < 0) { mfdestroy(mf); return NULL; } mrewind(mf); return mf; } /* * Takes a dirname possibly including % rules and appends the filename * to it. * * Returns expanded pathname or NULL for malloc failure. */ static char *expand_path(char *file, char *dirname) { size_t len = strlen(dirname); size_t lenf = strlen(file); char *cp, *path; path = malloc(len+lenf+2); // worst expansion DIR/FILE if (!path) return NULL; if (dirname[len-1] == '/') len--; /* Special case for "./" or absolute filenames */ if (*file == '/' || (len==1 && *dirname == '.')) { sprintf(path, "%s", file); } else { /* Handle %[0-9]*s expansions, if required */ char *path_end = path; *path = 0; while ((cp = strchr(dirname, '%'))) { char *endp; long l = strtol(cp+1, &endp, 10); if (*endp != 's') { strncpy(path_end, dirname, (endp+1)-dirname); path_end += (endp+1)-dirname; dirname = endp+1; continue; } strncpy(path_end, dirname, cp-dirname); path_end += cp-dirname; if (l) { strncpy(path_end, file, l); path_end += MIN(strlen(file), l); file += MIN(strlen(file), l); } else { strcpy(path_end, file); path_end += strlen(file); file += strlen(file); } len -= (endp+1) - dirname; dirname = endp+1; } strncpy(path_end, dirname, len); path_end += MIN(strlen(dirname), len); *path_end = 0; if (*file) { *path_end++ = '/'; strcpy(path_end, file); } } //fprintf(stderr, "*PATH=\"%s\"\n", path); return path; } /* * Searches for file in the directory 'dirname'. If it finds it, it opens * it. This also searches for compressed versions of the file in dirname * too. * * Returns mFILE pointer if found * NULL if not */ static mFILE *find_file_dir(char *file, char *dirname) { char *path; mFILE *mf = NULL; path = expand_path(file, dirname); if (is_file(path)) mf = mfopen(path, "rbm"); free(path); return mf; } /* * ------------------------------------------------------------------------ * Public functions below. */ /* * Opens a trace file named 'file'. This is initially looked for as a * pathname relative to a file named "relative_to". This may (for * example) be the name of an experiment file referencing the trace * file. In this case by passing relative_to as the experiment file * filename the trace file will be picked up in the same directory as * the experiment file. Relative_to may be supplied as NULL. * * 'file' is looked for at relative_to, then the current directory, and then * all of the locations listed in 'path' (which is a colon separated list). * If 'path' is NULL it uses the RAWDATA environment variable instead. * * Returns a mFILE pointer when found. * NULL otherwise. */ mFILE *open_path_mfile(char *file, char *path, char *relative_to) { char *newsearch; char *ele; mFILE *fp; /* Use path first */ if (!path) path = getenv("RAWDATA"); if (NULL == (newsearch = tokenise_search_path(path))) return NULL; /* * Step through the search path testing out each component. * We now look through each path element treating some prefixes as * special, otherwise we treat the element as a directory. */ for (ele = newsearch; *ele; ele += strlen(ele)+1) { char *ele2; /* * '|' prefixing a path component indicates that we do not * wish to perform the compression extension searching in that * location. * * NB: this has been removed from the htslib implementation. */ if (*ele == '|') { ele2 = ele+1; } else { ele2 = ele; } if (0 == strncmp(ele2, "URL=", 4)) { if ((fp = find_file_url(file, ele2+4))) { free(newsearch); return fp; } } else if (!strncmp(ele2, "http:", 5) || !strncmp(ele2, "ftp:", 4)) { if ((fp = find_file_url(file, ele2))) { free(newsearch); return fp; } } else if ((fp = find_file_dir(file, ele2))) { free(newsearch); return fp; } } free(newsearch); /* Look in the same location as the incoming 'relative_to' filename */ if (relative_to) { char *cp; char relative_path[PATH_MAX+1]; strcpy(relative_path, relative_to); if ((cp = strrchr(relative_path, '/'))) *cp = 0; if ((fp = find_file_dir(file, relative_path))) return fp; } return NULL; } /* * As per open_path_mfile, but searching only for local filenames. * This is useful as we may avoid doing a full mfopen and loading * the entire file into memory. * * Returns the expanded pathname if found. * NULL if not */ char *find_path(char *file, char *path) { char *newsearch; char *ele; char *outpath = NULL; /* Use path first */ if (!path) path = getenv("RAWDATA"); if (NULL == (newsearch = tokenise_search_path(path))) return NULL; for (ele = newsearch; *ele; ele += strlen(ele)+1) { char *ele2 = (*ele == '|') ? ele+1 : ele; if (!strncmp(ele2, "URL=", 4) || !strncmp(ele2, "http:", 5) || !strncmp(ele2, "ftp:", 4)) { continue; } else { outpath = expand_path(file, ele2); if (is_file(outpath)) { free(newsearch); return outpath; } else { free(outpath); } } } free(newsearch); return NULL; } stacks-1.44/htslib/cram/._open_trace_file.h000644 000765 000024 00000000265 12750743246 021432 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/open_trace_file.h000644 000765 000024 00000011641 12750743246 021215 0ustar00catchenstaff000000 000000 /* Author: James Bonfield Copyright (c) 2000-2001 MEDICAL RESEARCH COUNCIL All rights reserved Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: . Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. . Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. . Neither the name of the MEDICAL RESEARCH COUNCIL, THE LABORATORY OF MOLECULAR BIOLOGY nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /* Copyright (c) 2008, 2009, 2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #ifndef _OPEN_TRACE_FILE_H_ #define _OPEN_TRACE_FILE_H_ #include "cram/mFILE.h" #ifdef __cplusplus extern "C" { #endif /* * Tokenises the search path splitting on colons (unix) or semicolons * (windows). * We also explicitly add a "./" to the end of the search path * * Returns: A new search path with items separated by nul chars. Two nul * chars in a row represent the end of the tokenised path. * Returns NULL for a failure. * * The returned data has been malloced. It is up to the caller to free this * memory. */ char *tokenise_search_path(char *searchpath); /* * Opens a trace file named 'file'. This is initially looked for as a * pathname relative to a file named "relative_to". This may (for * example) be the name of an experiment file referencing the trace * file. In this case by passing relative_to as the experiment file * filename the trace file will be picked up in the same directory as * the experiment file. Relative_to may be supplied as NULL. * * 'file' is looked for at relative_to, then the current directory, and then * all of the locations listed in 'path' (which is a colon separated list). * If 'path' is NULL it uses the RAWDATA environment variable instead. * * Returns a mFILE pointer when found. * NULL otherwise. */ mFILE *open_path_mfile(char *file, char *path, char *relative_to); /* * Returns a mFILE containing the entire contents of the url; * NULL on failure. */ mFILE *find_file_url(char *file, char *url); /* * As per open_path_mfile, but searching only for local filenames. * This is useful as we may avoid doing a full mfopen and loading * the entire file into memory. * * Returns the expanded pathname if found. * NULL if not */ char *find_path(char *file, char *path); #ifdef __cplusplus } #endif #endif /* _OPEN_TRACE_FILE_H_ */ stacks-1.44/htslib/cram/._os.h000644 000765 000024 00000000265 12750743246 016735 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/os.h000644 000765 000024 00000022254 12750743246 016522 0ustar00catchenstaff000000 000000 /* Copyright (c) 1993, 1995-2002 MEDICAL RESEARCH COUNCIL All rights reserved Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1 Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2 Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3 Neither the name of the MEDICAL RESEARCH COUNCIL, THE LABORATORY OF MOLECULAR BIOLOGY nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /* Copyright (c) 2004, 2006, 2009-2011, 2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /* * File: os.h * * Author: * MRC Laboratory of Molecular Biology * Hills Road * Cambridge CB2 2QH * United Kingdom * * Description: operating system specific type definitions * */ #ifndef _OS_H_ #define _OS_H_ #include #include #ifdef __cplusplus extern "C" { #endif /*----------------------------------------------------------------------------- * Detection of endianness. The main part of this is done in autoconf, but * for the case of MacOS FAT binaries we fall back on auto-sensing based on * processor type too. */ /* Set by autoconf */ #define SP_LITTLE_ENDIAN /* Mac FAT binaries or unknown. Auto detect based on CPU type */ #if !defined(SP_BIG_ENDIAN) && !defined(SP_LITTLE_ENDIAN) /* * x86 equivalents */ #if defined(__i386__) || defined(__i386) || defined(__amd64__) || defined(__amd64) || defined(__x86_64__) || defined(__x86_64) || defined(__i686__) || defined(__i686) # if defined(SP_BIG_ENDIAN) # undef SP_BIG_ENDIAN # endif # define SP_LITTLE_ENDIAN #endif /* * DEC Alpha */ #if defined(__alpha__) || defined(__alpha) # if defined(SP_LITTLE_ENDIAN) # undef SP_LITTLE_ENDIAN # endif # define SP_BIG_ENDIAN #endif /* * SUN Sparc */ #if defined(__sparc__) || defined(__sparc) # if defined(SP_LITTLE_ENDIAN) # undef SP_LITTLE_ENDIAN # endif # define SP_BIG_ENDIAN #endif /* * PowerPC */ #if defined(__ppc__) || defined(__ppc) # if defined(SP_LITTLE_ENDIAN) # undef SP_LITTLE_ENDIAN # endif # define SP_BIG_ENDIAN #endif /* Some catch-alls */ #if defined(__LITTLE_ENDIAN__) || defined(__LITTLEENDIAN__) # define SP_LITTLE_ENDIAN #endif #if defined(__BIG_ENDIAN__) || defined(__BIGENDIAN__) # define SP_BIG_ENDIAN #endif #if defined(SP_BIG_ENDIAN) && defined(SP_LITTLE_ENDIAN) # error Both BIG and LITTLE endian defined. Fix os.h and/or Makefile #endif #if !defined(SP_BIG_ENDIAN) && !defined(SP_LITTLE_ENDIAN) # error Neither BIG nor LITTLE endian defined. Fix os.h and/or Makefile #endif #endif /*----------------------------------------------------------------------------- * Allow for unaligned memory access. This is used in BAM code as the packed * structure has 4-byte cigar ints after the variable length name. * * Consider using AX_CHECK_ALIGNED_ACCESS_REQUIRED in autoconf. */ #if defined(__i386__) || defined(__i386) || defined(__amd64__) || defined(__amd64) || defined(__x86_64__) || defined(__x86_64) || defined(__i686__) || defined(__i686) # define ALLOW_UAC #endif /*----------------------------------------------------------------------------- * Byte swapping macros */ /* * Our new swap runs at the same speed on Ultrix, but substantially faster * (300% for swap_int4, ~50% for swap_int2) on an Alpha (due to the lack of * decent 'char' support). * * They also have the ability to swap in situ (src == dst). Newer code now * relies on this so don't change back! */ #define iswap_int8(x) \ (((x & 0x00000000000000ffLL) << 56) + \ ((x & 0x000000000000ff00LL) << 40) + \ ((x & 0x0000000000ff0000LL) << 24) + \ ((x & 0x00000000ff000000LL) << 8) + \ ((x & 0x000000ff00000000LL) >> 8) + \ ((x & 0x0000ff0000000000LL) >> 24) + \ ((x & 0x00ff000000000000LL) >> 40) + \ ((x & 0xff00000000000000LL) >> 56)) #define iswap_int4(x) \ (((x & 0x000000ff) << 24) + \ ((x & 0x0000ff00) << 8) + \ ((x & 0x00ff0000) >> 8) + \ ((x & 0xff000000) >> 24)) #define iswap_int2(x) \ (((x & 0x00ff) << 8) + \ ((x & 0xff00) >> 8)) /* * Linux systems may use byteswap.h to get assembly versions of byte-swap * on intel systems. This can be as trivial as the bswap opcode, which works * out at over 2-times faster than iswap_int4 above. */ #if 0 #if defined(__linux__) # include # undef iswap_int8 # undef iswap_int4 # undef iswap_int2 # define iswap_int8 bswap_64 # define iswap_int4 bswap_32 # define iswap_int2 bswap_16 #endif #endif /* * Macros to specify that data read in is of a particular endianness. * The macros here swap to the appropriate order for the particular machine * running the macro and return the new answer. These may also be used when * writing to a file to specify that we wish to write in (eg) big endian * format. * * This leads to efficient code as most of the time these macros are * trivial. */ #ifdef SP_BIG_ENDIAN #define le_int4(x) iswap_int4((x)) #define le_int2(x) iswap_int2((x)) #endif #ifdef SP_LITTLE_ENDIAN #define le_int4(x) (x) #define le_int2(x) (x) #endif /*----------------------------------------------------------------------------- * definitions, incase they're not present */ #ifndef PRId64 #define __PRI64__ "l" #define PRId64 __PRI64__ "d" #define PRId32 "d" #define PRId16 "d" #define PRId8 "d" #define PRIu64 __PRI64__ "u" #define PRIu32 "u" #define PRIu16 "u" #define PRIu8 "u" #endif /*----------------------------------------------------------------------------- * Operating system specifics. * These ought to be done by autoconf, but are legacy code. */ /* * SunOS 4.x * Even though we use the ANSI gcc, we make use the the standard SunOS 4.x * libraries and include files, which are non-ansi */ #if defined(__sun__) && !defined(__svr4__) #define SEEK_SET 0 #define SEEK_CUR 1 #define SEEK_END 2 #endif /* * Microsoft Visual C++ * Windows */ #if defined(_MSC_VER) #define popen _popen #define pclose _pclose #define ftruncate(fd,len) _chsize(fd,len) #endif /* * Microsoft Windows running MinGW */ #if defined(__MINGW32__) /* #define mkdir(filename,mode) mkdir((filename)) */ #define sysconf(x) 512 #define ftruncate(fd,len) _chsize(fd,len) #endif /* Generic WIN32 API issues */ #ifdef _WIN32 # ifndef HAVE_FSEEKO # if __MSVCRT_VERSION__ >= 0x800 /* if you have MSVCR80 installed then you can use these definitions: */ # define off_t __int64 # define fseeko _fseeki64 # define ftello _ftelli64 # else /* otherwise we're stuck with 32-bit file support */ # define off_t long # define fseeko fseek # define ftello ftell # endif # endif /* !HAVE_FSEEKO */ #endif /* _WIN32 */ #ifdef __cplusplus } #endif #endif /*_OS_H_*/ stacks-1.44/htslib/cram/._pooled_alloc.c000644 000765 000024 00000000265 12750743246 020743 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/pooled_alloc.c000644 000765 000024 00000010751 12750743246 020527 0ustar00catchenstaff000000 000000 /* Copyright (c) 2009 Genome Research Ltd. Author: Rob Davies Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #include #include #include #include #include "cram/pooled_alloc.h" #include "cram/misc.h" //#define TEST_MAIN #define PSIZE 1024*1024 // credit to http://graphics.stanford.edu/~seander/bithacks.html static int next_power_2(unsigned int v) { v--; v |= v >> 1; v |= v >> 2; v |= v >> 4; v |= v >> 8; v |= v >> 16; v++; return v; } /* * Creates a pool. * Pool allocations are approx minimum of 1024*dsize or PSIZE. * (Assumes we're not trying to use pools for >= 2Gb or more) */ pool_alloc_t *pool_create(size_t dsize) { pool_alloc_t *p; if (NULL == (p = (pool_alloc_t *)malloc(sizeof(*p)))) return NULL; /* Minimum size is a pointer, for free list */ dsize = (dsize + sizeof(void *) - 1) & ~(sizeof(void *)-1); if (dsize < sizeof(void *)) dsize = sizeof(void *); p->dsize = dsize; p->psize = MIN(PSIZE, next_power_2(p->dsize*1024)); p->npools = 0; p->pools = NULL; p->free = NULL; return p; } static pool_t *new_pool(pool_alloc_t *p) { size_t n = p->psize / p->dsize; pool_t *pool; pool = realloc(p->pools, (p->npools + 1) * sizeof(*p->pools)); if (NULL == pool) return NULL; p->pools = pool; pool = &p->pools[p->npools]; pool->pool = malloc(n * p->dsize); if (NULL == pool->pool) return NULL; pool->used = 0; p->npools++; return pool; } void pool_destroy(pool_alloc_t *p) { size_t i; for (i = 0; i < p->npools; i++) { free(p->pools[i].pool); } free(p->pools); free(p); } void *pool_alloc(pool_alloc_t *p) { pool_t *pool; void *ret; /* Look on free list */ if (NULL != p->free) { ret = p->free; p->free = *((void **)p->free); return ret; } /* Look for space in the last pool */ if (p->npools) { pool = &p->pools[p->npools - 1]; if (pool->used + p->dsize < p->psize) { ret = ((char *) pool->pool) + pool->used; pool->used += p->dsize; return ret; } } /* Need a new pool */ pool = new_pool(p); if (NULL == pool) return NULL; pool->used = p->dsize; return pool->pool; } void pool_free(pool_alloc_t *p, void *ptr) { *(void **)ptr = p->free; p->free = ptr; } #ifdef TEST_MAIN typedef struct { int x, y, z; } xyz; #define NP 10000 int main(void) { int i; xyz *item; xyz **items; pool_alloc_t *p = pool_create(sizeof(xyz)); items = (xyz **)malloc(NP * sizeof(*items)); for (i = 0; i < NP; i++) { item = pool_alloc(p); item->x = i; item->y = i+1; item->z = i+2; items[i] = item; } for (i = 0; i < NP; i++) { item = items[i]; if (i % 3) pool_free(p, item); } for (i = 0; i < NP; i++) { item = pool_alloc(p); item->x = 1000000+i; item->y = 1000000+i+1; item->z = 1000000+i+2; } for (i = 0; i < NP; i++) { item = items[i]; printf("%d\t%d\t%d\t%d\n", i, item->x, item->y, item->z); pool_free(p, item); } return 0; } #endif stacks-1.44/htslib/cram/._pooled_alloc.h000644 000765 000024 00000000265 12750743246 020750 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/pooled_alloc.h000644 000765 000024 00000004163 12750743246 020534 0ustar00catchenstaff000000 000000 /* Copyright (c) 2009 Genome Research Ltd. Author: Rob Davies Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #ifndef _POOLED_ALLOC_H_ #define _POOLED_ALLOC_H_ #ifdef __cplusplus extern "C" { #endif /* * Implements a pooled block allocator where all items are the same size, * but we need many of them. */ typedef struct { void *pool; size_t used; } pool_t; typedef struct { size_t dsize; size_t psize; size_t npools; pool_t *pools; void *free; } pool_alloc_t; pool_alloc_t *pool_create(size_t dsize); void pool_destroy(pool_alloc_t *p); void *pool_alloc(pool_alloc_t *p); void pool_free(pool_alloc_t *p, void *ptr); #ifdef __cplusplus } #endif #endif /*_POOLED_ALLOC_H_*/ stacks-1.44/htslib/cram/._rANS_byte.h000644 000765 000024 00000000265 12750743246 020142 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/rANS_byte.h000644 000765 000024 00000027064 12750743246 017733 0ustar00catchenstaff000000 000000 /* rans_byte.h originally from https://github.com/rygorous/ryg_rans * * This is a public-domain implementation of several rANS variants. rANS is an * entropy coder from the ANS family, as described in Jarek Duda's paper * "Asymmetric numeral systems" (http://arxiv.org/abs/1311.2540). */ /*-------------------------------------------------------------------------- */ // Simple byte-aligned rANS encoder/decoder - public domain - Fabian 'ryg' Giesen 2014 // // Not intended to be "industrial strength"; just meant to illustrate the general // idea. #ifndef RANS_BYTE_HEADER #define RANS_BYTE_HEADER #include #ifdef assert #define RansAssert assert #else #define RansAssert(x) #endif // READ ME FIRST: // // This is designed like a typical arithmetic coder API, but there's three // twists you absolutely should be aware of before you start hacking: // // 1. You need to encode data in *reverse* - last symbol first. rANS works // like a stack: last in, first out. // 2. Likewise, the encoder outputs bytes *in reverse* - that is, you give // it a pointer to the *end* of your buffer (exclusive), and it will // slowly move towards the beginning as more bytes are emitted. // 3. Unlike basically any other entropy coder implementation you might // have used, you can interleave data from multiple independent rANS // encoders into the same bytestream without any extra signaling; // you can also just write some bytes by yourself in the middle if // you want to. This is in addition to the usual arithmetic encoder // property of being able to switch models on the fly. Writing raw // bytes can be useful when you have some data that you know is // incompressible, and is cheaper than going through the rANS encode // function. Using multiple rANS coders on the same byte stream wastes // a few bytes compared to using just one, but execution of two // independent encoders can happen in parallel on superscalar and // Out-of-Order CPUs, so this can be *much* faster in tight decoding // loops. // // This is why all the rANS functions take the write pointer as an // argument instead of just storing it in some context struct. // -------------------------------------------------------------------------- // L ('l' in the paper) is the lower bound of our normalization interval. // Between this and our byte-aligned emission, we use 31 (not 32!) bits. // This is done intentionally because exact reciprocals for 31-bit uints // fit in 32-bit uints: this permits some optimizations during encoding. #define RANS_BYTE_L (1u << 23) // lower bound of our normalization interval // State for a rANS encoder. Yep, that's all there is to it. typedef uint32_t RansState; // Initialize a rANS encoder. static inline void RansEncInit(RansState* r) { *r = RANS_BYTE_L; } // Renormalize the encoder. Internal function. static inline RansState RansEncRenorm(RansState x, uint8_t** pptr, uint32_t freq, uint32_t scale_bits) { uint32_t x_max = ((RANS_BYTE_L >> scale_bits) << 8) * freq; // this turns into a shift. if (x >= x_max) { uint8_t* ptr = *pptr; do { *--ptr = (uint8_t) (x & 0xff); x >>= 8; } while (x >= x_max); *pptr = ptr; } return x; } // Encodes a single symbol with range start "start" and frequency "freq". // All frequencies are assumed to sum to "1 << scale_bits", and the // resulting bytes get written to ptr (which is updated). // // NOTE: With rANS, you need to encode symbols in *reverse order*, i.e. from // beginning to end! Likewise, the output