commit: 4c3880bfe25f6b81461523e3f1f38b43614d51b9 Author: Martin Mokrejš <mmokrejs <AT> fold <DOT> natur <DOT> cuni <DOT> cz> AuthorDate: Sat May 23 21:51:58 2015 +0000 Commit: Martin Mokrejs <mmokrejs <AT> fold <DOT> natur <DOT> cuni <DOT> cz> CommitDate: Sat May 23 21:51:58 2015 +0000 URL: https://gitweb.gentoo.org/proj/sci.git/commit/?id=4c3880bf
sci-biology/amos: bring in upstream patch to cope with secondary alignments from 'bwa mem -M' Package-Manager: portage-2.2.18 sci-biology/abyss/ChangeLog | 17 ++++++++ sci-biology/abyss/abyss-1.3.3-r1.ebuild | 42 +++++++++++++++++++ sci-biology/abyss/abyss-1.3.4-r1.ebuild | 47 +++++++++++++++++++++ sci-biology/abyss/abyss-1.3.6-r1.ebuild | 48 ++++++++++++++++++++++ .../abyss/files/abyss-1.3.3-ac_prog_ar.patch | 18 ++++++++ sci-biology/abyss/files/abyss-1.3.3-gcc-4.7.patch | 15 +++++++ sci-biology/abyss/files/abyss-1.3.3-samtoafg.patch | 40 ++++++++++++++++++ sci-biology/abyss/files/abyss-1.3.4-gcc-4.7.patch | 15 +++++++ sci-biology/abyss/files/abyss-1.3.4-samtoafg.patch | 40 ++++++++++++++++++ .../abyss/files/abyss-1.3.6-ac_prog_ar.patch | 18 ++++++++ sci-biology/abyss/files/abyss-1.3.6-gcc-4.7.patch | 15 +++++++ sci-biology/abyss/files/abyss-1.3.6-samtoafg.patch | 40 ++++++++++++++++++ sci-biology/abyss/metadata.xml | 5 +++ 13 files changed, 360 insertions(+) diff --git a/sci-biology/abyss/ChangeLog b/sci-biology/abyss/ChangeLog new file mode 100644 index 0000000..c680959 --- /dev/null +++ b/sci-biology/abyss/ChangeLog @@ -0,0 +1,17 @@ +# ChangeLog for sci-biology/abyss +# Copyright 1999-2015 Gentoo Foundation; Distributed under the GPL v2 +# $Header: $ + +*abyss-1.3.3-r1 (23 May 2015) +*abyss-1.3.4-r1 (23 May 2015) +*abyss-1.3.6-r1 (23 May 2015) + + 23 May 2015; Martin Mokrejs <[email protected]> + +abyss-1.3.3-r1.ebuild, +abyss-1.3.4-r1.ebuild, +abyss-1.3.6-r1.ebuild, + +files/abyss-1.3.3-ac_prog_ar.patch, +files/abyss-1.3.3-gcc-4.7.patch, + +files/abyss-1.3.3-samtoafg.patch, +files/abyss-1.3.4-gcc-4.7.patch, + +files/abyss-1.3.4-samtoafg.patch, +files/abyss-1.3.6-ac_prog_ar.patch, + +files/abyss-1.3.6-gcc-4.7.patch, +files/abyss-1.3.6-samtoafg.patch, + +metadata.xml: + sci-biology/amos: bring in upstream patch to cope with secondary alignments + from 'bwa mem -M' diff --git a/sci-biology/abyss/abyss-1.3.3-r1.ebuild b/sci-biology/abyss/abyss-1.3.3-r1.ebuild new file mode 100644 index 0000000..b727861 --- /dev/null +++ b/sci-biology/abyss/abyss-1.3.3-r1.ebuild @@ -0,0 +1,42 @@ +# Copyright 1999-2015 Gentoo Foundation +# Distributed under the terms of the GNU General Public License v2 +# $Header: /var/cvsroot/gentoo-x86/sci-biology/abyss/abyss-1.3.3.ebuild,v 1.5 2013/08/11 14:55:22 jlec Exp $ + +EAPI="4" + +inherit autotools eutils toolchain-funcs + +DESCRIPTION="Assembly By Short Sequences - a de novo, parallel, paired-end sequence assembler" +HOMEPAGE="http://www.bcgsc.ca/platform/bioinfo/software/abyss/"; +SRC_URI="http://www.bcgsc.ca/downloads/abyss/${P}.tar.gz"; + +LICENSE="abyss" +SLOT="0" +IUSE="+mpi openmp" +KEYWORDS="~amd64 ~x86" + +DEPEND=" + dev-cpp/sparsehash + mpi? ( virtual/mpi )" +RDEPEND="${DEPEND}" + +# todo: --enable-maxk=N configure option +# todo: fix automagic mpi toggling + +src_prepare() { + tc-export AR + epatch \ + "${FILESDIR}"/${P}-gcc-4.7.patch \ + "${FILESDIR}"/${P}-ac_prog_ar.patch \ + "${FILESDIR}"/${P}-samtoafg.patch + + sed -i -e "s/-Werror//" configure.ac || die #365195 + sed -i -e "/dist_pkgdoc_DATA/d" Makefile.am || die + eautoreconf +} + +src_configure() { + econf \ + --docdir="${EPREFIX}/usr/share/doc/${PF}" \ + $(use_enable openmp) +} diff --git a/sci-biology/abyss/abyss-1.3.4-r1.ebuild b/sci-biology/abyss/abyss-1.3.4-r1.ebuild new file mode 100644 index 0000000..48d7671 --- /dev/null +++ b/sci-biology/abyss/abyss-1.3.4-r1.ebuild @@ -0,0 +1,47 @@ +# Copyright 1999-2015 Gentoo Foundation +# Distributed under the terms of the GNU General Public License v2 +# $Header: /var/cvsroot/gentoo-x86/sci-biology/abyss/abyss-1.3.4.ebuild,v 1.3 2013/08/11 14:55:22 jlec Exp $ + +EAPI=5 + +AUTOTOOLS_AUTORECONF=true + +inherit autotools-utils + +DESCRIPTION="Assembly By Short Sequences - a de novo, parallel, paired-end sequence assembler" +HOMEPAGE="http://www.bcgsc.ca/platform/bioinfo/software/abyss/"; +SRC_URI="http://www.bcgsc.ca/downloads/abyss/${P}.tar.gz"; + +LICENSE="abyss" +SLOT="0" +IUSE="+mpi openmp" +KEYWORDS="~amd64 ~x86" + +DEPEND=" + dev-cpp/sparsehash + mpi? ( virtual/mpi )" +RDEPEND="${DEPEND}" + +# todo: --enable-maxk=N configure option +# todo: fix automagic mpi toggling + +PATCHES=( + "${FILESDIR}"/${P}-gcc-4.7.patch + "${FILESDIR}"/${PN}-1.3.3-ac_prog_ar.patch + "${FILESDIR}"/${P}-samtoafg.patch + ) + +src_prepare() { + tc-export AR + sed -i -e "s/-Werror//" configure.ac || die #365195 + sed -i -e "/dist_pkgdoc_DATA/d" Makefile.am || die + autotools-utils_src_prepare +} + +src_configure() { + local myeconfargs=( + --docdir="${EPREFIX}/usr/share/doc/${PF}" + $(use_enable openmp) + ) + autotools-utils_src_configure +} diff --git a/sci-biology/abyss/abyss-1.3.6-r1.ebuild b/sci-biology/abyss/abyss-1.3.6-r1.ebuild new file mode 100644 index 0000000..d5d019e --- /dev/null +++ b/sci-biology/abyss/abyss-1.3.6-r1.ebuild @@ -0,0 +1,48 @@ +# Copyright 1999-2015 Gentoo Foundation +# Distributed under the terms of the GNU General Public License v2 +# $Header: /var/cvsroot/gentoo-x86/sci-biology/abyss/abyss-1.3.6.ebuild,v 1.2 2014/07/06 11:12:45 jlec Exp $ + +EAPI=5 + +AUTOTOOLS_AUTORECONF=true + +inherit autotools-utils + +DESCRIPTION="Assembly By Short Sequences - a de novo, parallel, paired-end sequence assembler" +HOMEPAGE="http://www.bcgsc.ca/platform/bioinfo/software/abyss/"; +SRC_URI="http://www.bcgsc.ca/downloads/abyss/${P}.tar.gz"; + +LICENSE="abyss" +SLOT="0" +IUSE="+mpi openmp" +KEYWORDS="~amd64 ~x86" + +DEPEND=" + dev-cpp/sparsehash + dev-libs/boost + mpi? ( virtual/mpi )" +RDEPEND="${DEPEND}" + +# todo: --enable-maxk=N configure option +# todo: fix automagic mpi toggling + +PATCHES=( + "${FILESDIR}"/${P}-gcc-4.7.patch + "${FILESDIR}"/${P}-ac_prog_ar.patch + "${FILESDIR}"/${P}-samtoafg.patch + ) + +src_prepare() { + tc-export