Mercurial > repos > drosofff > ged_bowtie
changeset 2:22353286c3af draft default tip
Deleted selected files
| author | drosofff |
|---|---|
| date | Mon, 12 May 2014 12:04:45 -0400 |
| parents | eed2a141eb0c |
| children | |
| files | sRbowtie_wrapper.py sRbowtie_wrapper.xml |
| diffstat | 2 files changed, 0 insertions(+), 289 deletions(-) [+] |
line wrap: on
line diff
--- a/sRbowtie_wrapper.py Sun May 11 18:18:25 2014 -0400 +++ /dev/null Thu Jan 01 00:00:00 1970 +0000 @@ -1,106 +0,0 @@ -#!/usr/bin/env python -# generic bowtie wrapper for bowtie, small RNA oriented -# version 1 17-1-2014 -# Usage sRbowtie_wrapper.py <1 input_fasta_file> <2 alignment method> <3 -v mismatches> <4 out_type> <5 buildIndexIfHistory> <6 fasta/bowtie index> <7 bowtie output> <8 ali_fasta> <9 unali_fasta> - -import sys, re, os, subprocess, shlex, tempfile, shutil - -def stop_err( msg ): - sys.stderr.write( '%s\n' % msg ) - sys.exit() - -def bowtieCommandLiner (alignment_method, v_mis, out_type, aligned, unaligned, input, index, output): - if alignment_method=="RNA": - x = "-v %s -M 1 --best --strata -p 12 --norc --suppress 2,6,7,8" % v_mis - elif alignment_method=="unique": - x = "-v %s -m 1 -p 12 --suppress 6,7,8" % v_mis - elif alignment_method=="multiple": - x = "-v %s -M 1 --best --strata -p 12 --suppress 6,7,8" % v_mis - elif alignment_method=="k_option": - x = "-v %s -k 1 --best -p 12 --suppress 6,7,8" % v_mis - elif alignment_method=="n_option": - x = "-n %s -M 1 --best -p 12 --suppress 6,7,8" % v_mis - elif alignment_method=="a_option": - x = "-v %s -a --best -p 12 --suppress 6,7,8" % v_mis - if aligned == "None" and unaligned == "None": fasta_command = "" - elif aligned != "None" and unaligned == "None": fasta_command= " --al %s" % aligned - elif aligned == "None" and unaligned != "None": fasta_command = " --un %s" % unaligned - else: fasta_command = " --al %s --un %s" % (aligned, unaligned) - x = x + fasta_command - if out_type == "tabular": - return "bowtie %s %s -f %s > %s" % (x, index, input, output) - elif out_type=="sam": - return "bowtie %s -S %s -f %s > %s" % (x, index, input, output) - elif out_type=="bam": - return "bowtie %s -S %s -f %s |samtools view -bS - > %s" % (x, index, input, output) - -def bowtie_squash(fasta): - # make temp directory for placement of indices and copy reference file there if necessary - tmp_index_dir = tempfile.mkdtemp() - ref_file = tempfile.NamedTemporaryFile( dir=tmp_index_dir ) - ref_file_name = ref_file.name - ref_file.close() # by default, this action delete the temporary file ! - os.symlink( fasta, ref_file_name ) # now there is a symlink between the fasta source file and the deleted ref_file name - cmd1 = 'bowtie-build -f %s %s' % (ref_file_name, ref_file_name ) # this will work with the bowtie command line but we have to change the working dir ! - try: - FNULL = open(os.devnull, 'w') - tmp = tempfile.NamedTemporaryFile( dir=tmp_index_dir ).name # a full path temp file to tmp_index_dir is created. just a string - tmp_stderr = open( tmp, 'wb' ) # creates and open the file handler - proc = subprocess.Popen( args=cmd1, shell=True, cwd=tmp_index_dir, stderr=FNULL, stdout=FNULL ) # fileno() method return the file descriptor number of tmp_stderr / stderr=tmp_stderr.fileno() - # here I played a while a finish by redirecting everythin in dev/null. Clean later tmp_stderr calls - returncode = proc.wait() - tmp_stderr.close() - FNULL.close() - sys.stdout.write(cmd1 + "\n") - except Exception, e: - # clean up temp dir - if os.path.exists( tmp_index_dir ): - shutil.rmtree( tmp_index_dir ) - stop_err( 'Error indexing reference sequence\n' + str( e ) ) - # no Cleaning if no Exception, to be cleaned later after bowtie alignment - index_full_path = os.path.join(tmp_index_dir, ref_file_name) # bowtie fashion path (without extention) ... - return tmp_index_dir, index_full_path - -def bowtie_alignment(command_line, flyPreIndexed=''): - # make temp directory just for stderr - tmp_index_dir = tempfile.mkdtemp() - tmp = tempfile.NamedTemporaryFile( dir=tmp_index_dir ).name - tmp_stderr = open( tmp, 'wb' ) - # conditional statement for sorted bam generation viewable in Trackster - if "samtools" in command_line: - target_file = command_line.split()[-1] # recover the final output file name - path_to_unsortedBam = os.path.join(tmp_index_dir, "unsorted.bam") - path_to_sortedBam = os.path.join(tmp_index_dir, "unsorted.bam.sorted") - first_command_line = " ".join(command_line.split()[:-3]) + " -o " + path_to_unsortedBam + " - " # example: bowtie -v 0 -M 1 --best --strata -p 12 --suppress 6,7,8 -S /home/galaxy/galaxy-dist/bowtie/Dmel/dmel-all-chromosome-r5.49 -f /home/galaxy/galaxy-dist/database/files/003/dataset_3460.dat |samtools view -bS -o /tmp/tmp_PgMT0/unsorted.bam - - second_command_line = "samtools sort %s %s" % (path_to_unsortedBam, path_to_sortedBam) # Be carreful : this indeed will generate an unsorted.bam.sorted.bam file, NOT a unsorted.bam.sorted file - p = subprocess.Popen(args=first_command_line, cwd=tmp_index_dir, shell=True, stderr=tmp_stderr.fileno()) - returncode = p.wait() - sys.stdout.write("%s\n" % first_command_line + str(returncode)) - p = subprocess.Popen(args=second_command_line, cwd=tmp_index_dir, shell=True, stderr=tmp_stderr.fileno()) - returncode = p.wait() - sys.stdout.write("\n%s\n" % second_command_line + str(returncode)) - if os.path.isfile(path_to_sortedBam + ".bam"): - shutil.copy2(path_to_sortedBam + ".bam", target_file) - else: - p = subprocess.Popen(args=command_line, shell=True, stderr=tmp_stderr.fileno()) - returncode = p.wait() - sys.stdout.write(command_line + "\n") - tmp_stderr.close() - ## cleaning if the index was created in the fly - if os.path.exists( flyPreIndexed ): - shutil.rmtree( flyPreIndexed ) - # cleaning tmp files and directories - if os.path.exists( tmp_index_dir ): - shutil.rmtree( tmp_index_dir ) - return - -def __main__(): - F = open (sys.argv[-3], "w") - if sys.argv[5] == "history": - tmp_dir, index_path = bowtie_squash(sys.argv[6]) - else: - tmp_dir, index_path = "dummy/dymmy", sys.argv[6] - command_line = bowtieCommandLiner(sys.argv[2], sys.argv[3], sys.argv[4], sys.argv[-2], sys.argv[-1], sys.argv[1], index_path, sys.argv[7]) - bowtie_alignment(command_line, flyPreIndexed=tmp_dir) - F.close() -if __name__=="__main__": __main__()
--- a/sRbowtie_wrapper.xml Sun May 11 18:18:25 2014 -0400 +++ /dev/null Thu Jan 01 00:00:00 1970 +0000 @@ -1,183 +0,0 @@ -<tool id="sRbowtie" name="GED bowtie" version="1.0.0"> - <description>small RNA oriented</description> - <requirements><requirement type='package'>bowtie</requirement></requirements> - <parallelism method="basic"></parallelism> - <command interpreter="python"> sRbowtie_wrapper.py $input - $method - $v_mismatches - $output_type - $refGenomeSource.genomeSource - ## the very source of the index (indexed or fasta file) - #if $refGenomeSource.genomeSource == "history": - $refGenomeSource.ownFile - #else: - $refGenomeSource.index - #end if - ## - $output - $aligned - $unaligned - </command> - <inputs> - <param name="input" type="data" format="fasta" label="Input fasta file: reads clipped from their adapter" help="Only with clipped, raw fasta files"/> -<!-- which method will be used --> - <param name="method" type="select" label="What kind of matching do you want to do?" help="bowtie parameters adjusted to the type of matching. RNA option match to only one strand"> - <option value="RNA">Match on sense strand RNA reference index, multiple mappers randomly matched at a single position</option> - <option value="unique">Match unique mappers on DNA reference index</option> - <option value="multiple" selected="true">Match on DNA, multiple mappers randomly matched at a single position</option> - <option value="k_option">Match on DNA as fast as possible, without taking care of mapping issues (for raw annotation of reads)</option> - <option value="n_option">Match on DNA - RNAseq mode (-n bowtie option)</option> - <option value="a_option">Match and report all valid alignments</option> - </param> - -<!-- END of which method will be used --> - - <param name="v_mismatches" type="select" label="Number of mismatches allowed" help="specify the -v bowtie option"> - <option value="0">0</option> - <option value="1" selected="true">1</option> - <option value="2">2</option> - <option value="3">3</option> - </param> - - -<!-- nouvel index in dev below --> - <conditional name="refGenomeSource"> - <param name="genomeSource" type="select" label="Will you select a reference genome from your history or use a built-in index?" help="Built-ins were indexed using default options"> - <option value="indexed">Use a built-in index</option> - <option value="history">Use one from the history</option> - </param> - - <when value="indexed"> - <param name="index" type="select" label="Select a DNA reference index" help="if your genome of interest is not listed - contact GED team"> - <options from_data_table="ged_bowtie_indexes"/> - </param> - </when> - <when value="history"> - <param name="ownFile" type="data" format="fasta" label="Select a fasta file, to serve as index reference" /> - </when> - </conditional> -<!-- nouvel index input FIN --> - <param name="output_type" type="select" label="Select output format" help="Note that the BAM will be viewable in trackster only if you choose a full genome referenced for Trackster usage. see the doc below"> - <option value="tabular" select="true">tabular</option> - <option value="sam">sam</option> - <option value="bam">bam</option> - </param> - - <param name="additional_fasta" type="select" label="additional fasta output" help="to get aligned and unaligned reads in fasta format"> - <option value="No" select="true">No</option> - <option value="al">aligned</option> - <option value="unal">unaligned</option> - <option value="al_and_unal">both aligned and unaligned</option> - </param> - - </inputs> - <outputs> - <data format="tabular" name="output" label="Bowtie Output"> - <change_format> - <when input="output_type" value="sam" format="sam" /> - <when input="output_type" value="bam" format="bam" /> - </change_format> - </data> - - <data format="fasta" name="aligned" label="Matched reads"> - <filter>additional_fasta == "al" or additional_fasta == "al_and_unal"</filter> - </data> - <data format="fasta" name="unaligned" label ="Unmatched reads"> - <filter>additional_fasta == "unal" or additional_fasta == "al_and_unal"</filter> - </data> - </outputs> - - <help> - -**What it does** - -Bowtie_ is a short read aligner designed to be ultrafast and memory-efficient. It is developed by Ben Langmead and Cole Trapnell. Please cite: Langmead B, Trapnell C, Pop M, Salzberg SL. Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biology 10:R25. - -.. _Bowtie: http://bowtie-bio.sourceforge.net/index.shtml - -A generic "Map with Bowtie for Illumina" Galaxy tool is available in the main Galaxy distribution. - -However, this useful Bowtie wrapper tool only takes as inputs FASTQ files. - -Our sRbowtie wrapper is intented to work specifically with short reads FASTA inputs and to serve downstream small RNA sequencing analyses - ------- - -**OPTIONS** - -.. class:: infomark - -This script uses Bowtie to match reads on a reference index. - -Depending on the type of matching, different bowtie options are used: - -**Match on sense strand RNA reference index, multiple mappers randomly matched at a single position** - -Match on RNA reference, SENSE strand, randomly attributing multiple mapper to target with least mismatches, the polarity column is suppressed in the bowtie tabular report: - -*-v [0,1,2,3] -M 1 --best --strata -p 12 --norc --suppress 2,6,7,8* - -**Match unique mappers on DNA reference index** - -Match ONLY unique mappers on DNA reference index - -*-v [0,1,2,3] -m 1 -p 12 --suppress 6,7,8* - -Note that using this option with -v values other than 0 is questionnable... - -**Match on DNA, multiple mappers randomly matched at a single position** - -Match multiple mappers, randomly attributing multiple mapper to target with least mismatches, number of mismatch allowed specified by -v option: - -*-v [0,1,2,3] -M 1 --best --strata -p 12 --suppress 6,7,8* - -**Match on DNA as fast as possible, without taking care of mapping issues (for raw annotation of reads)** - -Match with highest speed, not guaranteeing best hit for speed gain: - -*-v [0,1,2,3] -k 1 --best -p 12 --suppress 6,7,8* - - ------ - -**Input formats** - -.. class:: warningmark - -*The only accepted format for the script is a raw fasta list of reads, clipped from their adapter* - ------ - -**OUTPUTS** - -If you choose tabular as the output format, you will obtain the matched reads in standard bowtie output format, having the following columns:: - - Column Description - -------- -------------------------------------------------------- - 1 FastaID fasta identifier - 2 polarity + or - depending whether the match was reported on the forward or reverse strand - 3 target name of the matched target - 4 Offset O-based coordinate of the miR on the miRBase pre-miR sequence - 5 Seq sequence of the matched Read - -If you choose SAM, you will get the output in unordered SAM format. - -.. class:: warningmark - -if you choose BAM, the output will be in sorted BAM format. -To be viewable in Trackster, several condition must be fulfilled: - -.. class:: infomark - -Reads must have been matched to a FULL genome whose chromosome names are compatible with Trackster genome indexes - -.. class:: infomark - -the database/Build (dbkey) which is indicated for the dataset (Pencil - Database/Build field) must match a Trackster genome index. - -Please contact the GED galaxy team is your reference genome is not referenced properly in GED galaxy - -**Optionnal matched and unmatched fasta reads can be obtained, for further annotations** - - </help> -</tool>