bytestream is written *backwards*: // ptr starts pointing at the end of the output buffer and keeps decrementing. static inline void RansEncPut(RansState* r, uint8_t** pptr, uint32_t start, uint32_t freq, uint32_t scale_bits) { // renormalize RansState x = RansEncRenorm(*r, pptr, freq, scale_bits); // x = C(s,x) *r = ((x / freq) << scale_bits) + (x % freq) + start; } // Flushes the rANS encoder. static inline void RansEncFlush(RansState* r, uint8_t** pptr) { uint32_t x = *r; uint8_t* ptr = *pptr; ptr -= 4; ptr[0] = (uint8_t) (x >> 0); ptr[1] = (uint8_t) (x >> 8); ptr[2] = (uint8_t) (x >> 16); ptr[3] = (uint8_t) (x >> 24); *pptr = ptr; } // Initializes a rANS decoder. // Unlike the encoder, the decoder works forwards as you'd expect. static inline void RansDecInit(RansState* r, uint8_t** pptr) { uint32_t x; uint8_t* ptr = *pptr; x = ptr[0] << 0; x |= ptr[1] << 8; x |= ptr[2] << 16; x |= ptr[3] << 24; ptr += 4; *pptr = ptr; *r = x; } // Returns the current cumulative frequency (map it to a symbol yourself!) static inline uint32_t RansDecGet(RansState* r, uint32_t scale_bits) { return *r & ((1u << scale_bits) - 1); } // Advances in the bit stream by "popping" a single symbol with range start // "start" and frequency "freq". All frequencies are assumed to sum to "1 << scale_bits", // and the resulting bytes get written to ptr (which is updated). static inline void RansDecAdvance(RansState* r, uint8_t** pptr, uint32_t start, uint32_t freq, uint32_t scale_bits) { uint32_t mask = (1u << scale_bits) - 1; // s, x = D(x) uint32_t x = *r; x = freq * (x >> scale_bits) + (x & mask) - start; // renormalize if (x < RANS_BYTE_L) { uint8_t* ptr = *pptr; do x = (x << 8) | *ptr++; while (x < RANS_BYTE_L); *pptr = ptr; } *r = x; } // -------------------------------------------------------------------------- // That's all you need for a full encoder; below here are some utility // functions with extra convenience or optimizations. // Encoder symbol description // This (admittedly odd) selection of parameters was chosen to make // RansEncPutSymbol as cheap as possible. typedef struct { uint32_t x_max; // (Exclusive) upper bound of pre-normalization interval uint32_t rcp_freq; // Fixed-point reciprocal frequency uint32_t bias; // Bias uint16_t cmpl_freq; // Complement of frequency: (1 << scale_bits) - freq uint16_t rcp_shift; // Reciprocal shift } RansEncSymbol; // Decoder symbols are straightforward. typedef struct { uint16_t start; // Start of range. uint16_t freq; // Symbol frequency. } RansDecSymbol; // Initializes an encoder symbol to start "start" and frequency "freq" static inline void RansEncSymbolInit(RansEncSymbol* s, uint32_t start, uint32_t freq, uint32_t scale_bits) { RansAssert(scale_bits <= 16); RansAssert(start <= (1u << scale_bits)); RansAssert(freq <= (1u << scale_bits) - start); // Say M := 1 << scale_bits. // // The original encoder does: // x_new = (x/freq)*M + start + (x%freq) // // The fast encoder does (schematically): // q = mul_hi(x, rcp_freq) >> rcp_shift (division) // r = x - q*freq (remainder) // x_new = q*M + bias + r (new x) // plugging in r into x_new yields: // x_new = bias + x + q*(M - freq) // =: bias + x + q*cmpl_freq (*) // // and we can just precompute cmpl_freq. Now we just need to // set up our parameters such that the original encoder and // the fast encoder agree. s->x_max = ((RANS_BYTE_L >> scale_bits) << 8) * freq; s->cmpl_freq = (uint16_t) ((1 << scale_bits) - freq); if (freq < 2) { // freq=0 symbols are never valid to encode, so it doesn't matter what // we set our values to. // // freq=1 is tricky, since the reciprocal of 1 is 1; unfortunately, // our fixed-point reciprocal approximation can only multiply by values // smaller than 1. // // So we use the "next best thing": rcp_freq=0xffffffff, rcp_shift=0. // This gives: // q = mul_hi(x, rcp_freq) >> rcp_shift // = mul_hi(x, (1<<32) - 1)) >> 0 // = floor(x - x/(2^32)) // = x - 1 if 1 <= x < 2^32 // and we know that x>0 (x=0 is never in a valid normalization interval). // // So we now need to choose the other parameters such that // x_new = x*M + start // plug it in: // x*M + start (desired result) // = bias + x + q*cmpl_freq (*) // = bias + x + (x - 1)*(M - 1) (plug in q=x-1, cmpl_freq) // = bias + 1 + (x - 1)*M // = x*M + (bias + 1 - M) // // so we have start = bias + 1 - M, or equivalently // bias = start + M - 1. s->rcp_freq = ~0u; s->rcp_shift = 0; s->bias = start + (1 << scale_bits) - 1; } else { // Alverson, "Integer Division using reciprocals" // shift=ceil(log2(freq)) uint32_t shift = 0; while (freq > (1u << shift)) shift++; s->rcp_freq = (uint32_t) (((1ull << (shift + 31)) + freq-1) / freq); s->rcp_shift = shift - 1; // With these values, 'q' is the correct quotient, so we // have bias=start. s->bias = start; } s->rcp_shift += 32; // Avoid the extra >>32 in RansEncPutSymbol } // Initialize a decoder symbol to start "start" and frequency "freq" static inline void RansDecSymbolInit(RansDecSymbol* s, uint32_t start, uint32_t freq) { RansAssert(start <= (1 << 16)); RansAssert(freq <= (1 << 16) - start); s->start = (uint16_t) start; s->freq = (uint16_t) freq; } // Encodes a given symbol. This is faster than straight RansEnc since we can do // multiplications instead of a divide. // // See RansEncSymbolInit for a description of how this works. static inline void RansEncPutSymbol(RansState* r, uint8_t** pptr, RansEncSymbol const* sym) { RansAssert(sym->x_max != 0); // can't encode symbol with freq=0 // renormalize uint32_t x = *r; uint32_t x_max = sym->x_max; if (x >= x_max) { uint8_t* ptr = *pptr; do { *--ptr = (uint8_t) (x & 0xff); x >>= 8; } while (x >= x_max); *pptr = ptr; } // x = C(s,x) // NOTE: written this way so we get a 32-bit "multiply high" when // available. If you're on a 64-bit platform with cheap multiplies // (e.g. x64), just bake the +32 into rcp_shift. //uint32_t q = (uint32_t) (((uint64_t)x * sym->rcp_freq) >> 32) >> sym->rcp_shift; // The extra >>32 has already been added to RansEncSymbolInit uint32_t q = (uint32_t) (((uint64_t)x * sym->rcp_freq) >> sym->rcp_shift); *r = x + sym->bias + q * sym->cmpl_freq; } // Equivalent to RansDecAdvance that takes a symbol. static inline void RansDecAdvanceSymbol(RansState* r, uint8_t** pptr, RansDecSymbol const* sym, uint32_t scale_bits) { RansDecAdvance(r, pptr, sym->start, sym->freq, scale_bits); } // Advances in the bit stream by "popping" a single symbol with range start // "start" and frequency "freq". All frequencies are assumed to sum to "1 << scale_bits". // No renormalization or output happens. static inline void RansDecAdvanceStep(RansState* r, uint32_t start, uint32_t freq, uint32_t scale_bits) { uint32_t mask = (1u << scale_bits) - 1; // s, x = D(x) uint32_t x = *r; *r = freq * (x >> scale_bits) + (x & mask) - start; } // Equivalent to RansDecAdvanceStep that takes a symbol. static inline void RansDecAdvanceSymbolStep(RansState* r, RansDecSymbol const* sym, uint32_t scale_bits) { RansDecAdvanceStep(r, sym->start, sym->freq, scale_bits); } // Renormalize. static inline void RansDecRenorm(RansState* r, uint8_t** pptr) { // renormalize uint32_t x = *r; if (x < RANS_BYTE_L) { uint8_t* ptr = *pptr; do x = (x << 8) | *ptr++; while (x < RANS_BYTE_L); *pptr = ptr; } *r = x; } #endif // RANS_BYTE_HEADER stacks-1.44/htslib/cram/._rANS_static.c000644 000765 000024 00000000265 12750743246 020461 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/rANS_static.c000644 000765 000024 00000051104 12750743246 020242 0ustar00catchenstaff000000 000000 /* * Copyright (c) 2014 Genome Research Ltd. * Author(s): James Bonfield * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are met: * * 1. Redistributions of source code must retain the above copyright notice, * this list of conditions and the following disclaimer. * * 2. Redistributions in binary form must reproduce the above * copyright notice, this list of conditions and the following * disclaimer in the documentation and/or other materials provided * with the distribution. * * 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger * Institute nor the names of its contributors may be used to endorse * or promote products derived from this software without specific * prior written permission. * * THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS * IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A * PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH * LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /* * Author: James Bonfield, Wellcome Trust Sanger Institute. 2014 */ #include #include #include #include #include #include #include #include #include "cram/rANS_static.h" #include "cram/rANS_byte.h" #define TF_SHIFT 12 #define TOTFREQ (1<0?(a):-(a)) #ifndef BLK_SIZE # define BLK_SIZE 1024*1024 #endif // Room to allow for expanded BLK_SIZE on worst case compression. #define BLK_SIZE2 ((int)(1.05*BLK_SIZE)) /*----------------------------------------------------------------------------- * Memory to memory compression functions. * * These are original versions without any manual loop unrolling. They * are easier to understand, but can be up to 2x slower. */ unsigned char *rans_compress_O0(unsigned char *in, unsigned int in_size, unsigned int *out_size) { unsigned char *out_buf = malloc(1.05*in_size + 257*257*3 + 9); unsigned char *cp, *out_end; RansEncSymbol syms[256]; RansState rans0, rans1, rans2, rans3; uint8_t* ptr; int F[256] = {0}, i, j, tab_size, rle, x, fsum = 0; int m = 0, M = 0; uint64_t tr; if (!out_buf) return NULL; ptr = out_end = out_buf + (int)(1.05*in_size) + 257*257*3 + 9; // Compute statistics for (i = 0; i < in_size; i++) { F[in[i]]++; } tr = ((uint64_t)TOTFREQ<<31)/in_size + (1<<30)/in_size; // Normalise so T[i] == TOTFREQ for (m = M = j = 0; j < 256; j++) { if (!F[j]) continue; if (m < F[j]) m = F[j], M = j; if ((F[j] = (F[j]*tr)>>31) == 0) F[j] = 1; fsum += F[j]; } fsum++; if (fsum < TOTFREQ) F[M] += TOTFREQ-fsum; else F[M] -= fsum-TOTFREQ; //printf("F[%d]=%d\n", M, F[M]); assert(F[M]>0); // Encode statistics. cp = out_buf+9; for (x = rle = j = 0; j < 256; j++) { if (F[j]) { // j if (rle) { rle--; } else { *cp++ = j; if (!rle && j && F[j-1]) { for(rle=j+1; rle<256 && F[rle]; rle++) ; rle -= j+1; *cp++ = rle; } //fprintf(stderr, "%d: %d %d\n", j, rle, N[j]); } // F[j] if (F[j]<128) { *cp++ = F[j]; } else { *cp++ = 128 | (F[j]>>8); *cp++ = F[j]&0xff; } RansEncSymbolInit(&syms[j], x, F[j], TF_SHIFT); x += F[j]; } } *cp++ = 0; //write(1, out_buf+4, cp-(out_buf+4)); tab_size = cp-out_buf; RansEncInit(&rans0); RansEncInit(&rans1); RansEncInit(&rans2); RansEncInit(&rans3); switch (i=(in_size&3)) { case 3: RansEncPutSymbol(&rans2, &ptr, &syms[in[in_size-(i-2)]]); case 2: RansEncPutSymbol(&rans1, &ptr, &syms[in[in_size-(i-1)]]); case 1: RansEncPutSymbol(&rans0, &ptr, &syms[in[in_size-(i-0)]]); case 0: break; } for (i=(in_size &~3); i>0; i-=4) { RansEncSymbol *s3 = &syms[in[i-1]]; RansEncSymbol *s2 = &syms[in[i-2]]; RansEncSymbol *s1 = &syms[in[i-3]]; RansEncSymbol *s0 = &syms[in[i-4]]; RansEncPutSymbol(&rans3, &ptr, s3); RansEncPutSymbol(&rans2, &ptr, s2); RansEncPutSymbol(&rans1, &ptr, s1); RansEncPutSymbol(&rans0, &ptr, s0); } RansEncFlush(&rans3, &ptr); RansEncFlush(&rans2, &ptr); RansEncFlush(&rans1, &ptr); RansEncFlush(&rans0, &ptr); // Finalise block size and return it *out_size = (out_end - ptr) + tab_size; cp = out_buf; *cp++ = 0; // order *cp++ = ((*out_size-9)>> 0) & 0xff; *cp++ = ((*out_size-9)>> 8) & 0xff; *cp++ = ((*out_size-9)>>16) & 0xff; *cp++ = ((*out_size-9)>>24) & 0xff; *cp++ = (in_size>> 0) & 0xff; *cp++ = (in_size>> 8) & 0xff; *cp++ = (in_size>>16) & 0xff; *cp++ = (in_size>>24) & 0xff; memmove(out_buf + tab_size, ptr, out_end-ptr); return out_buf; } typedef struct { struct { int F; int C; } fc[256]; unsigned char *R; } ari_decoder; unsigned char *rans_uncompress_O0(unsigned char *in, unsigned int in_size, unsigned int *out_size) { /* Load in the static tables */ unsigned char *cp = in + 9; int i, j, x, out_sz, in_sz, rle; char *out_buf; ari_decoder D; RansDecSymbol syms[256]; memset(&D, 0, sizeof(D)); if (*in++ != 0) // Order-0 check return NULL; in_sz = ((in[0])<<0) | ((in[1])<<8) | ((in[2])<<16) | ((in[3])<<24); out_sz = ((in[4])<<0) | ((in[5])<<8) | ((in[6])<<16) | ((in[7])<<24); if (in_sz != in_size-9) return NULL; out_buf = malloc(out_sz); if (!out_buf) return NULL; //fprintf(stderr, "out_sz=%d\n", out_sz); // Precompute reverse lookup of frequency. rle = x = 0; j = *cp++; do { if ((D.fc[j].F = *cp++) >= 128) { D.fc[j].F &= ~128; D.fc[j].F = ((D.fc[j].F & 127) << 8) | *cp++; } D.fc[j].C = x; RansDecSymbolInit(&syms[j], D.fc[j].C, D.fc[j].F); /* Build reverse lookup table */ if (!D.R) D.R = (unsigned char *)malloc(TOTFREQ); memset(&D.R[x], j, D.fc[j].F); x += D.fc[j].F; if (!rle && j+1 == *cp) { j = *cp++; rle = *cp++; } else if (rle) { rle--; j++; } else { j = *cp++; } } while(j); assert(x < TOTFREQ); RansState rans0, rans1, rans2, rans3; uint8_t *ptr = cp; RansDecInit(&rans0, &ptr); RansDecInit(&rans1, &ptr); RansDecInit(&rans2, &ptr); RansDecInit(&rans3, &ptr); int out_end = (out_sz&~3); RansState R[4]; R[0] = rans0; R[1] = rans1; R[2] = rans2; R[3] = rans3; uint32_t mask = (1u << TF_SHIFT)-1; for (i=0; i < out_end; i+=4) { uint32_t m[4] = {R[0] & mask, R[1] & mask, R[2] & mask, R[3] & mask}; uint8_t c[4] = {D.R[m[0]], D.R[m[1]], D.R[m[2]], D.R[m[3]]}; out_buf[i+0] = c[0]; out_buf[i+1] = c[1]; out_buf[i+2] = c[2]; out_buf[i+3] = c[3]; // RansDecAdvanceSymbolStep(&R[0], &syms[c[0]], TF_SHIFT); // RansDecAdvanceSymbolStep(&R[1], &syms[c[1]], TF_SHIFT); // RansDecAdvanceSymbolStep(&R[2], &syms[c[2]], TF_SHIFT); // RansDecAdvanceSymbolStep(&R[3], &syms[c[3]], TF_SHIFT); R[0] = syms[c[0]].freq * (R[0]>>TF_SHIFT); R[1] = syms[c[1]].freq * (R[1]>>TF_SHIFT); R[2] = syms[c[2]].freq * (R[2]>>TF_SHIFT); R[3] = syms[c[3]].freq * (R[3]>>TF_SHIFT); R[0] += m[0] - syms[c[0]].start; R[1] += m[1] - syms[c[1]].start; R[2] += m[2] - syms[c[2]].start; R[3] += m[3] - syms[c[3]].start; RansDecRenorm(&R[0], &ptr); RansDecRenorm(&R[1], &ptr); RansDecRenorm(&R[2], &ptr); RansDecRenorm(&R[3], &ptr); } rans0 = R[0]; rans1 = R[1]; rans2 = R[2]; rans3 = R[3]; switch(out_sz&3) { unsigned char c; case 0: break; case 1: c = D.R[RansDecGet(&rans0, TF_SHIFT)]; RansDecAdvanceSymbol(&rans0, &ptr, &syms[c], TF_SHIFT); out_buf[out_end] = c; break; case 2: c = D.R[RansDecGet(&rans0, TF_SHIFT)]; RansDecAdvanceSymbol(&rans0, &ptr, &syms[c], TF_SHIFT); out_buf[out_end] = c; c = D.R[RansDecGet(&rans1, TF_SHIFT)]; RansDecAdvanceSymbol(&rans1, &ptr, &syms[c], TF_SHIFT); out_buf[out_end+1] = c; break; case 3: c = D.R[RansDecGet(&rans0, TF_SHIFT)]; RansDecAdvanceSymbol(&rans0, &ptr, &syms[c], TF_SHIFT); out_buf[out_end] = c; c = D.R[RansDecGet(&rans1, TF_SHIFT)]; RansDecAdvanceSymbol(&rans1, &ptr, &syms[c], TF_SHIFT); out_buf[out_end+1] = c; c = D.R[RansDecGet(&rans2, TF_SHIFT)]; RansDecAdvanceSymbol(&rans2, &ptr, &syms[c], TF_SHIFT); out_buf[out_end+2] = c; break; } *out_size = out_sz; if (D.R) free(D.R); return (unsigned char *)out_buf; } unsigned char *rans_compress_O1(unsigned char *in, unsigned int in_size, unsigned int *out_size) { unsigned char *out_buf = NULL, *out_end, *cp; unsigned int last_i, tab_size, rle_i, rle_j; RansEncSymbol (*syms)[256] = NULL; /* syms[256][256] */ int (*F)[256] = NULL; /* F[256][256] */ int *T = NULL; /* T[256] */ int i, j; unsigned char c; if (in_size < 4) return rans_compress_O0(in, in_size, out_size); syms = malloc(256 * sizeof(*syms)); if (!syms) goto cleanup; F = calloc(256, sizeof(*F)); if (!F) goto cleanup; T = calloc(256, sizeof(*T)); if (!T) goto cleanup; out_buf = malloc(1.05*in_size + 257*257*3 + 9); if (!out_buf) goto cleanup; out_end = out_buf + (int)(1.05*in_size) + 257*257*3 + 9; cp = out_buf+9; //for (last = 0, i=in_size-1; i>=0; i--) { // F[last][c = in[i]]++; // T[last]++; // last = c; //} for (last_i=i=0; i>2)]]++; F[0][in[2*(in_size>>2)]]++; F[0][in[3*(in_size>>2)]]++; T[0]+=3; // Normalise so T[i] == TOTFREQ for (rle_i = i = 0; i < 256; i++) { int t2, m, M; unsigned int x; if (T[i] == 0) continue; //uint64_t p = (TOTFREQ * TOTFREQ) / t; double p = ((double)TOTFREQ)/T[i]; for (t2 = m = M = j = 0; j < 256; j++) { if (!F[i][j]) continue; if (m < F[i][j]) m = F[i][j], M = j; //if ((F[i][j] = (F[i][j] * p) / TOTFREQ) == 0) if ((F[i][j] *= p) == 0) F[i][j] = 1; t2 += F[i][j]; } t2++; if (t2 < TOTFREQ) F[i][M] += TOTFREQ-t2; else F[i][M] -= t2-TOTFREQ; // Store frequency table // i if (rle_i) { rle_i--; } else { *cp++ = i; // FIXME: could use order-0 statistics to observe which alphabet // symbols are present and base RLE on that ordering instead. if (i && T[i-1]) { for(rle_i=i+1; rle_i<256 && T[rle_i]; rle_i++) ; rle_i -= i+1; *cp++ = rle_i; } } int *F_i_ = F[i]; x = 0; rle_j = 0; for (j = 0; j < 256; j++) { if (F_i_[j]) { //fprintf(stderr, "F[%d][%d]=%d, x=%d\n", i, j, F_i_[j], x); // j if (rle_j) { rle_j--; } else { *cp++ = j; if (!rle_j && j && F_i_[j-1]) { for(rle_j=j+1; rle_j<256 && F_i_[rle_j]; rle_j++) ; rle_j -= j+1; *cp++ = rle_j; } } // F_i_[j] if (F_i_[j]<128) { *cp++ = F_i_[j]; } else { *cp++ = 128 | (F_i_[j]>>8); *cp++ = F_i_[j]&0xff; } RansEncSymbolInit(&syms[i][j], x, F_i_[j], TF_SHIFT); x += F_i_[j]; } } *cp++ = 0; } *cp++ = 0; //write(1, out_buf+4, cp-(out_buf+4)); tab_size = cp - out_buf; assert(tab_size < 257*257*3); RansState rans0, rans1, rans2, rans3; RansEncInit(&rans0); RansEncInit(&rans1); RansEncInit(&rans2); RansEncInit(&rans3); uint8_t* ptr = out_end; int isz4 = in_size>>2; int i0 = 1*isz4-2; int i1 = 2*isz4-2; int i2 = 3*isz4-2; int i3 = 4*isz4-2; unsigned char l0 = in[i0+1]; unsigned char l1 = in[i1+1]; unsigned char l2 = in[i2+1]; unsigned char l3 = in[i3+1]; // Deal with the remainder l3 = in[in_size-1]; for (i3 = in_size-2; i3 > 4*isz4-2; i3--) { unsigned char c3 = in[i3]; RansEncPutSymbol(&rans3, &ptr, &syms[c3][l3]); l3 = c3; } for (; i0 >= 0; i0--, i1--, i2--, i3--) { unsigned char c0, c1, c2, c3; RansEncSymbol *s3 = &syms[c3 = in[i3]][l3]; RansEncSymbol *s2 = &syms[c2 = in[i2]][l2]; RansEncSymbol *s1 = &syms[c1 = in[i1]][l1]; RansEncSymbol *s0 = &syms[c0 = in[i0]][l0]; RansEncPutSymbol(&rans3, &ptr, s3); RansEncPutSymbol(&rans2, &ptr, s2); RansEncPutSymbol(&rans1, &ptr, s1); RansEncPutSymbol(&rans0, &ptr, s0); l0 = c0; l1 = c1; l2 = c2; l3 = c3; } RansEncPutSymbol(&rans3, &ptr, &syms[0][l3]); RansEncPutSymbol(&rans2, &ptr, &syms[0][l2]); RansEncPutSymbol(&rans1, &ptr, &syms[0][l1]); RansEncPutSymbol(&rans0, &ptr, &syms[0][l0]); RansEncFlush(&rans3, &ptr); RansEncFlush(&rans2, &ptr); RansEncFlush(&rans1, &ptr); RansEncFlush(&rans0, &ptr); *out_size = (out_end - ptr) + tab_size; cp = out_buf; *cp++ = 1; // order *cp++ = ((*out_size-9)>> 0) & 0xff; *cp++ = ((*out_size-9)>> 8) & 0xff; *cp++ = ((*out_size-9)>>16) & 0xff; *cp++ = ((*out_size-9)>>24) & 0xff; *cp++ = (in_size>> 0) & 0xff; *cp++ = (in_size>> 8) & 0xff; *cp++ = (in_size>>16) & 0xff; *cp++ = (in_size>>24) & 0xff; memmove(out_buf + tab_size, ptr, out_end-ptr); cleanup: free(syms); free(F); free(T); return out_buf; } unsigned char *rans_uncompress_O1(unsigned char *in, unsigned int in_size, unsigned int *out_size) { /* Load in the static tables */ unsigned char *cp = in + 9; int i, j = -999, x, out_sz, in_sz, rle_i, rle_j; char *out_buf = NULL; ari_decoder *D = NULL; /* D[256] */ RansDecSymbol (*syms)[256] = NULL; /* syms[256][256] */ if (*in++ != 1) // Order-1 check return NULL; in_sz = ((in[0])<<0) | ((in[1])<<8) | ((in[2])<<16) | ((in[3])<<24); out_sz = ((in[4])<<0) | ((in[5])<<8) | ((in[6])<<16) | ((in[7])<<24); if (in_sz != in_size-9) return NULL; D = calloc(256, sizeof(*D)); if (!D) goto cleanup; syms = malloc(256 * sizeof(*syms)); if (!syms) goto cleanup; //fprintf(stderr, "out_sz=%d\n", out_sz); //i = *cp++; rle_i = 0; i = *cp++; do { rle_j = x = 0; j = *cp++; do { if ((D[i].fc[j].F = *cp++) >= 128) { D[i].fc[j].F &= ~128; D[i].fc[j].F = ((D[i].fc[j].F & 127) << 8) | *cp++; } D[i].fc[j].C = x; //fprintf(stderr, "i=%d j=%d F=%d C=%d\n", i, j, D[i].fc[j].F, D[i].fc[j].C); if (!D[i].fc[j].F) D[i].fc[j].F = TOTFREQ; RansDecSymbolInit(&syms[i][j], D[i].fc[j].C, D[i].fc[j].F); /* Build reverse lookup table */ if (!D[i].R) { D[i].R = (unsigned char *)malloc(TOTFREQ); if (!D[i].R) goto cleanup; } memset(&D[i].R[x], j, D[i].fc[j].F); x += D[i].fc[j].F; assert(x <= TOTFREQ); if (!rle_j && j+1 == *cp) { j = *cp++; rle_j = *cp++; } else if (rle_j) { rle_j--; j++; } else { j = *cp++; } } while(j); if (!rle_i && i+1 == *cp) { i = *cp++; rle_i = *cp++; } else if (rle_i) { rle_i--; i++; } else { i = *cp++; } } while (i); // Precompute reverse lookup of frequency. RansState rans0, rans1, rans2, rans3; uint8_t *ptr = cp; RansDecInit(&rans0, &ptr); RansDecInit(&rans1, &ptr); RansDecInit(&rans2, &ptr); RansDecInit(&rans3, &ptr); int isz4 = out_sz>>2; int l0 = 0; int l1 = 0; int l2 = 0; int l3 = 0; int i4[] = {0*isz4, 1*isz4, 2*isz4, 3*isz4}; RansState R[4]; R[0] = rans0; R[1] = rans1; R[2] = rans2; R[3] = rans3; /* Allocate output buffer */ out_buf = malloc(out_sz); if (!out_buf) goto cleanup; for (; i4[0] < isz4; i4[0]++, i4[1]++, i4[2]++, i4[3]++) { uint32_t m[4] = {R[0] & ((1u << TF_SHIFT)-1), R[1] & ((1u << TF_SHIFT)-1), R[2] & ((1u << TF_SHIFT)-1), R[3] & ((1u << TF_SHIFT)-1)}; uint8_t c[4] = {D[l0].R[m[0]], D[l1].R[m[1]], D[l2].R[m[2]], D[l3].R[m[3]]}; out_buf[i4[0]] = c[0]; out_buf[i4[1]] = c[1]; out_buf[i4[2]] = c[2]; out_buf[i4[3]] = c[3]; //RansDecAdvanceSymbolStep(&R[0], &syms[l0][c[0]], TF_SHIFT); //RansDecAdvanceSymbolStep(&R[1], &syms[l1][c[1]], TF_SHIFT); //RansDecAdvanceSymbolStep(&R[2], &syms[l2][c[2]], TF_SHIFT); //RansDecAdvanceSymbolStep(&R[3], &syms[l3][c[3]], TF_SHIFT); R[0] = syms[l0][c[0]].freq * (R[0]>>TF_SHIFT); R[1] = syms[l1][c[1]].freq * (R[1]>>TF_SHIFT); R[2] = syms[l2][c[2]].freq * (R[2]>>TF_SHIFT); R[3] = syms[l3][c[3]].freq * (R[3]>>TF_SHIFT); R[0] += m[0] - syms[l0][c[0]].start; R[1] += m[1] - syms[l1][c[1]].start; R[2] += m[2] - syms[l2][c[2]].start; R[3] += m[3] - syms[l3][c[3]].start; RansDecRenorm(&R[0], &ptr); RansDecRenorm(&R[1], &ptr); RansDecRenorm(&R[2], &ptr); RansDecRenorm(&R[3], &ptr); l0 = c[0]; l1 = c[1]; l2 = c[2]; l3 = c[3]; } rans0 = R[0]; rans1 = R[1]; rans2 = R[2]; rans3 = R[3]; // Remainder for (; i4[3] < out_sz; i4[3]++) { unsigned char c3 = D[l3].R[RansDecGet(&rans3, TF_SHIFT)]; out_buf[i4[3]] = c3; RansDecAdvanceSymbol(&rans3, &ptr, &syms[l3][c3], TF_SHIFT); l3 = c3; } *out_size = out_sz; cleanup: if (D) { for (i = 0; i < 256; i++) if (D[i].R) free(D[i].R); free(D); } free(syms); return (unsigned char *)out_buf; } /*----------------------------------------------------------------------------- * Simple interface to the order-0 vs order-1 encoders and decoders. */ unsigned char *rans_compress(unsigned char *in, unsigned int in_size, unsigned int *out_size, int order) { return order ? rans_compress_O1(in, in_size, out_size) : rans_compress_O0(in, in_size, out_size); } unsigned char *rans_uncompress(unsigned char *in, unsigned int in_size, unsigned int *out_size) { /* Both rans_uncompress functions need to be able to read at least 9 bytes. */ if (in_size < 9) return NULL; return in[0] ? rans_uncompress_O1(in, in_size, out_size) : rans_uncompress_O0(in, in_size, out_size); } #ifdef TEST_MAIN /*----------------------------------------------------------------------------- * Main. * * This is a simple command line tool for testing order-0 and order-1 * compression using the rANS codec. Simply compile with * * gcc -DTEST_MAIN -O3 -I. cram/rANS_static.c -o cram/rANS_static * * Usage: cram/rANS_static -o0 < file > file.o0 * cram/rANS_static -d < file.o0 > file2 * * cram/rANS_static -o1 < file > file.o1 * cram/rANS_static -d < file.o1 > file2 */ int main(int argc, char **argv) { int opt, order = 0; unsigned char in_buf[BLK_SIZE2+257*257*3]; int decode = 0; FILE *infp = stdin, *outfp = stdout; struct timeval tv1, tv2; size_t bytes = 0; extern char *optarg; extern int optind; while ((opt = getopt(argc, argv, "o:d")) != -1) { switch (opt) { case 'o': order = atoi(optarg); break; case 'd': decode = 1; break; } } order = order ? 