def make_call_from_reads(queue,
idx,
calls,
refrfile,
ksize=31,
delta=50,
seedsize=31,
maxdiff=None,
match=1,
mismatch=2,
gapopen=5,
gapextend=0,
min_ikmers=None,
refrseqs=None,
logstream=sys.stderr):
while True:
if queue.empty():
sleep(3)
continue
reads = queue.get()
ccmatch = re.search(r'kvcc=(\d+)', reads[0].name)
cc = ccmatch.group(1) if ccmatch else None
message = '[kevlar::alac::make_call_from_reads'
message += ' (thread={:d})]'.format(idx)
message += ' grabbed partition={} from queue,'.format(cc)
message += ' queue size now {:d}'.format(queue.qsize())
print(message, file=sys.stderr)
# Assemble partitioned reads into contig(s)
contigs = list(assemble_fml_asm(reads, logstream=logstream))
if min_ikmers is not None:
# Apply min ikmer filter if it's set
contigs = [c for c in contigs if len(c.annotations) >= min_ikmers]
if len(contigs) == 0:
queue.task_done()
continue
# Identify the genomic region(s) associated with each contig
localizer = localize(contigs,
refrfile,
seedsize,
delta=delta,
maxdiff=maxdiff,
refrseqs=refrseqs,
logstream=logstream)
targets = list(localizer)
if len(targets) == 0:
queue.task_done()
continue
# Align contigs to genomic targets to make variant calls
caller = call(targets, contigs, match, mismatch, gapopen, gapextend,
ksize, refrfile)
for varcall in caller:
if cc:
varcall.annotate('PART', cc)
calls.append(varcall)
queue.task_done()
def test_call_pico_indel(ccid, varcall):
qfile = data_file('pico' + ccid + '.contig.augfasta')
tfile = data_file('pico' + ccid + '.gdna.fa')
qinstream = kevlar.parse_augmented_fastx(kevlar.open(qfile, 'r'))
queryseqs = [record for record in qinstream]
targetseqs = [record for record in khmer.ReadParser(tfile)]
calls = list(call(targetseqs, queryseqs))
assert len(calls) == 1
assert str(calls[0]) == varcall
def test_call_pico_indel(ccid, varcall):
qfile = data_file('pico' + ccid + '.contig.augfasta')
tfile = data_file('pico' + ccid + '.gdna.fa')
qinstream = kevlar.parse_augmented_fastx(kevlar.open(qfile, 'r'))
queries = [record for record in qinstream]
tinstream = kevlar.reference.load_refr_cutouts(kevlar.open(tfile, 'r'))
targets = [record for record in tinstream]
calls = list(call(targets, queries))
assert len(calls) == 1
assert str(calls[0]) == varcall
def test_snv_dedup():
contigfile = data_file('bee-dupl.contigs.augfasta')
contigstream = kevlar.parse_augmented_fastx(kevlar.open(contigfile, 'r'))
contigs = list(contigstream)
gdnafile = data_file('bee-dupl.gdna.fa')
gdnastream = kevlar.reference.load_refr_cutouts(kevlar.open(gdnafile, 'r'))
targets = list(gdnastream)
calls = list(call(targets, contigs, ksize=27))
assert len(calls) == 1
assert calls[0].seqid == 'linkagegroup5'
assert calls[0].position == 8174 - 1
def alac(pstream, refrfile, ksize=31, delta=25, maxdiff=10000, match=1,
mismatch=2, gapopen=5, gapextend=0, greedy=False,
logstream=sys.stderr):
assembler = assemble_greedy if greedy else assemble_fml_asm
for partition in pstream:
contigs = [c for c in assembler(partition, logstream=logstream)]
targets = [t for t in localize(contigs, refrfile, ksize, delta=delta)]
caller = call(
targets, contigs, match, mismatch, gapopen, gapextend, ksize
)
for varcall in caller:
yield varcall
def test_variant_kmers():
# variant here---------------|
window = 'TTATTTTTAACAAAGGAGCAAAGGAGCAAAGGGCAAATACAATGAGGCAAAGATAGTCTCT'
qfile = data_file('ssc223.contig.augfasta')
tfile = data_file('ssc223.gdna.fa')
qinstream = kevlar.parse_augmented_fastx(kevlar.open(qfile, 'r'))
queryseqs = [record for record in qinstream]
targetseqs = [record for record in khmer.ReadParser(tfile)]
calls = list(call(targetseqs, queryseqs))
assert len(calls) == 1
assert calls[0].window == window
def test_perfect_match():
