galaxy/tools/sr_mapping/bfast_wrapper.py

#!/usr/bin/env python
"""
Runs BFAST on single-end or paired-end data.
TODO: more documentation

TODO:
    - auto-detect gzip or bz2
    - split options (?)
    - queue lengths (?)
    - assumes reference always has been indexed
    - main and secondary indexes
    - scoring matrix file ?
    - read group file ?

usage: bfast_wrapper.py [options]
    -r, --ref=r: The reference genome to use or index
    -f, --fastq=f: The fastq file to use for the mapping
    -F, --output=u: The file to save the output (SAM format)
    -s, --fileSource=s: Whether to use a previously indexed reference sequence or one from history (indexed or history)
    -p, --params=p: Parameter setting to use (pre_set or full)
    -n, --numThreads=n: The number of threads to use
    -A, --space=A: The encoding space (0: base 1: color)
    -o, --offsets=o: The offsets for 'match'
    -l, --loadAllIndexes=l: Load all indexes into memory
    -k, --keySize=k: truncate key size in 'match'
    -K, --maxKeyMatches=K: the maximum number of matches to allow before a key is ignored
    -M, --maxNumMatches=M: the maximum number of matches to allow before the read is discarded
    -w, --whichStrand=w: the strands to consider (0: both 1: forward 2: reverse)
    -t, --timing=t: output timing information to stderr
    -u, --ungapped=u: performed ungapped local alignment
    -U, --unconstrained=U: performed local alignment without mask constraints
    -O, --offset=O: the number of bases before and after each hit to consider in local alignment
    -q, --avgMismatchQuality=q: average mismatch quality
    -a, --algorithm=a: post processing algorithm (0: no filtering, 1: all passing filters, 2: unique, 3: best scoring unique, 4: best score all)
    -P, --disallowPairing=P: do not choose alignments based on pairing
    -R, --reverse=R: paired end reads are given on reverse strands
    -z, --random=z: output a random best scoring alignment
    -D, --dbkey=D: Dbkey for reference genome
    -H, --suppressHeader=H: Suppress the sam header
"""

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


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


def __main__():
    parser = optparse.OptionParser()
    parser.add_option('-r', '--ref', dest='ref', help='The reference genome to index and use')
    parser.add_option('-f', '--fastq', dest='fastq', help='The fastq file to use for the mapping')
    parser.add_option('-F', '--output', dest='output', help='The file to save the output (SAM format)')
    parser.add_option('-A', '--space', dest='space', type="choice", default='0', choices=('0', '1'), help='The encoding space (0: base 1: color)')
    parser.add_option('-H', '--suppressHeader', action="store_true", dest='suppressHeader', default=False, help='Suppress header')
    parser.add_option('-n', '--numThreads', dest='numThreads', type="int", default="1", help='The number of threads to use')
    parser.add_option('-t', '--timing', action="store_true", default=False, dest='timing', help='output timming information to stderr')
    parser.add_option('-l', '--loadAllIndexes', action="store_true", default=False, dest='loadAllIndexes', help='Load all indexes into memory')
    parser.add_option('-m', '--indexMask', dest='indexMask', help='String containing info on how to build custom indexes')
    parser.add_option("-b", "--buildIndex", action="store_true", dest="buildIndex", default=False, help='String containing info on how to build custom indexes')
    parser.add_option("--indexRepeatMasker", action="store_true", dest="indexRepeatMasker", default=False, help='Do not index lower case sequences. Such as those created by RepeatMasker')
    parser.add_option('--indexContigOptions', dest='indexContigOptions', default="", help='The contig range options to use for the indexing')
    parser.add_option('--indexExonsFileName', dest='indexExonsFileName', default="", help='The exons file to use for the indexing')

    parser.add_option('-o', '--offsets', dest='offsets', default="", help='The offsets for \'match\'')
    parser.add_option('-k', '--keySize', dest='keySize', type="int", default="-1", help='truncate key size in \'match\'')
    parser.add_option('-K', '--maxKeyMatches', dest='maxKeyMatches', type="int", default="-1", help='the maximum number of matches to allow before a key is ignored')
    parser.add_option('-M', '--maxNumMatches', dest='maxNumMatches', type="int", default="-1", help='the maximum number of matches to allow bfore the read is discarded')
    parser.add_option('-w', '--whichStrand', dest='whichStrand', type="choice", default='0', choices=('0', '1', '2'), help='the strands to consider (0: both 1: forward 2: reverse)')