1 : 0; // Only support O(0) and O(1) if (optind < argc) { if (!(infp = fopen(argv[optind], "rb"))) { perror(argv[optind]); return 1; } optind++; } if (optind < argc) { if (!(outfp = fopen(argv[optind], "wb"))) { perror(argv[optind]); return 1; } optind++; } gettimeofday(&tv1, NULL); if (decode) { // Only used in some test implementations of RC_GetFreq() //RC_init(); //RC_init2(); for (;;) { uint32_t in_size, out_size; unsigned char *out; if (9 != fread(in_buf, 1, 9, infp)) break; in_size = *(int *)&in_buf[1]; if (in_size != fread(in_buf+9, 1, in_size, infp)) { fprintf(stderr, "Truncated input\n"); exit(1); } out = rans_uncompress(in_buf, in_size+9, &out_size); if (!out) abort(); fwrite(out, 1, out_size, outfp); free(out); bytes += out_size; } } else { for (;;) { uint32_t in_size, out_size; unsigned char *out; in_size = fread(in_buf, 1, BLK_SIZE, infp); if (in_size <= 0) break; out = rans_compress(in_buf, in_size, &out_size, order); fwrite(out, 1, out_size, outfp); free(out); bytes += in_size; } } gettimeofday(&tv2, NULL); fprintf(stderr, "Took %ld microseconds, %5.1f MB/s\n", (long)(tv2.tv_sec - tv1.tv_sec)*1000000 + tv2.tv_usec - tv1.tv_usec, (double)bytes / ((long)(tv2.tv_sec - tv1.tv_sec)*1000000 + tv2.tv_usec - tv1.tv_usec)); return 0; } #endif stacks-1.44/htslib/cram/._rANS_static.h000644 000765 000024 00000000265 12750743246 020466 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/rANS_static.h000644 000765 000024 00000003761 12750743246 020255 0ustar00catchenstaff000000 000000 /* * Copyright (c) 2014 Genome Research Ltd. * Author(s): James Bonfield * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are met: * * 1. Redistributions of source code must retain the above copyright notice, * this list of conditions and the following disclaimer. * * 2. Redistributions in binary form must reproduce the above * copyright notice, this list of conditions and the following * disclaimer in the documentation and/or other materials provided * with the distribution. * * 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger * Institute nor the names of its contributors may be used to endorse * or promote products derived from this software without specific * prior written permission. * * THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS * IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A * PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH * LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #ifndef RANS_STATIC_H #define RANS_STATIC_H #ifdef __cplusplus extern "C" { #endif unsigned char *rans_compress(unsigned char *in, unsigned int in_size, unsigned int *out_size, int order); unsigned char *rans_uncompress(unsigned char *in, unsigned int in_size, unsigned int *out_size); #ifdef __cplusplus } #endif #endif /* RANS_STATIC_H */ stacks-1.44/htslib/cram/._sam_header.c000644 000765 000024 00000000265 12750743246 020377 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/sam_header.c000644 000765 000024 00000071757 12750743246 020200 0ustar00catchenstaff000000 000000 /* Copyright (c) 2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #include #include #include #include "cram/sam_header.h" #include "cram/string_alloc.h" static void sam_hdr_error(char *msg, char *line, int len, int lno) { int j; for (j = 0; j < len && line[j] != '\n'; j++) ; fprintf(stderr, "%s at line %d: \"%.*s\"\n", msg, lno, j, line); } void sam_hdr_dump(SAM_hdr *hdr) { khint_t k; int i; printf("===DUMP===\n"); for (k = kh_begin(hdr->h); k != kh_end(hdr->h); k++) { SAM_hdr_type *t1, *t2; char c[2]; if (!kh_exist(hdr->h, k)) continue; t1 = t2 = kh_val(hdr->h, k); c[0] = kh_key(hdr->h, k)>>8; c[1] = kh_key(hdr->h, k)&0xff; printf("Type %.2s, count %d\n", c, t1->prev->order+1); do { SAM_hdr_tag *tag; printf(">>>%d ", t1->order); for (tag = t1->tag; tag; tag=tag->next) { printf("\"%.2s\":\"%.*s\"\t", tag->str, tag->len-3, tag->str+3); } putchar('\n'); t1 = t1->next; } while (t1 != t2); } /* Dump out PG chains */ printf("\n@PG chains:\n"); for (i = 0; i < hdr->npg_end; i++) { int j; printf(" %d:", i); for (j = hdr->pg_end[i]; j != -1; j = hdr->pg[j].prev_id) { printf("%s%d(%.*s)", j == hdr->pg_end[i] ? " " : "->", j, hdr->pg[j].name_len, hdr->pg[j].name); } printf("\n"); } puts("===END DUMP==="); } /* Updates the hash tables in the SAM_hdr structure. * * Returns 0 on success; * -1 on failure */ static int sam_hdr_update_hashes(SAM_hdr *sh, int type, SAM_hdr_type *h_type) { /* Add to reference hash? */ if ((type>>8) == 'S' && (type&0xff) == 'Q') { SAM_hdr_tag *tag; SAM_SQ *new_ref; int nref = sh->nref; new_ref = realloc(sh->ref, (sh->nref+1)*sizeof(*sh->ref)); if (!new_ref) return -1; sh->ref = new_ref; tag = h_type->tag; sh->ref[nref].name = NULL; sh->ref[nref].len = 0; sh->ref[nref].ty = h_type; sh->ref[nref].tag = tag; while (tag) { if (tag->str[0] == 'S' && tag->str[1] == 'N') { if (!(sh->ref[nref].name = malloc(tag->len))) return -1; strncpy(sh->ref[nref].name, tag->str+3, tag->len-3); sh->ref[nref].name[tag->len-3] = 0; } else if (tag->str[0] == 'L' && tag->str[1] == 'N') { sh->ref[nref].len = atoi(tag->str+3); } tag = tag->next; } if (sh->ref[nref].name) { khint_t k; int r; k = kh_put(m_s2i, sh->ref_hash, sh->ref[nref].name, &r); if (-1 == r) return -1; kh_val(sh->ref_hash, k) = nref; } else { return -1; // SN should be present, according to spec. } sh->nref++; } /* Add to read-group hash? */ if ((type>>8) == 'R' && (type&0xff) == 'G') { SAM_hdr_tag *tag; SAM_RG *new_rg; int nrg = sh->nrg; new_rg = realloc(sh->rg, (sh->nrg+1)*sizeof(*sh->rg)); if (!new_rg) return -1; sh->rg = new_rg; tag = h_type->tag; sh->rg[nrg].name = NULL; sh->rg[nrg].name_len = 0; sh->rg[nrg].ty = h_type; sh->rg[nrg].tag = tag; sh->rg[nrg].id = nrg; while (tag) { if (tag->str[0] == 'I' && tag->str[1] == 'D') { if (!(sh->rg[nrg].name = malloc(tag->len))) return -1; strncpy(sh->rg[nrg].name, tag->str+3, tag->len-3); sh->rg[nrg].name[tag->len-3] = 0; sh->rg[nrg].name_len = strlen(sh->rg[nrg].name); } tag = tag->next; } if (sh->rg[nrg].name) { khint_t k; int r; k = kh_put(m_s2i, sh->rg_hash, sh->rg[nrg].name, &r); if (-1 == r) return -1; kh_val(sh->rg_hash, k) = nrg; } else { return -1; // ID should be present, according to spec. } sh->nrg++; } /* Add to program hash? */ if ((type>>8) == 'P' && (type&0xff) == 'G') { SAM_hdr_tag *tag; SAM_PG *new_pg; int npg = sh->npg; new_pg = realloc(sh->pg, (sh->npg+1)*sizeof(*sh->pg)); if (!new_pg) return -1; sh->pg = new_pg; tag = h_type->tag; sh->pg[npg].name = NULL; sh->pg[npg].name_len = 0; sh->pg[npg].ty = h_type; sh->pg[npg].tag = tag; sh->pg[npg].id = npg; sh->pg[npg].prev_id = -1; while (tag) { if (tag->str[0] == 'I' && tag->str[1] == 'D') { if (!(sh->pg[npg].name = malloc(tag->len))) return -1; strncpy(sh->pg[npg].name, tag->str+3, tag->len-3); sh->pg[npg].name[tag->len-3] = 0; sh->pg[npg].name_len = strlen(sh->pg[npg].name); } else if (tag->str[0] == 'P' && tag->str[1] == 'P') { // Resolve later if needed khint_t k; char tmp = tag->str[tag->len]; tag->str[tag->len] = 0; k = kh_get(m_s2i, sh->pg_hash, tag->str+3); tag->str[tag->len] = tmp; if (k != kh_end(sh->pg_hash)) { int p_id = kh_val(sh->pg_hash, k); sh->pg[npg].prev_id = sh->pg[p_id].id; /* Unmark previous entry as a PG termination */ if (sh->npg_end > 0 && sh->pg_end[sh->npg_end-1] == p_id) { sh->npg_end--; } else { int i; for (i = 0; i < sh->npg_end; i++) { if (sh->pg_end[i] == p_id) { memmove(&sh->pg_end[i], &sh->pg_end[i+1], (sh->npg_end-i-1)*sizeof(*sh->pg_end)); sh->npg_end--; } } } } else { sh->pg[npg].prev_id = -1; } } tag = tag->next; } if (sh->pg[npg].name) { khint_t k; int r; k = kh_put(m_s2i, sh->pg_hash, sh->pg[npg].name, &r); if (-1 == r) return -1; kh_val(sh->pg_hash, k) = npg; } else { return -1; // ID should be present, according to spec. } /* Add to npg_end[] array. Remove later if we find a PP line */ if (sh->npg_end >= sh->npg_end_alloc) { int *new_pg_end; int new_alloc = sh->npg_end_alloc ? sh->npg_end_alloc*2 : 4; new_pg_end = realloc(sh->pg_end, new_alloc * sizeof(int)); if (!new_pg_end) return -1; sh->npg_end_alloc = new_alloc; sh->pg_end = new_pg_end; } sh->pg_end[sh->npg_end++] = npg; sh->npg++; } return 0; } /* * Appends a formatted line to an existing SAM header. * Line is a full SAM header record, eg "@SQ\tSN:foo\tLN:100", with * optional new-line. If it contains more than 1 line then multiple lines * will be added in order. * * Input text is of maximum length len or as terminated earlier by a NUL. * Len may be 0 if unknown, in which case lines must be NUL-terminated. * * Returns 0 on success * -1 on failure */ int sam_hdr_add_lines(SAM_hdr *sh, const char *lines, int len) { int i, lno, text_offset; char *hdr; if (!len) len = strlen(lines); text_offset = ks_len(&sh->text); if (EOF == kputsn(lines, len, &sh->text)) return -1; hdr = ks_str(&sh->text) + text_offset; for (i = 0, lno = 1; i < len && hdr[i] != '\0'; i++, lno++) { khint32_t type; khint_t k; int l_start = i, new; SAM_hdr_type *h_type; SAM_hdr_tag *h_tag, *last; if (hdr[i] != '@') { int j; for (j = i; j < len && hdr[j] != '\0' && hdr[j] != '\n'; j++) ; sam_hdr_error("Header line does not start with '@'", &hdr[l_start], len - l_start, lno); return -1; } type = (hdr[i+1]<<8) | hdr[i+2]; if (hdr[i+1] < 'A' || hdr[i+1] > 'z' || hdr[i+2] < 'A' || hdr[i+2] > 'z') { sam_hdr_error("Header line does not have a two character key", &hdr[l_start], len - l_start, lno); return -1; } i += 3; if (hdr[i] == '\n') continue; // Add the header line type if (!(h_type = pool_alloc(sh->type_pool))) return -1; if (-1 == (k = kh_put(sam_hdr, sh->h, type, &new))) return -1; // Form the ring, either with self or other lines of this type if (!new) { SAM_hdr_type *t = kh_val(sh->h, k), *p; p = t->prev; assert(p->next == t); p->next = h_type; h_type->prev = p; t->prev = h_type; h_type->next = t; h_type->order = p->order+1; } else { kh_val(sh->h, k) = h_type; h_type->prev = h_type->next = h_type; h_type->order = 0; } // Parse the tags on this line last = NULL; if ((type>>8) == 'C' && (type&0xff) == 'O') { int j; if (hdr[i] != '\t') { sam_hdr_error("Missing tab", &hdr[l_start], len - l_start, lno); return -1; } for (j = ++i; j < len && hdr[j] != '\0' && hdr[j] != '\n'; j++) ; if (!(h_type->tag = h_tag = pool_alloc(sh->tag_pool))) return -1; h_tag->str = string_ndup(sh->str_pool, &hdr[i], j-i); h_tag->len = j-i; h_tag->next = NULL; if (!h_tag->str) return -1; i = j; } else { do { int j; if (hdr[i] != '\t') { sam_hdr_error("Missing tab", &hdr[l_start], len - l_start, lno); return -1; } for (j = ++i; j < len && hdr[j] != '\0' && hdr[j] != '\n' && hdr[j] != '\t'; j++) ; if (!(h_tag = pool_alloc(sh->tag_pool))) return -1; h_tag->str = string_ndup(sh->str_pool, &hdr[i], j-i); h_tag->len = j-i; h_tag->next = NULL; if (!h_tag->str) return -1; if (h_tag->len < 3 || h_tag->str[2] != ':') { sam_hdr_error("Malformed key:value pair", &hdr[l_start], len - l_start, lno); return -1; } if (last) last->next = h_tag; else h_type->tag = h_tag; last = h_tag; i = j; } while (i < len && hdr[i] != '\0' && hdr[i] != '\n'); } /* Update RG/SQ hashes */ if (-1 == sam_hdr_update_hashes(sh, type, h_type)) return -1; } return 0; } /* * Adds a single line to a SAM header. * Specify type and one or more key,value pairs, ending with the NULL key. * Eg. sam_hdr_add(h, "SQ", "ID", "foo", "LN", "100", NULL). * * Returns index for specific entry on success (eg 2nd SQ, 4th RG) * -1 on failure */ int sam_hdr_add(SAM_hdr *sh, const char *type, ...) { va_list args; va_start(args, type); return sam_hdr_vadd(sh, type, args, NULL); } /* * sam_hdr_add with a va_list interface. * * Note: this function invokes va_arg at least once, making the value * of ap indeterminate after the return. The caller should call * va_start/va_end before/after calling this function or use va_copy. */ int sam_hdr_vadd(SAM_hdr *sh, const char *type, va_list ap, ...) { va_list args; SAM_hdr_type *h_type; SAM_hdr_tag *h_tag, *last; int new; khint32_t type_i = (type[0]<<8) | type[1], k; if (EOF == kputc_('@', &sh->text)) return -1; if (EOF == kputsn(type, 2, &sh->text)) return -1; if (!(h_type = pool_alloc(sh->type_pool))) return -1; if (-1 == (k = kh_put(sam_hdr, sh->h, type_i, &new))) return -1; // Form the ring, either with self or other lines of this type if (!new) { SAM_hdr_type *t = kh_val(sh->h, k), *p; p = t->prev; assert(p->next == t); p->next = h_type; h_type->prev = p; t->prev = h_type; h_type->next = t; h_type->order = p->order + 1; } else { kh_val(sh->h, k) = h_type; h_type->prev = h_type->next = h_type; h_type->order = 0; } last = NULL; // Any ... varargs va_start(args, ap); for (;;) { char *k, *v; int idx; if (!(k = (char *)va_arg(args, char *))) break; v = va_arg(args, char *); if (EOF == kputc_('\t', &sh->text)) return -1; if (!(h_tag = pool_alloc(sh->tag_pool))) return -1; idx = ks_len(&sh->text); if (EOF == kputs(k, &sh->text)) return -1; if (EOF == kputc_(':', &sh->text)) return -1; if (EOF == kputs(v, &sh->text)) return -1; h_tag->len = ks_len(&sh->text) - idx; h_tag->str = string_ndup(sh->str_pool, ks_str(&sh->text) + idx, h_tag->len); h_tag->next = NULL; if (!h_tag->str) return -1; if (last) last->next = h_tag; else h_type->tag = h_tag; last = h_tag; } va_end(args); // Plus the specified va_list params for (;;) { char *k, *v; int idx; if (!(k = (char *)va_arg(ap, char *))) break; v = va_arg(ap, char *); if (EOF == kputc_('\t', &sh->text)) return -1; if (!(h_tag = pool_alloc(sh->tag_pool))) return -1; idx = ks_len(&sh->text); if (EOF == kputs(k, &sh->text)) return -1; if (EOF == kputc_(':', &sh->text)) return -1; if (EOF == kputs(v, &sh->text)) return -1; h_tag->len = ks_len(&sh->text) - idx; h_tag->str = string_ndup(sh->str_pool, ks_str(&sh->text) + idx, h_tag->len); h_tag->next = NULL; if (!h_tag->str) return -1; if (last) last->next = h_tag; else h_type->tag = h_tag; last = h_tag; } va_end(ap); if (EOF == kputc('\n', &sh->text)) return -1; int itype = (type[0]<<8) | type[1]; if (-1 == sam_hdr_update_hashes(sh, itype, h_type)) return -1; return h_type->order; } /* * Returns the first header item matching 'type'. If ID is non-NULL it checks * for the tag ID: and compares against the specified ID. * * Returns NULL if no type/ID is found */ SAM_hdr_type *sam_hdr_find(SAM_hdr *hdr, char *type, char *ID_key, char *ID_value) { SAM_hdr_type *t1, *t2; int itype = (type[0]<<8)|(type[1]); khint_t k; /* Special case for types we have prebuilt hashes on */ if (ID_key) { if (type[0] == 'S' && type[1] == 'Q' && ID_key[0] == 'S' && ID_key[1] == 'N') { k = kh_get(m_s2i, hdr->ref_hash, ID_value); return k != kh_end(hdr->ref_hash) ? hdr->ref[kh_val(hdr->ref_hash, k)].ty : NULL; } if (type[0] == 'R' && type[1] == 'G' && ID_key[0] == 'I' && ID_key[1] == 'D') { k = kh_get(m_s2i, hdr->rg_hash, ID_value); return k != kh_end(hdr->rg_hash) ? hdr->rg[kh_val(hdr->rg_hash, k)].ty : NULL; } if (type[0] == 'P' && type[1] == 'G' && ID_key[0] == 'I' && ID_key[1] == 'D') { k = kh_get(m_s2i, hdr->pg_hash, ID_value); return k != kh_end(hdr->pg_hash) ? hdr->pg[kh_val(hdr->pg_hash, k)].ty : NULL; } } k = kh_get(sam_hdr, hdr->h, itype); if (k == kh_end(hdr->h)) return NULL; if (!ID_key) return kh_val(hdr->h, k); t1 = t2 = kh_val(hdr->h, k); do { SAM_hdr_tag *tag; for (tag = t1->tag; tag; tag = tag->next) { if (tag->str[0] == ID_key[0] && tag->str[1] == ID_key[1]) { char *cp1 = tag->str+3; char *cp2 = ID_value; while (*cp1 && *cp1 == *cp2) cp1++, cp2++; if (*cp2 || *cp1) continue; return t1; } } t1 = t1->next; } while (t1 != t2); return NULL; } /* * As per SAM_hdr_type, but returns a complete line of formatted text * for a specific head type/ID combination. If ID is NULL then it returns * the first line of the specified type. * * The returned string is malloced and should be freed by the calling * function with free(). * * Returns NULL if no type/ID is found. */ char *sam_hdr_find_line(SAM_hdr *hdr, char *type, char *ID_key, char *ID_value) { SAM_hdr_type *ty = sam_hdr_find(hdr, type, ID_key, ID_value); kstring_t ks = KS_INITIALIZER; SAM_hdr_tag *tag; int r = 0; if (!ty) return NULL; // Paste together the line from the hashed copy r |= (kputc_('@', &ks) == EOF); r |= (kputs(type, &ks) == EOF); for (tag = ty->tag; tag; tag = tag->next) { r |= (kputc_('\t', &ks) == EOF); r |= (kputsn(tag->str, tag->len, &ks) == EOF); } if (r) { KS_FREE(&ks); return NULL; } return ks_str(&ks); } /* * Looks for a specific key in a single sam header line. * If prev is non-NULL it also fills this out with the previous tag, to * permit use in key removal. *prev is set to NULL when the tag is the first * key in the list. When a tag isn't found, prev (if non NULL) will be the last * tag in the existing list. * * Returns the tag pointer on success * NULL on failure */ SAM_hdr_tag *sam_hdr_find_key(SAM_hdr *sh, SAM_hdr_type *type, char *key, SAM_hdr_tag **prev) { SAM_hdr_tag *tag, *p = NULL; for (tag = type->tag; tag; p = tag, tag = tag->next) { if (tag->str[0] == key[0] && tag->str[1] == key[1]) { if (prev) *prev = p; return tag; } } if (prev) *prev = p; return NULL; } /* * Adds or updates tag key,value pairs in a header line. * Eg for adding M5 tags to @SQ lines or updating sort order for the * @HD line (although use the sam_hdr_sort_order() function for * HD manipulation, which is a wrapper around this funuction). * * Specify multiple key,value pairs ending in NULL. * * Returns 0 on success * -1 on failure */ int sam_hdr_update(SAM_hdr *hdr, SAM_hdr_type *type, ...) { va_list ap; va_start(ap, type); for (;;) { char *k, *v; int idx; SAM_hdr_tag *tag, *prev; if (!(k = (char *)va_arg(ap, char *))) break; v = va_arg(ap, char *); tag = sam_hdr_find_key(hdr, type, k, &prev); if (!tag) { if (!