PageRenderTime 33ms CodeModel.GetById 13ms app.highlight 16ms RepoModel.GetById 1ms app.codeStats 0ms

/tools/ngs_rna/cufflinks_wrapper.py

https://bitbucket.org/cistrome/cistrome-harvard/
Python | 155 lines | 140 code | 8 blank | 7 comment | 4 complexity | f52cfd4115ad255e14618715479eb39d MD5 | raw file
  1#!/usr/bin/env python
  2
  3import optparse, os, shutil, subprocess, sys, tempfile
  4
  5def stop_err( msg ):
  6    sys.stderr.write( "%s\n" % msg )
  7    sys.exit()
  8    
  9# Copied from sam_to_bam.py:
 10def check_seq_file( dbkey, cached_seqs_pointer_file ):
 11    seq_path = ''
 12    for line in open( cached_seqs_pointer_file ):
 13        line = line.rstrip( '\r\n' )
 14        if line and not line.startswith( '#' ) and line.startswith( 'index' ):
 15            fields = line.split( '\t' )
 16            if len( fields ) < 3:
 17                continue
 18            if fields[1] == dbkey:
 19                seq_path = fields[2].strip()
 20                break
 21    return seq_path
 22	
 23def __main__():
 24    #Parse Command Line
 25    parser = optparse.OptionParser()
 26    parser.add_option( '-1', '--input', dest='input', help=' file of RNA-Seq read alignments in the SAM format. SAM is a standard short read alignment, that allows aligners to attach custom tags to individual alignments, and Cufflinks requires that the alignments you supply have some of these tags. Please see Input formats for more details.' )
 27    parser.add_option( '-s', '--inner-dist-std-dev', dest='inner_dist_std_dev', help='The standard deviation for the distribution on inner distances between mate pairs. The default is 20bp.' )
 28    parser.add_option( '-I', '--max-intron-length', dest='max_intron_len', help='The minimum intron length. Cufflinks will not report transcripts with introns longer than this, and will ignore SAM alignments with REF_SKIP CIGAR operations longer than this. The default is 300,000.' )
 29    parser.add_option( '-F', '--min-isoform-fraction', dest='min_isoform_fraction', help='After calculating isoform abundance for a gene, Cufflinks filters out transcripts that it believes are very low abundance, because isoforms expressed at extremely low levels often cannot reliably be assembled, and may even be artifacts of incompletely spliced precursors of processed transcripts. This parameter is also used to filter out introns that have far fewer spliced alignments supporting them. The default is 0.05, or 5% of the most abundant isoform (the major isoform) of the gene.' )
 30    parser.add_option( '-j', '--pre-mrna-fraction', dest='pre_mrna_fraction', help='Some RNA-Seq protocols produce a significant amount of reads that originate from incompletely spliced transcripts, and these reads can confound the assembly of fully spliced mRNAs. Cufflinks uses this parameter to filter out alignments that lie within the intronic intervals implied by the spliced alignments. The minimum depth of coverage in the intronic region covered by the alignment is divided by the number of spliced reads, and if the result is lower than this parameter value, the intronic alignments are ignored. The default is 5%.' )
 31    parser.add_option( '-p', '--num-threads', dest='num_threads', help='Use this many threads to align reads. The default is 1.' )
 32    parser.add_option( '-m', '--inner-mean-dist', dest='inner_mean_dist', help='This is the expected (mean) inner distance between mate pairs. \
 33                                                                                For, example, for paired end runs with fragments selected at 300bp, \
 34                                                                                where each end is 50bp, you should set -r to be 200. The default is 45bp.')
 35    parser.add_option( '-G', '--GTF', dest='GTF', help='Tells Cufflinks to use the supplied reference annotation to estimate isoform expression. It will not assemble novel transcripts, and the program will ignore alignments not structurally compatible with any reference transcript.' )
 36    parser.add_option( '-g', '--GTF-guide', dest='GTFguide', help='use reference transcript annotation to guide assembly' )
 37    
 38    # Normalization options.
 39    parser.add_option( "-N", "--quartile-normalization", dest="do_normalization", action="store_true" )
 40
 41    # Wrapper / Galaxy options.
 42    parser.add_option( '-A', '--assembled-isoforms-output', dest='assembled_isoforms_output_file', help='Assembled isoforms output file; formate is GTF.' )
 43
 44    # Advanced Options:	
 45    parser.add_option( '--num-importance-samples', dest='num_importance_samples', help='Sets the number of importance samples generated for each locus during abundance estimation. Default: 1000' )
 46    parser.add_option( '--max-mle-iterations', dest='max_mle_iterations', help='Sets the number of iterations allowed during maximum likelihood estimation of abundances. Default: 5000' )
 47
 48    # Bias correction options.
 49    parser.add_option( '-b', dest='do_bias_correction', action="store_true", help='Providing Cufflinks with a multifasta file via this option instructs it to run our new bias detection and correction algorithm which can significantly improve accuracy of transcript abundance estimates.')