    parser.add_option('--scoringMatrixFileName', dest='scoringMatrixFileName', help='Scoring Matrix file used to score the alignments')
    parser.add_option('-u', '--ungapped', dest='ungapped', action="store_true", default=False, help='performed ungapped local alignment')
    parser.add_option('-U', '--unconstrained', dest='unconstrained', action="store_true", default=False, help='performed local alignment without mask constraints')
    parser.add_option('-O', '--offset', dest='offset', type="int", default="0", help='the number of bases before and after each hit to consider in local alignment')
    parser.add_option('-q', '--avgMismatchQuality', type="int", default="-1", dest='avgMismatchQuality', help='average mismatch quality')

    parser.add_option('-a', '--algorithm', dest='algorithm', default='0', type="choice", choices=('0', '1', '2', '3', '4'), help='post processing algorithm (0: no filtering, 1: all passing filters, 2: unique, 3: best scoring unique, 4: best score all')
    parser.add_option('--unpaired', dest='unpaired', action="store_true", default=False, help='do not choose alignments based on pairing')
    parser.add_option('--reverseStrand', dest='reverseStrand', action="store_true", default=False, help='paired end reads are given on reverse strands')
    parser.add_option('--pairedEndInfer', dest='pairedEndInfer', action="store_true", default=False, help='break ties when one end of a paired end read by estimating the insert size distribution')
    parser.add_option('--randomBest', dest='randomBest', action="store_true", default=False, help='output a random best scoring alignment')

    (options, args) = parser.parse_args()

    # output version # of tool
    try:
        tmp = tempfile.NamedTemporaryFile().name
        tmp_stdout = open(tmp, 'wb')
        proc = subprocess.Popen(args='bfast 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('%s\n' % stdout)
        else:
            raise Exception
    except Exception:
        sys.stdout.write('Could not determine BFAST version\n')

    buffsize = 1048576

    # make temp directory for bfast, requires trailing slash
    tmp_dir = '%s/' % tempfile.mkdtemp()

    # 'generic' options used in all bfast commands here
    if options.timing:
        all_cmd_options = "-t"
    else:
        all_cmd_options = ""

    try:
        if options.buildIndex:
            reference_filepath = tempfile.NamedTemporaryFile(dir=tmp_dir, suffix='.fa').name
            # build bfast indexes
            os.symlink(options.ref, reference_filepath)

            # bfast fast2brg
            try:
                nuc_space = ["0"]
                if options.space == "1":
                    # color space localalign appears to require nuc space brg
                    nuc_space.append("1")
                for space in nuc_space:
                    cmd = 'bfast fasta2brg -f "%s" -A "%s" %s' % (reference_filepath, space, all_cmd_options)
                    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 as e:
                raise Exception('Error in \'bfast fasta2brg\'.\n' + str(e))

            # bfast index
            try:
                all_index_cmds = 'bfast index %s -f "%s" -A "%s" -n "%s"' % (all_cmd_options, reference_filepath, options.space, options.numThreads)

                if options.indexRepeatMasker:
                    all_index_cmds += " -R"

                if options.indexContigOptions:
                    index_contig_options = [int(_) for _ in options.indexContigOptions.split(',')]
                    if index_contig_options[0] >= 0:
                        all_index_cmds += ' -s "%s"' % index_contig_options[0]
                    if index_contig_options[1] >= 0:
                        all_index_cmds += ' -S "%s"' % index_contig_options[1]
                    if index_contig_options[2] >= 0:
                        all_index_cmds += ' -e "%s"' % index_contig_options[2]
                    if index_contig_options[3] >= 0:
                        all_index_cmds += ' -E "%s"' % index_contig_options[3]
                elif options.indexExonsFileName:
                    all_index_cmds += ' -x "%s"' % options.indexExonsFileName

                index_count = 1
                for mask, hash_width in [mask.split(':') for mask in options.indexMask.split(',')]:
                    cmd = '%s -m "%s" -w "%s" -i "%i"' % (all_index_cmds, mask, hash_width, index_count)
                    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)
                    index_count += 1
            except Exception as e:
                raise Exception('Error in \'bfast index\'.\n' + str(e))

        else:
            reference_filepath = options.ref
        assert reference_filepath and os.path.exists(reference_filepath), 'A valid genome reference was not provided.'