AR + sed -i -e "s/-Werror//" configure.ac || die #365195 + sed -i -e "/dist_pkgdoc_DATA/d" Makefile.am || die + autotools-utils_src_prepare +} + +src_configure() { + local myeconfargs=( + --docdir="${EPREFIX}/usr/share/doc/${PF}" + $(use_enable openmp) + ) + autotools-utils_src_configure +} diff --git a/sci-biology/abyss/files/abyss-1.3.3-ac_prog_ar.patch b/sci-biology/abyss/files/abyss-1.3.3-ac_prog_ar.patch new file mode 100644 index 0000000..300868f --- /dev/null +++ b/sci-biology/abyss/files/abyss-1.3.3-ac_prog_ar.patch @@ -0,0 +1,18 @@ + configure.ac | 4 ++++ + 1 file changed, 4 insertions(+) + +diff --git a/configure.ac b/configure.ac +index 5c6cb92..b99bedd 100644 +--- a/configure.ac ++++ b/configure.ac +@@ -12,6 +12,10 @@ AC_PROG_CPP + AC_PROG_CXX + AC_PROG_INSTALL + AC_PROG_RANLIB ++AN_MAKEVAR([AR], [AC_PROG_AR]) ++AN_PROGRAM([ar], [AC_PROG_AR]) ++AC_DEFUN([AC_PROG_AR], [AC_CHECK_TOOL(AR, ar, :)]) ++AC_PROG_AR + + # Checks for header files. + AC_CHECK_HEADERS([dlfcn.h fcntl.h float.h limits.h \ diff --git a/sci-biology/abyss/files/abyss-1.3.3-gcc-4.7.patch b/sci-biology/abyss/files/abyss-1.3.3-gcc-4.7.patch new file mode 100644 index 0000000..42066f7 --- /dev/null +++ b/sci-biology/abyss/files/abyss-1.3.3-gcc-4.7.patch @@ -0,0 +1,15 @@ + ParseAligns/abyss-fixmate.cc | 1 + + 1 files changed, 1 insertions(+), 0 deletions(-) + +diff --git a/ParseAligns/abyss-fixmate.cc b/ParseAligns/abyss-fixmate.cc +index 506ea0c..a0a403c 100644 +--- a/ParseAligns/abyss-fixmate.cc ++++ b/ParseAligns/abyss-fixmate.cc +@@ -15,6 +15,7 @@ + #include <iterator> + #include <sstream> + #include <string> ++#include <unistd.h> + + using namespace std; + diff --git a/sci-biology/abyss/files/abyss-1.3.3-samtoafg.patch b/sci-biology/abyss/files/abyss-1.3.3-samtoafg.patch new file mode 100644 index 0000000..cb7dfb0 --- /dev/null +++ b/sci-biology/abyss/files/abyss-1.3.3-samtoafg.patch @@ -0,0 +1,40 @@ +Hello, + +I see Shaun Jackmans' (abyss-)sam2afg script is available within the +Amos source repository. I find this very useful for coercing output from +assemblers which which don't track read location into amos by remapping +reads against the assembly and converting to afg - not ideal but better +than nothing when you don't have the 'real' read locations to work with. + +This fails, however when reads are aligned with 'bwa mem', which can +output multi-part alignments. sam2afg checks for reuse of the same read +id (presumably to prevent the generation of non-unique eid values), +consequently encountering multiple alignments for a read causes it to die. + +The following one-line patch allows sam2afg to skip these secondary +alignments present in 'bwa mem' output, provided bwa mem has been run +with the '-M' argument which sets the SAM 'secondary alignment' flag on +the alignments in question. + +Hopefully this will also be of use to others... + +Best Regards, +James + +-- +Dr. James Abbott +Lead Bioinformatician +Bioinformatics Support Service +Imperial College, London + + +--- abyss-1.3.6/bin/abyss-samtoafg.ori 2015-05-23 23:43:46.797747928 +0200 ++++ abyss-1.3.6/bin/abyss-samtoafg 2015-05-23 23:44:09.227747743 +0200 +@@ -105,6 +105,7 @@ + die unless defined $qqual; + + $tstart--; # convert to zero-based coordinate ++ next if $flag & 0x100; # secondary alignment + $qid .