Exemplo n.º 1
0
 def convertReadAlignment(self, read):
     """
     Convert a pysam ReadAlignment to a GA4GH ReadAlignment
     """
     # TODO fill out remaining fields
     # TODO refine in tandem with code in converters module
     ret = protocol.ReadAlignment()
     ret.alignedQuality = list(read.query_qualities)
     ret.alignedSequence = read.query_sequence
     ret.alignment = protocol.LinearAlignment()
     ret.alignment.mappingQuality = read.mapping_quality
     ret.alignment.position = protocol.Position()
     ret.alignment.position.referenceName = self._samFile.getrname(
         read.reference_id)
     ret.alignment.position.position = read.reference_start
     ret.alignment.position.strand = \
         protocol.Strand.POS_STRAND  # TODO fix this!
     ret.alignment.cigar = []
     for operation, length in read.cigar:
         gaCigarUnit = protocol.CigarUnit()
         gaCigarUnit.operation = SamCigar.int2ga(operation)
         gaCigarUnit.operationLength = length
         gaCigarUnit.referenceSequence = None  # TODO fix this!
         ret.alignment.cigar.append(gaCigarUnit)
     ret.duplicateFragment = SamFlags.isFlagSet(read.flag,
                                                SamFlags.DUPLICATE_FRAGMENT)
     ret.failedVendorQualityChecks = SamFlags.isFlagSet(
         read.flag, SamFlags.FAILED_VENDOR_QUALITY_CHECKS)
     ret.fragmentLength = read.template_length
     ret.fragmentName = read.query_name
     ret.id = "{}:{}".format(self._id, read.query_name)
     ret.info = {key: [str(value)] for key, value in read.tags}
     ret.nextMatePosition = None
     if read.next_reference_id != -1:
         ret.nextMatePosition = protocol.Position()
         ret.nextMatePosition.referenceName = self._samFile.getrname(
             read.next_reference_id)
         ret.nextMatePosition.position = read.next_reference_start
         ret.nextMatePosition.strand = \
             protocol.Strand.POS_STRAND  # TODO fix this!
     # TODO Is this the correct mapping between numberReads and
     # sam flag 0x1? What about the mapping between numberReads
     # and 0x40 and 0x80?
     ret.numberReads = None
     ret.readNumber = None
     if SamFlags.isFlagSet(read.flag, SamFlags.NUMBER_READS):
         ret.numberReads = 2
         if SamFlags.isFlagSet(read.flag, SamFlags.READ_NUMBER_ONE):
             ret.readNumber = 0
         elif SamFlags.isFlagSet(read.flag, SamFlags.READ_NUMBER_TWO):
             ret.readNumber = 1
     ret.properPlacement = SamFlags.isFlagSet(read.flag,
                                              SamFlags.PROPER_PLACEMENT)
     ret.readGroupId = self._id
     ret.secondaryAlignment = SamFlags.isFlagSet(
         read.flag, SamFlags.SECONDARY_ALIGNMENT)
     ret.supplementaryAlignment = SamFlags.isFlagSet(
         read.flag, SamFlags.SUPPLEMENTARY_ALIGNMENT)
     return ret
Exemplo n.º 2
0
 def convertReadAlignment(self, read):
     """
     Convert a pysam ReadAlignment to a GA4GH ReadAlignment
     """
     samFile = self._parentContainer.getFileHandle(self._parentSamFilePath)
     # TODO fill out remaining fields
     # TODO refine in tandem with code in converters module
     ret = protocol.ReadAlignment()
     ret.fragmentId = 'TODO'
     if read.query_qualities is None:
         ret.alignedQuality = []
     else:
         ret.alignedQuality = list(read.query_qualities)
     ret.alignedSequence = read.query_sequence
     if SamFlags.isFlagSet(read.flag, SamFlags.READ_UNMAPPED):
         ret.alignment = None
     else:
         ret.alignment = protocol.LinearAlignment()
         ret.alignment.mappingQuality = read.mapping_quality
         ret.alignment.position = protocol.Position()
         ret.alignment.position.referenceName = samFile.getrname(
             read.reference_id)
         ret.alignment.position.position = read.reference_start
         ret.alignment.position.strand = protocol.Strand.POS_STRAND
         if SamFlags.isFlagSet(read.flag, SamFlags.READ_REVERSE_STRAND):
             ret.alignment.position.strand = protocol.Strand.NEG_STRAND
         ret.alignment.cigar = []
         for operation, length in read.cigar:
             gaCigarUnit = protocol.CigarUnit()
             gaCigarUnit.operation = SamCigar.int2ga(operation)
             gaCigarUnit.operationLength = length
             gaCigarUnit.referenceSequence = None  # TODO fix this!
             ret.alignment.cigar.append(gaCigarUnit)
     ret.duplicateFragment = SamFlags.isFlagSet(read.flag,
                                                SamFlags.DUPLICATE_READ)
     ret.failedVendorQualityChecks = SamFlags.isFlagSet(
         read.flag, SamFlags.FAILED_QUALITY_CHECK)
     ret.fragmentLength = read.template_length
     ret.fragmentName = read.query_name
     ret.info = {key: [str(value)] for key, value in read.tags}
     if SamFlags.isFlagSet(read.flag, SamFlags.MATE_UNMAPPED):
         ret.nextMatePosition = None
     else:
         ret.nextMatePosition = protocol.Position()
         if read.next_reference_id != -1:
             ret.nextMatePosition.referenceName = samFile.getrname(
                 read.next_reference_id)
         else:
             ret.nextMatePosition.referenceName = ""
         ret.nextMatePosition.position = read.next_reference_start
         ret.nextMatePosition.strand = protocol.Strand.POS_STRAND
         if SamFlags.isFlagSet(read.flag, SamFlags.MATE_REVERSE_STRAND):
             ret.nextMatePosition.strand = protocol.Strand.NEG_STRAND
     if SamFlags.isFlagSet(read.flag, SamFlags.READ_PAIRED):
         ret.numberReads = 2
     else:
         ret.numberReads = 1
     ret.readNumber = None
     if SamFlags.isFlagSet(read.flag, SamFlags.FIRST_IN_PAIR):
         if SamFlags.isFlagSet(read.flag, SamFlags.SECOND_IN_PAIR):
             ret.readNumber = 2
         else:
             ret.readNumber = 0
     elif SamFlags.isFlagSet(read.flag, SamFlags.SECOND_IN_PAIR):
         ret.readNumber = 1
     ret.properPlacement = SamFlags.isFlagSet(read.flag,
                                              SamFlags.READ_PROPER_PAIR)
     ret.readGroupId = self.getId()
     ret.secondaryAlignment = SamFlags.isFlagSet(
         read.flag, SamFlags.SECONDARY_ALIGNMENT)
     ret.supplementaryAlignment = SamFlags.isFlagSet(
         read.flag, SamFlags.SUPPLEMENTARY_ALIGNMENT)
     ret.id = self.getReadAlignmentId(ret)
     return ret