Mercurial > repos > msjeon > abyss_tool
changeset 0:ce99b6666d1f draft
Uploaded
author | msjeon |
---|---|
date | Mon, 11 Jun 2012 14:48:15 -0400 |
parents | |
children | f1a33991bbbb |
files | abyss_toolsuite-92636934a189/abyss/abyss-pe.xml abyss_toolsuite-92636934a189/abyss/abyss-pe_wrapper.pl abyss_toolsuite-92636934a189/abyss/abyss.xml abyss_toolsuite-92636934a189/abyss/abyss_wrapper.sh abyss_toolsuite-92636934a189/abyss/suite_config.xml |
diffstat | 5 files changed, 198 insertions(+), 0 deletions(-) [+] |
line wrap: on
line diff
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/abyss_toolsuite-92636934a189/abyss/abyss-pe.xml Mon Jun 11 14:48:15 2012 -0400 @@ -0,0 +1,58 @@ +<tool id="abyss-pe" name="Abyss Paired-End" version="1.0.0"> +<description>Assemble short paired reads</description> +<command interpreter='perl'>abyss-pe_wrapper.pl +$k +$n +$outfile.extra_files_path +$outfile +$contigs_outfile +$sam_outfile +$coverage_histogram_outfile +#for $i in $infiles +${i.infile} +#end for +</command> + +<inputs> + <repeat name="infiles" title="Paired Reads Files"> + <param name="infile" type="data" format="fasta,fastq,fastqsanger,fastqillumina" label="Paired read sequences"/> + </repeat> + <param name="k" type="integer" value="41" label="[-k] K-mer size" help="Try multiple sizes, starting around 2/3 read length" /> + <param name="n" type="integer" value="10" label="[-n] Min. num. pairs for scaffolding" help="Requirement for joining contigs into a scaffold" /> +</inputs> + +<outputs> + <data name="outfile" format="txt" /> + <data name="contigs_outfile" format="fasta" label="Contigs" /> + <data name="sam_outfile" format="sam" label="Read aligments (Sam)" /> + <data name="coverage_histogram_outfile" format="txt" label="Coverage histogram" /> +</outputs> + +<help> +**What it does** + +ABySS is a de novo, paired-end sequence assembler that is designed for short reads. + +**Input** + +The suffix of the read identifier for a pair of reads must be one of '1' and '2', or 'A' and 'B', or 'F' and 'R', or 'F3' and 'R3', or 'forward' and 'reverse'. The reads may be interleaved in the same file or found in different files; however, interleaved mates will use less memory. + +**Description** + +This tool performs the following commands: + +ABYSS - the single-end assembler +AdjList - finds overlaps of length k-1 between contigs +KAligner** - aligns reads to contigs +ParseAligns** - finds pairs of reads in alignments +DistanceEst** - estimates distances between contigs +Overlap - find overlaps between blunt contigs +SimpleGraph - finds paths between pairs of contigs +MergePaths - merges consistent paths +Consensus - for a colour-space assembly, convert the colour-space contigs to nucleotide contigs + +**Reference** + +http://www.bcgsc.ca/platform/bioinfo/software/abyss +</help> +</tool>
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/abyss_toolsuite-92636934a189/abyss/abyss-pe_wrapper.pl Mon Jun 11 14:48:15 2012 -0400 @@ -0,0 +1,90 @@ +#!/usr/bin/env perl + +use strict; +use warnings; +use File::Copy; + +die("ERROR: Expected at least 8 arguments; got: @ARGV\n") unless @ARGV >= 8; + +# GET ARGS +my $kmer_size=shift @ARGV; +my $min_num_pairs=shift @ARGV; +my $outdir=shift @ARGV; +my $outfile=shift @ARGV; +my $contigs_outfile=shift @ARGV; +my $sam_outfile=shift @ARGV; +my $hist_outfile=shift @ARGV; + +# ALL FILES GO IN THIS extra_files_path +unless (-d $outdir) { + mkdir $outdir or die("Unable to make dir, $outdir\n"); +} +chdir $outdir; + +# RUN COMMAND +`abyss-pe k=$kmer_size n=$min_num_pairs in='@ARGV' name=abyss 2> $outdir/abyss.stderr > $outdir/abyss.stdout`; +if ($? != 0) { + unless ( -s "$outdir/abyss-3.hist") { print STDERR "NO CONTIGS WERE PRODUCED!