(tag = pool_alloc(hdr->tag_pool))) return -1; if (prev) prev->next = tag; else type->tag = tag; tag->next = NULL; } idx = ks_len(&hdr->text); if (ksprintf(&hdr->text, "%2.2s:%s", k, v) < 0) return -1; tag->len = ks_len(&hdr->text) - idx; tag->str = string_ndup(hdr->str_pool, ks_str(&hdr->text) + idx, tag->len); if (!tag->str) return -1; } va_end(ap); return 0; } #define K(a) (((a)[0]<<8)|((a)[1])) /* * Returns the sort order: */ enum sam_sort_order sam_hdr_sort_order(SAM_hdr *hdr) { return hdr->sort_order; } static enum sam_sort_order sam_hdr_parse_sort_order(SAM_hdr *hdr) { khint_t k; enum sam_sort_order so; so = ORDER_UNKNOWN; k = kh_get(sam_hdr, hdr->h, K("HD")); if (k != kh_end(hdr->h)) { SAM_hdr_type *ty = kh_val(hdr->h, k); SAM_hdr_tag *tag; for (tag = ty->tag; tag; tag = tag->next) { if (tag->str[0] == 'S' && tag->str[1] == 'O') { if (strcmp(tag->str+3, "unsorted") == 0) so = ORDER_UNSORTED; else if (strcmp(tag->str+3, "queryname") == 0) so = ORDER_NAME; else if (strcmp(tag->str+3, "coordinate") == 0) so = ORDER_COORD; else if (strcmp(tag->str+3, "unknown") != 0) fprintf(stderr, "Unknown sort order field: %s\n", tag->str+3); } } } return so; } /* * Reconstructs the kstring from the header hash table. * Returns 0 on success * -1 on failure */ int sam_hdr_rebuild(SAM_hdr *hdr) { /* Order: HD then others */ kstring_t ks = KS_INITIALIZER; khint_t k; k = kh_get(sam_hdr, hdr->h, K("HD")); if (k != kh_end(hdr->h)) { SAM_hdr_type *ty = kh_val(hdr->h, k); SAM_hdr_tag *tag; if (EOF == kputs("@HD", &ks)) return -1; for (tag = ty->tag; tag; tag = tag->next) { if (EOF == kputc_('\t', &ks)) return -1; if (EOF == kputsn_(tag->str, tag->len, &ks)) return -1; } if (EOF == kputc('\n', &ks)) return -1; } for (k = kh_begin(hdr->h); k != kh_end(hdr->h); k++) { SAM_hdr_type *t1, *t2; if (!kh_exist(hdr->h, k)) continue; if (kh_key(hdr->h, k) == K("HD")) continue; t1 = t2 = kh_val(hdr->h, k); do { SAM_hdr_tag *tag; char c[2]; if (EOF == kputc_('@', &ks)) return -1; c[0] = kh_key(hdr->h, k)>>8; c[1] = kh_key(hdr->h, k)&0xff; if (EOF == kputsn_(c, 2, &ks)) return -1; for (tag = t1->tag; tag; tag=tag->next) { if (EOF == kputc_('\t', &ks)) return -1; if (EOF == kputsn_(tag->str, tag->len, &ks)) return -1; } if (EOF == kputc('\n', &ks)) return -1; t1 = t1->next; } while (t1 != t2); } if (ks_str(&hdr->text)) KS_FREE(&hdr->text); hdr->text = ks; return 0; } /* * Creates an empty SAM header, ready to be populated. * * Returns a SAM_hdr struct on success (free with sam_hdr_free()) * NULL on failure */ SAM_hdr *sam_hdr_new() { SAM_hdr *sh = calloc(1, sizeof(*sh)); if (!sh) return NULL; sh->h = kh_init(sam_hdr); if (!sh->h) goto err; sh->ID_cnt = 1; sh->ref_count = 1; sh->nref = 0; sh->ref = NULL; if (!(sh->ref_hash = kh_init(m_s2i))) goto err; sh->nrg = 0; sh->rg = NULL; if (!(sh->rg_hash = kh_init(m_s2i))) goto err; sh->npg = 0; sh->pg = NULL; sh->npg_end = sh->npg_end_alloc = 0; sh->pg_end = NULL; if (!(sh->pg_hash = kh_init(m_s2i))) goto err; KS_INIT(&sh->text); if (!(sh->tag_pool = pool_create(sizeof(SAM_hdr_tag)))) goto err; if (!(sh->type_pool = pool_create(sizeof(SAM_hdr_type)))) goto err; if (!(sh->str_pool = string_pool_create(8192))) goto err; return sh; err: if (sh->h) kh_destroy(sam_hdr, sh->h); if (sh->tag_pool) pool_destroy(sh->tag_pool); if (sh->type_pool) pool_destroy(sh->type_pool); if (sh->str_pool) string_pool_destroy(sh->str_pool); free(sh); return NULL; } /* * Tokenises a SAM header into a hash table. * Also extracts a few bits on specific data types, such as @RG lines. * * Returns a SAM_hdr struct on success (free with sam_hdr_free()) * NULL on failure */ SAM_hdr *sam_hdr_parse_(const char *hdr, int len) { /* Make an empty SAM_hdr */ SAM_hdr *sh; sh = sam_hdr_new(); if (NULL == sh) return NULL; if (NULL == hdr) return sh; // empty header is permitted /* Parse the header, line by line */ if (-1 == sam_hdr_add_lines(sh, hdr, len)) { sam_hdr_free(sh); return NULL; } /* Obtain sort order */ sh->sort_order = sam_hdr_parse_sort_order(sh); //sam_hdr_dump(sh); //sam_hdr_add(sh, "RG", "ID", "foo", "SM", "bar", NULL); //sam_hdr_rebuild(sh); //printf(">>%s<<", ks_str(sh->text)); //parse_references(sh); //parse_read_groups(sh); sam_hdr_link_pg(sh); //sam_hdr_dump(sh); return sh; } /* * Produces a duplicate copy of hdr and returns it. * Returns NULL on failure */ SAM_hdr *sam_hdr_dup(SAM_hdr *hdr) { if (-1 == sam_hdr_rebuild(hdr)) return NULL; return sam_hdr_parse_(sam_hdr_str(hdr), sam_hdr_length(hdr)); } /*! Increments a reference count on hdr. * * This permits multiple files to share the same header, all calling * sam_hdr_free when done, without causing errors for other open files. */ void sam_hdr_incr_ref(SAM_hdr *hdr) { hdr->ref_count++; } /*! Increments a reference count on hdr. * * This permits multiple files to share the same header, all calling * sam_hdr_free when done, without causing errors for other open files. * * If the reference count hits zero then the header is automatically * freed. This makes it a synonym for sam_hdr_free(). */ void sam_hdr_decr_ref(SAM_hdr *hdr) { sam_hdr_free(hdr); } /*! Deallocates all storage used by a SAM_hdr struct. * * This also decrements the header reference count. If after decrementing * it is still non-zero then the header is assumed to be in use by another * caller and the free is not done. * * This is a synonym for sam_hdr_dec_ref(). */ void sam_hdr_free(SAM_hdr *hdr) { if (!hdr) return; if (--hdr->ref_count > 0) return; if (ks_str(&hdr->text)) KS_FREE(&hdr->text); if (hdr->h) kh_destroy(sam_hdr, hdr->h); if (hdr->ref_hash) kh_destroy(m_s2i, hdr->ref_hash); if (hdr->ref) { int i; for (i = 0; i < hdr->nref; i++) if (hdr->ref[i].name) free(hdr->ref[i].name); free(hdr->ref); } if (hdr->rg_hash) kh_destroy(m_s2i, hdr->rg_hash); if (hdr->rg) { int i; for (i = 0; i < hdr->nrg; i++) if (hdr->rg[i].name) free(hdr->rg[i].name); free(hdr->rg); } if (hdr->pg_hash) kh_destroy(m_s2i, hdr->pg_hash); if (hdr->pg) { int i; for (i = 0; i < hdr->npg; i++) if (hdr->pg[i].name) free(hdr->pg[i].name); free(hdr->pg); } if (hdr->pg_end) free(hdr->pg_end); if (hdr->type_pool) pool_destroy(hdr->type_pool); if (hdr->tag_pool) pool_destroy(hdr->tag_pool); if (hdr->str_pool) string_pool_destroy(hdr->str_pool); free(hdr); } int sam_hdr_length(SAM_hdr *hdr) { return ks_len(&hdr->text); } char *sam_hdr_str(SAM_hdr *hdr) { return ks_str(&hdr->text); } /* * Looks up a reference sequence by name and returns the numerical ID. * Returns -1 if unknown reference. */ int sam_hdr_name2ref(SAM_hdr *hdr, const char *ref) { khint_t k = kh_get(m_s2i, hdr->ref_hash, ref); return k == kh_end(hdr->ref_hash) ? -1 : kh_val(hdr->ref_hash, k); } /* * Looks up a read-group by name and returns a pointer to the start of the * associated tag list. * * Returns NULL on failure */ SAM_RG *sam_hdr_find_rg(SAM_hdr *hdr, const char *rg) { khint_t k = kh_get(m_s2i, hdr->rg_hash, rg); return k == kh_end(hdr->rg_hash) ? NULL : &hdr->rg[kh_val(hdr->rg_hash, k)]; } /* * Fixes any PP links in @PG headers. * If the entries are in order then this doesn't need doing, but incase * our header is out of order this goes through the sh->pg[] array * setting the prev_id field. * * Note we can have multiple complete chains. This code should identify the * tails of these chains as these are the entries we have to link to in * subsequent PP records. * * Returns 0 on sucess * -1 on failure (indicating broken PG/PP records) */ int sam_hdr_link_pg(SAM_hdr *hdr) { int i, j, ret = 0; hdr->npg_end_alloc = hdr->npg; hdr->pg_end = realloc(hdr->pg_end, hdr->npg * sizeof(*hdr->pg_end)); if (!hdr->pg_end) return -1; for (i = 0; i < hdr->npg; i++) hdr->pg_end[i] = i; for (i = 0; i < hdr->npg; i++) { khint_t k; SAM_hdr_tag *tag; char tmp; for (tag = hdr->pg[i].tag; tag; tag = tag->next) { if (tag->str[0] == 'P' && tag->str[1] == 'P') break; } if (!tag) { /* Chain start points */ continue; } tmp = tag->str[tag->len]; tag->str[tag->len] = 0; k = kh_get(m_s2i, hdr->pg_hash, tag->str+3); tag->str[tag->len] = tmp; if (k == kh_end(hdr->pg_hash)) { ret = -1; continue; } hdr->pg[i].prev_id = hdr->pg[kh_val(hdr->pg_hash, k)].id; hdr->pg_end[kh_val(hdr->pg_hash, k)] = -1; } for (i = j = 0; i < hdr->npg; i++) { if (hdr->pg_end[i] != -1) hdr->pg_end[j++] = hdr->pg_end[i]; } hdr->npg_end = j; return ret; } /* * Returns a unique ID from a base name. * * The value returned is valid until the next call to * this function. */ const char *sam_hdr_PG_ID(SAM_hdr *sh, const char *name) { khint_t k = kh_get(m_s2i, sh->pg_hash, name); if (k == kh_end(sh->pg_hash)) return name; do { sprintf(sh->ID_buf, "%.1000s.%d", name, sh->ID_cnt++); k = kh_get(m_s2i, sh->pg_hash, sh->ID_buf); } while (k != kh_end(sh->pg_hash)); return sh->ID_buf; } /* * Add an @PG line. * * If we wish complete control over this use sam_hdr_add() directly. This * function uses that, but attempts to do a lot of tedious house work for * you too. * * - It will generate a suitable ID if the supplied one clashes. * - It will generate multiple @PG records if we have multiple PG chains. * * Call it as per sam_hdr_add() with a series of key,value pairs ending * in NULL. * * Returns 0 on success * -1 on failure */ int sam_hdr_add_PG(SAM_hdr *sh, const char *name, ...) { va_list args; if (sh->npg_end) { /* Copy ends array to avoid us looping while modifying it */ int *end = malloc(sh->npg_end * sizeof(int)); int i, nends = sh->npg_end; if (!end) return -1; memcpy(end, sh->pg_end, nends * sizeof(*end)); for (i = 0; i < nends; i++) { va_start(args, name); if (-1 == sam_hdr_vadd(sh, "PG", args, "ID", sam_hdr_PG_ID(sh, name), "PN", name, "PP", sh->pg[end[i]].name, NULL)) { free(end); return -1; } va_end(args); } free(end); } else { va_start(args, name); if (-1 == sam_hdr_vadd(sh, "PG", args, "ID", sam_hdr_PG_ID(sh, name), "PN", name, NULL)) return -1; va_end(args); } //sam_hdr_dump(sh); return 0; } /* * A function to help with construction of CL tags in @PG records. * Takes an argc, argv pair and returns a single space-separated string. * This string should be deallocated by the calling function. * * Returns malloced char * on success * NULL on failure */ char *stringify_argv(int argc, char *argv[]) { char *str, *cp; size_t nbytes = 1; int i, j; /* Allocate */ for (i = 0; i < argc; i++) { nbytes += strlen(argv[i]) + 1; } if (!(str = malloc(nbytes))) return NULL; /* Copy */ cp = str; for (i = 0; i < argc; i++) { j = 0; while (argv[i][j]) { if (argv[i][j] == '\t') *cp++ = ' '; else *cp++ = argv[i][j]; j++; } *cp++ = ' '; } *cp++ = 0; return str; } stacks-1.44/htslib/cram/._sam_header.h000644 000765 000024 00000000265 12750743246 020404 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/sam_header.h000644 000765 000024 00000033330 12750743246 020166 0ustar00catchenstaff000000 000000 /* Copyright (c) 2013-2014 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /*! \file * SAM header parsing. * * These functions can be shared between SAM, BAM and CRAM file * formats as all three internally use the same string encoding for * header fields. */ /* * TODO. * * - Sort order (parse to struct, enum type, updating funcs) * - Removal of lines. * - Updating of lines */ #ifndef _SAM_HDR_H_ #define _SAM_HDR_H_ #include #include "cram/string_alloc.h" #include "cram/pooled_alloc.h" #include "htslib/khash.h" #include "htslib/kstring.h" #ifdef __cplusplus extern "C" { #endif // For structure assignment. Eg kstring_t s = KS_INITIALIZER; #define KS_INITIALIZER {0,0,0} // For initialisation elsewhere. Eg KS_INIT(x->str); #define KS_INIT(ks) ((ks)->l = 0, (ks)->m = 0, (ks)->s = NULL) // Frees the string subfield only. Assumes 's' itself is static. #define KS_FREE(ks) do { if ((ks)->s) free((ks)->s); } while(0) /* * Proposed new SAM header parsing 1 @SQ ID:foo LN:100 2 @SQ ID:bar LN:200 3 @SQ ID:ram LN:300 UR:xyz 4 @RG ID:r ... 5 @RG ID:s ... Hash table for 2-char @keys without dup entries. If dup lines, we form a circular linked list. Ie hash keys = {RG, SQ}. HASH("SQ")--\ | (3) <-> 1 <-> 2 <-> 3 <-> (1) HASH("RG")--\ | (5) <-> 4 <-> 5 <-> (4) Items stored in the hash values also form their own linked lists: Ie SQ->ID(foo)->LN(100) SQ->ID(bar)->LN(200) SQ->ID(ram)->LN(300)->UR(xyz) RG->ID(r) */ /*! A single key:value pair on a header line * * These form a linked list and hold strings. The strings are * allocated from a string_alloc_t pool referenced in the master * SAM_hdr structure. Do not attempt to free, malloc or manipulate * these strings directly. */ typedef struct SAM_hdr_tag_s { struct SAM_hdr_tag_s *next; char *str; int len; } SAM_hdr_tag; /*! The parsed version of the SAM header string. * * Each header type (SQ, RG, HD, etc) points to its own SAM_hdr_type * struct via the main hash table h in the SAM_hdr struct. * * These in turn consist of circular bi-directional linked lists (ie * rings) to hold the multiple instances of the same header type * code. For example if we have 5 \@SQ lines the primary hash table * will key on \@SQ pointing to the first SAM_hdr_type and that in turn * will be part of a ring of 5 elements. * * For each SAM_hdr_type structure we also point to a SAM_hdr_tag * structure which holds the tokenised attributes; the tab separated * key:value pairs per line. */ typedef struct SAM_hdr_item_s { struct SAM_hdr_item_s *next; // cirular struct SAM_hdr_item_s *prev; SAM_hdr_tag *tag; // first tag int order; // 0 upwards } SAM_hdr_type; /*! Parsed \@SQ lines */ typedef struct { char *name; uint32_t len; SAM_hdr_type *ty; SAM_hdr_tag *tag; } SAM_SQ; /*! Parsed \@RG lines */ typedef struct { char *name; SAM_hdr_type *ty; SAM_hdr_tag *tag; int name_len; int id; // numerical ID } SAM_RG; /*! Parsed \@PG lines */ typedef struct { char *name; SAM_hdr_type *ty; SAM_hdr_tag *tag; int name_len; int id; // numerical ID int prev_id; // -1 if none } SAM_PG; /*! Sort order parsed from @HD line */ enum sam_sort_order { ORDER_UNKNOWN =-1, ORDER_UNSORTED = 0, ORDER_NAME = 1, ORDER_COORD = 2, //ORDER_COLLATE = 3 // maybe one day! }; KHASH_MAP_INIT_INT(sam_hdr, SAM_hdr_type*) KHASH_MAP_INIT_STR(m_s2i, int) /*! Primary structure for header manipulation * * The initial header text is held in the text kstring_t, but is also * parsed out into SQ, RG and PG arrays. These have a hash table * associated with each to allow lookup by ID or SN fields instead of * their numeric array indices. Additionally PG has an array to hold * the linked list start points (the last in a PP chain). * * Use the appropriate sam_hdr_* functions to edit the header, and * call sam_hdr_rebuild() any time the textual form needs to be * updated again. */ typedef struct { kstring_t text; //!< concatenated text, indexed by SAM_hdr_tag khash_t(sam_hdr) *h; string_alloc_t *str_pool; //!< Pool of SAM_hdr_tag->str strings pool_alloc_t *type_pool;//!< Pool of SAM_hdr_type structs pool_alloc_t *tag_pool; //!< Pool of SAM_hdr_tag structs // @SQ lines / references int nref; //!< Number of \@SQ lines SAM_SQ *ref; //!< Array of parsed \@SQ lines khash_t(m_s2i) *ref_hash; //!< Maps SQ SN field to sq[] index // @RG lines / read-groups int nrg; //!< Number of \@RG lines SAM_RG *rg; //!< Array of parsed \@RG lines khash_t(m_s2i) *rg_hash; //!< Maps RG ID field to rg[] index // @PG lines / programs int npg; //!< Number of \@PG lines int npg_end; //!< Number of terminating \@PG lines int npg_end_alloc; //!< Size of pg_end field SAM_PG *pg; //!< Array of parsed \@PG lines khash_t(m_s2i) *pg_hash; //!< Maps PG ID field to pg[] index int *pg_end; //!< \@PG chain termination IDs // @HD data enum sam_sort_order sort_order; //!< @HD SO: field // @cond internal char ID_buf[1024]; // temporary buffer int ID_cnt; int ref_count; // number of uses of this SAM_hdr // @endcond } SAM_hdr; /*! Creates an empty SAM header, ready to be populated. * * @return * Returns a SAM_hdr struct on success (free with sam_hdr_free()) * NULL on failure */ SAM_hdr *sam_hdr_new(void); /*! Tokenises a SAM header into a hash table. * * Also extracts a few bits on specific data types, such as @RG lines. * * @return * Returns a SAM_hdr struct on success (free with sam_hdr_free()); * NULL on failure */ SAM_hdr *sam_hdr_parse_(const char *hdr, int len); /*! Produces a duplicate copy of hdr and returns it. * @return * Returns NULL on failure */ SAM_hdr *sam_hdr_dup(SAM_hdr *hdr); /*! Increments a reference count on hdr. * * This permits multiple files to share the same header, all calling * sam_hdr_free when done, without causing errors for other open files. */ void sam_hdr_incr_ref(SAM_hdr *hdr); /*! Increments a reference count on hdr. * * This permits multiple files to share the same header, all calling * sam_hdr_free when done, without causing errors for other open files. * * If the reference count hits zero then the header is automatically * freed. This makes it a synonym for sam_hdr_free(). */ void sam_hdr_decr_ref(SAM_hdr *hdr); /*! Deallocates all storage used by a SAM_hdr struct. * * This also decrements the header reference count. If after decrementing * it is still non-zero then the header is assumed to be in use by another * caller and the free is not done. * * This is a synonym for sam_hdr_dec_ref(). */ void sam_hdr_free(SAM_hdr *hdr); /*! Returns the current length of the SAM_hdr in text form. * * Call sam_hdr_rebuild() first if editing has taken place. */ int sam_hdr_length(SAM_hdr *hdr); /*! Returns the string form of the SAM_hdr. * * Call sam_hdr_rebuild() first if editing has taken place. */ char *sam_hdr_str(SAM_hdr *hdr); /*! Appends a formatted line to an existing SAM header. * * Line is a full SAM header record, eg "@SQ\tSN:foo\tLN:100", with * optional new-line. If it contains more than 1 line then multiple lines * will be added in order. * * Input text is of maximum length len or as terminated earlier by a NUL. * Len may be 0 if unknown, in which case lines must be NUL-terminated. * * @return * Returns 0 on success; * -1 on failure */ int sam_hdr_add_lines(SAM_hdr *sh, const char *lines, int len); /*! Adds a single line to a SAM header. * * Specify type and one or more key,value pairs, ending with the NULL key. * Eg. sam_hdr_add(h, "SQ", "ID", "foo", "LN", "100", NULL). * * @return * Returns 0 on success; * -1 on failure */ int sam_hdr_add(SAM_hdr *sh, const char *type, ...); /*! Adds a single line to a SAM header. * * This is much like sam_hdr_add() but with the additional va_list * argument. This is followed by specifying type and one or more * key,value pairs, ending with the NULL key. * * Eg. sam_hdr_vadd(h, "SQ", args, "ID", "foo", "LN", "100", NULL). * * The purpose of the additional va_list parameter is to permit other * varargs functions to call this while including their own additional * parameters; an example is in sam_hdr_add_PG(). * * Note: this function invokes va_arg at least once, making the value * of ap indeterminate after the return. The caller should call * va_start/va_end before/after calling this function or use va_copy. * * @return * Returns 0 on success; * -1 on failure */ int sam_hdr_vadd(SAM_hdr *sh, const char *type, va_list ap, ...); /*! * @return * Returns the first header item matching 'type'. If ID is non-NULL it checks * for the tag ID: and compares against the specified ID. * * Returns NULL if no type/ID is found */ SAM_hdr_type *sam_hdr_find(SAM_hdr *hdr, char *type, char *ID_key, char *ID_value); /*! * * As per SAM_hdr_type, but returns a complete line of formatted text * for a specific head type/ID combination. If ID is NULL then it returns * the first line of the specified type. * * The returned string is malloced and should be freed by the calling * function with free(). * * @return * Returns NULL if no type/ID is found. */ char *sam_hdr_find_line(SAM_hdr *hdr, char *type, char *ID_key, char *ID_value); /*! Looks for a specific key in a single sam header line. * * If prev is non-NULL it also fills this out with the previous tag, to * permit use in key removal. *prev is set to NULL when the tag is the first * key in the list. When a tag isn't found, prev (if non NULL) will be the last * tag in the existing list. * * @return * Returns the tag pointer on success; * NULL on failure */ SAM_hdr_tag *sam_hdr_find_key(SAM_hdr *sh, SAM_hdr_type *type, char *key, SAM_hdr_tag **prev); /*! Adds or updates tag key,value pairs in a header line. * * Eg for adding M5 tags to @SQ lines or updating sort order for the * @HD line (although use the sam_hdr_sort_order() function for * HD manipulation, which is a wrapper around this funuction). * * Specify multiple key,value pairs ending in NULL. * * @return * Returns 0 on success; * -1 on failure */ int sam_hdr_update(SAM_hdr *hdr, SAM_hdr_type *type, ...); /*! Returns the sort order from the @HD SO: field */ enum sam_sort_order sam_hdr_sort_order(SAM_hdr *hdr); /*! Reconstructs the kstring from the header hash table. * @return * Returns 0 on success; * -1 on failure */ int sam_hdr_rebuild(SAM_hdr *hdr); /*! Looks up a reference sequence by name and returns the numerical ID. * @return * Returns -1 if unknown reference. */ int sam_hdr_name2ref(SAM_hdr *hdr, const char *ref); /*! Looks up a read-group by name and returns a pointer to the start of the * associated tag list. * * @return * Returns NULL on failure */ SAM_RG *sam_hdr_find_rg(SAM_hdr *hdr, const char *rg); /*! Fixes any PP links in @PG headers. * * If the entries are in order then this doesn't need doing, but incase * our header is out of order this goes through the sh->pg[] array * setting the prev_id field. * * @return * Returns 0 on sucess; * -1 on failure (indicating broken PG/PP records) */ int sam_hdr_link_pg(SAM_hdr *hdr); /*! Add an @PG line. * * If we wish complete control over this use sam_hdr_add() directly. This * function uses that, but attempts to do a lot of tedious house work for * you too. * * - It will generate a suitable ID if the supplied one clashes. * - It will generate multiple @PG records if we have multiple PG chains. * * Call it as per sam_hdr_add() with a series of key,value pairs ending * in NULL. * * @return * Returns 0 on success; * -1 on failure */ int sam_hdr_add_PG(SAM_hdr *sh, const char *name, ...); /*! * A function to help with construction of CL tags in @PG records. * Takes an argc, argv pair and returns a single space-separated string. * This string should be deallocated by the calling function. * * @return * Returns malloced char * on success; * NULL on failure */ char *stringify_argv(int argc, char *argv[]); #ifdef __cplusplus } #endif #endif /* _SAM_HDR_H_ */ stacks-1.44/htslib/cram/._string_alloc.c000644 000765 000024 00000000265 12750743246 020767 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/string_alloc.c000644 