        # set up aligning and generate aligning command options
        # set up temp output files
        tmp_bmf = tempfile.NamedTemporaryFile(dir=tmp_dir)
        tmp_bmf_name = tmp_bmf.name
        tmp_bmf.close()
        tmp_baf = tempfile.NamedTemporaryFile(dir=tmp_dir)
        tmp_baf_name = tmp_baf.name
        tmp_baf.close()

        bfast_match_cmd = 'bfast match -f "%s" -r "%s" -n "%s" -A "%s" -T "%s" -w "%s" %s' % (reference_filepath, options.fastq, options.numThreads, options.space, tmp_dir, options.whichStrand, all_cmd_options)
        bfast_localalign_cmd = 'bfast localalign -f "%s" -m "%s" -n "%s" -A "%s" -o "%s" %s' % (reference_filepath, tmp_bmf_name, options.numThreads, options.space, options.offset, all_cmd_options)
        bfast_postprocess_cmd = 'bfast postprocess -O 1 -f "%s" -i "%s" -n "%s" -A "%s" -a "%s" %s' % (reference_filepath, tmp_baf_name, options.numThreads, options.space, options.algorithm, all_cmd_options)

        if options.offsets:
            bfast_match_cmd += ' -o "%s"' % options.offsets
        if options.keySize >= 0:
            bfast_match_cmd += ' -k "%s"' % options.keySize
        if options.maxKeyMatches >= 0:
            bfast_match_cmd += ' -K "%s"' % options.maxKeyMatches
        if options.maxNumMatches >= 0:
            bfast_match_cmd += ' -M "%s"' % options.maxNumMatches
            bfast_localalign_cmd += ' -M "%s"' % options.maxNumMatches
        if options.scoringMatrixFileName:
            bfast_localalign_cmd += ' -x "%s"' % options.scoringMatrixFileName
            bfast_postprocess_cmd += ' -x "%s"' % options.scoringMatrixFileName
        if options.ungapped:
            bfast_localalign_cmd += ' -u'
        if options.unconstrained:
            bfast_localalign_cmd += ' -U'
        if options.avgMismatchQuality >= 0:
            bfast_localalign_cmd += ' -q "%s"' % options.avgMismatchQuality
            bfast_postprocess_cmd += ' -q "%s"' % options.avgMismatchQuality
        if options.algorithm == 3:
            if options.pairedEndInfer:
                bfast_postprocess_cmd += ' -P'
            if options.randomBest:
                bfast_postprocess_cmd += ' -z'
        if options.unpaired:
            bfast_postprocess_cmd += ' -U'
        if options.reverseStrand:
            bfast_postprocess_cmd += ' -R'

        # instead of using temp files, should we stream through pipes?
        bfast_match_cmd += " > %s" % tmp_bmf_name
        bfast_localalign_cmd += " > %s" % tmp_baf_name
        bfast_postprocess_cmd += " > %s" % options.output

        # need to nest try-except in try-finally to handle 2.4
        try:
            # bfast 'match'
            try:
                tmp = tempfile.NamedTemporaryFile(dir=tmp_dir).name
                tmp_stderr = open(tmp, 'wb')
                proc = subprocess.Popen(args=bfast_match_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 as e:
                raise Exception('Error in \'bfast match\'. \n' + str(e))
            # bfast 'localalign'
            try:
                tmp = tempfile.NamedTemporaryFile(dir=tmp_dir).name
                tmp_stderr = open(tmp, 'wb')
                proc = subprocess.Popen(args=bfast_localalign_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 as e:
                raise Exception('Error in \'bfast localalign\'. \n' + str(e))
            # bfast 'postprocess'
            try:
                tmp = tempfile.NamedTemporaryFile(dir=tmp_dir).name
                tmp_stderr = open(tmp, 'wb')
                proc = subprocess.Popen(args=bfast_postprocess_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 as e:
                raise Exception('Error in \'bfast postprocess\'. \n' + str(e))
            # remove header if necessary
            if options.suppressHeader:
                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 as e:
                    raise Exception('Error moving output file before removing headers. \n' + str(e))
                fout = open(options.output, 'w')
                for line in open(tmp_out.name, 'r'):
                    if len(line) < 3 or line[0:3] not in ['@HD', '@SQ', '@RG', '@PG', '@CO']:
                        fout.write(line)
                fout.close()
            # check that there are results in the output file
            if os.path.getsize(options.output) > 0:
                if "0" == options.space:
                    sys.stdout.write('BFAST run on Base Space data')
                else:
                    sys.stdout.write('BFAST run on Color Space data')
            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 as e:
            stop_err('The alignment failed.\n' + str(e))
    finally:
        # clean up temp dir
        if os.path.exists(tmp_dir):
            shutil.rmtree(tmp_dir)


if __name__ == "__main__":
    __main__()