= "/1" if $flag & 0x40; #FREAD1 + $qid .= "/2" if $flag & 0x80; #FREAD2 + diff --git a/sci-biology/abyss/files/abyss-1.3.4-gcc-4.7.patch b/sci-biology/abyss/files/abyss-1.3.4-gcc-4.7.patch new file mode 100644 index 0000000..c2cc35c --- /dev/null +++ b/sci-biology/abyss/files/abyss-1.3.4-gcc-4.7.patch @@ -0,0 +1,15 @@ + ParseAligns/abyss-fixmate.cc | 1 + + 1 files changed, 1 insertions(+), 0 deletions(-) + +diff --git a/ParseAligns/abyss-fixmate.cc b/ParseAligns/abyss-fixmate.cc +index 1a169cf..36cc05b 100644 +--- a/ParseAligns/abyss-fixmate.cc ++++ b/ParseAligns/abyss-fixmate.cc +@@ -16,6 +16,7 @@ + #include <iterator> + #include <sstream> + #include <string> ++#include <unistd.h> + + using namespace std; + diff --git a/sci-biology/abyss/files/abyss-1.3.4-samtoafg.patch b/sci-biology/abyss/files/abyss-1.3.4-samtoafg.patch new file mode 100644 index 0000000..cb7dfb0 --- /dev/null +++ b/sci-biology/abyss/files/abyss-1.3.4-samtoafg.patch @@ -0,0 +1,40 @@ +Hello, + +I see Shaun Jackmans' (abyss-)sam2afg script is available within the +Amos source repository. I find this very useful for coercing output from +assemblers which which don't track read location into amos by remapping +reads against the assembly and converting to afg - not ideal but better +than nothing when you don't have the 'real' read locations to work with. + +This fails, however when reads are aligned with 'bwa mem', which can +output multi-part alignments. sam2afg checks for reuse of the same read +id (presumably to prevent the generation of non-unique eid values), +consequently encountering multiple alignments for a read causes it to die. + +The following one-line patch allows sam2afg to skip these secondary +alignments present in 'bwa mem' output, provided bwa mem has been run +with the '-M' argument which sets the SAM 'secondary alignment' flag on +the alignments in question. + +Hopefully this will also be of use to others... + +Best Regards, +James + +-- +Dr. James Abbott +Lead Bioinformatician +Bioinformatics Support Service +Imperial College, London + + +--- abyss-1.3.6/bin/abyss-samtoafg.ori 2015-05-23 23:43:46.797747928 +0200 ++++ abyss-1.3.6/bin/abyss-samtoafg 2015-05-23 23:44:09.227747743 +0200 +@@ -105,6 +105,7 @@ + die unless defined $qqual; + + $tstart--; # convert to zero-based coordinate ++ next if $flag & 0x100; # secondary alignment + $qid .= "/1" if $flag & 0x40; #FREAD1 + $qid .