000765 000024 00000010115 12750743246 020545 0ustar00catchenstaff000000 000000 /* Copyright (c) 2010 Genome Research Ltd. Author: Andrew Whitwham Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /* A pooled string allocator intended to cut down on the memory overhead of many small string allocations. Andrew Whitwham, September 2010. */ #include #include #include #include #include "string_alloc.h" #define MIN_STR_SIZE 1024 /* creates the string pool. max_length is the initial size a single string can be. Tha max_length can grow as needed */ string_alloc_t *string_pool_create(size_t max_length) { string_alloc_t *a_str; if (NULL == (a_str = (string_alloc_t *)malloc(sizeof(*a_str)))) { return NULL; } if (max_length < MIN_STR_SIZE) max_length = MIN_STR_SIZE; a_str->nstrings = 0; a_str->max_length = max_length; a_str->strings = NULL; return a_str; } /* internal function to do the actual memory allocation */ static string_t *new_string_pool(string_alloc_t *a_str) { string_t *str; str = realloc(a_str->strings, (a_str->nstrings + 1) * sizeof(*a_str->strings)); if (NULL == str) return NULL; a_str->strings = str; str = &a_str->strings[a_str->nstrings]; str->str = malloc(a_str->max_length);; if (NULL == str->str) return NULL; str->used = 0; a_str->nstrings++; return str; } /* free allocated memory */ void string_pool_destroy(string_alloc_t *a_str) { size_t i; for (i = 0; i < a_str->nstrings; i++) { free(a_str->strings[i].str); } free(a_str->strings); free(a_str); } /* allocate space for a string */ char *string_alloc(string_alloc_t *a_str, size_t length) { string_t *str; char *ret; if (length <= 0) return NULL; // add to last string pool if we have space if (a_str->nstrings) { str = &a_str->strings[a_str->nstrings - 1]; if (str->used + length < a_str->max_length) { ret = str->str + str->used; str->used += length; return ret; } } // increase the max length if needs be if (length > a_str->max_length) a_str->max_length = length; // need a new string pool str = new_string_pool(a_str); if (NULL == str) return NULL; str->used = length; return str->str; } /* equivalent to strdup */ char *string_dup(string_alloc_t *a_str, char *instr) { return string_ndup(a_str, instr, strlen(instr)); } char *string_ndup(string_alloc_t *a_str, char *instr, size_t len) { char *str = string_alloc(a_str, len + 1); if (NULL == str) return NULL; strncpy(str, instr, len); str[len] = 0; return str; } stacks-1.44/htslib/cram/._string_alloc.h000644 000765 000024 00000000265 12750743246 020774 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/string_alloc.h000644 000765 000024 00000004420 12750743246 020554 0ustar00catchenstaff000000 000000 /* Copyright (c) 2010 Genome Research Ltd. Author: Andrew Whitwham Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #ifndef _STRING_ALLOC_H_ #define _STRING_ALLOC_H_ #include #ifdef __cplusplus extern "C" { #endif /* * A pooled string allocator intended to cut down on the * memory overhead of many small string allocations. * * Andrew Whitwham, September 2010. */ typedef struct { char *str; size_t used; } string_t; typedef struct { size_t max_length; size_t nstrings; string_t *strings; } string_alloc_t; string_alloc_t *string_pool_create(size_t max_length); void string_pool_destroy(string_alloc_t *a_str); char *string_alloc(string_alloc_t *a_str, size_t length); char *string_dup(string_alloc_t *a_str, char *instr); char *string_ndup(string_alloc_t *a_str, char *instr, size_t len); #ifdef __cplusplus } #endif #endif stacks-1.44/htslib/cram/._thread_pool.c000644 000765 000024 00000000265 12750743246 020607 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/thread_pool.c000644 000765 000024 00000043020 12750743246 020366 0ustar00catchenstaff000000 000000 /* Copyright (c) 2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #include #include #include #include #include #include #include #include #include "cram/thread_pool.h" //#define DEBUG //#define DEBUG_TIME #define IN_ORDER #ifdef DEBUG static int worker_id(t_pool *p) { int i; pthread_t s = pthread_self(); for (i = 0; i < p->tsize; i++) { if (pthread_equal(s, p->t[i].tid)) return i; } return -1; } #endif /* ---------------------------------------------------------------------------- * A queue to hold results from the thread pool. * * Each thread pool may have jobs of multiple types being queued up and * interleaved, so we allow several results queue per pool. * * The jobs themselves are expected to push their results onto their * appropriate results queue. */ /* * Adds a result to the end of the result queue. * * Returns 0 on success; * -1 on failure */ static int t_pool_add_result(t_pool_job *j, void *data) { t_results_queue *q = j->q; t_pool_result *r; #ifdef DEBUG fprintf(stderr, "%d: Adding resulting to queue %p, serial %d\n", worker_id(j->p), q, j->serial); #endif /* No results queue is fine if we don't want any results back */ if (!q) return 0; if (!(r = malloc(sizeof(*r)))) return -1; r->next = NULL; r->data = data; r->serial = j->serial; pthread_mutex_lock(&q->result_m); if (q->result_tail) { q->result_tail->next = r; q->result_tail = r; } else { q->result_head = q->result_tail = r; } q->queue_len++; q->pending--; #ifdef DEBUG fprintf(stderr, "%d: Broadcasting result_avail (id %d)\n", worker_id(j->p), r->serial); #endif pthread_cond_signal(&q->result_avail_c); #ifdef DEBUG fprintf(stderr, "%d: Broadcast complete\n", worker_id(j->p)); #endif pthread_mutex_unlock(&q->result_m); return 0; } /* Core of t_pool_next_result() */ static t_pool_result *t_pool_next_result_locked(t_results_queue *q) { t_pool_result *r, *last; for (last = NULL, r = q->result_head; r; last = r, r = r->next) { if (r->serial == q->next_serial) break; } if (r) { if (q->result_head == r) q->result_head = r->next; else last->next = r->next; if (q->result_tail == r) q->result_tail = last; if (!q->result_head) q->result_tail = NULL; q->next_serial++; q->queue_len--; } return r; } /* * Pulls a result off the head of the result queue. Caller should * free it (and any internals as appropriate) after use. This doesn't * wait for a result to be present. * * Results will be returned in strict order. * * Returns t_pool_result pointer if a result is ready. * NULL if not. */ t_pool_result *t_pool_next_result(t_results_queue *q) { t_pool_result *r; #ifdef DEBUG fprintf(stderr, "Requesting next result on queue %p\n", q); #endif pthread_mutex_lock(&q->result_m); r = t_pool_next_result_locked(q); pthread_mutex_unlock(&q->result_m); #ifdef DEBUG fprintf(stderr, "(q=%p) Found %p\n", q, r); #endif return r; } t_pool_result *t_pool_next_result_wait(t_results_queue *q) { t_pool_result *r; #ifdef DEBUG fprintf(stderr, "Waiting for result %d...\n", q->next_serial); #endif pthread_mutex_lock(&q->result_m); while (!(r = t_pool_next_result_locked(q))) { /* Possible race here now avoided via _locked() call, but incase... */ struct timeval now; struct timespec timeout; gettimeofday(&now, NULL); timeout.tv_sec = now.tv_sec + 10; timeout.tv_nsec = now.tv_usec * 1000; pthread_cond_timedwait(&q->result_avail_c, &q->result_m, &timeout); } pthread_mutex_unlock(&q->result_m); return r; } /* * Returns true if there are no items on the finished results queue and * also none still pending. */ int t_pool_results_queue_empty(t_results_queue *q) { int empty; pthread_mutex_lock(&q->result_m); empty = q->queue_len == 0 && q->pending == 0; pthread_mutex_unlock(&q->result_m); return empty; } /* * Returns the number of completed jobs on the results queue. */ int t_pool_results_queue_len(t_results_queue *q) { int len; pthread_mutex_lock(&q->result_m); len = q->queue_len; pthread_mutex_unlock(&q->result_m); return len; } int t_pool_results_queue_sz(t_results_queue *q) { int len; pthread_mutex_lock(&q->result_m); len = q->queue_len + q->pending; pthread_mutex_unlock(&q->result_m); return len; } /* * Frees a result 'r' and if free_data is true also frees * the internal r->data result too. */ void t_pool_delete_result(t_pool_result *r, int free_data) { if (!r) return; if (free_data && r->data) free(r->data); free(r); } /* * Initialises a results queue. * * Results queue pointer on success; * NULL on failure */ t_results_queue *t_results_queue_init(void) { t_results_queue *q = malloc(sizeof(*q)); pthread_mutex_init(&q->result_m, NULL); pthread_cond_init(&q->result_avail_c, NULL); q->result_head = NULL; q->result_tail = NULL; q->next_serial = 0; q->curr_serial = 0; q->queue_len = 0; q->pending = 0; return q; } /* Deallocates memory for a results queue */ void t_results_queue_destroy(t_results_queue *q) { #ifdef DEBUG fprintf(stderr, "Destroying results queue %p\n", q); #endif if (!q) return; pthread_mutex_destroy(&q->result_m); pthread_cond_destroy(&q->result_avail_c); memset(q, 0xbb, sizeof(*q)); free(q); #ifdef DEBUG fprintf(stderr, "Destroyed results queue %p\n", q); #endif } /* ---------------------------------------------------------------------------- * The thread pool. */ #define TDIFF(t2,t1) ((t2.tv_sec-t1.tv_sec)*1000000 + t2.tv_usec-t1.tv_usec) /* * A worker thread. * * Each thread waits for the pool to be non-empty. * As soon as this applies, one of them succeeds in getting the lock * and then executes the job. */ static void *t_pool_worker(void *arg) { t_pool_worker_t *w = (t_pool_worker_t *)arg; t_pool *p = w->p; t_pool_job *j; #ifdef DEBUG_TIME struct timeval t1, t2, t3; #endif for (;;) { // Pop an item off the pool queue #ifdef DEBUG_TIME gettimeofday(&t1, NULL); #endif pthread_mutex_lock(&p->pool_m); #ifdef DEBUG_TIME gettimeofday(&t2, NULL); p->wait_time += TDIFF(t2,t1); w->wait_time += TDIFF(t2,t1); #endif // If there is something on the job list and a higher priority // thread waiting, let it handle this instead. // while (p->head && p->t_stack_top != -1 && p->t_stack_top < w->idx) { // pthread_mutex_unlock(&p->pool_m); // pthread_cond_signal(&p->t[p->t_stack_top].pending_c); // pthread_mutex_lock(&p->pool_m); // } while (!p->head && !p->shutdown) { p->nwaiting++; if (p->njobs == 0) pthread_cond_signal(&p->empty_c); #ifdef DEBUG_TIME gettimeofday(&t2, NULL); #endif #ifdef IN_ORDER // Push this thread to the top of the waiting stack if (p->t_stack_top == -1 || p->t_stack_top > w->idx) p->t_stack_top = w->idx; p->t_stack[w->idx] = 1; pthread_cond_wait(&w->pending_c, &p->pool_m); p->t_stack[w->idx] = 0; /* Find new t_stack_top */ { int i; p->t_stack_top = -1; for (i = 0; i < p->tsize; i++) { if (p->t_stack[i]) { p->t_stack_top = i; break; } } } #else pthread_cond_wait(&p->pending_c, &p->pool_m); #endif #ifdef DEBUG_TIME gettimeofday(&t3, NULL); p->wait_time += TDIFF(t3,t2); w->wait_time += TDIFF(t3,t2); #endif p->nwaiting--; } if (p->shutdown) { #ifdef DEBUG_TIME p->total_time += TDIFF(t3,t1); #endif #ifdef DEBUG fprintf(stderr, "%d: Shutting down\n", worker_id(p)); #endif pthread_mutex_unlock(&p->pool_m); pthread_exit(NULL); } j = p->head; if (!(p->head = j->next)) p->tail = NULL; if (p->njobs-- >= p->qsize) pthread_cond_signal(&p->full_c); if (p->njobs == 0) pthread_cond_signal(&p->empty_c); pthread_mutex_unlock(&p->pool_m); // We have job 'j' - now execute it. t_pool_add_result(j, j->func(j->arg)); #ifdef DEBUG_TIME pthread_mutex_lock(&p->pool_m); gettimeofday(&t3, NULL); p->total_time += TDIFF(t3,t1); pthread_mutex_unlock(&p->pool_m); #endif memset(j, 0xbb, sizeof(*j)); free(j); } return NULL; } /* * Creates a worker pool of length qsize with tsize worker threads. * * Returns pool pointer on success; * NULL on failure */ t_pool *t_pool_init(int qsize, int tsize) { int i; t_pool *p = malloc(sizeof(*p)); p->qsize = qsize; p->tsize = tsize; p->njobs = 0; p->nwaiting = 0; p->shutdown = 0; p->head = p->tail = NULL; p->t_stack = NULL; #ifdef DEBUG_TIME p->total_time = p->wait_time = 0; #endif p->t = malloc(tsize * sizeof(p->t[0])); pthread_mutex_init(&p->pool_m, NULL); pthread_cond_init(&p->empty_c, NULL); pthread_cond_init(&p->full_c, NULL); pthread_mutex_lock(&p->pool_m); #ifdef IN_ORDER if (!(p->t_stack = malloc(tsize * sizeof(*p->t_stack)))) return NULL; p->t_stack_top = -1; for (i = 0; i < tsize; i++) { t_pool_worker_t *w = &p->t[i]; p->t_stack[i] = 0; w->p = p; w->idx = i; w->wait_time = 0; pthread_cond_init(&w->pending_c, NULL); if (0 != pthread_create(&w->tid, NULL, t_pool_worker, w)) return NULL; } #else pthread_cond_init(&p->pending_c, NULL); for (i = 0; i < tsize; i++) { t_pool_worker_t *w = &p->t[i]; w->p = p; w->idx = i; pthread_cond_init(&w->pending_c, NULL); if (0 != pthread_create(&w->tid, NULL, t_pool_worker, w)) return NULL; } #endif pthread_mutex_unlock(&p->pool_m); return p; } /* * Adds an item to the work pool. * * FIXME: Maybe return 1,0,-1 and distinguish between job dispathed vs * result returned. Ie rather than blocking on full queue we're permitted * to return early on "result available" event too. * Caller would then have a while loop around t_pool_dispatch. * Or, return -1 and set errno to EAGAIN to indicate job not yet submitted. * * Returns 0 on success * -1 on failure */ int t_pool_dispatch(t_pool *p, t_results_queue *q, void *(*func)(void *arg), void *arg) { return t_pool_dispatch2(p, q, func, arg, 0); } /* * As above but optional non-block flag. * * nonblock 0 => block if input queue is full * nonblock +1 => don't block if input queue is full, but do not add task * nonblock -1 => add task regardless of whether queue is full (over-size) */ int t_pool_dispatch2(t_pool *p, t_results_queue *q, void *(*func)(void *arg), void *arg, int nonblock) { t_pool_job *j; #ifdef DEBUG fprintf(stderr, "Dispatching job for queue %p, serial %d\n", q, q->curr_serial); #endif pthread_mutex_lock(&p->pool_m); if (p->njobs >= p->qsize && nonblock == 1) { pthread_mutex_unlock(&p->pool_m); errno = EAGAIN; return -1; } if (!(j = malloc(sizeof(*j)))) return -1; j->func = func; j->arg = arg; j->next = NULL; j->p = p; j->q = q; if (q) { pthread_mutex_lock(&q->result_m); j->serial = q->curr_serial++; q->pending++; pthread_mutex_unlock(&q->result_m); } else { j->serial = 0; } // Check if queue is full if (nonblock == 0) while (p->njobs >= p->qsize) pthread_cond_wait(&p->full_c, &p->pool_m); p->njobs++; // if (q->curr_serial % 100 == 0) // fprintf(stderr, "p->njobs = %d p->qsize = %d\n", p->njobs, p->qsize); if (p->tail) { p->tail->next = j; p->tail = j; } else { p->head = p->tail = j; } #ifdef DEBUG fprintf(stderr, "Dispatched (serial %d)\n", j->serial); #endif // Let a worker know we have data. #ifdef IN_ORDER // Keep incoming queue at 1 per running thread, so there is always // something waiting when they end their current task. If we go above // this signal to start more threads (if available). This has the effect // of concentrating jobs to fewer cores when we are I/O bound, which in // turn benefits systems with auto CPU frequency scaling. if (p->t_stack_top >= 0 && p->njobs > p->tsize - p->nwaiting) pthread_cond_signal(&p->t[p->t_stack_top].pending_c); #else pthread_cond_signal(&p->pending_c); #endif pthread_mutex_unlock(&p->pool_m); return 0; } /* * Flushes the pool, but doesn't exit. This simply drains the queue and * ensures all worker threads have finished their current task. * * Returns 0 on success; * -1 on failure */ int t_pool_flush(t_pool *p) { int i; #ifdef DEBUG fprintf(stderr, "Flushing pool %p\n", p); #endif // Drains the queue pthread_mutex_lock(&p->pool_m); // Wake up everything for the final sprint! for (i = 0; i < p->tsize; i++) if (p->t_stack[i]) pthread_cond_signal(&p->t[i].pending_c); while (p->njobs || p->nwaiting != p->tsize) pthread_cond_wait(&p->empty_c, &p->pool_m); pthread_mutex_unlock(&p->pool_m); #ifdef DEBUG fprintf(stderr, "Flushed complete for pool %p, njobs=%d, nwaiting=%d\n", p, p->njobs, p->nwaiting); #endif return 0; } /* * Destroys a thread pool. If 'kill' is true the threads are terminated now, * otherwise they are joined into the main thread so they will finish their * current work load. * * Use t_pool_destroy(p,0) after a t_pool_flush(p) on a normal shutdown or * t_pool_destroy(p,1) to quickly exit after a fatal error. */ void t_pool_destroy(t_pool *p, int kill) { int i; #ifdef DEBUG fprintf(stderr, "Destroying pool %p, kill=%d\n", p, kill); #endif /* Send shutdown message to worker threads */ if (!kill) { pthread_mutex_lock(&p->pool_m); p->shutdown = 1; #ifdef DEBUG fprintf(stderr, "Sending shutdown request\n"); #endif #ifdef IN_ORDER for (i = 0; i < p->tsize; i++) pthread_cond_signal(&p->t[i].pending_c); #else pthread_cond_broadcast(&p->pending_c); #endif pthread_mutex_unlock(&p->pool_m); #ifdef DEBUG fprintf(stderr, "Shutdown complete\n"); #endif for (i = 0; i < p->tsize; i++) pthread_join(p->t[i].tid, NULL); } else { for (i = 0; i < p->tsize; i++) pthread_kill(p->t[i].tid, SIGINT); } pthread_mutex_destroy(&p->pool_m); pthread_cond_destroy(&p->empty_c); pthread_cond_destroy(&p->full_c); #ifdef IN_ORDER for (i = 0; i < p->tsize; i++) pthread_cond_destroy(&p->t[i].pending_c); #else pthread_cond_destroy(&p->pending_c); #endif #ifdef DEBUG_TIME fprintf(stderr, "Total time=%f\n", p->total_time / 1000000.0); fprintf(stderr, "Wait time=%f\n", p->wait_time / 1000000.0); fprintf(stderr, "%d%% utilisation\n", (int)(100 - ((100.0 * p->wait_time) / p->total_time + 0.5))); for (i = 0; i < p->tsize; i++) fprintf(stderr, "%d: Wait time=%f\n", i, p->t[i].wait_time / 1000000.0); #endif if (p->t_stack) free(p->t_stack); free(p->t); free(p); #ifdef DEBUG fprintf(stderr, "Destroyed pool %p\n", p); #endif } /*----------------------------------------------------------------------------- * Test app. */ #ifdef TEST_MAIN #include #include void *doit(void *arg) { int i, k, x = 0; int job = *(int *)arg; int *res; printf("Worker: execute job %d\n", job); usleep(random() % 1000000); // to coerce job completion out of order if (0) { for (k = 0; k < 100; k++) { for (i = 0; i < 100000; i++) { x++; x += x * sin(i); x += x * cos(x); } } x *= 100; x += job; } else { x = job*job; } printf("Worker: job %d terminating, x=%d\n", job, x); free(arg); res = malloc(sizeof(*res)); *res = x; return res; } #define NTHREADS 8 int main(int argc, char **argv) { t_pool *p = t_pool_init(NTHREADS*2, NTHREADS); t_results_queue *q = t_results_queue_init(); int i; t_pool_result *r; // Dispatch jobs for (i = 0; i < 20; i++) { int *ip = malloc(sizeof(*ip)); *ip = i; printf("Submitting %d\n", i); t_pool_dispatch(p, q, doit, ip); // Check for results if ((r = t_pool_next_result(q))) { printf("RESULT: %d\n", *(int *)r->data); t_pool_delete_result(r, 1); } } t_pool_flush(p); while ((r = t_pool_next_result(q))) { printf("RESULT: %d\n", *(int *)r->data); t_pool_delete_result(r, 1); } t_pool_destroy(p, 0); t_results_queue_destroy(q); return 0; } #endif stacks-1.44/htslib/cram/._thread_pool.h000644 000765 000024 00000000265 12750743246 020614 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/thread_pool.h000644 000765 000024 00000015002 12750743246 020372 0ustar00catchenstaff000000 000000 /* Copyright (c) 2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /* * This file implements a thread pool for multi-threading applications. * It consists of two distinct interfaces: thread pools an results queues. * * The pool of threads is given a function pointer and void* data to pass in. * This means the pool can run jobs of multiple types, albeit first come * first served with no job scheduling. * * Upon completion, the return value from the function pointer is added to * a results queue. We may have multiple queues in use for the one pool. * * An example: reading from BAM and writing to CRAM with 10 threads. We'll * have a pool of 10 threads and two results queues holding decoded BAM blocks * and encoded CRAM blocks respectively. */ #ifndef _THREAD_POOL_H_ #define _THREAD_POOL_H_ #include #ifdef __cplusplus extern "C" { #endif struct t_pool; struct t_results_queue; typedef struct t_pool_job { void *(*func)(void *arg); void *arg; struct t_pool_job *next; struct t_pool *p; struct t_results_queue *q; int serial; } t_pool_job; typedef struct t_res { struct t_res *next; int serial; // sequential number for ordering void *data; // result itself } t_pool_result; struct t_pool; typedef struct { struct t_pool *p; int idx; pthread_t tid; pthread_cond_t pending_c; long long wait_time; } t_pool_worker_t; typedef struct t_pool { int qsize; // size of queue int njobs; // pending job count int nwaiting; // how many workers waiting for new jobs int shutdown; // true if pool is being destroyed // queue of pending jobs t_pool_job *head, *tail; // threads int tsize; // maximum number of jobs t_pool_worker_t *t; // Mutexes pthread_mutex_t pool_m; // used when updating head/tail pthread_cond_t empty_c; pthread_cond_t pending_c; // not empty pthread_cond_t full_c; // array of worker IDs free int *t_stack, t_stack_top; // Debugging to check wait time long long total_time, wait_time; } t_pool; typedef struct t_results_queue { t_pool_result *result_head; t_pool_result *result_tail; int next_serial; int curr_serial; int queue_len; // number of items in queue int pending; // number of pending items (in progress or in pool list) pthread_mutex_t result_m; pthread_cond_t result_avail_c; } t_results_queue; /* * Creates a worker pool of length qsize with tsize worker threads. * * Returns pool pointer on success; * NULL on failure */ t_pool *t_pool_init(int qsize, int tsize); /* * Adds an item to the work pool. * * FIXME: Maybe return 1,0,-1 and distinguish between job dispathed vs * result returned. Ie rather than blocking on full queue we're permitted * to return early on "result available" event too. * Caller would then have a while loop around t_pool_dispatch. * Or, return -1 and set errno to E_AGAIN to indicate job not yet submitted. * * Returns 0 on success * -1 on failure */ int t_pool_dispatch(t_pool *p, t_results_queue *q, void *(*func)(void *arg), void *arg); int t_pool_dispatch2(t_pool *p, t_results_queue *q, void *(*func)(void *arg), void *arg, int nonblock); /* * Flushes the pool, but doesn't exit. This simply drains the queue and * ensures all worker threads have finished their current task. * * Returns 0 on success; * -1 on failure */ int t_pool_flush(t_pool *p); /* * Destroys a thread pool. If 'kill' is true the threads are terminated now, * otherwise they are joined into the main thread so they will finish their * current work load. * * Use t_pool_destroy(p,0) after a t_pool_flush(p) on a normal shutdown or * t_pool_destroy(p,1) to quickly exit after a fatal error. */ void t_pool_destroy(t_pool *p, int kill); /* * Pulls a result off the head of the result queue. Caller should * free it (and any internals as appropriate) after use. This doesn't * wait for a result to be present. * * Results will be returned in strict order. * * Returns t_pool_result pointer if a result is ready. * NULL if not. */ t_pool_result *t_pool_next_result(t_results_queue *q); t_pool_result *t_pool_next_result_wait(t_results_queue *q); /* * Frees a result 'r' and if free_data is true also frees * the internal r->data result too. */ void t_pool_delete_result(t_pool_result *r, int free_data); /* * Initialises a results queue. * * Results queue pointer on success; * NULL on failure */ t_results_queue *t_results_queue_init(void); /* Deallocates memory for a results queue */ void t_results_queue_destroy(t_results_queue *q); /* * Returns true if there are no items on the finished results queue and * also none still pending. */ int t_pool_results_queue_empty(t_results_queue *q); /* * Returns the number of completed jobs on the results queue. */ int t_pool_results_queue_len(t_results_queue *q); /* * Returns the number of completed jobs plus the number queued up to run. */ int t_pool_results_queue_sz(t_results_queue *q); #ifdef __cplusplus } #endif #endif /* _THREAD_POOL_H_ */ stacks-1.44/htslib/cram/._vlen.c000644 000765 000024 00000000265 12750743246 017253 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/vlen.c000644 000765 000024 00000027662 12750743246 017050 0ustar00catchenstaff000000 000000 /* Author: James Bonfield (jkb@sanger.ac.uk) Copyright (c) 1995-1996 MEDICAL RESEARCH COUNCIL All rights reserved Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1 Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2 Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3 Neither the name of the MEDICAL RESEARCH COUNCIL, THE LABORATORY OF MOLECULAR BIOLOGY nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /* Copyright (c) 2004, 2009, 2011-2012 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #include #include #include #include #include #include #include "cram/vlen.h" #include "cram/os.h" #ifndef MAX #define MAX(a,b) ((a)>(b)?