[gentoo-commits] proj/sci:master commit in: sci-biology/TransDecoder/files/, sci-biology/TransDecoder/

[Martin Mokrejs]

commit:     cb4121a23d10b5a0879682b261c0e55e483bc218
Author:     Martin Mokrejš <mmokrejs <AT> fold <DOT> natur <DOT> cuni <DOT> cz>
AuthorDate: Thu Apr 30 22:46:00 2015 +0000
Commit:     Martin Mokrejs <mmokrejs <AT> fold <DOT> natur <DOT> cuni <DOT> cz>
CommitDate: Thu Apr 30 22:46:00 2015 +0000
URL:        https://gitweb.gentoo.org/proj/sci.git/commit/?id=cb4121a2

sci-biology/TransDecoder: renamed package to keep original letter-casing

Package-Manager: portage-2.2.18

 sci-biology/TransDecoder/ChangeLog                 |  54 ++++++++
 sci-biology/TransDecoder/TransDecoder-2.0.1.ebuild |  69 +++++++++++
 sci-biology/TransDecoder/files/TransDecoder.patch  | 136 +++++++++++++++++++++
 .../TransDecoder/files/pfam_runner.pl.patch        |  20 +++
 sci-biology/TransDecoder/metadata.xml              |   9 ++
 5 files changed, 288 insertions(+)

diff --git a/sci-biology/TransDecoder/ChangeLog b/sci-biology/TransDecoder/ChangeLog
new file mode 100644
index 0000000..6fd8f5c
--- /dev/null
+++ b/sci-biology/TransDecoder/ChangeLog
@@ -0,0 +1,54 @@
+# ChangeLog for sci-biology/transdecoder
+# Copyright 1999-2015 Gentoo Foundation; Distributed under the GPL v2
+# $Header: $
+
+  27 Apr 2015; Martin Mokrejs <mmokrejs@×××××××××××××××.cz>
+  transdecoder-2.0.1.ebuild:
+  sci-biology/transdecoder: more ebuild cleanup
+
+  27 Apr 2015; Martin Mokrejs <mmokrejs@×××××××××××××××.cz>
+  transdecoder-2.0.1.ebuild:
+  sci-biology/transdecoder: ebuild cleanup
+
+  27 Apr 2015; Martin Mokrejs <mmokrejs@×××××××××××××××.cz>
+  -transdecoder-20140704.ebuild, transdecoder-2.0.1.ebuild:
+  sci-biology/transdecoder: drop old
+
+  17 Apr 2015; Martin Mokrejs <mmokrejs@×××××××××××××××.cz>
+  transdecoder-2.0.1.ebuild, transdecoder-20140704.ebuild:
+  sci-biology/transdecoder: fixed installation of perl-related files
+
+  19 Mar 2015; Martin Mokrejs <mmokrejs@×××××××××××××××.cz>
+  transdecoder-2.0.1.ebuild:
+  dropped hmmer dependency altogether, added pkg_postinst() with
+  usage/dependency info
+
+*transdecoder-2.0.1 (19 Mar 2015)
+
+  19 Mar 2015; Martin Mokrejs <mmokrejs@×××××××××××××××.cz>
+  +transdecoder-2.0.1.ebuild, transdecoder-20140704.ebuild:
+  removing KEYWORDS until the perl files are installed to some other place. At
+  the moment I get: 'perl-module.eclass: Suspicious environment values found.
+  PERL5LIB="/usr/lib64/perl5/vendor_perl/5.18.2/TransDecoder"'
+
+  15 Feb 2015; Martin Mokrejs <mmokrejs@×××××××××××××××.cz>
+  transdecoder-20140704.ebuild:
+  drop hmmer-3 dependency, it is an optional dependency only
+
+  10 Jan 2015; Martin Mokrejs <mmokrejs@×××××××××××××××.cz>
+  transdecoder-20140704.ebuild:
+  install *.pm into PERL5LIB/TransDecoder and pass that via env.d
+
+  09 Jan 2015; Martin Mokrejs <mmokrejs@×××××××××××××××.cz>
+  +files/TransDecoder.patch, +files/pfam_runner.pl.patch,
+  transdecoder-20140704.ebuild:
+  sci-biology/transdecoder: added patches so that we use PATH to loclate
+  binaries and not in a local subdirectory named 'util', drop sys-
+  cluster/openmpi requirement, it does not link against it all all, this is a
+  bunch of perl and shell scripts
+
+*transdecoder-20140704 (08 Jan 2015)
+
+  08 Jan 2015; Martin Mokrejs <mmokrejs@×××××××××××××××.cz> +metadata.xml,
+  +transdecoder-20140704.ebuild:
+  sci-biology/transdecoder: new package

