htsworkflow/pipelines/qseq2fastq.py

   1 #!/usr/bin/env python
   2 """Convert a collection of qseq or a tar file of qseq files to a fastq file
   3 """
   4 from __future__ import print_function, unicode_literals
   5 from glob import glob
   6 import os
   7 from optparse import OptionParser
   8 import numpy
   9 import sys
  10 import tarfile
  11
  12 from htsworkflow.util.version import version
  13 from htsworkflow.util.conversion import parse_slice
  14
  15
  16 def main(cmdline=None):
  17     """Command line driver: [None, '-i', 'tarfile', '-o', 'target.fastq']
  18     """
  19     parser = make_parser()
  20     opts, args = parser.parse_args(cmdline)
  21
  22     if opts.version:
  23         print(version())
  24         return 0
  25
  26     if opts.infile is not None:
  27         qseq_generator = tarfile_generator(opts.infile)
  28     elif len(args) > 0:
  29         qseq_generator = file_generator(args)
  30     else:
  31         qseq_generator = [sys.stdin]
  32
  33     if opts.output is not None:
  34         output = open(opts.output, 'w')
  35     else:
  36         output = sys.stdout
  37
  38     if opts.nopass_output is not None:
  39         nopass_output = open(opts.nopass_output, 'w')
  40     else:
  41         nopass_output = None
  42
  43     qseq_parser = Qseq2Fastq(qseq_generator, output, nopass_output)
  44     qseq_parser.fastq = not opts.fasta
  45     qseq_parser.flowcell_id = opts.flowcell
  46     qseq_parser.trim = parse_slice(opts.slice)
  47     qseq_parser.reportFilter = opts.pf
  48
  49     qseq_parser.run()
  50
  51
  52 def make_parser():
  53     """Return option parser"""
  54     usage = "%prog: [options] *_qseq.txt\nProduces Phred33 files by default"
  55     parser = OptionParser(usage)
  56     parser.add_option("-a", "--fasta", default=False, action="store_true",
  57                       help="produce fasta files instead of fastq files")
  58     parser.add_option("-f", "--flowcell", default=None,
  59                       help="Set flowcell ID for output file")
  60     parser.add_option("-i", "--infile", default=None,
  61                       help='source tar file (if reading from an archive '\
  62                            'instead of a directory)')
  63     parser.add_option("-o", "--output", default=None,
  64                       help="output fastq file")
  65     parser.add_option("-n", "--nopass-output", default=None,
  66                       help="if provided send files that failed "\
  67                            "illumina filter to a differentfile")
  68     parser.add_option("-s", "--slice",
  69                       help="specify python slice, e.g. 0:75, 0:-1",
  70                       default=None)
  71     parser.add_option("--pf", help="report pass filter flag", default=False,
  72                       action="store_true")
  73     parser.add_option("--version", default=False, action="store_true",
  74                       help="report software version")
  75
  76     return parser
  77
  78
  79 def file_generator(pattern_list):
  80     """Given a list of glob patterns yield open streams for matching files"""
  81     for pattern in pattern_list:
  82         for filename in glob(pattern):
  83             yield open(filename, "r")
  84
  85
  86 def tarfile_generator(tarfilename):
  87     """Yield open streams for files inside a tarfile"""
  88     archive = tarfile.open(tarfilename, 'r|*')
  89     for tarinfo in archive:
  90         yield archive.extractfile(tarinfo)
  91
  92
  93 class Qseq2Fastq(object):
  94     """
  95     Convert qseq files to fastq (or fasta) files.
  96     """
  97     def __init__(self, sources, pass_destination, nopass_destination=None):
  98         self.sources = sources
  99         self.pass_destination = pass_destination
 100         if nopass_destination is not None:
 101             self.nopass_destination = nopass_destination
 102         else:
 103             self.nopass_destination = pass_destination
 104
 105         self.fastq = True
 106         self.flowcell_id = None
 107         self.trim = slice(None)
 108         self.report_filter = False
 109
 110     def _format_flowcell_id(self):
 111         """
 112         Return formatted flowcell ID
 113         """
 114         if self.flowcell_id is not None:
 115             return self.flowcell_id + "_"
 116         else:
 117             return ""
 118
 119     def run(self):
 120         """Run conversion
 121         (Used to match threading/multiprocessing API)
 122         """
 123         if self.fastq:
 124             header_template = '@'
 125         else:
 126             # fasta case
 127             header_template = '>'
 128         header_template += self._format_flowcell_id() + \
 129                            '%s_%s:%s:%s:%s:%s/%s%s%s'
 130
 131         for qstream in self.sources:
 132             for line in qstream:
 133                 # parse line
 134                 record = line.rstrip().split('\t')
 135                 machine_name = record[0]
 136                 run_number = record[1]
 137                 lane_number = record[2]
 138                 tile = record[3]
 139                 x = record[4]
 140                 y = record[5]
 141                 #index = record[6]
 142                 read = record[7]
 143                 sequence = record[8].replace('.', 'N')
 144                 quality = convert_illumina_quality(record[9])
 145
 146                 # add pass qc filter if we want it
 147                 pass_qc = int(record[10])
 148                 if self.report_filter:
 149                     pass_qc_msg = " pf=%s" % (pass_qc)
 150                 else:
 151                     pass_qc_msg = ""
 152
 153                 header = header_template % ( \
 154                     machine_name,
 155                     run_number,
 156                     lane_number,
 157                     tile,
 158                     x,
 159                     y,
 160                     read,
 161                     pass_qc_msg,
 162                     os.linesep)
 163
 164                 # if we passed the filter write to the "good" file
 165                 if pass_qc:
 166                     destination = self.pass_destination
 167                 else:
 168                     destination = self.nopass_destination
 169
 170                 destination.write(header)
 171                 destination.write(sequence[self.trim])
 172                 destination.write(os.linesep)
 173                 if self.fastq:
 174                     destination.write('+')
 175                     destination.write(os.linesep)
 176                     destination.write(quality[self.trim].tostring())
 177                     destination.write(os.linesep)
 178
 179 def convert_illumina_quality(illumina_quality):
 180     """Convert an Illumina quality score to a Phred ASCII quality score.
 181     """
 182     # Illumina scores are Phred + 64
 183     # Fastq scores are Phread + 33
 184     # the following code grabs the string, converts to short ints and
 185     # subtracts 31 (64-33) to convert between the two score formats.
 186     # The numpy solution is twice as fast as some of my other
 187     # ideas for the conversion.
 188     # sorry about the uglyness in changing from character, to 8-bit int
 189     # and back to a character array
 190     quality = numpy.asarray(illumina_quality, 'c')
 191     quality.dtype = numpy.uint8
 192     quality -= 31
 193      # I'd like to know what the real numpy char type is
 194     quality.dtype = '|S1'
 195     return quality
 196
 197
 198 if __name__ == "__main__":
 199     main()