= "/2" if $flag & 0x80; #FREAD2 + diff --git a/sci-biology/abyss/files/abyss-1.3.6-ac_prog_ar.patch b/sci-biology/abyss/files/abyss-1.3.6-ac_prog_ar.patch new file mode 100644 index 0000000..158e9b1 --- /dev/null +++ b/sci-biology/abyss/files/abyss-1.3.6-ac_prog_ar.patch @@ -0,0 +1,18 @@ + configure.ac | 4 ++++ + 1 file changed, 4 insertions(+) + +diff --git a/configure.ac b/configure.ac +index 9d4bb66..aa94364 100644 +--- a/configure.ac ++++ b/configure.ac +@@ -12,6 +12,10 @@ AC_PROG_CPP + AC_PROG_CXX + AC_PROG_INSTALL + AC_PROG_RANLIB ++AN_MAKEVAR([AR], [AC_PROG_AR]) ++AN_PROGRAM([ar], [AC_PROG_AR]) ++AC_DEFUN([AC_PROG_AR], [AC_CHECK_TOOL(AR, ar, :)]) ++AC_PROG_AR + AC_CHECK_TOOL(GHC, ghc) + AM_CONDITIONAL([HAVE_GHC], [test "$GHC"]) + diff --git a/sci-biology/abyss/files/abyss-1.3.6-gcc-4.7.patch b/sci-biology/abyss/files/abyss-1.3.6-gcc-4.7.patch new file mode 100644 index 0000000..c2cc35c --- /dev/null +++ b/sci-biology/abyss/files/abyss-1.3.6-gcc-4.7.patch @@ -0,0 +1,15 @@ + ParseAligns/abyss-fixmate.cc | 1 + + 1 files changed, 1 insertions(+), 0 deletions(-) + +diff --git a/ParseAligns/abyss-fixmate.cc b/ParseAligns/abyss-fixmate.cc +index 1a169cf..36cc05b 100644 +--- a/ParseAligns/abyss-fixmate.cc ++++ b/ParseAligns/abyss-fixmate.cc +@@ -16,6 +16,7 @@ + #include <iterator> + #include <sstream> + #include <string> ++#include <unistd.h> + + using namespace std; + diff --git a/sci-biology/abyss/files/abyss-1.3.6-samtoafg.patch b/sci-biology/abyss/files/abyss-1.3.6-samtoafg.patch new file mode 100644 index 0000000..cb7dfb0 --- /dev/null +++ b/sci-biology/abyss/files/abyss-1.3.6-samtoafg.patch @@ -0,0 +1,40 @@ +Hello, + +I see Shaun Jackmans' (abyss-)sam2afg script is available within the +Amos source repository. I find this very useful for coercing output from +assemblers which which don't track read location into amos by remapping +reads against the assembly and converting to afg - not ideal but better +than nothing when you don't have the 'real' read locations to work with. + +This fails, however when reads are aligned with 'bwa mem', which can +output multi-part alignments. sam2afg checks for reuse of the same read +id (presumably to prevent the generation of non-unique eid values), +consequently encountering multiple alignments for a read causes it to die. + +The following one-line patch allows sam2afg to skip these secondary +alignments present in 'bwa mem' output, provided bwa mem has been run +with the '-M' argument which sets the SAM 'secondary alignment' flag on +the alignments in question. + +Hopefully this will also be of use to others... + +Best Regards, +James + +-- +Dr. James Abbott +Lead Bioinformatician +Bioinformatics Support Service +Imperial College, London + + +--- abyss-1.3.6/bin/abyss-samtoafg.ori 2015-05-23 23:43:46.797747928 +0200 ++++ abyss-1.3.6/bin/abyss-samtoafg 2015-05-23 23:44:09.227747743 +0200 +@@ -105,6 +105,7 @@ + die unless defined $qqual; + + $tstart--; # convert to zero-based coordinate ++ next if $flag & 0x100; # secondary alignment + $qid .= "/1" if $flag & 0x40; #FREAD1 + $qid .= "/2" if $flag & 0x80; #FREAD2 + diff --git a/sci-biology/abyss/metadata.xml b/sci-biology/abyss/metadata.xml new file mode 100644 index 0000000..f17a827 --- /dev/null +++ b/sci-biology/abyss/metadata.xml @@ -0,0 +1,5 @@ +<?xml version="1.0" encoding="UTF-8"?> +<!DOCTYPE pkgmetadata SYSTEM "http://www.gentoo.org/dtd/metadata.dtd";> +<pkgmetadata> + <herd>sci-biology</herd> +</pkgmetadata>
