def test_pileup():
pe = PileupElement('A', 40)
pe2 = PileupElement('A', 20)
pe3 = PileupElement('G', None)
pileup = Pileup([pe, pe2, pe3])
assert pileup.bases() == ['A', 'A', 'G']
assert pileup.quals() == [40, 20, None]
assert pileup.most_common_base() == 'A'
def merge(reads, max_qual=46, fraction_agree=0.75):
"""
:type reads: list[pysam.AlignedSegment]
:type max_qual: int
:type fraction_agree: float
:return: pysam.AlignedSegment
"""
logging.debug("Merging {} read(s).".format(len(reads)))
if len(reads) == 1:
return reads[0]
merged = reads[0]
max_read_len = max([len(r.seq) for r in reads])
new_seq = []
new_quals = []
for k in range(max_read_len):
pileup = Pileup.from_reads(reads, k)
merged_pileup_element = pileup.merge(max_qual, fraction_agree)
new_seq.append(merged_pileup_element.base)
new_quals.append(merged_pileup_element.phredqual())
logging.debug("new seq is {}".format("".join(new_seq)))
logging.debug("new qual is {}".format("".join(new_quals)))
merged.seq = "".join(new_seq)
merged.qual = "".join(new_quals)
return merged
def test_merge():
merged_pileup_elements = []
""" :type: list[PileupElement] """
pileups = []
""" :type: list[Pileup] """
reads = [my_read, my_read2, my_read3, my_read4]
for k in range(0, 4):
pile = Pileup.from_reads(reads, k)
pileups.append(pile)
merged_pileup_elements.append(pile.merge(max_qual=45, fraction_agree=0.75))
# test that ties gives N
assert merged_pileup_elements[0].base == 'N' # tie, 2xA 2xG
# test that majority vote works
assert merged_pileup_elements[1].base == 'G' # G wins from 3xG 1xA
# test all 4 bases that were merged
assert "".join([pe.base for pe in merged_pileup_elements]) == 'NGCT'
# test that qualities were properly merged
assert [pe.qual for pe in merged_pileup_elements] == [2, 45, 45, 45]
# test that the size of each pileup works
assert [len(p.bases()) for p in pileups] == [4, 4, 4, 3]
assert ReadMerger.key(a) == ReadMerger.key(b)
assert Pileup.from_reads([a, b], 0).merge(45, .75).base == 'N'
assert Pileup.from_reads([a, b], 0).merge(45, .75).qual == 2
assert Pileup.from_reads([a, b], 1).merge(45, .75).base == 'G'
tmpfilename = tempfile.mkstemp('.bam')[1]
outfile = pysam.AlignmentFile(tmpfilename, "wb", header=header)
readmerger = ReadMerger([a, b], outfile, fraction_agree=0.75, max_qual=45, reads_between_logs=1)
exitcode = readmerger.do_work()
assert exitcode == 0
metrics = readmerger.metrics.dict()
assert 'L1' in metrics
os.remove(tmpfilename)