(a):(b)) #endif #ifndef ABS #define ABS(a) ((a)>0?(a):-(a)) #endif /* #define DEBUG_printf(a,n) printf(a,n) */ #define DEBUG_printf(a,n) /* * vlen: 27/10/95 written by James Bonfield, jkb@mrc-lmb.cam.ac.uk * * Given sprintf style of arguments this routine returns the maximum * size of buffer needed to allocate to use with sprintf. It errs on * the side of caution by being simplistic in its approach: we assume * all numbers are of maximum length. * * Handles the usual type conversions (%[%diuaxXcfeEgGpns]), but not * the 'wide' character conversions (%C and %S). * Precision is handled in the correct formats, including %*.* * notations. * Additionally, some of the more dubious (but probably illegal) cases * are supported (eg "%10%" will expand to " %" on many * systems). * * We also assume that the largest integer and larger pointer are 64 * bits, which at least covers the machines we'll need it for. */ int flen(char *fmt, ...) { va_list args; va_start(args, fmt); return vflen(fmt, args); } int vflen(char *fmt, va_list ap) { int len = 0; char *cp, c; long long l; int i; double d; /* * This code modifies 'ap', but we do not know if va_list is a structure * or a pointer to an array so we do not know if it is a local variable * or not. * C99 gets around this by defining va_copy() to make copies of ap, but * this does not exist on all systems. * For now, I just assume that when va_list is a pointer the system also * provides a va_copy macro to work around this problem. The only system * I have seen needing this so far was Linux on AMD64. */ #if defined(HAVE_VA_COPY) va_list ap_local; va_copy(ap_local, ap); # define ap ap_local #endif for(cp = fmt; *cp; cp++) { switch(*cp) { /* A format specifier */ case '%': { char *endp; long conv_len1=0, conv_len2=0, conv_len=0; signed int arg_size; /* Firstly, strip the modifier flags (+-#0 and [space]) */ for(; (c=*++cp);) { if ('#' == c) len+=2; /* Worst case of "0x" */ else if ('-' == c || '+' == c || ' ' == c) len++; else break; } /* Width specifier */ l = strtol(cp, &endp, 10); if (endp != cp) { cp = endp; conv_len = conv_len1 = l; } else if (*cp == '*') { conv_len = conv_len1 = (int)va_arg(ap, int); cp++; } /* Precision specifier */ if ('.' == *cp) { cp++; conv_len2 = strtol(cp, &endp, 10); if (endp != cp) { cp = endp; } else if (*cp == '*') { conv_len2 = (int)va_arg(ap, int); cp++; } conv_len = MAX(conv_len1, conv_len2); } /* Short/long identifier */ if ('h' == *cp) { arg_size = -1; /* short */ cp++; } else if ('l' == *cp) { arg_size = 1; /* long */ cp++; if ('l' == *cp) { arg_size = 2; /* long long */ cp++; } } else { arg_size = 0; /* int */ } /* The actual type */ switch (*cp) { case '%': /* * Not real ANSI I suspect, but we'll allow for the * completely daft "%10%" example. */ len += MAX(conv_len1, 1); break; case 'd': case 'i': case 'u': case 'a': case 'x': case 'X': /* Remember: char and short are sent as int on the stack */ if (arg_size == -1) l = (long)va_arg(ap, int); else if (arg_size == 1) l = va_arg(ap, long); else if (arg_size == 2) l = va_arg(ap, long long); else l = (long)va_arg(ap, int); DEBUG_printf("%d", l); /* * No number can be more than 24 characters so we'll take * the max of conv_len and 24 (23 is len(2^64) in octal). * All that work above and we then go and estimate ;-), * but it's needed incase someone does %500d. */ len += MAX(conv_len, 23); break; case 'c': i = va_arg(ap, int); DEBUG_printf("%c", i); /* * Note that %10c and %.10c act differently. * Besides, I think precision is not really allowed for %c. */ len += MAX(conv_len1, i>=0x80 ?MB_CUR_MAX :1); break; case 'f': d = va_arg(ap, double); DEBUG_printf("%f", d); /* * Maybe "Inf" or "NaN", but we'll not worry about that. * Again, err on side of caution and take max of conv_len * and max length of a double. The worst case I can * think of is 317 characters (-1[308 zeros].000000) * without using precision codes. That's horrid. I * cheat and either use 317 or 15 depending on how * large the number is as I reckon 99% of floats * aren't that long. */ l = (ABS(d) > 1000000) ? 317 : 15; l = MAX(l, conv_len1 + 2); if (conv_len2) l += conv_len2 - 6; len += l; break; case 'e': case 'E': case 'g': case 'G': d = va_arg(ap, double); DEBUG_printf("%g", d); /* * Maybe "Inf" or "NaN", but we'll not worry about that * Again, err on side of caution and take max of conv_len * and max length of a double (which defaults to only * '-' + 6 + '.' + 'E[+-]xxx' == 13. */ len += MAX(conv_len, 13); break; case 'p': l = (long)va_arg(ap, void *); /* * Max pointer is 64bits == 16 chars (on alpha), * == 20 with + "0x". */ DEBUG_printf("%p", (void *)l); len += MAX(conv_len, 20); break; case 'n': /* produces no output */ break; case 's': { char *s = (char *)va_arg(ap, char *); DEBUG_printf("%s", s); if (!conv_len2) { len += MAX(conv_len, (int)strlen(s)); } else { len += conv_len; } break; } default: /* wchar_t types of 'C' and 'S' aren't supported */ DEBUG_printf("Arg is %c\n", *cp); } } case '\0': break; default: DEBUG_printf("%c", *cp); len++; } } va_end(ap); return len+1; /* one for the null character */ } #if 0 int main() { int l; char buf[10000]; sprintf(buf, "d: %d\n", 500); l = flen("d: %d\n", 500); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, ""); l = flen(""); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%s\n","test"); l = flen("%s\n", "test"); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%c\n", 'a'); l = flen("%c\n", 'a'); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%31.30f\n", -9999.99); l = flen("%31.30f\n", -9999.99); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%f\n", -1e308); l = flen("%f\n", -1e308); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%.9f\n", -1e308); l = flen("%.9f\n", -1e308); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%10.20f\n", -1.999222333); l = flen("%10.20f\n", -1.999222333); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%#g\n", -3.14159265358e-222); l = flen("%#g\n", -3.1415927e-222); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%e\n", -123456789123456789.1); l = flen("%e\n", -123456789123456789.1); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%c %f %d %s %c %g %ld %s\n", 'a', 3.1, 9, "one", 'b', 4.2, 9, "two"); l = flen("%c %f %d %s %c %g %ld %s\n", 'a', 3.1, 9, "one", 'b', 4.2, 9, "two"); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%*.*e %*c\n", 10, 5, 9.0, 20, 'x'); l = flen("%*.*e %*c\n", 10, 5, 9.0, 20, 'x'); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%10c\n", 'z'); l = flen("%10c\n", 'z'); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%.10c\n", 'z'); l = flen("%.10c\n", 'z'); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%10d\n", 'z'); l = flen("%10d\n", 'z'); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%.10d\n", 'z'); l = flen("%.10d\n", 'z'); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%10%\n"); l = flen("%10%\n"); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%.10%\n"); l = flen("%.10%\n"); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%s\n", "0123456789"); l = flen("%s\n", "0123456789"); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%5s\n", "0123456789"); l = flen("%5s\n", "0123456789"); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%50s\n", "0123456789"); l = flen("%50s\n", "0123456789"); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%.5s\n", "0123456789"); l = flen("%.5s\n", "0123456789"); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%.50s\n", "0123456789"); l = flen("%.50s\n", "0123456789"); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%5.50s\n", "0123456789"); l = flen("%5.50s\n", "0123456789"); printf("%d %d\n\n", strlen(buf), l); sprintf(buf, "%50.5s\n", "0123456789"); l = flen("%50.5s\n", "0123456789"); printf("%d %d\n\n", strlen(buf), l); return 0; } #endif stacks-1.44/htslib/cram/._vlen.h000644 000765 000024 00000000265 12750743246 017260 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/vlen.h000644 000765 000024 00000003427 12750743246 017046 0ustar00catchenstaff000000 000000 /* Author: James Bonfield (jkb@sanger.ac.uk) Copyright (c) 1995-1996 MEDICAL RESEARCH COUNCIL All rights reserved Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1 Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2 Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3 Neither the name of the MEDICAL RESEARCH COUNCIL, THE LABORATORY OF MOLECULAR BIOLOGY nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #ifndef _VLEN_H_ #define _VLEN_H_ #ifdef __cplusplus extern "C" { #endif extern int vflen(char *fmt, va_list ap); extern int flen(char *fmt, ...); #ifdef __cplusplus } #endif #endif /* _VLEN_H_ */ stacks-1.44/htslib/cram/._zfio.c000644 000765 000024 00000000265 12750743246 017256 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/zfio.c000644 000765 000024 00000011460 12750743246 017040 0ustar00catchenstaff000000 000000 /* Copyright (c) 2009-2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #include #include #include #include "cram/os.h" #include "cram/zfio.h" /* ------------------------------------------------------------------------ */ /* Some wrappers around FILE * vs gzFile *, allowing for either */ /* * gzopen() works on both compressed and uncompressed data, but it has * a significant performance hit even for uncompressed data (tested as * 25s using FILE* to 46s via gzOpen and 66s via gzOpen when gzipped). * * Hence we use our own wrapper 'zfp' which is a FILE* when uncompressed * and gzFile* when compressed. This also means we could hide bzopen in * there too if desired. */ off_t zftello(zfp *zf) { return zf->fp ? ftello(zf->fp) : -1; } int zfseeko(zfp *zf, off_t offset, int whence) { return zf->fp ? fseeko(zf->fp, offset, whence) : -1; } /* * A wrapper for either fgets or gzgets depending on what has been * opened. */ char *zfgets(char *line, int size, zfp *zf) { if (zf->fp) return fgets(line, size, zf->fp); else return gzgets(zf->gz, line, size); } /* * A wrapper for either fputs or gzputs depending on what has been * opened. */ int zfputs(char *line, zfp *zf) { if (zf->fp) return fputs(line, zf->fp); else return gzputs(zf->gz, line) ? 0 : EOF; } /* * Peeks at and returns the next character without consuming it from the * input. (Ie a combination of getc and ungetc). */ int zfpeek(zfp *zf) { int c; if (zf->fp) { c = getc(zf->fp); if (c != EOF) ungetc(c, zf->fp); } else { c = gzgetc(zf->gz); if (c != EOF) gzungetc(c, zf->gz); } return c; } /* A replacement for either feof of gzeof */ int zfeof(zfp *zf) { return zf->fp ? feof(zf->fp) : gzeof(zf->gz); } /* A replacement for either fopen or gzopen */ zfp *zfopen(const char *path, const char *mode) { char path2[1024]; zfp *zf; if (!(zf = (zfp *)malloc(sizeof(*zf)))) return NULL; zf->fp = NULL; zf->gz = NULL; /* Try normal fopen */ if (mode[0] != 'z' && mode[1] != 'z' && NULL != (zf->fp = fopen(path, mode))) { unsigned char magic[2]; if (2 != fread(magic, 1, 2, zf->fp)) { free(zf); return NULL; } if (!(magic[0] == 0x1f && magic[1] == 0x8b)) { fseeko(zf->fp, 0, SEEK_SET); return zf; } fclose(zf->fp); zf->fp = NULL; } #ifdef HAVE_POPEN /* * I've no idea why, by gzgets is VERY slow, maybe because it handles * arbitrary seeks. * popen to gzip -cd is 3 times faster though. */ if (*mode == 'w') { } else { if (access(path, R_OK) == 0) { sprintf(path2, "gzip -cd < %.*s", 1000, path); if (NULL != (zf->fp = popen(path2, "r"))) return zf; } sprintf(path2, "gzip -cd < %.*s.gz", 1000, path); if (NULL != (zf->fp = popen(path2, "r"))) return zf; printf("Failed on %s\n", path); } else { sprintf(path2, "gzip > %.*s", 1000, path); if (NULL != (zf->fp = popen(path2, "w"))) return zf; } printf("Failed on %s\n", path); } #else /* Gzopen instead */ if ((zf->gz = gzopen(path, mode))) return zf; sprintf(path2, "%.*s.gz", 1020, path); if ((zf->gz = gzopen(path2, mode))) return zf; #endif perror(path); free(zf); return NULL; } int zfclose(zfp *zf) { int r = (zf->fp) ? fclose(zf->fp) : gzclose(zf->gz); free(zf); return r; } stacks-1.44/htslib/cram/._zfio.h000644 000765 000024 00000000265 12750743246 017263 0ustar00catchenstaff000000 000000 Mac OS X  2µATTRµcom.apple.quarantineq/0001;57a394a3;Firefox.app;stacks-1.44/htslib/cram/zfio.h000644 000765 000024 00000004052 12750743246 017044 0ustar00catchenstaff000000 000000 /* Copyright (c) 2009-2013 Genome Research Ltd. Author: James Bonfield Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the names Genome Research Ltd and Wellcome Trust Sanger Institute nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY GENOME RESEARCH LTD AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GENOME RESEARCH LTD OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #ifndef _ZFIO_H_ #define _ZFIO_H_ #include #include #ifdef __cplusplus extern "C" { #endif /* * Either a gzFile or a FILE. */ typedef struct { FILE *fp; gzFile gz; } zfp; off_t zftello(zfp *zf); int zfseeko(zfp *zf, off_t offset, int whence); char *zfgets(char *line, int size, zfp *zf); int zfputs(char *line, zfp *zf); zfp *zfopen(const char *path, const char *mode); int zfclose(zfp *zf); int zfpeek(zfp *zf); int zfeof(zfp *zf); #ifdef __cplusplus } #endif #endif /* _ZFIO_H_ */ stacks-1.44/config/compile000755 000765 000024 00000016245 12750771222 016342 0ustar00catchenstaff000000 000000 #! /bin/sh # Wrapper for compilers which do not understand '-c -o'. scriptversion=2012-10-14.11; # UTC # Copyright (C) 1999-2014 Free Software Foundation, Inc. # Written by Tom Tromey . # # This program is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2, or (at your option) # any later version. # # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this program. If not, see . # As a special exception to the GNU General Public License, if you # distribute this file as part of a program that contains a # configuration script generated by Autoconf, you may include it under # the same distribution terms that you use for the rest of that program. # This file is maintained in Automake, please report # bugs to or send patches to # . nl=' ' # We need space, tab and new line, in precisely that order. Quoting is # there to prevent tools from complaining about whitespace usage. IFS=" "" $nl" file_conv= # func_file_conv build_file lazy # Convert a $build file to $host form and store it in $file # Currently only supports Windows hosts. If the determined conversion # type is listed in (the comma separated) LAZY, no conversion will # take place. func_file_conv () { file=$1 case $file in / | /[!/]*) # absolute file, and not a UNC file if test -z "$file_conv"; then # lazily determine how to convert abs files case `uname -s` in MINGW*) file_conv=mingw ;; CYGWIN*) file_conv=cygwin ;; *) file_conv=wine ;; esac fi case $file_conv/,$2, in *,$file_conv,*) ;; mingw/*) file=`cmd //C echo "$file " | sed -e 's/"\(.*\) " *$/\1/'` ;; cygwin/*) file=`cygpath -m "$file" || echo "$file"` ;; wine/*) file=`winepath -w "$file" || echo "$file"` ;; esac ;; esac } # func_cl_dashL linkdir # Make cl look for libraries in LINKDIR func_cl_dashL () { func_file_conv "$1" if test -z "$lib_path"; then lib_path=$file else lib_path="$lib_path;$file" fi linker_opts="$linker_opts -LIBPATH:$file" } # func_cl_dashl library # Do a library search-path lookup for cl func_cl_dashl () { lib=$1 found=no save_IFS=$IFS IFS=';' for dir in $lib_path $LIB do IFS=$save_IFS if $shared && test -f "$dir/$lib.dll.lib"; then found=yes lib=$dir/$lib.dll.lib break fi if test -f "$dir/$lib.lib"; then found=yes lib=$dir/$lib.lib break fi if test -f "$dir/lib$lib.a"; then found=yes lib=$dir/lib$lib.a break fi done IFS=$save_IFS if test "$found" != yes; then lib=$lib.lib fi } # func_cl_wrapper cl arg... # Adjust compile command to suit cl func_cl_wrapper () { # Assume a capable shell lib_path= shared=: linker_opts= for arg do if test -n "$eat"; then eat= else case $1 in -o) # configure might choose to run compile as 'compile cc -o foo foo.c'. eat=1 case $2 in *.o | *.[oO][bB][jJ]) func_file_conv "$2" set x "$@" -Fo"$file" shift ;; *) func_file_conv "$2" set x "$@" -Fe"$file" shift ;; esac ;; -I) eat=1 func_file_conv "$2" mingw set x "$@" -I"$file" shift ;; -I*) func_file_conv "${1#-I}" mingw set x "$@" -I"$file" shift ;; -l) eat=1 func_cl_dashl "$2" set x "$@" "$lib" shift ;; -l*) func_cl_dashl "${1#-l}" set x "$@" "$lib" shift ;; -L) eat=1 func_cl_dashL "$2" ;; -L*) func_cl_dashL "${1#-L}" ;; -static) shared=false ;; -Wl,*) arg=${1#-Wl,} save_ifs="$IFS"; IFS=',' for flag in $arg; do IFS="$save_ifs" linker_opts="$linker_opts $flag" done IFS="$save_ifs" ;; -Xlinker) eat=1 linker_opts="$linker_opts $2" ;; -*) set x "$@" "$1" shift ;; *.cc | *.CC | *.cxx | *.CXX | *.[cC]++) func_file_conv "$1" set x "$@" -Tp"$file" shift ;; *.c | *.cpp | *.CPP | *.lib | *.LIB | *.Lib | *.OBJ | *.obj | *.[oO]) func_file_conv "$1" mingw set x "$@" "$file" shift ;; *) set x "$@" "$1" shift ;; esac fi shift done if test -n "$linker_opts"; then linker_opts="-link$linker_opts" fi exec "$@" $linker_opts exit 1 } eat= case $1 in '') echo "$0: No command. Try '$0 --help' for more information." 1>&2 exit 1; ;; -h | --h*) cat <<\EOF Usage: compile [--help] [--version] PROGRAM [ARGS] Wrapper for compilers which do not understand '-c -o'. Remove '-o dest.o' from ARGS, run PROGRAM with the remaining arguments, and rename the output as expected. If you are trying to build a whole package this is not the right script to run: please start by reading the file 'INSTALL'. Report bugs to . EOF exit $? ;; -v | --v*) echo "compile $scriptversion" exit $? ;; cl | *[/\\]cl | cl.exe | *[/\\]cl.exe ) func_cl_wrapper "$@" # Doesn't return... ;; esac ofile= cfile= for arg do if test -n "$eat"; then eat= else case $1 in -o) # configure might choose to run compile as 'compile cc -o foo foo.c'. # So we strip '-o arg' only if arg is an object. eat=1 case $2 in *.o | *.obj) ofile=$2 ;; *) set x "$@" -o "$2" shift ;; esac ;; *.c) cfile=$1 set x "$@" "$1" shift ;; *) set x "$@" "$1" shift ;; esac fi shift done if test -z "$ofile" || test -z "$cfile"; then # If no '-o' option was seen then we might have been invoked from a # pattern rule where we don't need one. That is ok -- this is a # normal compilation that the losing compiler can handle. If no # '.c' file was seen then we are probably linking. That is also # ok. exec "$@" fi # Name of file we expect compiler to create. cofile=`echo "$cfile" | sed 's|^.