annotate bwa_mem.py @ 0:6820983ba5d5 draft

Uploaded

# -*- coding: utf-8 -*-
#!/usr/bin/env python
## [e-mail] 22/07/2013
## Copyright © 2014 CRS4 Srl. http://www.crs4.it/
## Modified by:
## Nicola Soranzo <[e-mail]>

"""
Runs BWA on single-end or paired-end data.
Produces a SAM file containing the mappings.
Works with BWA version >= 0.7.5.
NOTICE: In this wrapper, we only use 'mem' for mapping step.

usage: bwa_mem.py [options]

See below for options
"""

import optparse, os, shutil, subprocess, sys, tempfile

def stop_err( msg ):
    sys.stderr.write( '%s\n' % msg )
    sys.exit()

def check_is_double_encoded( fastq ):
    # check that first read is bases, not one base followed by numbers
    bases = [ 'A', 'C', 'G', 'T', 'a', 'c', 'g', 't', 'N' ]
    nums = [ '0', '1', '2', '3' ]
    for line in file( fastq, 'rb'):
        if not line.strip() or line.startswith( '@' ):
            continue
        if len( [ b for b in line.strip() if b in nums ] ) > 0:
            return False
        elif line.strip()[0] in bases and len( [ b for b in line.strip() if b in bases ] ) == len( line.strip() ):
            return True
        else:
            raise Exception, 'First line in first read does not appear to be a valid FASTQ read in either base-space or color-space'
    raise Exception, 'There is no non-comment and non-blank line in your FASTQ file'

def __main__():
    descr = "bwa_mem.py: Map (long length) reads against a reference genome with BWA-MEM."
    parser = optparse.OptionParser(description=descr)
    parser.add_option( '-t', '--threads', default=1, help='The number of threads to use [1]' )
    parser.add_option( '--ref', help='The reference genome to use or index' )
    parser.add_option( '-f', '--fastq', help='The (forward) fastq file to use for the mapping' )
    parser.add_option( '-F', '--rfastq', help='The reverse fastq file to use for mapping if paired-end data' )
    parser.add_option( '-u', '--output', help='The file to save the output (SAM format)' )
    parser.add_option( '-g', '--genAlignType', help='The type of pairing (single or paired)' )
    parser.add_option( '--params', help='Parameter setting to use (pre_set or full)' )
    parser.add_option( '-s', '--fileSource', help='Whether to use a previously indexed reference sequence or one form history (indexed or history)' )
    parser.add_option( '-D', '--dbkey', help='Dbkey for reference genome' )

    parser.add_option( '-k', '--minEditDistSeed', default=19, type=int, help='Minimum edit distance to the seed [19]' )
    parser.add_option( '-w', '--bandWidth', default=100, type=int, help='Band width for banded alignment [100]' )
    parser.add_option( '-d', '--offDiagonal', default=100, type=int, help='off-diagonal X-dropoff [100]' )
    parser.add_option( '-r', '--internalSeeds', default=1.5, type=float, help='look for internal seeds inside a seed longer than {-k} * FLOAT [1.5]' )
    parser.add_option( '-c', '--seedsOccurrence', default=10000, type=int, help='skip seeds with more than INT occurrences [10000]' )
    parser.add_option( '-S', '--mateRescue', default=False, help='skip mate rescue' )
    parser.add_option( '-P', '--skipPairing', default=False, help='skpe pairing, mate rescue performed unless -S also in use' )
    parser.add_option( '-A', '--seqMatch', default=1, type=int, help='score of a sequence match' )
    parser.add_option( '-B', '--mismatch', default=4, type=int, help='penalty for a mismatch' )
    parser.add_option( '-O', '--gapOpen', default=6, type=int, help='gap open penalty' )
    parser.add_option( '-E', '--gapExtension', default=None, help='gap extension penalty; a gap of size k cost {-O} + {-E}*k [1]' )
    parser.add_option( '-L', '--clipping', default=5, type=int, help='penalty for clipping [5]' )
    parser.add_option( '-U', '--unpairedReadpair', default=17, type=int, help='penalty for an unpaired read pair [17]' )
    parser.add_option( '-p', '--interPairEnd', default=False, help='first query file consists of interleaved paired-end sequences' )
    parser.add_option( '--rgid', help='Read group identifier' )
    parser.add_option( '--rgsm', help='Sample' )
    parser.add_option( '--rgpl', choices=[ 'CAPILLARY', 'LS454', 'ILLUMINA', 'SOLID', 'HELICOS', 'IONTORRENT', 'PACBIO' ], help='Platform/technology used to produce the reads' )
    parser.add_option( '--rglb', help='Library name' )
    parser.add_option( '--rgpu', help='Platform unit (e.g. flowcell-barcode.lane for Illumina or slide for SOLiD)' )
    parser.add_option( '--rgcn', help='Sequencing center that produced the read' )
    parser.add_option( '--rgds', help='Description' )
    parser.add_option( '--rgdt', help='Date that run was produced (ISO8601 format date or date/time, like YYYY-MM-DD)' )
    parser.add_option( '--rgfo', help='Flow order' )
    parser.add_option( '--rgks', help='The array of nucleotide bases that correspond to the key sequence of each read' )
    parser.add_option( '--rgpg', help='Programs used for processing the read group' )
    parser.add_option( '--rgpi', help='Predicted median insert size' )
    parser.add_option( '-T', '--minScore', default=30, type=int, help='minimum score to output [30]' )
    parser.add_option( '-M', '--mark', default=False, help='mark shorter split hits as secondary (for Picard/GATK compatibility)' )
    parser.add_option( '-H', '--suppressHeader', dest='suppressHeader', help='Suppress header' )
    (options, args) = parser.parse_args()
    if len(args) > 0:
        parser.error('Wrong number of arguments')

