view tools/smalt_wrapper.py @ 0:87f4f05b44bb draft

Uploaded

#!/usr/bin/env python

"""
Runs Smalt on single-end or paired-end data.
Produces a SAM file containing the mappings.
Works with Smalt version 0.7.1.

usage: smalt_wrapper.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 __main__():
    #Parse Command Line
    parser = optparse.OptionParser()
    parser.add_option( '-t', '--threads', dest='threads', help='The number of threads to use' )
    parser.add_option( '-r', '--ref', dest='ref', help='The reference genome to use or index' )
    parser.add_option( '-f', '--input1', dest='fastq', help='The (forward) fastq file to use for the mapping' )
    parser.add_option( '-F', '--input2', dest='rfastq', help='The reverse fastq file to use for mapping if paired-end data' )
    parser.add_option( '-u', '--output', dest='output', help='The file to save the output (SAM format)' )
    parser.add_option( '-g', '--genAlignType', dest='genAlignType', help='The type of pairing (single or paired)' )
    parser.add_option( '-p', '--params', dest='params', help='Parameter setting to use (pre_set or full)' )
    parser.add_option( '-s', '--fileSource', dest='fileSource', help='Whether to use a previously indexed reference sequence or one form history (indexed or history)' )
    parser.add_option( '-x', '--exhaustiveSearch', dest='exhaustiveSearch', help='This flag triggers a more exhaustive search for alignments at the cost of decreased speed' )
    parser.add_option( '-c', '--minCover', dest='minCover', help='Only consider mappings where the k-mer word seeds cover the query read to a minimum extent' )
    parser.add_option( '-d', '--scorDiff', dest='scorDiff', help='Set a threshold of the Smith-Waterman alignment score relative to the maximum score' )
    parser.add_option( '-i', '--insertMax', dest='insertMax', help='Maximum insert size (Only in paired-end mode)' )
    parser.add_option( '-j', '--insertMin', dest='insertMin', help='Minimum insert size (Only in paired-end mode)' )
    parser.add_option( '-l', '--pairTyp', dest='pairTyp', help='Type of read pair library, can be either pe, mp or pp' )
    parser.add_option( '-m', '--minScor', dest='minScor', help='Sets an absolute threshold of the Smith-Waterman scores' )
    parser.add_option( '-a', '--partialAlignments', dest='partialAlignments', help='Report partial alignments if they are complementary on the read (split reads)' )
    parser.add_option( '-q', '--minBasq', dest='minBasq', help='Sets a base quality threshold (0 <= minbasq <= 10, default 0)' )
    parser.add_option( '-e', '--seed', dest='seed', help='If <seed> >= 0 report an alignment selected at random where there are multiple mappings with the same best alignment score. With <seed> = 0 (default) a seed is derived from the current calendar time. If <seed> < 0 reads with multiple best mappings are reported as "not mapped".' )
    parser.add_option( '-w', '--complexityWeighted', dest='complexityWeighted', help='Smith-Waterman scores are complexity weighted' )
    parser.add_option( '-y', '--minId', dest='minId', help='Sets an identity threshold for a mapping to be reported' )
    parser.add_option( '-D', '--dbkey', dest='dbkey', help='Dbkey for reference genome' )
    parser.add_option( '-X', '--do_not_build_index', dest='do_not_build_index', action='store_true', help="Don't build index" )
    parser.add_option( '-H', '--suppressHeader', dest='suppressHeader', help='Suppress header' )
    (options, args) = parser.parse_args()

    # output version # of tool
    try:
        tmp = tempfile.NamedTemporaryFile().name
        tmp_stdout = open( tmp, 'wb' )
        proc = subprocess.Popen( args='smalt 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( 'SMALT %s\n' % stdout )
        else:
            raise Exception
    except:
        sys.stdout.write( 'Could not determine SMALT 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' and not options.do_not_build_index:
        ref_file = tempfile.NamedTemporaryFile( dir=tmp_index_dir )
        ref_file_name = ref_file.name
        ref_file.close()
        os.symlink( options.ref, ref_file_name )
        cmd1 = 'smalt index %s %s' % ( ref_file_name, 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

    # set up aligning and generate aligning command options
    if options.params == 'pre_set':
        aligning_cmds = '-n %s ' % ( options.threads )
        gen_alignment_cmds = ''
    else:
        if options.exhaustiveSearch == 'true':
            exhaustiveSearch = '-x'
            minCover = '-c %s' % options.minCover
        else:
            exhaustiveSearch = ''
            minCover = ''
        if options.partialAlignments == 'true':
            partialAlignments = '-x'
        else:
            partialAlignments = ''
        if options.complexityWeighted == 'true':
            complexityWeighted = '-w'
        else:
            complexityWeighted = ''
        aligning_cmds = '-d %s -m %s -q %s -r %s -y %s' % \
                        ( options.scorDiff, options.minScor, options.minBasq, options.seed, options.minId )
        if options.genAlignType == 'paired':
            gen_alignment_cmds = '-i %s -j %s -l %s' % ( options.insertMax, options.insertMin, options.pairTyp )
        else:
            gen_alignment_cmds = ''
    # prepare actual aligning and generate aligning commands
    if options.genAlignType == 'paired':
        cmd = 'smalt map %s %s -o %s %s %s ' % ( aligning_cmds, gen_alignment_cmds, options.output, ref_file_name, fastq, rfastq )
    else:
        cmd = 'smalt map %s -o %s %s %s ' % ( aligning_cmds, options.output, ref_file_name, fastq )
    # perform alignments
    buffsize = 1048576
    try:
        # need to nest try-except in try-finally to handle 2.4
        try:
            # align
            try:
                tmp = tempfile.NamedTemporaryFile( dir=tmp_dir ).name
                tmp_stderr = open( tmp, 'wb' )
                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 aligning sequence. ' + 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( 'SMALT 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__()