diff --git a/sci-biology/TransDecoder/TransDecoder-2.0.1.ebuild b/sci-biology/TransDecoder/TransDecoder-2.0.1.ebuild
new file mode 100644
index 0000000..e8f5134
--- /dev/null
+++ b/sci-biology/TransDecoder/TransDecoder-2.0.1.ebuild
@@ -0,0 +1,69 @@
+# Copyright 1999-2015 Gentoo Foundation
+# Distributed under the terms of the GNU General Public License v2
+# $Header: $
+
+EAPI=5
+
+PERL_EXPORT_PHASE_FUNCTIONS=no
+inherit perl-module eutils toolchain-funcs
+
+DESCRIPTION="Extract ORF/CDS regions from FASTA sequences"
+HOMEPAGE="http://sourceforge.net/projects/transdecoder/"
+SRC_URI="https://github.com/TransDecoder/TransDecoder/archive/"${PV}".tar.gz -> ${P}.tar.gz"
+
+LICENSE="BSD-BroadInstitute"
+SLOT="0"
+KEYWORDS="~amd64"
+IUSE=""
+
+DEPEND=""
+RDEPEND="${DEPEND}
+	sci-biology/cd-hit
+	sci-biology/hmmer
+	sci-biology/parafly
+	sci-biology/ffindex"
+# cdhit-4.6.1 is a real dependency, at least hmmer is optional (also ncbi-tools++ is now used for ORF searches)
+
+S="${WORKDIR}"/TransDecoder-2.0.1
+
+##src_prepare(){
+#	#mv Makefile Makefile.old
+#	#epatch "${FILESDIR}"/TransDecoder.patch
+#	#epatch "${FILESDIR}"/pfam_runner.pl.patch
+#}
+
+# avoid fetching 1.5TB "${S}"/pfam/Pfam-AB.hmm.bin, see
+# "Re: [Transdecoder-users] Announcement: Transdecoder release r20140704" thread in archives
+#
+# you cna get it from http://downloads.sourceforge.net/project/transdecoder/Pfam-AB.hmm.bin
+
+src_install(){
+	dobin TransDecoder.Predict TransDecoder.LongOrfs
+	insinto /usr/share/${PN}/util
+	dobin util/*.pl
+	# zap the bundled cdhit binaries copied from transdecoder_plugins/cdhit/ to util/bin
+	rm -rf util/bin
+	#
+	#  * sci-biology/trinityrnaseq-20140413:0::science
+	# *      /usr/bin/Fasta_reader.pm
+	# *      /usr/bin/GFF3_utils.pm
+	# *      /usr/bin/Gene_obj.pm
+	# *      /usr/bin/Gene_obj_indexer.pm
+	# *      /usr/bin/Longest_orf.pm
+	# *      /usr/bin/Nuc_translator.pm
+	# *      /usr/bin/TiedHash.pm
+	#
+	perl_set_version
+	insinto ${VENDOR_LIB}/${PN}
+	dobin PerlLib/*.pm # BUG: install into /usr/bin but wanted to have it readable and executable in ${VENDOR_LIB}/${PN} instead
+	einfo "Fetch on your own:"
+	einfo "wget --mirror -nH -nd http://downloads.sourceforge.net/project/transdecoder/Pfam-AB.hmm.bin"
+	einfo "hmmpress Pfam-AB.hmm.bin"
+}
+
+pkg_postinst(){
+	einfo "It is recommended to use TransDecoder with hmmer-3 or at least NCBI blast"
+	einfo "from either sci-biology/ncbi-blast+ (released more often) or"
+	einfo "from sci-biology/ncbi-toolkit++ (huge bundle with releases and less frequent bugfixes)"
+	einfo "Author says the minimum requirement is sci-biology/cd-hit"
+}