    # output version # of tool
    try:
        tmp = tempfile.NamedTemporaryFile().name
        tmp_stdout = open( tmp, 'wb' )
        proc = subprocess.Popen( args='bwa 2>&1', shell=True, stdout=tmp_stdout )
        tmp_stdout.close()
        returncode = proc.wait()
        stdout = None
        for line in open( tmp_stdout.name, 'rb' ):
            if line.lower().find( 'version' ) >= 0:
                stdout = line.strip()
                break
        if stdout:
            sys.stdout.write( 'BWA %s\n' % stdout )
        else:
            raise Exception
    except:
        sys.stdout.write( 'Could not determine BWA version\n' )

    fastq = options.fastq
    if options.rfastq:
        rfastq = options.rfastq

    # make temp directory for placement of indices
    tmp_index_dir = tempfile.mkdtemp()
    tmp_dir = tempfile.mkdtemp()
    # index if necessary
    if options.fileSource == 'history':
        ref_file = tempfile.NamedTemporaryFile( dir=tmp_index_dir )
        ref_file_name = ref_file.name
        ref_file.close()
        os.symlink( options.ref, ref_file_name )
        # determine which indexing algorithm to use, based on size
        try:
            size = os.stat( options.ref ).st_size
            if size <= 2**30:
                indexingAlg = 'is'
            else:
                indexingAlg = 'bwtsw'
        except:
            indexingAlg = 'is'
        indexing_cmds = '-a %s' % indexingAlg
        cmd1 = 'bwa index %s %s' % ( indexing_cmds, ref_file_name )
        try:
            tmp = tempfile.NamedTemporaryFile( dir=tmp_index_dir ).name
            tmp_stderr = open( tmp, 'wb' )
            proc = subprocess.Popen( args=cmd1, shell=True, cwd=tmp_index_dir, stderr=tmp_stderr.fileno() )
            returncode = proc.wait()
            tmp_stderr.close()
            # get stderr, allowing for case where it's very large
            tmp_stderr = open( tmp, 'rb' )
            stderr = ''
            buffsize = 1048576
            try:
                while True:
                    stderr += tmp_stderr.read( buffsize )
                    if not stderr or len( stderr ) % buffsize != 0:
                        break
            except OverflowError:
                pass
            tmp_stderr.close()
            if returncode != 0:
                raise Exception, stderr
        except Exception, e:
            # clean up temp dirs
            if os.path.exists( tmp_index_dir ):
                shutil.rmtree( tmp_index_dir )
            if os.path.exists( tmp_dir ):
                shutil.rmtree( tmp_dir )
            stop_err( 'Error indexing reference sequence. ' + str( e ) )
    else:
        ref_file_name = options.ref
    # if options.illumina13qual:
    #     illumina_quals = "-I"
    # else:
    #     illumina_quals = ""