contigfile = data_file('nodiff.contig.fa')
contigstream = kevlar.parse_augmented_fastx(kevlar.open(contigfile, 'r'))
contigs = list(contigstream)
gdnafile = data_file('nodiff.gdna.fa')
gdnastream = kevlar.reference.load_refr_cutouts(kevlar.open(gdnafile, 'r'))
targets = list(gdnastream)
calls = list(call(targets, contigs))
assert len(calls) == 1
assert calls[0].seqid == 'chr99'
assert calls[0].position == 2899377
assert calls[0].filterstr == 'PerfectMatch'
def test_funky_cigar_deletion():
contigfile = data_file('funkycigar/deletion.contig.fa')
contigstream = kevlar.parse_augmented_fastx(kevlar.open(contigfile, 'r'))
contigs = list(contigstream)
gdnafile = data_file('funkycigar/deletion.gdna.fa')
gdnastream = kevlar.reference.load_refr_cutouts(kevlar.open(gdnafile, 'r'))
targets = list(gdnastream)
calls = list(call(targets, contigs))
assert len(calls) == 1
assert calls[0].seqid == 'chr42'
assert calls[0].position == 53644
assert calls[0]._refr == 'ATGTCTGTTTTCTTAACCT'
assert calls[0]._alt == 'A'
def test_funky_cigar(part, coord, window):
contigfile = data_file('funkycigar/part.cc{:d}.contig.fa.gz'.format(part))
contigstream = kevlar.parse_augmented_fastx(kevlar.open(contigfile, 'r'))
contigs = list(contigstream)
gdnafile = data_file('funkycigar/part.cc{:d}.gdna.fa.gz'.format(part))
gdnastream = kevlar.reference.load_refr_cutouts(kevlar.open(gdnafile, 'r'))
targets = list(gdnastream)
calls = list(call(targets, contigs))
assert len(calls) == 1
print('DEBUG', calls[0].vcf, file=sys.stderr)
assert calls[0].seqid == '17'
assert calls[0].position == coord - 1
assert calls[0].attribute('ALTWINDOW') == window
def test_call_ssc_isolated_snv(ccid, cigar, varcall):
"""
Ensure isolated SNVs are called correctly.
SNVs that are well separated from other variants have a distinct alignment
signature as reflected in the CIGAR string reported by ksw2. They are
either of the form "delete-match-delete" or "delete-match-delete-match",
where the second match is very short (and spurious).
"""
qfile = data_file('ssc' + ccid + '.contig.augfasta')
tfile = data_file('ssc' + ccid + '.gdna.fa')
qinstream = kevlar.parse_augmented_fastx(kevlar.open(qfile, 'r'))
queryseqs = [record for record in qinstream]
targetseqs = [record for record in khmer.ReadParser(tfile)]
calls = list(call(targetseqs, queryseqs))
assert len(calls) == 1
assert str(calls[0]) == varcall
def test_multibest_revcom():
contigfile = data_file('multibestrc.contig.fa')
contigstream = kevlar.parse_augmented_fastx(kevlar.open(contigfile, 'r'))
contigs = list(contigstream)
gdnafile = data_file('multibestrc.gdna.fa')
gdnastream = kevlar.reference.load_refr_cutouts(kevlar.open(gdnafile, 'r'))
targets = list(gdnastream)
calls = list(call(targets, contigs))
assert len(calls) == 4
coordinates = [c.position + 1 for c in calls]
assert coordinates == [34495786, 34583830, 58088279, 60344854]
for c in calls:
assert c._refr == 'A'
assert c._alt == 'G'
assert c.window == ('CCTGAGCCCTCTCAAGTCGGGTCCTGGCCCGGTCTGCCCATGAGGCTGG'
'GCCTGAGCCCCA')
def test_cigar_filter_regression():
contigfile = data_file('14153.cc5463.contig.augfasta.gz')
contigstream = kevlar.parse_augmented_fastx(kevlar.open(contigfile, 'r'))
contigs = list(contigstream)
gdnafile = data_file('14153.cc5463.gdna.augfasta.gz')
gdnastream = kevlar.reference.load_refr_cutouts(kevlar.open(gdnafile, 'r'))
targets = list(gdnastream)
calls = sorted(call(targets, contigs), key=lambda c: c.position)
assert len(calls) == 2
assert calls[1].seqid == '6'
# Equally valid calls from equally optimal alignments
c1 = ('AGAAA', 'A', 154734241)
c2 = ('GAAGA', 'G', 154734239)
varcall = (calls[1]._refr, calls[1]._alt, calls[1].position)
assert varcall in (c1, c2)