 50    parser.add_option( '', '--dbkey', dest='dbkey', help='The build of the reference dataset' )
 51    parser.add_option( '', '--index_dir', dest='index_dir', help='GALAXY_DATA_INDEX_DIR' )
 52    parser.add_option( '', '--ref_file', dest='ref_file', help='The reference dataset from the history' )
 53    
 54    (options, args) = parser.parse_args()
 55    
 56    # output version # of tool
 57    try:
 58        tmp = tempfile.NamedTemporaryFile().name
 59        tmp_stdout = open( tmp, 'wb' )
 60        proc = subprocess.Popen( args='cufflinks --no-update-check 2>&1', shell=True, stdout=tmp_stdout )
 61        tmp_stdout.close()
 62        returncode = proc.wait()
 63        stdout = None
 64        for line in open( tmp_stdout.name, 'rb' ):
 65            if line.lower().find( 'cufflinks v' ) >= 0:
 66                stdout = line.strip()
 67                break
 68        if stdout:
 69            sys.stdout.write( '%s\n' % stdout )
 70        else:
 71            raise Exception
 72    except:
 73        sys.stdout.write( 'Could not determine Cufflinks version\n' )
 74    
 75    # If doing bias correction, set/link to sequence file.
 76    if options.do_bias_correction:
 77        cached_seqs_pointer_file = os.path.join( options.index_dir, 'sam_fa_indices.loc' )
 78        if not os.path.exists( cached_seqs_pointer_file ):
 79            stop_err( 'The required file (%s) does not exist.' % cached_seqs_pointer_file )
 80        # If found for the dbkey, seq_path will look something like /galaxy/data/equCab2/sam_index/equCab2.fa,
 81        # and the equCab2.fa file will contain fasta sequences.
 82        seq_path = check_seq_file( options.dbkey, cached_seqs_pointer_file )
 83        if options.ref_file != 'None':
 84            # Create symbolic link to ref_file so that index will be created in working directory.
 85            seq_path = "ref.fa"
 86            os.symlink( options.ref_file, seq_path  )
 87    
 88    # Build command.
 89    
 90    # Base; always use quiet mode to avoid problems with storing log output.
 91    cmd = "cufflinks -q --no-update-check"
 92    
 93    # Add options.
 94    if options.inner_dist_std_dev:
 95        cmd += ( " -s %i" % int ( options.inner_dist_std_dev ) )
 96    if options.max_intron_len:
 97        cmd += ( " -I %i" % int ( options.max_intron_len ) )
 98    if options.min_isoform_fraction:
 99        cmd += ( " -F %f" % float ( options.min_isoform_fraction ) )
100    if options.pre_mrna_fraction:
101        cmd += ( " -j %f" % float ( options.pre_mrna_fraction ) )    
102    if options.num_threads:
103        cmd += ( " -p %i" % int ( options.num_threads ) )
104    if options.inner_mean_dist:
105        cmd += ( " -m %i" % int ( options.inner_mean_dist ) )
106    if options.GTF:
107        cmd += ( " -G %s" % options.GTF )
108    if options.GTFguide:
109	cmd += ( " -g %s" % options.GTFguide )
110    if options.num_importance_samples:
111        cmd += ( " --num-importance-samples %i" % int ( options.num_importance_samples ) )
112    if options.max_mle_iterations:
113        cmd += ( " --max-mle-iterations %i" % int ( options.max_mle_iterations ) )
114    if options.do_normalization:
115        cmd += ( " -N" )
116    if options.do_bias_correction:
117        cmd += ( " -b %s" % seq_path )
118        
119    # Debugging.
120    print cmd
121        
122    # Add input files.
123    cmd += " " + options.input
124    
125    # Run command.
126    try:
127        tmp_name = tempfile.NamedTemporaryFile( dir="." ).name
128        tmp_stderr = open( tmp_name, 'wb' )
129        proc = subprocess.Popen( args=cmd, shell=True, stderr=tmp_stderr.fileno() )
130        returncode = proc.wait()
131        tmp_stderr.close()
132        
133        # Get stderr, allowing for case where it's very large.
134        tmp_stderr = open( tmp_name, 'rb' )
135        stderr = ''
136        buffsize = 1048576
137        try:
138            while True:
139                stderr += tmp_stderr.read( buffsize )
140                if not stderr or len( stderr ) % buffsize != 0:
141                    break
142        except OverflowError:
143            pass
144        tmp_stderr.close()
145
146        # Copy outputs.
147        shutil.copyfile( "transcripts.gtf" , options.assembled_isoforms_output_file )
148        
149        # Error checking.
150        if returncode != 0:
151            raise Exception, stderr            
152    except Exception, e:
153        stop_err( 'Error running cufflinks. ' + str( e ) )
154
155if __name__=="__main__": __main__()