diff --git a/sci-biology/TransDecoder/files/TransDecoder.patch b/sci-biology/TransDecoder/files/TransDecoder.patch
new file mode 100644
index 0000000..c0cff94
--- /dev/null
+++ b/sci-biology/TransDecoder/files/TransDecoder.patch
@@ -0,0 +1,136 @@
+--- /usr/bin/TransDecoder	2015-01-09 11:22:55.000000000 +0100
++++ TransDecoder	2015-01-09 14:31:44.095839522 +0100
+@@ -48,7 +48,7 @@
+  --prepare_pfam                         Prepare data for PFAM search and then quit (for running PFAM on HPC/computing cluster
+                                          with or without MPI )
+ 
+- --CPU <int>                            number of threads to use; (default: 2)
++ --CPU <int>                            number of threads to use; (default: 1)
+ 
+  --MPI                                  use MPI w/ execution of hmmscan
+ 
+@@ -76,7 +76,7 @@
+ 
+ =head1 PFAM 
+ 
+-You will need hmmer installed. Use hmmpress to prepare the database for hmmer.
++You will need hmmer installed. Use hmmpress from >=hmmer-3.0 to prepare the database for hmmer.
+  L<See|https://sourceforge.net/projects/transdecoder/files/Pfam-AB.hmm.bin> for downloading the database.
+ 
+ =head1 CD-HIT
+@@ -105,7 +105,6 @@
+ use Longest_orf;
+ 
+ my $UTIL_DIR = "$FindBin::RealBin/util";
+-$ENV{PATH} = "$UTIL_DIR/bin:$ENV{PATH}";
+ $ENV{LD_LIBRARY_PATH} .= ":$FindBin::RealBin/util/lib64";
+ 
+ my ($cd_hit_est_exec) = &check_program('cd-hit-est');
+@@ -124,7 +123,7 @@
+ my $verbose;
+ my $search_pfam = "";
+ my ($reuse,$pfam_out);
+-my $CPU = 2;
++my $CPU = 1;
+ my $RETAIN_LONG_ORFS = 900;
+ my $MPI = 0;
+ 
+@@ -330,15 +329,15 @@
+ my $top_cds_file = $train_file && -s $train_file ? $train_file : "$cds_file.top_${top_ORFs_train}_longest";
+ if (!-s $top_cds_file) {
+     # get longest entries
+-    my $cmd = "$UTIL_DIR/get_top_longest_fasta_entries.pl $cds_file $top_ORFs_train > $top_cds_file";
++    my $cmd = "get_top_longest_fasta_entries.pl $cds_file $top_ORFs_train > $top_cds_file";
+     
+     unless ($reuse && -s $top_cds_file){
+         if ($cd_hit_est_exec){
+             # to speed things up only check for redundancy up to 4x the number of entries we want
+             my $red_num = $top_ORFs_train * 4 ;
+-            &process_cmd("$UTIL_DIR/get_top_longest_fasta_entries.pl $cds_file $red_num > $workdir/redundant_top");
++            &process_cmd("get_top_longest_fasta_entries.pl $cds_file $red_num > $workdir/redundant_top");
+             &process_cmd("$cd_hit_est_exec -r 1 -i $workdir/redundant_top -o $workdir/redundant_top.nr90 -M 0 -T $CPU >/dev/null 2>/dev/null");
+-            &process_cmd("$UTIL_DIR/get_top_longest_fasta_entries.pl $workdir/redundant_top.nr90 $top_ORFs_train > $top_cds_file");
++            &process_cmd("get_top_longest_fasta_entries.pl $workdir/redundant_top.nr90 $top_ORFs_train > $top_cds_file");
+             unlink("$workdir/redundant_top");
+             unlink("$workdir/redundant_top.nr90");
+             unlink("$workdir/redundant_top.nr90.bak.clstr");
+@@ -349,20 +348,20 @@
+     }
+ }
+ 
+-$cmd = "$UTIL_DIR/compute_base_probs.pl $transcripts_file $TOP_STRAND_ONLY > $workdir/base_freqs.dat";
++$cmd = "compute_base_probs.pl $transcripts_file $TOP_STRAND_ONLY > $workdir/base_freqs.dat";
+ &process_cmd($cmd) unless $reuse && -s "$workdir/base_freqs.dat";
+ 
+ 
+ # get hexamer scores
+-#$cmd = "$UTIL_DIR/seq_n_background_to_logliklihood_vals.pl $top_cds_file $transcripts_file.random > hexamer.scores";
++#$cmd = "seq_n_background_to_logliklihood_vals.pl $top_cds_file $transcripts_file.random > hexamer.scores";
+ #&process_cmd($cmd) unless ($reuse && -s "hexamer.scores");
+ 
+-$cmd = "$UTIL_DIR/seq_n_baseprobs_to_logliklihood_vals.pl $top_cds_file $workdir/base_freqs.dat > $workdir/hexamer.scores";
++$cmd = "seq_n_baseprobs_to_logliklihood_vals.pl $top_cds_file $workdir/base_freqs.dat > $workdir/hexamer.scores";
+ &process_cmd($cmd) unless $reuse && -s "$workdir/hexamer.scores";
+ 
+ 
+ # score all cds entries
+-$cmd = "$UTIL_DIR/score_CDS_liklihood_all_6_frames.pl $cds_file $workdir/hexamer.scores > $cds_file.scores";
++$cmd = "score_CDS_liklihood_all_6_frames.pl $cds_file $workdir/hexamer.scores > $cds_file.scores";
+ &process_cmd($cmd) unless ($reuse && -s "$cds_file.scores");
+ 
+ 
+@@ -440,18 +439,18 @@
+ }
+ 
+ # index the current gff file:
+-$cmd = "$UTIL_DIR/index_gff3_files_by_isoform.pl $gff3_file";
++$cmd = "index_gff3_files_by_isoform.pl $gff3_file";
+ &process_cmd($cmd);
+ 
+ # retrieve the best entries:
+-$cmd = "$UTIL_DIR/gene_list_to_gff.pl $acc_file $gff3_file.inx > $cds_file.best_candidates.gff3";
++$cmd = "gene_list_to_gff.pl $acc_file $gff3_file.inx > $cds_file.best_candidates.gff3";
+ &process_cmd($cmd);
+ 
+ {
+     my $final_output_prefix = basename($transcripts_file) . ".transdecoder";
+     
+     # exclude shadow orfs (smaller orfs in different reading frame that are eclipsed by longer orfs)
+-    $cmd = "$UTIL_DIR/remove_eclipsed_ORFs.pl $cds_file.best_candidates.gff3 > $final_output_prefix.gff3";
++    $cmd = "remove_eclipsed_ORFs.pl $cds_file.best_candidates.gff3 > $final_output_prefix.gff3";
+     &process_cmd($cmd);
+     
+ 
+@@ -462,14 +461,14 @@
+     my $gff3_file = "$final_output_prefix.gff3";
+     my $bed_file = $gff3_file;
+     $bed_file =~ s/\.gff3$/\.bed/;
+-    $cmd = "$UTIL_DIR/gff3_file_to_bed.pl $gff3_file > $bed_file";
++    $cmd = "gff3_file_to_bed.pl $gff3_file > $bed_file";
+     &process_cmd($cmd);
+     
+     
+     # make a peptide file:
+     my $best_pep_file = $gff3_file;
+     $best_pep_file =~ s/\.gff3$/\.pep/;
+-    $cmd = "$UTIL_DIR/gff3_file_to_proteins.pl $gff3_file $transcripts_file > $best_pep_file";
++    $cmd = "gff3_file_to_proteins.pl $gff3_file $transcripts_file > $best_pep_file";
+     &process_cmd($cmd);
+ 
+ 
+@@ -477,13 +476,13 @@
+     # make a CDS file:
+     my $best_cds_file = $best_pep_file;
+     $best_cds_file =~ s/\.pep$/\.cds/;
+-    $cmd = "$UTIL_DIR/gff3_file_to_proteins.pl $gff3_file $transcripts_file CDS > $best_cds_file";
++    $cmd = "gff3_file_to_proteins.pl $gff3_file $transcripts_file CDS > $best_cds_file";
+     &process_cmd($cmd);
+ 
+     # make a CDS file:
+     my $best_cdna_file = $best_pep_file;
+     $best_cdna_file =~ s/\.pep$/\.mRNA/;
+-    $cmd = "$UTIL_DIR/gff3_file_to_proteins.pl $gff3_file $transcripts_file cDNA > $best_cdna_file";
++    $cmd = "gff3_file_to_proteins.pl $gff3_file $transcripts_file cDNA > $best_cdna_file";
+     &process_cmd($cmd);
+     
+ }