*[\\/]||; s|^[a-zA-Z]:||; s/\.c$/.o/'` # Create the lock directory. # Note: use '[/\\:.-]' here to ensure that we don't use the same name # that we are using for the .o file. Also, base the name on the expected # object file name, since that is what matters with a parallel build. lockdir=`echo "$cofile" | sed -e 's|[/\\:.-]|_|g'`.d while true; do if mkdir "$lockdir" >/dev/null 2>&1; then break fi sleep 1 done # FIXME: race condition here if user kills between mkdir and trap. trap "rmdir '$lockdir'; exit 1" 1 2 15 # Run the compile. "$@" ret=$? if test -f "$cofile"; then test "$cofile" = "$ofile" || mv "$cofile" "$ofile" elif test -f "${cofile}bj"; then test "${cofile}bj" = "$ofile" || mv "${cofile}bj" "$ofile" fi rmdir "$lockdir" exit $ret # Local Variables: # mode: shell-script # sh-indentation: 2 # eval: (add-hook 'write-file-hooks 'time-stamp) # time-stamp-start: "scriptversion=" # time-stamp-format: "%:y-%02m-%02d.%02H" # time-stamp-time-zone: "UTC" # time-stamp-end: "; # UTC" # End: stacks-1.44/config/depcomp000755 000765 000024 00000056016 12750771222 016341 0ustar00catchenstaff000000 000000 #! /bin/sh # depcomp - compile a program generating dependencies as side-effects scriptversion=2013-05-30.07; # UTC # Copyright (C) 1999-2014 Free Software Foundation, Inc. # This program is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2, or (at your option) # any later version. # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # You should have received a copy of the GNU General Public License # along with this program. If not, see . # As a special exception to the GNU General Public License, if you # distribute this file as part of a program that contains a # configuration script generated by Autoconf, you may include it under # the same distribution terms that you use for the rest of that program. # Originally written by Alexandre Oliva . case $1 in '') echo "$0: No command. Try '$0 --help' for more information." 1>&2 exit 1; ;; -h | --h*) cat <<\EOF Usage: depcomp [--help] [--version] PROGRAM [ARGS] Run PROGRAMS ARGS to compile a file, generating dependencies as side-effects. Environment variables: depmode Dependency tracking mode. source Source file read by 'PROGRAMS ARGS'. object Object file output by 'PROGRAMS ARGS'. DEPDIR directory where to store dependencies. depfile Dependency file to output. tmpdepfile Temporary file to use when outputting dependencies. libtool Whether libtool is used (yes/no). Report bugs to . EOF exit $? ;; -v | --v*) echo "depcomp $scriptversion" exit $? ;; esac # Get the directory component of the given path, and save it in the # global variables '$dir'. Note that this directory component will # be either empty or ending with a '/' character. This is deliberate. set_dir_from () { case $1 in */*) dir=`echo "$1" | sed -e 's|/[^/]*$|/|'`;; *) dir=;; esac } # Get the suffix-stripped basename of the given path, and save it the # global variable '$base'. set_base_from () { base=`echo "$1" | sed -e 's|^.*/||' -e 's/\.[^.]*$//'` } # If no dependency file was actually created by the compiler invocation, # we still have to create a dummy depfile, to avoid errors with the # Makefile "include basename.Plo" scheme. make_dummy_depfile () { echo "#dummy" > "$depfile" } # Factor out some common post-processing of the generated depfile. # Requires the auxiliary global variable '$tmpdepfile' to be set. aix_post_process_depfile () { # If the compiler actually managed to produce a dependency file, # post-process it. if test -f "$tmpdepfile"; then # Each line is of the form 'foo.o: dependency.h'. # Do two passes, one to just change these to # $object: dependency.h # and one to simply output # dependency.h: # which is needed to avoid the deleted-header problem. { sed -e "s,^.*\.[$lower]*:,$object:," < "$tmpdepfile" sed -e "s,^.*\.[$lower]*:[$tab ]*,," -e 's,$,:,' < "$tmpdepfile" } > "$depfile" rm -f "$tmpdepfile" else make_dummy_depfile fi } # A tabulation character. tab=' ' # A newline character. nl=' ' # Character ranges might be problematic outside the C locale. # These definitions help. upper=ABCDEFGHIJKLMNOPQRSTUVWXYZ lower=abcdefghijklmnopqrstuvwxyz digits=0123456789 alpha=${upper}${lower} if test -z "$depmode" || test -z "$source" || test -z "$object"; then echo "depcomp: Variables source, object and depmode must be set" 1>&2 exit 1 fi # Dependencies for sub/bar.o or sub/bar.obj go into sub/.deps/bar.Po. depfile=${depfile-`echo "$object" | sed 's|[^\\/]*$|'${DEPDIR-.deps}'/&|;s|\.\([^.]*\)$|.P\1|;s|Pobj$|Po|'`} tmpdepfile=${tmpdepfile-`echo "$depfile" | sed 's/\.\([^.]*\)$/.T\1/'`} rm -f "$tmpdepfile" # Avoid interferences from the environment. gccflag= dashmflag= # Some modes work just like other modes, but use different flags. We # parameterize here, but still list the modes in the big case below, # to make depend.m4 easier to write. Note that we *cannot* use a case # here, because this file can only contain one case statement. if test "$depmode" = hp; then # HP compiler uses -M and no extra arg. gccflag=-M depmode=gcc fi if test "$depmode" = dashXmstdout; then # This is just like dashmstdout with a different argument. dashmflag=-xM depmode=dashmstdout fi cygpath_u="cygpath -u -f -" if test "$depmode" = msvcmsys; then # This is just like msvisualcpp but w/o cygpath translation. # Just convert the backslash-escaped backslashes to single forward # slashes to satisfy depend.m4 cygpath_u='sed s,\\\\,/,g' depmode=msvisualcpp fi if test "$depmode" = msvc7msys; then # This is just like msvc7 but w/o cygpath translation. # Just convert the backslash-escaped backslashes to single forward # slashes to satisfy depend.m4 cygpath_u='sed s,\\\\,/,g' depmode=msvc7 fi if test "$depmode" = xlc; then # IBM C/C++ Compilers xlc/xlC can output gcc-like dependency information. gccflag=-qmakedep=gcc,-MF depmode=gcc fi case "$depmode" in gcc3) ## gcc 3 implements dependency tracking that does exactly what ## we want. Yay! Note: for some reason libtool 1.4 doesn't like ## it if -MD -MP comes after the -MF stuff. Hmm. ## Unfortunately, FreeBSD c89 acceptance of flags depends upon ## the command line argument order; so add the flags where they ## appear in depend2.am. Note that the slowdown incurred here ## affects only configure: in makefiles, %FASTDEP% shortcuts this. for arg do case $arg in -c) set fnord "$@" -MT "$object" -MD -MP -MF "$tmpdepfile" "$arg" ;; *) set fnord "$@" "$arg" ;; esac shift # fnord shift # $arg done "$@" stat=$? if test $stat -ne 0; then rm -f "$tmpdepfile" exit $stat fi mv "$tmpdepfile" "$depfile" ;; gcc) ## Note that this doesn't just cater to obsosete pre-3.x GCC compilers. ## but also to in-use compilers like IMB xlc/xlC and the HP C compiler. ## (see the conditional assignment to $gccflag above). ## There are various ways to get dependency output from gcc. Here's ## why we pick this rather obscure method: ## - Don't want to use -MD because we'd like the dependencies to end ## up in a subdir. Having to rename by hand is ugly. ## (We might end up doing this anyway to support other compilers.) ## - The DEPENDENCIES_OUTPUT environment variable makes gcc act like ## -MM, not -M (despite what the docs say). Also, it might not be ## supported by the other compilers which use the 'gcc' depmode. ## - Using -M directly means running the compiler twice (even worse ## than renaming). if test -z "$gccflag"; then gccflag=-MD, fi "$@" -Wp,"$gccflag$tmpdepfile" stat=$? if test $stat -ne 0; then rm -f "$tmpdepfile" exit $stat fi rm -f "$depfile" echo "$object : \\" > "$depfile" # The second -e expression handles DOS-style file names with drive # letters. sed -e 's/^[^:]*: / /' \ -e 's/^['$alpha']:\/[^:]*: / /' < "$tmpdepfile" >> "$depfile" ## This next piece of magic avoids the "deleted header file" problem. ## The problem is that when a header file which appears in a .P file ## is deleted, the dependency causes make to die (because there is ## typically no way to rebuild the header). We avoid this by adding ## dummy dependencies for each header file. Too bad gcc doesn't do ## this for us directly. ## Some versions of gcc put a space before the ':'. On the theory ## that the space means something, we add a space to the output as ## well. hp depmode also adds that space, but also prefixes the VPATH ## to the object. Take care to not repeat it in the output. ## Some versions of the HPUX 10.20 sed can't process this invocation ## correctly. Breaking it into two sed invocations is a workaround. tr ' ' "$nl" < "$tmpdepfile" \ | sed -e 's/^\\$//' -e '/^$/d' -e "s|.*$object$||" -e '/:$/d' \ | sed -e 's/$/ :/' >> "$depfile" rm -f "$tmpdepfile" ;; hp) # This case exists only to let depend.m4 do its work. It works by # looking at the text of this script. This case will never be run, # since it is checked for above. exit 1 ;; sgi) if test "$libtool" = yes; then "$@" "-Wp,-MDupdate,$tmpdepfile" else "$@" -MDupdate "$tmpdepfile" fi stat=$? if test $stat -ne 0; then rm -f "$tmpdepfile" exit $stat fi rm -f "$depfile" if test -f "$tmpdepfile"; then # yes, the sourcefile depend on other files echo "$object : \\" > "$depfile" # Clip off the initial element (the dependent). Don't try to be # clever and replace this with sed code, as IRIX sed won't handle # lines with more than a fixed number of characters (4096 in # IRIX 6.2 sed, 8192 in IRIX 6.5). We also remove comment lines; # the IRIX cc adds comments like '#:fec' to the end of the # dependency line. tr ' ' "$nl" < "$tmpdepfile" \ | sed -e 's/^.*\.o://' -e 's/#.*$//' -e '/^$/ d' \ | tr "$nl" ' ' >> "$depfile" echo >> "$depfile" # The second pass generates a dummy entry for each header file. tr ' ' "$nl" < "$tmpdepfile" \ | sed -e 's/^.*\.o://' -e 's/#.*$//' -e '/^$/ d' -e 's/$/:/' \ >> "$depfile" else make_dummy_depfile fi rm -f "$tmpdepfile" ;; xlc) # This case exists only to let depend.m4 do its work. It works by # looking at the text of this script. This case will never be run, # since it is checked for above. exit 1 ;; aix) # The C for AIX Compiler uses -M and outputs the dependencies # in a .u file. In older versions, this file always lives in the # current directory. Also, the AIX compiler puts '$object:' at the # start of each line; $object doesn't have directory information. # Version 6 uses the directory in both cases. set_dir_from "$object" set_base_from "$object" if test "$libtool" = yes; then tmpdepfile1=$dir$base.u tmpdepfile2=$base.u tmpdepfile3=$dir.libs/$base.u "$@" -Wc,-M else tmpdepfile1=$dir$base.u tmpdepfile2=$dir$base.u tmpdepfile3=$dir$base.u "$@" -M fi stat=$? if test $stat -ne 0; then rm -f "$tmpdepfile1" "$tmpdepfile2" "$tmpdepfile3" exit $stat fi for tmpdepfile in "$tmpdepfile1" "$tmpdepfile2" "$tmpdepfile3" do test -f "$tmpdepfile" && break done aix_post_process_depfile ;; tcc) # tcc (Tiny C Compiler) understand '-MD -MF file' since version 0.9.26 # FIXME: That version still under development at the moment of writing. # Make that this statement remains true also for stable, released # versions. # It will wrap lines (doesn't matter whether long or short) with a # trailing '\', as in: # # foo.o : \ # foo.c \ # foo.h \ # # It will put a trailing '\' even on the last line, and will use leading # spaces rather than leading tabs (at least since its commit 0394caf7 # "Emit spaces for -MD"). "$@" -MD -MF "$tmpdepfile" stat=$? if test $stat -ne 0; then rm -f "$tmpdepfile" exit $stat fi rm -f "$depfile" # Each non-empty line is of the form 'foo.o : \' or ' dep.h \'. # We have to change lines of the first kind to '$object: \'. sed -e "s|.*:|$object :|" < "$tmpdepfile" > "$depfile" # And for each line of the second kind, we have to emit a 'dep.h:' # dummy dependency, to avoid the deleted-header problem. sed -n -e 's|^ *\(.*\) *\\$|\1:|p' < "$tmpdepfile" >> "$depfile" rm -f "$tmpdepfile" ;; ## The order of this option in the case statement is important, since the ## shell code in configure will try each of these formats in the order ## listed in this file. A plain '-MD' option would be understood by many ## compilers, so we must ensure this comes after the gcc and icc options. pgcc) # Portland's C compiler understands '-MD'. # Will always output deps to 'file.d' where file is the root name of the # source file under compilation, even if file resides in a subdirectory. # The object file name does not affect the name of the '.d' file. # pgcc 10.2 will output # foo.o: sub/foo.c sub/foo.h # and will wrap long lines using '\' : # foo.o: sub/foo.c ... \ # sub/foo.h ... \ # ... set_dir_from "$object" # Use the source, not the object, to determine the base name, since # that's sadly what pgcc will do too. set_base_from "$source" tmpdepfile=$base.d # For projects that build the same source file twice into different object # files, the pgcc approach of using the *source* file root name can cause # problems in parallel builds. Use a locking strategy to avoid stomping on # the same $tmpdepfile. lockdir=$base.d-lock trap " echo '$0: caught signal, cleaning up...' >&2 rmdir '$lockdir' exit 1 " 1 2 13 15 numtries=100 i=$numtries while test $i -gt 0; do # mkdir is a portable test-and-set. if mkdir "$lockdir" 2>/dev/null; then # This process acquired the lock. "$@" -MD stat=$? # Release the lock. rmdir "$lockdir" break else # If the lock is being held by a different process, wait # until the winning process is done or we timeout. while test -d "$lockdir" && test $i -gt 0; do sleep 1 i=`expr $i - 1` done fi i=`expr $i - 1` done trap - 1 2 13 15 if test $i -le 0; then echo "$0: failed to acquire lock after $numtries attempts" >&2 echo "$0: check lockdir '$lockdir'" >&2 exit 1 fi if test $stat -ne 0; then rm -f "$tmpdepfile" exit $stat fi rm -f "$depfile" # Each line is of the form `foo.o: dependent.h', # or `foo.o: dep1.h dep2.h \', or ` dep3.h dep4.h \'. # Do two passes, one to just change these to # `$object: dependent.h' and one to simply `dependent.h:'. sed "s,^[^:]*:,$object :," < "$tmpdepfile" > "$depfile" # Some versions of the HPUX 10.20 sed can't process this invocation # correctly. Breaking it into two sed invocations is a workaround. sed 's,^[^:]*: \(.*\)$,\1,;s/^\\$//;/^$/d;/:$/d' < "$tmpdepfile" \ | sed -e 's/$/ :/' >> "$depfile" rm -f "$tmpdepfile" ;; hp2) # The "hp" stanza above does not work with aCC (C++) and HP's ia64 # compilers, which have integrated preprocessors. The correct option # to use with these is +Maked; it writes dependencies to a file named # 'foo.d', which lands next to the object file, wherever that # happens to be. # Much of this is similar to the tru64 case; see comments there. set_dir_from "$object" set_base_from "$object" if test "$libtool" = yes; then tmpdepfile1=$dir$base.d tmpdepfile2=$dir.libs/$base.d "$@" -Wc,+Maked else tmpdepfile1=$dir$base.d tmpdepfile2=$dir$base.d "$@" +Maked fi stat=$? if test $stat -ne 0; then rm -f "$tmpdepfile1" "$tmpdepfile2" exit $stat fi for tmpdepfile in "$tmpdepfile1" "$tmpdepfile2" do test -f "$tmpdepfile" && break done if test -f "$tmpdepfile"; then sed -e "s,^.*\.[$lower]*:,$object:," "$tmpdepfile" > "$depfile" # Add 'dependent.h:' lines. sed -ne '2,${ s/^ *// s/ \\*$// s/$/:/ p }' "$tmpdepfile" >> "$depfile" else make_dummy_depfile fi rm -f "$tmpdepfile" "$tmpdepfile2" ;; tru64) # The Tru64 compiler uses -MD to generate dependencies as a side # effect. 'cc -MD -o foo.o ...' puts the dependencies into 'foo.o.d'. # At least on Alpha/Redhat 6.1, Compaq CCC V6.2-504 seems to put # dependencies in 'foo.d' instead, so we check for that too. # Subdirectories are respected. set_dir_from "$object" set_base_from "$object" if test "$libtool" = yes; then # Libtool generates 2 separate objects for the 2 libraries. These # two compilations output dependencies in $dir.libs/$base.o.d and # in $dir$base.o.d. We have to check for both files, because # one of the two compilations can be disabled. We should prefer # $dir$base.o.d over $dir.libs/$base.o.d because the latter is # automatically cleaned when .libs/ is deleted, while ignoring # the former would cause a distcleancheck panic. tmpdepfile1=$dir$base.o.d # libtool 1.5 tmpdepfile2=$dir.libs/$base.o.d # Likewise. tmpdepfile3=$dir.libs/$base.d # Compaq CCC V6.2-504 "$@" -Wc,-MD else tmpdepfile1=$dir$base.d tmpdepfile2=$dir$base.d tmpdepfile3=$dir$base.d "$@" -MD fi stat=$? if test $stat -ne 0; then rm -f "$tmpdepfile1" "$tmpdepfile2" "$tmpdepfile3" exit $stat fi for tmpdepfile in "$tmpdepfile1" "$tmpdepfile2" "$tmpdepfile3" do test -f "$tmpdepfile" && break done # Same post-processing that is required for AIX mode. aix_post_process_depfile ;; msvc7) if test "$libtool" = yes; then showIncludes=-Wc,-showIncludes else showIncludes=-showIncludes fi "$@" $showIncludes > "$tmpdepfile" stat=$? grep -v '^Note: including file: ' "$tmpdepfile" if test $stat -ne 0; then rm -f "$tmpdepfile" exit $stat fi rm -f "$depfile" echo "$object : \\" > "$depfile" # The first sed program below extracts the file names and escapes # backslashes for cygpath. The second sed program outputs the file # name when reading, but also accumulates all include files in the # hold buffer in order to output them again at the end. This only # works with sed implementations that can handle large buffers. sed < "$tmpdepfile" -n ' /^Note: including file: *\(.*\)/ { s//\1/ s/\\/\\\\/g p }' | $cygpath_u | sort -u | sed -n ' s/ /\\ /g s/\(.*\)/'"$tab"'\1 \\/p s/.\(.*\) \\/\1:/ H $ { s/.*/'"$tab"'/ G p }' >> "$depfile" echo >> "$depfile" # make sure the fragment doesn't end with a backslash rm -f "$tmpdepfile" ;; msvc7msys) # This case exists only to let depend.m4 do its work. It works by # looking at the text of this script. This case will never be run, # since it is checked for above. exit 1 ;; #nosideeffect) # This comment above is used by automake to tell side-effect # dependency tracking mechanisms from slower ones. dashmstdout) # Important note: in order to support this mode, a compiler *must* # always write the preprocessed file to stdout, regardless of -o. "$@" || exit $? # Remove the call to Libtool. if test "$libtool" = yes; then while test "X$1" != 'X--mode=compile'; do shift done shift fi # Remove '-o $object'. IFS=" " for arg do case $arg in -o) shift ;; $object) shift ;; *) set fnord "$@" "$arg" shift # fnord shift # $arg ;; esac done test -z "$dashmflag" && dashmflag=-M # Require at least two characters before searching for ':' # in the target name. This is to cope with DOS-style filenames: # a dependency such as 'c:/foo/bar' could be seen as target 'c' otherwise. "$@" $dashmflag | sed "s|^[$tab ]*[^:$tab ][^:][^:]*:[$tab ]*|$object: |" > "$tmpdepfile" rm -f "$depfile" cat < "$tmpdepfile" > "$depfile" # Some versions of the HPUX 10.20 sed can't process this sed invocation # correctly. Breaking it into two sed invocations is a workaround. tr ' ' "$nl" < "$tmpdepfile" \ | sed -e 's/^\\$//' -e '/^$/d' -e '/:$/d' \ | sed -e 's/$/ :/' >> "$depfile" rm -f "$tmpdepfile" ;; dashXmstdout) # This case only exists to satisfy depend.m4. It is never actually # run, as this mode is specially recognized in the preamble. exit 1 ;; makedepend) "$@" || exit $? # Remove any Libtool call if test "$libtool" = yes; then while test "X$1" != 'X--mode=compile'; do shift done shift fi # X makedepend shift cleared=no eat=no for arg do case $cleared in no) set ""; shift cleared=yes ;; esac if test $eat = yes; then eat=no continue fi case "$arg" in -D*|-I*) set fnord "$@" "$arg"; shift ;; # Strip any option that makedepend may not understand. Remove # the object too, otherwise makedepend will parse it as a source file. -arch) eat=yes ;; -*|$object) ;; *) set fnord "$@" "$arg"; shift ;; esac done obj_suffix=`echo "$object" | sed 's/^.*\././'` touch "$tmpdepfile" ${MAKEDEPEND-makedepend} -o"$obj_suffix" -f"$tmpdepfile" "$@" rm -f "$depfile" # makedepend may prepend the VPATH from the source file name to the object. # No need to regex-escape $object, excess matching of '.' is harmless. sed "s|^.*\($object *:\)|\1|" "$tmpdepfile" > "$depfile" # Some versions of the HPUX 10.20 sed can't process the last invocation # correctly. Breaking it into two sed invocations is a workaround. sed '1,2d' "$tmpdepfile" \ | tr ' ' "$nl" \ | sed -e 's/^\\$//' -e '/^$/d' -e '/:$/d' \ | sed -e 's/$/ :/' >> "$depfile" rm -f "$tmpdepfile" "$tmpdepfile".bak ;; cpp) # Important note: in order to support this mode, a compiler *must* # always write the preprocessed file to stdout. "$@" || exit $? # Remove the call to Libtool. if test "$libtool" = yes; then while test "X$1" != 'X--mode=compile'; do shift done shift fi # Remove '-o $object'. IFS=" " for arg do case $arg in -o) shift ;; $object) shift ;; *) set fnord "$@" "$arg" shift # fnord shift # $arg ;; esac done "$@" -E \ | sed -n -e '/^# [0-9][0-9]* "\([^"]*\)".*/ s:: \1 \\:p' \ -e '/^#line [0-9][0-9]* "\([^"]*\)".*/ s:: \1 \\:p' \ | sed '$ s: \\$::' > "$tmpdepfile" rm -f "$depfile" echo "$object : \\" > "$depfile" cat < "$tmpdepfile" >> "$depfile" sed < "$tmpdepfile" '/^$/d;s/^ //;s/ \\$//;s/$/ :/' >> "$depfile" rm -f "$tmpdepfile" ;; msvisualcpp) # Important note: in order to support this mode, a compiler *must* # always write the preprocessed file to stdout. "$@" || exit $? # Remove the call to Libtool. if test "$libtool" = yes; then while test "X$1" != 'X--mode=compile'; do shift done shift fi IFS=" " for arg do case "$arg" in -o) shift ;; $object) shift ;; "-Gm"|"/Gm"|"-Gi"|"/Gi"|"-ZI"|"/ZI") set fnord "$@" shift shift ;; *) set fnord "$@" "$arg" shift shift ;; esac done "$@" -E 2>/dev/null | sed -n '/^#line [0-9][0-9]* "\([^"]*\)"/ s::\1:p' | $cygpath_u | sort -u > "$tmpdepfile" rm -f "$depfile" echo "$object : \\" > "$depfile" sed < "$tmpdepfile" -n -e 's% %\\ %g' -e '/^\(.