    # set up aligning and generate aligning command args
    start_cmds = '-t %s ' % options.threads
    if options.params == 'pre_set':
        # aligning_cmds = '-t %s %s' % ( options.threads, illumina_quals )
        #start_cmds = '-t %s ' % options.threads
        end_cmds = ' '
        print start_cmds, end_cmds

    else:
        end_cmds = '-k %s -w %s -d %s -r %s -c %s -A %s -B %s -O %s -L %s -U %s -T %s ' % (options.minEditDistSeed, options.bandWidth, options.offDiagonal, options.internalSeeds, options.seedsOccurrence, options.seqMatch, options.mismatch, options.gapOpen, options.clipping, options.unpairedReadpair, options.minScore)
        if options.mateRescue:
            end_cmds += '-S '
            if options.skipPairing:
                end_cmds += '-P '
        else:
            if options.skipPairing:
                print "Option Error and will not be considered, you should also choose 'skip mate rescue -S' option! "
        if options.gapExtension != None:
            end_cmds += '-E %s ' % options.gapExtension

        if options.rgid:
            if not options.rglb or not options.rgpl or not options.rgsm or not options.rglb:
                stop_err( 'If you want to specify read groups, you must include the ID, LB, PL, and SM tags.' )
            # readGroup = '@RG\tID:%s\tLB:%s\tPL:%s\tSM:%s' % ( options.rgid, options.rglb, options.rgpl, options.rgsm )
            readGroup = '@RG\tID:%s\tLB:%s\tPL:%s\tSM:%s' % ( options.rgid, options.rglb, options.rgpl, options.rgsm )
            if options.rgpu:
                readGroup += '\tPU:%s' % options.rgpu
            if options.rgcn:
                readGroup += '\tCN:%s' % options.rgcn
            if options.rgds:
                readGroup += '\tDS:%s' % options.rgds
            if options.rgdt:
                readGroup += '\tDT:%s' % options.rgdt
            if options.rgfo:
                readGroup += '\tFO:%s' % options.rgfo
            if options.rgks:
                readGroup += '\tKS:%s' % options.rgks
            if options.rgpg:
                readGroup += '\tPG:%s' % options.rgpg
            if options.rgpi:
                readGroup += '\tPI:%s' % options.rgpi
            end_cmds += ' -R "%s" ' % readGroup

        if options.interPairEnd:
            end_cmds += '-p %s ' % options.interPairEnd
        if options.mark:
            end_cmds += '-M '


    if options.genAlignType == 'paired':
        cmd = 'bwa mem %s %s %s %s %s > %s' % ( start_cmds, ref_file_name, fastq, rfastq, end_cmds, options.output )
    else:
        cmd = 'bwa mem %s %s %s %s > %s' % ( start_cmds, ref_file_name, fastq, end_cmds, options.output )

  # perform alignments
    buffsize = 1048576
    try:
        # need to nest try-except in try-finally to handle 2.4
        try:
            try:
                tmp = tempfile.NamedTemporaryFile( dir=tmp_dir ).name
                tmp_stderr = open( tmp, 'wb' )
                print "The cmd is %s" % cmd
                proc = subprocess.Popen( args=cmd, shell=True, cwd=tmp_dir, stderr=tmp_stderr.fileno() )
                returncode = proc.wait()
                tmp_stderr.close()
                # get stderr, allowing for case where it's very large
                tmp_stderr = open( tmp, 'rb' )
                stderr = ''
                try:
                    while True:
                        stderr += tmp_stderr.read( buffsize )
                        if not stderr or len( stderr ) % buffsize != 0:
                            break
                except OverflowError:
                    pass
                tmp_stderr.close()
                if returncode != 0:
                    raise Exception, stderr
            except Exception, e:
                raise Exception, 'Error generating alignments. ' + str( e )
            # remove header if necessary
            if options.suppressHeader == 'true':
                tmp_out = tempfile.NamedTemporaryFile( dir=tmp_dir)
                tmp_out_name = tmp_out.name
                tmp_out.close()
                try:
                    shutil.move( options.output, tmp_out_name )
                except Exception, e:
                    raise Exception, 'Error moving output file before removing headers. ' + str( e )
                fout = file( options.output, 'w' )
                for line in file( tmp_out.name, 'r' ):
                    if not ( line.startswith( '@HD' ) or line.startswith( '@SQ' ) or line.startswith( '@RG' ) or line.startswith( '@PG' ) or line.startswith( '@CO' ) ):
                        fout.write( line )
                fout.close()
            # check that there are results in the output file
            if os.path.getsize( options.output ) > 0:
                sys.stdout.write( 'BWA run on %s-end data' % options.genAlignType )
            else:
                raise Exception, 'The output file is empty. You may simply have no matches, or there may be an error with your input file or settings.'
        except Exception, e:
            stop_err( 'The alignment failed.\n' + str( e ) )
    finally:
        # clean up temp dir
        if os.path.exists( tmp_index_dir ):
            shutil.rmtree( tmp_index_dir )
        if os.path.exists( tmp_dir ):
            shutil.rmtree( tmp_dir )

if __name__ == "__main__":
    __main__()