diff --git a/sci-biology/TransDecoder/files/pfam_runner.pl.patch b/sci-biology/TransDecoder/files/pfam_runner.pl.patch
new file mode 100644
index 0000000..7809b1a
--- /dev/null
+++ b/sci-biology/TransDecoder/files/pfam_runner.pl.patch
@@ -0,0 +1,20 @@
+--- /usr/bin/pfam_runner.pl	2015-01-09 11:22:55.000000000 +0100
++++ pfam_runner.pl	2015-01-09 14:25:43.385838579 +0100
+@@ -24,7 +24,7 @@
+ my $workdir;
+ my $verbose;
+ my ($reuse,$pfam_out);
+-my $CPU = 2;
++my $CPU = 1;
+ 
+ my $usage =  <<_EOH_;
+ 
+@@ -59,7 +59,7 @@
+ # -h                                     print this option menu and quit
+ # -v                                     verbose
+ #
+-# --CPU <int>                            number of threads to use; (default: 2)
++# --CPU <int>                            number of threads to use; (default: 1)
+ #
+ # --MPI                                  use MPI (via ffindex_apply_mpi)
+ #

diff --git a/sci-biology/TransDecoder/metadata.xml b/sci-biology/TransDecoder/metadata.xml
new file mode 100644
index 0000000..2bc8930
--- /dev/null
+++ b/sci-biology/TransDecoder/metadata.xml
@@ -0,0 +1,9 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE pkgmetadata SYSTEM "http://www.gentoo.org/dtd/metadata.dtd">
+<pkgmetadata>
+  <herd>sci-biology</herd>
+  <maintainer>
+    <email>mmokrejs@×××××××××××××××.cz</email>
+    <name>Martin Mokrejs</name>
+  </maintainer>
+</pkgmetadata>