\n" } + open(IN, "<$outdir/abyss.stdout") or die($!); + while (<IN>) { print STDERR $_ } + close IN; + die("ABORTING\n"); +} + +# FILTER HISTOGRAM +open (IN, "<$outdir/abyss.stderr") or die($!); +open (OUT, ">$outfile") or die($!); +while (my $line=<IN>) { + my @chars=split(//, $line); + my $filter=0; + foreach my $char (@chars) { + if (ord($char) >= 129) { + $filter=1; + last; + } + } + print OUT $line unless $filter; +} +close IN; +close OUT; +unlink("$outdir/abyss.stderr"); + +# OUTPUT INFO LINE TEXT +open(IN, "<$outdir/abyss.stdout") or die($!); +while (<IN>) { + if (/^Assembled \d+ k\-mer in \d+ contigs/) { + print; + last; + } +} +close IN; + +# GALAXY DOESN'T WANT GZIPPED DATAFILES +run("gunzip $outdir/abyss-3.sam.gz"); + +# MOVE OUTFILES +mv_outfile("$outdir/abyss-contigs.fa", $contigs_outfile); +mv_outfile("$outdir/abyss-3.sam", $sam_outfile); +mv_outfile("$outdir/coverage.hist", $hist_outfile); +exit; + +sub run { + my $cmd=shift; + my $output=`$cmd 2>&1`; + if ($? != 0) { + print STDERR "ERROR RUNNING COMMAND: $cmd\n"; + die($output); + } + return $output; +} + + +sub mv_outfile { + my ($src,$dest)=@_; + # if dest defined and src exist, then move outfiles to galaxy-specified location + if ( $dest ne 'None' and -f $src ) { + unlink($dest); + move($src,$dest); + } +} +__END__
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/abyss_toolsuite-92636934a189/abyss/abyss.xml Mon Jun 11 14:48:15 2012 -0400 @@ -0,0 +1,25 @@ +<tool id="abyss" name="Abyss" version="1.0.0"> +<description>Assemble short unpaired reads</description> +<command interpreter='bash'>abyss_wrapper.sh $k $infile $outfile</command> + +<inputs> + <param name="infile" type="data" format="fasta|fastq" label="Unpaired read sequences" /> + <param name="k" type="integer" value="41" label="K-mer size" help="Try multiple sizes, starting around 2/3 read length" /> +</inputs> + +<outputs> + <data name="outfile" format="fasta" /> +</outputs> + +<help> +**What it does** + +ABySS is a de novo sequence assembler that is designed for short reads. + +.. image:: http://www.bcgsc.ca/platform/bioinfo/software/abyss/screenshot + +**Reference** + +http://www.bcgsc.ca/platform/bioinfo/software/abyss +</help> +</tool>
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/abyss_toolsuite-92636934a189/abyss/abyss_wrapper.sh Mon Jun 11 14:48:15 2012 -0400 @@ -0,0 +1,16 @@ +#!/bin/bash + +if [ $# -ne 3 ] +then + echo "ERROR: Expected exactly 3 arguments; got: $*" 1>&2 + exit 1 +fi + +CMD=`ABYSS -k$1 $2 -o $3 2>&1` +if [ $? -ne 0 ] +then +{ + echo "COMMAND FAILURE: $CMD" 1>&2 + exit $? +} +fi
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/abyss_toolsuite-92636934a189/abyss/suite_config.xml Mon Jun 11 14:48:15 2012 -0400 @@ -0,0 +1,9 @@ +<suite id='abyss_toolsuite' name='Suite of Abyss assembler tools' version="1.0.0"> + <description>This suite contains Abyss and Abyss-PE config files and wrappers for Galaxy</description> + <tool id="abyss" name="Abyss" version="1.0.0"> + <description>Abyss (single-end) assembler</description> + </tool> + <tool id="abyss-pe" name="Abyss Paired-End" version="1.0.0"> + <description>Abyss-PE (paired-end) assembler</description> + </tool> +</suite>