*\)$/ s::'"$tab"'\1 \\:p' >> "$depfile" echo "$tab" >> "$depfile" sed < "$tmpdepfile" -n -e 's% %\\ %g' -e '/^\(.*\)$/ s::\1\::p' >> "$depfile" rm -f "$tmpdepfile" ;; msvcmsys) # This case exists only to let depend.m4 do its work. It works by # looking at the text of this script. This case will never be run, # since it is checked for above. exit 1 ;; none) exec "$@" ;; *) echo "Unknown depmode $depmode" 1>&2 exit 1 ;; esac exit 0 # Local Variables: # mode: shell-script # sh-indentation: 2 # eval: (add-hook 'write-file-hooks 'time-stamp) # time-stamp-start: "scriptversion=" # time-stamp-format: "%:y-%02m-%02d.%02H" # time-stamp-time-zone: "UTC" # time-stamp-end: "; # UTC" # End: stacks-1.44/config/install-sh000755 000765 000024 00000034523 12750771222 016767 0ustar00catchenstaff000000 000000 #!/bin/sh # install - install a program, script, or datafile scriptversion=2013-12-25.23; # UTC # This originates from X11R5 (mit/util/scripts/install.sh), which was # later released in X11R6 (xc/config/util/install.sh) with the # following copyright and license. # # Copyright (C) 1994 X Consortium # # Permission is hereby granted, free of charge, to any person obtaining a copy # of this software and associated documentation files (the "Software"), to # deal in the Software without restriction, including without limitation the # rights to use, copy, modify, merge, publish, distribute, sublicense, and/or # sell copies of the Software, and to permit persons to whom the Software is # furnished to do so, subject to the following conditions: # # The above copyright notice and this permission notice shall be included in # all copies or substantial portions of the Software. # # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE # X CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN # AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNEC- # TION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. # # Except as contained in this notice, the name of the X Consortium shall not # be used in advertising or otherwise to promote the sale, use or other deal- # ings in this Software without prior written authorization from the X Consor- # tium. # # # FSF changes to this file are in the public domain. # # Calling this script install-sh is preferred over install.sh, to prevent # 'make' implicit rules from creating a file called install from it # when there is no Makefile. # # This script is compatible with the BSD install script, but was written # from scratch. tab=' ' nl=' ' IFS=" $tab$nl" # Set DOITPROG to "echo" to test this script. doit=${DOITPROG-} doit_exec=${doit:-exec} # Put in absolute file names if you don't have them in your path; # or use environment vars. chgrpprog=${CHGRPPROG-chgrp} chmodprog=${CHMODPROG-chmod} chownprog=${CHOWNPROG-chown} cmpprog=${CMPPROG-cmp} cpprog=${CPPROG-cp} mkdirprog=${MKDIRPROG-mkdir} mvprog=${MVPROG-mv} rmprog=${RMPROG-rm} stripprog=${STRIPPROG-strip} posix_mkdir= # Desired mode of installed file. mode=0755 chgrpcmd= chmodcmd=$chmodprog chowncmd= mvcmd=$mvprog rmcmd="$rmprog -f" stripcmd= src= dst= dir_arg= dst_arg= copy_on_change=false is_target_a_directory=possibly usage="\ Usage: $0 [OPTION]... [-T] SRCFILE DSTFILE or: $0 [OPTION]... SRCFILES... DIRECTORY or: $0 [OPTION]... -t DIRECTORY SRCFILES... or: $0 [OPTION]... -d DIRECTORIES... In the 1st form, copy SRCFILE to DSTFILE. In the 2nd and 3rd, copy all SRCFILES to DIRECTORY. In the 4th, create DIRECTORIES. Options: --help display this help and exit. --version display version info and exit. -c (ignored) -C install only if different (preserve the last data modification time) -d create directories instead of installing files. -g GROUP $chgrpprog installed files to GROUP. -m MODE $chmodprog installed files to MODE. -o USER $chownprog installed files to USER. -s $stripprog installed files. -t DIRECTORY install into DIRECTORY. -T report an error if DSTFILE is a directory. Environment variables override the default commands: CHGRPPROG CHMODPROG CHOWNPROG CMPPROG CPPROG MKDIRPROG MVPROG RMPROG STRIPPROG " while test $# -ne 0; do case $1 in -c) ;; -C) copy_on_change=true;; -d) dir_arg=true;; -g) chgrpcmd="$chgrpprog $2" shift;; --help) echo "$usage"; exit $?;; -m) mode=$2 case $mode in *' '* | *"$tab"* | *"$nl"* | *'*'* | *'?'* | *'['*) echo "$0: invalid mode: $mode" >&2 exit 1;; esac shift;; -o) chowncmd="$chownprog $2" shift;; -s) stripcmd=$stripprog;; -t) is_target_a_directory=always dst_arg=$2 # Protect names problematic for 'test' and other utilities. case $dst_arg in -* | [=\(\)!]) dst_arg=./$dst_arg;; esac shift;; -T) is_target_a_directory=never;; --version) echo "$0 $scriptversion"; exit $?;; --) shift break;; -*) echo "$0: invalid option: $1" >&2 exit 1;; *) break;; esac shift done # We allow the use of options -d and -T together, by making -d # take the precedence; this is for compatibility with GNU install. if test -n "$dir_arg"; then if test -n "$dst_arg"; then echo "$0: target directory not allowed when installing a directory." >&2 exit 1 fi fi if test $# -ne 0 && test -z "$dir_arg$dst_arg"; then # When -d is used, all remaining arguments are directories to create. # When -t is used, the destination is already specified. # Otherwise, the last argument is the destination. Remove it from $@. for arg do if test -n "$dst_arg"; then # $@ is not empty: it contains at least $arg. set fnord "$@" "$dst_arg" shift # fnord fi shift # arg dst_arg=$arg # Protect names problematic for 'test' and other utilities. case $dst_arg in -* | [=\(\)!]) dst_arg=./$dst_arg;; esac done fi if test $# -eq 0; then if test -z "$dir_arg"; then echo "$0: no input file specified." >&2 exit 1 fi # It's OK to call 'install-sh -d' without argument. # This can happen when creating conditional directories. exit 0 fi if test -z "$dir_arg"; then if test $# -gt 1 || test "$is_target_a_directory" = always; then if test ! -d "$dst_arg"; then echo "$0: $dst_arg: Is not a directory." >&2 exit 1 fi fi fi if test -z "$dir_arg"; then do_exit='(exit $ret); exit $ret' trap "ret=129; $do_exit" 1 trap "ret=130; $do_exit" 2 trap "ret=141; $do_exit" 13 trap "ret=143; $do_exit" 15 # Set umask so as not to create temps with too-generous modes. # However, 'strip' requires both read and write access to temps. case $mode in # Optimize common cases. *644) cp_umask=133;; *755) cp_umask=22;; *[0-7]) if test -z "$stripcmd"; then u_plus_rw= else u_plus_rw='% 200' fi cp_umask=`expr '(' 777 - $mode % 1000 ')' $u_plus_rw`;; *) if test -z "$stripcmd"; then u_plus_rw= else u_plus_rw=,u+rw fi cp_umask=$mode$u_plus_rw;; esac fi for src do # Protect names problematic for 'test' and other utilities. case $src in -* | [=\(\)!]) src=./$src;; esac if test -n "$dir_arg"; then dst=$src dstdir=$dst test -d "$dstdir" dstdir_status=$? else # Waiting for this to be detected by the "$cpprog $src $dsttmp" command # might cause directories to be created, which would be especially bad # if $src (and thus $dsttmp) contains '*'. if test ! -f "$src" && test ! -d "$src"; then echo "$0: $src does not exist." >&2 exit 1 fi if test -z "$dst_arg"; then echo "$0: no destination specified." >&2 exit 1 fi dst=$dst_arg # If destination is a directory, append the input filename; won't work # if double slashes aren't ignored. if test -d "$dst"; then if test "$is_target_a_directory" = never; then echo "$0: $dst_arg: Is a directory" >&2 exit 1 fi dstdir=$dst dst=$dstdir/`basename "$src"` dstdir_status=0 else dstdir=`dirname "$dst"` test -d "$dstdir" dstdir_status=$? fi fi obsolete_mkdir_used=false if test $dstdir_status != 0; then case $posix_mkdir in '') # Create intermediate dirs using mode 755 as modified by the umask. # This is like FreeBSD 'install' as of 1997-10-28. umask=`umask` case $stripcmd.$umask in # Optimize common cases. *[2367][2367]) mkdir_umask=$umask;; .*0[02][02] | .[02][02] | .[02]) mkdir_umask=22;; *[0-7]) mkdir_umask=`expr $umask + 22 \ - $umask % 100 % 40 + $umask % 20 \ - $umask % 10 % 4 + $umask % 2 `;; *) mkdir_umask=$umask,go-w;; esac # With -d, create the new directory with the user-specified mode. # Otherwise, rely on $mkdir_umask. if test -n "$dir_arg"; then mkdir_mode=-m$mode else mkdir_mode= fi posix_mkdir=false case $umask in *[123567][0-7][0-7]) # POSIX mkdir -p sets u+wx bits regardless of umask, which # is incompatible with FreeBSD 'install' when (umask & 300) != 0. ;; *) tmpdir=${TMPDIR-/tmp}/ins$RANDOM-$$ trap 'ret=$?; rmdir "$tmpdir/d" "$tmpdir" 2>/dev/null; exit $ret' 0 if (umask $mkdir_umask && exec $mkdirprog $mkdir_mode -p -- "$tmpdir/d") >/dev/null 2>&1 then if test -z "$dir_arg" || { # Check for POSIX incompatibilities with -m. # HP-UX 11.23 and IRIX 6.5 mkdir -m -p sets group- or # other-writable bit of parent directory when it shouldn't. # FreeBSD 6.1 mkdir -m -p sets mode of existing directory. ls_ld_tmpdir=`ls -ld "$tmpdir"` case $ls_ld_tmpdir in d????-?r-*) different_mode=700;; d????-?--*) different_mode=755;; *) false;; esac && $mkdirprog -m$different_mode -p -- "$tmpdir" && { ls_ld_tmpdir_1=`ls -ld "$tmpdir"` test "$ls_ld_tmpdir" = "$ls_ld_tmpdir_1" } } then posix_mkdir=: fi rmdir "$tmpdir/d" "$tmpdir" else # Remove any dirs left behind by ancient mkdir implementations. rmdir ./$mkdir_mode ./-p ./-- 2>/dev/null fi trap '' 0;; esac;; esac if $posix_mkdir && ( umask $mkdir_umask && $doit_exec $mkdirprog $mkdir_mode -p -- "$dstdir" ) then : else # The umask is ridiculous, or mkdir does not conform to POSIX, # or it failed possibly due to a race condition. Create the # directory the slow way, step by step, checking for races as we go. case $dstdir in /*) prefix='/';; [-=\(\)!]*) prefix='./';; *) prefix='';; esac oIFS=$IFS IFS=/ set -f set fnord $dstdir shift set +f IFS=$oIFS prefixes= for d do test X"$d" = X && continue prefix=$prefix$d if test -d "$prefix"; then prefixes= else if $posix_mkdir; then (umask=$mkdir_umask && $doit_exec $mkdirprog $mkdir_mode -p -- "$dstdir") && break # Don't fail if two instances are running concurrently. test -d "$prefix" || exit 1 else case $prefix in *\'*) qprefix=`echo "$prefix" | sed "s/'/'\\\\\\\\''/g"`;; *) qprefix=$prefix;; esac prefixes="$prefixes '$qprefix'" fi fi prefix=$prefix/ done if test -n "$prefixes"; then # Don't fail if two instances are running concurrently. (umask $mkdir_umask && eval "\$doit_exec \$mkdirprog $prefixes") || test -d "$dstdir" || exit 1 obsolete_mkdir_used=true fi fi fi if test -n "$dir_arg"; then { test -z "$chowncmd" || $doit $chowncmd "$dst"; } && { test -z "$chgrpcmd" || $doit $chgrpcmd "$dst"; } && { test "$obsolete_mkdir_used$chowncmd$chgrpcmd" = false || test -z "$chmodcmd" || $doit $chmodcmd $mode "$dst"; } || exit 1 else # Make a couple of temp file names in the proper directory. dsttmp=$dstdir/_inst.$$_ rmtmp=$dstdir/_rm.$$_ # Trap to clean up those temp files at exit. trap 'ret=$?; rm -f "$dsttmp" "$rmtmp" && exit $ret' 0 # Copy the file name to the temp name. (umask $cp_umask && $doit_exec $cpprog "$src" "$dsttmp") && # and set any options; do chmod last to preserve setuid bits. # # If any of these fail, we abort the whole thing. If we want to # ignore errors from any of these, just make sure not to ignore # errors from the above "$doit $cpprog $src $dsttmp" command. # { test -z "$chowncmd" || $doit $chowncmd "$dsttmp"; } && { test -z "$chgrpcmd" || $doit $chgrpcmd "$dsttmp"; } && { test -z "$stripcmd" || $doit $stripcmd "$dsttmp"; } && { test -z "$chmodcmd" || $doit $chmodcmd $mode "$dsttmp"; } && # If -C, don't bother to copy if it wouldn't change the file. if $copy_on_change && old=`LC_ALL=C ls -dlL "$dst" 2>/dev/null` && new=`LC_ALL=C ls -dlL "$dsttmp" 2>/dev/null` && set -f && set X $old && old=:$2:$4:$5:$6 && set X $new && new=:$2:$4:$5:$6 && set +f && test "$old" = "$new" && $cmpprog "$dst" "$dsttmp" >/dev/null 2>&1 then rm -f "$dsttmp" else # Rename the file to the real destination. $doit $mvcmd -f "$dsttmp" "$dst" 2>/dev/null || # The rename failed, perhaps because mv can't rename something else # to itself, or perhaps because mv is so ancient that it does not # support -f. { # Now remove or move aside any old file at destination location. # We try this two ways since rm can't unlink itself on some # systems and the destination file might be busy for other # reasons. In this case, the final cleanup might fail but the new # file should still install successfully. { test ! -f "$dst" || $doit $rmcmd -f "$dst" 2>/dev/null || { $doit $mvcmd -f "$dst" "$rmtmp" 2>/dev/null && { $doit $rmcmd -f "$rmtmp" 2>/dev/null; :; } } || { echo "$0: cannot unlink or rename $dst" >&2 (exit 1); exit 1 } } && # Now rename the file to the real destination. $doit $mvcmd "$dsttmp" "$dst" } fi || exit 1 trap '' 0 fi done # Local variables: # eval: (add-hook 'write-file-hooks 'time-stamp) # time-stamp-start: "scriptversion=" # time-stamp-format: "%:y-%02m-%02d.%02H" # time-stamp-time-zone: "UTC" # time-stamp-end: "; # UTC" # End: stacks-1.44/config/missing000755 000765 000024 00000015330 12750771222 016355 0ustar00catchenstaff000000 000000 #! /bin/sh # Common wrapper for a few potentially missing GNU programs. scriptversion=2013-10-28.13; # UTC # Copyright (C) 1996-2014 Free Software Foundation, Inc. # Originally written by Fran,cois Pinard , 1996. # This program is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2, or (at your option) # any later version. # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # You should have received a copy of the GNU General Public License # along with this program. If not, see . # As a special exception to the GNU General Public License, if you # distribute this file as part of a program that contains a # configuration script generated by Autoconf, you may include it under # the same distribution terms that you use for the rest of that program. if test $# -eq 0; then echo 1>&2 "Try '$0 --help' for more information" exit 1 fi case $1 in --is-lightweight) # Used by our autoconf macros to check whether the available missing # script is modern enough. exit 0 ;; --run) # Back-compat with the calling convention used by older automake. shift ;; -h|--h|--he|--hel|--help) echo "\ $0 [OPTION]... PROGRAM [ARGUMENT]... Run 'PROGRAM [ARGUMENT]...', returning a proper advice when this fails due to PROGRAM being missing or too old. Options: -h, --help display this help and exit -v, --version output version information and exit Supported PROGRAM values: aclocal autoconf autoheader autom4te automake makeinfo bison yacc flex lex help2man Version suffixes to PROGRAM as well as the prefixes 'gnu-', 'gnu', and 'g' are ignored when checking the name. Send bug reports to ." exit $? ;; -v|--v|--ve|--ver|--vers|--versi|--versio|--version) echo "missing $scriptversion (GNU Automake)" exit $? ;; -*) echo 1>&2 "$0: unknown '$1' option" echo 1>&2 "Try '$0 --help' for more information" exit 1 ;; esac # Run the given program, remember its exit status. "$@"; st=$? # If it succeeded, we are done. test $st -eq 0 && exit 0 # Also exit now if we it failed (or wasn't found), and '--version' was # passed; such an option is passed most likely to detect whether the # program is present and works. case $2 in --version|--help) exit $st;; esac # Exit code 63 means version mismatch. This often happens when the user # tries to use an ancient version of a tool on a file that requires a # minimum version. if test $st -eq 63; then msg="probably too old" elif test $st -eq 127; then # Program was missing. msg="missing on your system" else # Program was found and executed, but failed. Give up. exit $st fi perl_URL=http://www.perl.org/ flex_URL=http://flex.sourceforge.net/ gnu_software_URL=http://www.gnu.org/software program_details () { case $1 in aclocal|automake) echo "The '$1' program is part of the GNU Automake package:" echo "<$gnu_software_URL/automake>" echo "It also requires GNU Autoconf, GNU m4 and Perl in order to run:" echo "<$gnu_software_URL/autoconf>" echo "<$gnu_software_URL/m4/>" echo "<$perl_URL>" ;; autoconf|autom4te|autoheader) echo "The '$1' program is part of the GNU Autoconf package:" echo "<$gnu_software_URL/autoconf/>" echo "It also requires GNU m4 and Perl in order to run:" echo "<$gnu_software_URL/m4/>" echo "<$perl_URL>" ;; esac } give_advice () { # Normalize program name to check for. normalized_program=`echo "$1" | sed ' s/^gnu-//; t s/^gnu//; t s/^g//; t'` printf '%s\n' "'$1' is $msg." configure_deps="'configure.ac' or m4 files included by 'configure.ac'" case $normalized_program in autoconf*) echo "You should only need it if you modified 'configure.ac'," echo "or m4 files included by it." program_details 'autoconf' ;; autoheader*) echo "You should only need it if you modified 'acconfig.h' or" echo "$configure_deps." program_details 'autoheader' ;; automake*) echo "You should only need it if you modified 'Makefile.am' or" echo "$configure_deps." program_details 'automake' ;; aclocal*) echo "You should only need it if you modified 'acinclude.m4' or" echo "$configure_deps." program_details 'aclocal' ;; autom4te*) echo "You might have modified some maintainer files that require" echo "the 'autom4te' program to be rebuilt." program_details 'autom4te' ;; bison*|yacc*) echo "You should only need it if you modified a '.y' file." echo "You may want to install the GNU Bison package:" echo "<$gnu_software_URL/bison/>" ;; lex*|flex*) echo "You should only need it if you modified a '.l' file." echo "You may want to install the Fast Lexical Analyzer package:" echo "<$flex_URL>" ;; help2man*) echo "You should only need it if you modified a dependency" \ "of a man page." echo "You may want to install the GNU Help2man package:" echo "<$gnu_software_URL/help2man/>" ;; makeinfo*) echo "You should only need it if you modified a '.texi' file, or" echo "any other file indirectly affecting the aspect of the manual." echo "You might want to install the Texinfo package:" echo "<$gnu_software_URL/texinfo/>" echo "The spurious makeinfo call might also be the consequence of" echo "using a buggy 'make' (AIX, DU, IRIX), in which case you might" echo "want to install GNU make:" echo "<$gnu_software_URL/make/>" ;; *) echo "You might have modified some files without having the proper" echo "tools for further handling them. Check the 'README' file, it" echo "often tells you about the needed prerequisites for installing" echo "this package. You may also peek at any GNU archive site, in" echo "case some other package contains this missing '$1' program." ;; esac } give_advice "$1" | sed -e '1s/^/WARNING: /' \ -e '2,$s/^/ /' >&2 # Propagate the correct exit status (expected to be 127 for a program # not found, 63 for a program that failed due to version mismatch). exit $st # Local variables: # eval: (add-hook 'write-file-hooks 'time-stamp) # time-stamp-start: "scriptversion=" # time-stamp-format: "%:y-%02m-%02d.%02H" # time-stamp-time-zone: "UTC" # time-stamp-end: "; # UTC" # End: stacks-1.44/config/test-driver000755 000765 000024 00000011040 12750771222 017146 0ustar00catchenstaff000000 000000 #! /bin/sh # test-driver - basic testsuite driver script. scriptversion=2013-07-13.22; # UTC # Copyright (C) 2011-2014 Free Software Foundation, Inc. # # This program is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2, or (at your option) # any later version. # # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this program. If not, see . # As a special exception to the GNU General Public License, if you # distribute this file as part of a program that contains a # configuration script generated by Autoconf, you may include it under # the same distribution terms that you use for the rest of that program. # This file is maintained in Automake, please report # bugs to or send patches to # . # Make unconditional expansion of undefined variables an error. This # helps a lot in preventing typo-related bugs. set -u usage_error () { echo "$0: $*" >&2 print_usage >&2 exit 2 } print_usage () { cat <$log_file 2>&1 estatus=$? if test $enable_hard_errors = no && test $estatus -eq 99; then tweaked_estatus=1 else tweaked_estatus=$estatus fi case $tweaked_estatus:$expect_failure in 0:yes) col=$red res=XPASS recheck=yes gcopy=yes;; 0:*) col=$grn res=PASS recheck=no gcopy=no;; 77:*) col=$blu res=SKIP recheck=no gcopy=yes;; 99:*) col=$mgn res=ERROR recheck=yes gcopy=yes;; *:yes) col=$lgn res=XFAIL recheck=no gcopy=yes;; *:*) col=$red res=FAIL recheck=yes gcopy=yes;; esac # Report the test outcome and exit status in the logs, so that one can # know whether the test passed or failed simply by looking at the '.log' # file, without the need of also peaking into the corresponding '.trs' # file (automake bug#11814). echo "$res $test_name (exit status: $estatus)" >>$log_file # Report outcome to console. echo "${col}${res}${std}: $test_name" # Register the test result, and other relevant metadata. echo ":test-result: $res" > $trs_file echo ":global-test-result: $res" >> $trs_file echo ":recheck: $recheck" >> $trs_file echo ":copy-in-global-log: $gcopy" >> $trs_file # Local Variables: # mode: shell-script # sh-indentation: 2 # eval: (add-hook 'write-file-hooks 'time-stamp) # time-stamp-start: "scriptversion=" # time-stamp-format: "%:y-%02m-%02d.%02H" # time-stamp-time-zone: "UTC" # time-stamp-end: "; # UTC" # End: