changeset 0:f2310e26012f draft

Uploaded

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/htseq_bams_to_count_matrix/htseqsams2mx.py	Thu Jun 06 08:18:03 2013 -0400
@@ -0,0 +1,359 @@
+# May 2013
+# Change to htseq as the counting engine - wrap so arbitrary number of columns created  
+# borged Simon Anders' "count.py" since we need a vector of counts rather than a new sam file as output
+# note attribution for htseq and count.py :
+## Written by Simon Anders ([e-mail]), European Molecular Biology
+## Laboratory (EMBL). (c) 2010. Released under the terms of the GNU General
+## Public License v3. Part of the 'HTSeq' framework, version HTSeq-0.5.4p3
+# updated ross lazarus august 2011 to NOT include region and to finesse the name as the region for bed3 format inputs
+# also now sums all duplicate named regions and provides a summary of any collapsing as the info
+# updated ross lazarus july 26 to respect the is_duplicate flag rather than try to second guess
+# note Heng Li argues that removing dupes is a bad idea for RNA seq
+# updated ross lazarus july 22 to count reads OUTSIDE each bed region during the processing of each bam
+# added better sorting with decoration of a dict key later sorted and undecorated.
+# code cleaned up and galaxified ross lazarus july 18 et seq.
+# bams2mx.py -turns a series of bam and a bed file into a matrix of counts Usage bams2mx.py <halfwindow> <bedfile.bed> <bam1.bam> 
+# <bam2.bam>
+# uses pysam to read and count bam reads over each bed interval for each sample for speed
+# still not so fast
+# TODO options -shift -unique
+#
+"""
+how this gets run:
+
+(vgalaxy)galaxy@iaas1-int:~$ cat database/job_working_directory/027/27014/galaxy_27014.sh
+#!/bin/sh
+GALAXY_LIB="/data/extended/galaxy/lib"
+if [ "$GALAXY_LIB" != "None" ]; then
+    if [ -n "$PYTHONPATH" ]; then
+        PYTHONPATH="$GALAXY_LIB:$PYTHONPATH"
+    else
+        PYTHONPATH="$GALAXY_LIB"
+    fi
+    export PYTHONPATH
+fi
+
+cd /data/extended/galaxy/database/job_working_directory/027/27014
+python /data/extended/galaxy/tools/rgenetics/htseqsams2mx.py -g "/data/extended/galaxy/database/files/034/dataset_34115.dat" -o "/data/extended/galaxy/database/files/034/dataset_34124.dat" -m "union" --id_attribute "gene_id" --feature_type "exon"     --samf "'/data/extended/galaxy/database/files/033/dataset_33980.dat','T5A_C1PPHACXX_AGTTCC_L003_R1.fastq_bwa.sam'"     --samf "'/data/extended/galaxy/database/files/033/dataset_33975.dat','T5A_C1PPHACXX_AGTTCC_L002_R1.fastq_bwa.sam'"; cd /data/extended/galaxy; /data/extended/galaxy/set_metadata.sh ./database/files /data/extended/galaxy/database/job_working_directory/027/27014 . /data/extended/galaxy/universe_wsgi.ini /data/tmp/tmpmwsElH /data/extended/galaxy/database/job_working_directory/027/27014/galaxy.json /data/extended/galaxy/database/job_working_directory/027/27014/metadata_in_HistoryDatasetAssociation_45202_sfOMGa,/data/extended/galaxy/database/job_working_directory/027/27014/metadata_kwds_HistoryDatasetAssociation_45202_gaMnxa,/data/extended/galaxy/database/job_working_directory/027/27014/metadata_out_HistoryDatasetAssociation_45202_kZPsZO,/data/extended/galaxy/database/job_working_directory/027/27014/metadata_results_HistoryDatasetAssociation_45202_bXU7IU,,/data/extended/galaxy/database/job_working_directory/027/27014/metadata_override_HistoryDatasetAssociation_45202_hyLAvh
+echo $? > /data/extended/galaxy/database/job_working_directory/027/27014/galaxy_27014.ec
+
+"""
+
+import os
+import re
+import sys 
+import HTSeq.scripts.count as htcount
+import optparse 
+import tempfile 
+import shutil
+import operator
+import subprocess
+import itertools
+import warnings
+import traceback
+import HTSeq
+import time
+
+
+class Xcpt(Exception):
+    def __init__(self, msg):
+        self.msg = msg
+
+
+def keynat(s=None):
+     '''
+     borrowed from http://code.activestate.com/recipes/285264-natural-string-sorting/
+     A natural sort helper function for sort() and sorted()
+     without using regular expressions or exceptions.
+     >>> items = ('Z', 'a', '10th', '1st', '9') sorted(items)
+     ['10th', '1st', '9', 'Z', 'a']
+     >>> sorted(items, key=keynat)
+     ['1st', '9', '10th', 'a', 'Z']
+     '''
+     if type(s) == type([]) or type(s) == type(()) :
+         s = s[0]
+     it = type(1)
+     r = []
+     for c in s:
+         if c.isdigit():
+             d = int(c)
+             if r and type( r[-1] ) == it:
+                 r[-1] = r[-1] * 10 + d
+             else:
+                 r.append(d)
+         else:
+             r.append(c.lower())
+     return r
+
+
+def sort_table(table, cols):
+    """ sort a table by multiple columns
+        table: a list of lists (or tuple of tuples) where each inner list 
+               represents a row
+        cols:  a list (or tuple) specifying the column numbers to sort by
+               e.g. (1,0) would sort by column 1, then by column 0
+    """
+    for col in reversed(cols):
+        table = sorted(table, key=operator.itemgetter(col))
+    return table
+
+
+
+def htseqMX(gff_filename,sam_filenames,colnames,opts):
+
+    class UnknownChrom( Exception ):
+       pass
+       
+    def my_showwarning( message, category, filename, lineno = None, line = None ):
+       sys.stderr.write( "Warning: %s\n" % message )      
+        
+    def invert_strand( iv ):
+       iv2 = iv.copy()
+       if iv2.strand == "+":
+          iv2.strand = "-"
+       elif iv2.strand == "-":
+          iv2.strand = "+"
+       else:
+          raise ValueError, "Illegal strand"
+       return iv2
+
+    def count_reads_in_features( sam_filenames, colnames, gff_filename, opts ):
+       """ Hacked version of htseq count.py
+       """
+       if opts.quiet:
+          warnings.filterwarnings( action="ignore", module="HTSeq" ) 
+       features = HTSeq.GenomicArrayOfSets( "auto", opts.stranded != "no" )
+       mapqMin = int(opts.mapqMin)       
+       counts = {}
+       empty = 0
+       ambiguous = 0
+       notaligned = 0
+       lowqual = 0
+       nonunique = 0          
+       gff = HTSeq.GFF_Reader( gff_filename )   
+       try:
+          for i,f in enumerate(gff):
+             if f.type == opts.feature_type:
+                try:
+                   feature_id = f.attr[ opts.id_attribute ]
+                except KeyError:
+                   sys.exit( "Feature at row %d %s does not contain a '%s' attribute" % 
+                      ( (i+1), f.name, opts.id_attribute ) )
+                if opts.stranded != "no" and f.iv.strand == ".":
+                   sys.exit( "Feature %s at %s does not have strand information but you are "
+                      "running htseq-count in stranded mode. Use '--stranded=no'." % 
+                      ( f.name, f.iv ) )
+                features[ f.iv ] += feature_id
+                counts[ f.attr[ opts.id_attribute ] ] = [0 for x in colnames] # we use sami as an index here to bump counts later
+       except:
+          sys.stderr.write( "Error occured in %s.\n" % gff.get_line_number_string() )
+          raise
+          
+       if not opts.quiet:
+          sys.stdout.write( "%d GFF lines processed.\n" % i )
+          
+       if len( counts ) == 0 and not opts.quiet:
+          sys.stdout.write( "Warning: No features of type '%s' found.\n" % opts.feature_type )
+       for sami,sam_filename in enumerate(sam_filenames):
+           colname = colnames[sami]
+           try:
+              read_seq = HTSeq.SAM_Reader( sam_filename )
+              first_read = iter(read_seq).next()
+              pe_mode = first_read.paired_end
+           except:
+              sys.stderr.write( "Error occured when reading first line of sam file %s\n" % sam_filename )
+              raise
+
+           try:
+              if pe_mode:
+                 read_seq_pe_file = read_seq
+                 read_seq = HTSeq.pair_SAM_alignments( read_seq )
+              for seqi,r in enumerate(read_seq):
+                 if not pe_mode:
+                    if not r.aligned:
+                       notaligned += 1
+                       continue
+                    try:
+                       if r.optional_field( "NH" ) > 1:
+                          nonunique += 1
+                          continue
+                    except KeyError:
+                       pass
+                    if r.aQual < mapqMin:
+                       lowqual += 1
+                       continue
+                    if opts.stranded != "reverse":
+                       iv_seq = ( co.ref_iv for co in r.cigar if co.type == "M" and co.size > 0 )
+                    else:
+                       iv_seq = ( invert_strand( co.ref_iv ) for co in r.cigar if co.type == "M" and co.size > 0 )            
+                 else:
+                    if r[0] is not None and r[0].aligned:
+                       if opts.stranded != "reverse":
+                          iv_seq = ( co.ref_iv for co in r[0].cigar if co.type == "M" and co.size > 0 )
+                       else:
+                          iv_seq = ( invert_strand( co.ref_iv ) for co in r[0].cigar if co.type == "M" and co.size > 0 )
+                    else:
+                       iv_seq = tuple()
+                    if r[1] is not None and r[1].aligned:            
+                       if opts.stranded != "reverse":
+                          iv_seq = itertools.chain( iv_seq, 
+                             ( invert_strand( co.ref_iv ) for co in r[1].cigar if co.type == "M" and co.size > 0 ) )
+                       else:
+                          iv_seq = itertools.chain( iv_seq, 
+                             ( co.ref_iv for co in r[1].cigar if co.type == "M" and co.size > 0 ) )
+                    else:
+                       if ( r[0] is None ) or not ( r[0].aligned ):
+                          notaligned += 1
+                          continue         
+                    try:
+                       if ( r[0] is not None and r[0].optional_field( "NH" ) > 1 ) or \
+                             ( r[1] is not None and r[1].optional_field( "NH" ) > 1 ):
+                          nonunique += 1
+                          continue
+                    except KeyError:
+                       pass
+                    if ( r[0] and r[0].aQual < mapqMin ) or ( r[1] and r[1].aQual < mapqMin ):
+                       lowqual += 1
+                       continue         
+                 
+                 try:
+                    if opts.mode == "union":
+                       fs = set()
+                       for iv in iv_seq:
+                          if iv.chrom not in features.chrom_vectors:
+                             raise UnknownChrom
+                          for iv2, fs2 in features[ iv ].steps():
+                             fs = fs.union( fs2 )
+                    elif opts.mode == "intersection-strict" or opts.mode == "intersection-nonempty":
+                       fs = None
+                       for iv in iv_seq:
+                          if iv.chrom not in features.chrom_vectors:
+                             raise UnknownChrom
+                          for iv2, fs2 in features[ iv ].steps():
+                             if len(fs2) > 0 or opts.mode == "intersection-strict":
+                                if fs is None:
+                                   fs = fs2.copy()
+                                else:
+                                   fs = fs.intersection( fs2 )
+                    else:
+                       sys.exit( "Illegal overlap mode %s" % opts.mode )
+                    if fs is None or len( fs ) == 0:
+                       empty += 1
+                    elif len( fs ) > 1:
+                       ambiguous += 1
+                    else:
+                       ck = list(fs)[0]  
+                       counts[ck][sami] += 1 # end up with counts for each sample as a list
+                 except UnknownChrom:
+                    if not pe_mode:
+                       rr = r 
+                    else: 
+                       rr = r[0] if r[0] is not None else r[1]
+                    empty += 1
+                    #if not quiet:
+                    #   sys.stderr.write( ( "Warning: Skipping read '%s', because chromosome " +
+                    #      "'%s', to which it has been aligned, did not appear in the GFF file.\n" ) % 
+                    #      ( rr.read.name, iv.chrom ) )
+           except:
+              if not pe_mode:
+                 sys.stderr.write( "Error occured in %s.\n" % read_seq.get_line_number_string() )
+              else:
+                 sys.stderr.write( "Error occured in %s.\n" % read_seq_pe_file.get_line_number_string() )
+              raise
+
+           if not opts.quiet:
+              sys.stdout.write( "%d sam %s processed.\n" % ( sami, "lines " if not pe_mode else "line pairs" ) )
+       return counts,empty,ambiguous,lowqual,notaligned,nonunique
+
+    warnings.showwarning = my_showwarning
+    assert os.path.isfile(gff_filename),'## unable to open supplied gff file %s' % gff_filename
+    try:
+        counts,empty,ambiguous,lowqual,notaligned,nonunique = count_reads_in_features( sam_filenames, colnames, gff_filename,opts)
+    except:
+        sys.stderr.write( "Error: %s\n" % str( sys.exc_info()[1] ) )
+        sys.stderr.write( "[Exception type: %s, raised in %s:%d]\n" % 
+         ( sys.exc_info()[1].__class__.__name__, 
+           os.path.basename(traceback.extract_tb( sys.exc_info()[2] )[-1][0]), 
+           traceback.extract_tb( sys.exc_info()[2] )[-1][1] ) )
+        sys.exit( 1 )
+    return counts,empty,ambiguous,lowqual,notaligned,nonunique
+
+
+def usage():
+        print >> sys.stderr, """Usage: python htseqsams2mx.py -w <halfwindowsize> -g <gfffile.gff> -o <outfilename> [-i] [-c] --samf "<sam1.sam>,<sam1.column_header>" --samf "...<samN.column_header>" """
+        sys.exit(1)
+
+if __name__ == "__main__":
+    """  
+    <command interpreter="python">
+    htseqsams2mx.py -w "$halfwin" -g "$gfffile" -o "$outfile" -m "union"
+    #for $s in $samfiles:
+    --samf "'${s.samf}','${s.samf.name}'"
+    #end for
+    </command>
+    """
+    if len(sys.argv) < 2:
+        usage()
+        sys.exit(1)
+    starttime = time.time()
+    op = optparse.OptionParser()
+    # All tools
+    op.add_option('-w', '--halfwindow', default="0")
+    op.add_option('-m', '--mode', default="union")
+    op.add_option('-s', '--stranded', default="no")
+    op.add_option('-y', '--feature_type', default="exon")
+    op.add_option('-g', '--gff_file', default=None)
+    op.add_option('-o', '--outfname', default=None)
+    op.add_option('-f','--forceName', default="false")
+    op.add_option('--samf', default=[], action="append")
+    op.add_option('--mapqMin', default='0')
+    op.add_option( "-t", "--type", type="string", dest="featuretype",
+          default = "exon", help = "feature type (3rd column in GFF file) to be used, " +
+             "all features of other type are ignored (default, suitable for Ensembl " +
+             "GTF files: exon)" )
+
+    op.add_option( "-i", "--id_attribute", type="string", dest="id_attribute",
+          default = "gene_name", help = "GTF attribute to be used as feature ID (default, " +
+          "suitable for Ensembl GTF files: gene_id)" )
+
+    op.add_option( "-q", "--quiet", action="store_true", dest="quiet", default = False,
+          help = "suppress progress report and warnings" )    
+    opts, args = op.parse_args()
+    halfwindow = int(opts.halfwindow)
+    gff_file = opts.gff_file
+    assert os.path.isfile(gff_file),'##ERROR htseqsams2mx: Supplied input GFF file "%s" not found' % gff_file
+    outfname = opts.outfname
+    sam_filenames = []
+    colnames = []
+    samdat = opts.samf
+    samf = [x.split(',')[0].replace("'",'').replace('"','') for x in samdat] # get rid of wrapper supplied quotes
+    assert len(set(samf)) == len(samf),'## ERROR sams2mx: Duplicate input sam file in %s' % ','.join(samf)
+    scolnames = [x.split(',')[1].replace("'",'').replace('"','') for x in samdat]
+    assert len(samf) == len(scolnames), '##ERROR sams2mx: Count of sam/cname not consistent - %d/%d' % (len(samf),len(scolname))
+    for i,b in enumerate(samf):
+        assert os.path.isfile(b),'## Supplied input sam file "%s" not found' % b
+        sam_filenames.append(b)
+        sampName = scolnames[i] # better be unique
+        sampName = sampName.replace('#','') # for R
+        sampName = sampName.replace('(','') # for R
+        sampName = sampName.replace(')','') # for R
+        sampName = sampName.replace(' ','_') # for R
+        colnames.append(sampName)
+    counts,empty,ambiguous,lowqual,notaligned,nonunique = htseqMX(gff_file, sam_filenames,colnames,opts)
+    heads = '\t'.join(['Contig',] + colnames)
+    res = [heads,]
+    contigs = counts.keys()
+    contigs.sort()
+    totalc = 0
+    for contig in contigs:
+        totalc += sum(counts[contig])
+        crow = [contig,] + ['%d' % x for x in counts[contig]]
+        res.append('\t'.join(crow))
+    outf = open(opts.outfname,'w')
+    outf.write('\n'.join(res))
+    outf.write('\n')
+    outf.close()
+    walltime = int(time.time() - starttime)
+    accumulatornames = ('walltimeseconds','contigs','emptyread','ambiguous','lowqual','notaligned','nonunique')
+    notes = ['%s=%d' % (accumulatornames[i],x) for i,x in enumerate((len(contigs),empty,ambiguous,lowqual,notaligned,nonunique))]
+    print >> sys.stdout, ','.join(notes)

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/htseq_bams_to_count_matrix/htseqsams2mx.xml	Thu Jun 06 08:18:03 2013 -0400
@@ -0,0 +1,89 @@
+<tool id="htseqsams2mx" name="Multiple SAMs to count matrix" version="0.2">
+    <description>for DGE</description>
+    <requirements>
+      <requirement type="package" version="0.5.4">package_htseq_0_5_4</requirement>
+    </requirements>
+    <command interpreter="python">
+    htseqsams2mx.py -g "$gfffile" -o "$outfile" -m "$model" --id_attribute "$id_attr" --feature_type "$feature_type"
+    --samf "'$firstsamf','${firstsamf.name}'"
+    #for $s in $samfiles:
+    --samf "'${s.samf}','${s.samf.name}'"
+    #end for
+    </command>
+  <inputs>
+    <param format="gtf" name="gfffile" type="data" label="Gene model (GFF) file to count reads over from your current history" size="100" />
+    <param name="mapqMin" label="Filter reads with mapq below than this value" 
+    help="0 to count any mapping quality read. Otherwise only reads at or above specified mapq will be counted" 
+    type="integer" value="5"/>
+    <param name="title" label="Name for this job's output file" type="text" size="80" value="bams to DGE matrix"/>
+    <param name="forceName" value="false" type="boolean" label="Force replacement to chr:start-offset as the name for each contig in the output"
+      truevalue="true" falsevalue="false" checked="no" help="Leave as false to use the contig names as supplied in your bed file" />
+    <param name="model"  type="select" label="Model for counting reads over the supplied gene model- see HTSeq docs"
+        help="If in doubt, union is a reasonable default but intersection-strict avoids double counting over overlapping exons">
+        <option value="union" selected="true">union</option>
+        <option value="intersection-strict">intersection-strict</option>
+        <option value="intersection-nonempty">intersection-nonempty</option>
+    </param>   
+    <param name="id_attr" type="select" label="GTF attribute to output as the name for each contig - see HTSeq docs"
+        help="If in doubt, gene name is the only option right now">
+        <option value="gene_name" selected="true">gene name</option>
+        <option value="gene_id">gene id</option>
+        <option value="transcript_id">transcript id</option>
+        <option value="transcript_name">transcript name</option>
+    </param>   
+    <param name="feature_type" type="select" label="GTF feature type for counting reads over the supplied gene model- see HTSeq docs"
+        help="exon is all">
+        <option value="exon" selected="true">exon</option>
+        <option value="CDS">CDS</option>
+        <option value="UTR">UTR</option>
+        <option value="transcript">transcript</option>
+    </param>   
+    <param name="firstsamf" type="data" label="SAM file from your history to count reads overlapping gene model regions" format="sam" />
+    <repeat name="samfiles" title="Additional SAM files from your history to count reads overlapping gene model regions">
+        <param name="samf" type="data" label="Additional SAM file from your history" format="sam" size="100"/>
+    </repeat>
+  </inputs>
+  <outputs>
+    <data format="tabular" name="outfile" label="${title}_htseqsams2mx.xls" />
+  </outputs>
+  <help>
+
+**Warning**
+
+This code will count reads overlapping contigs supplied in the gff file.
+
+
+**Note**
+
+htseqsams2mx is an experimental tool currently under test
+
+There is much discussion about whether to count optical/pcr duplicates. If you set the ignore flag to True, any reads in the input BAM files marked as
+duplicates by upstream processing such as the Picard MarkDuplicates tool will NOT be counted. The 'right' setting depends on your data and coverage. For extremely deep 
+coverage, true duplicate reads are inevitable and ignoring them may be throwing away useful real data. In most cases, counting them is probably a reasonable
+choice - any induced bias is likely to be non-differential between samples, whereas it's not at all clear whether that's the case if they are ignored. 
+
+----
+
+**What this tool does**
+
+Counts reads in multiple sample aligned sam format files using HTSeq counting over a gene model supplied as a GFF file
+
+The output is a tabular file containing the count matrix and suitable for downstream processing.
+
+----
+
+**Attribution**
+
+
+This Galaxy wrapper was written for a revised version by Ross Lazarus and is licensed under the LGPL_ like other rgenetics artefacts
+
+.. _LGPL: http://www.gnu.org/copyleft/lesser.html
+
+  </help>
+
+</tool>
+
+
+
+
+

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/htseq_bams_to_count_matrix/tool_dependencies.xml	Thu Jun 06 08:18:03 2013 -0400
@@ -0,0 +1,6 @@
+<?xml version="1.0"?>
+<tool_dependency>
+    <package name="package_htseq_0_5_4" version="0.5.4">
+       <repository name="package_htseq_0_5_4" owner="fubar" toolshed="http://testtoolshed.g2.bx.psu.edu"/>
+    </package>
+</tool_dependency>