def test_haplotag_10X_2():
with TemporaryDirectory() as tempdir:
outbam = tempdir + '/output.bam'
run_haplotag(variant_file='tests/data/haplotag.10X_2.vcf.gz', alignment_file='tests/data/haplotag.10X.bam', output=outbam)
for a1, a2 in zip(pysam.AlignmentFile('tests/data/haplotag.10X.bam'), pysam.AlignmentFile(outbam)):
assert a1.query_name == a2.query_name
if a1.has_tag('HP') and a2.has_tag('HP'):
assert a1.get_tag('HP') == a2.get_tag('HP')
def test_haplotag_sample_given():
with TemporaryDirectory() as tempdir:
outbam = tempdir + '/output.bam'
run_haplotag(variant_file='tests/data/haplotag_sample.vcf.gz', alignment_file='tests/data/haplotag_sample.bam', given_samples=['mother'], output=outbam)
for alignment in pysam.AlignmentFile(outbam):
if alignment.get_tag('RG') == 'mother':
assert alignment.has_tag('HP')
else:
assert not alignment.has_tag('HP')
def test_haplotag_missing_chromosome():
with TemporaryDirectory() as tempdir:
outbam = tempdir + '/output.bam'
# input BAM contains a chromosom for which there is no variant in the input VCF
run_haplotag(variant_file='tests/data/haplotag.missing_chr.vcf.gz', alignment_file='tests/data/haplotag.large.bam', output=outbam)
ps_count = 0
for alignment in pysam.AlignmentFile(outbam):
if alignment.has_tag('PS'):
ps_count += 1
assert ps_count > 0
def test_haplotag():
with TemporaryDirectory() as tempdir:
outvcf = tempdir + '/output.vcf'
outbam = tempdir + '/output.bam'
run_whatshap(phase_input_files=[recombination_breaks_bamfile], variant_file='tests/data/quartet.vcf', output=outvcf, ped='tests/data/recombination_breaks.ped')
run_haplotag(variant_file=outvcf, alignment_file=recombination_breaks_bamfile, output=outbam)
ps_count = 0
for alignment in pysam.AlignmentFile(outbam):
if alignment.has_tag('PS'):
ps_count += 1
assert ps_count > 0
def haplotag_different_sorting():
with TemporaryDirectory() as tempdir:
outbam1 = tempdir + '/output1.bam'
outbam2 = tempdir + '/output2.bam'
# both VCFs contain the same positions, but chromosomes are sorted differently
run_haplotag(variant_file='tests/data/haplotag.large.vcf.gz', alignment_file='tests/data/haplotag.large.bam', output=outbam1)
run_haplotag(variant_file='tests/data/haplotag.large.2.vcf.gz', alignment_file='tests/data/haplotag.large.bam', output=outbam2)
for a1, a2 in zip(pysam.AlignmentFile(outbam1), pysam.AlignmentFile(outbam2)):
assert a1.query_name == a2.query_name
if a1.has_tag('HP'):
assert a2.has_tag('HP')
assert a1.get_tag('HP') == a2.get_tag('HP')
def test_haplotag_no_readgroups1():
with TemporaryDirectory() as tempdir:
outbam1 = tempdir + '/output1.bam'
outbam2 = tempdir + '/output2.bam'
# run haplotag with/without --ignore-read-groups, results should be identical since files contain only data for one sample
run_haplotag(variant_file='tests/data/haplotag_1.vcf.gz', alignment_file='tests/data/haplotag.bam', output=outbam1)
run_haplotag(variant_file='tests/data/haplotag_1.vcf.gz', alignment_file='tests/data/haplotag_noRG.bam', output=outbam2, ignore_read_groups=True)
for a1, a2 in zip(pysam.AlignmentFile(outbam1), pysam.AlignmentFile(outbam2)):
assert a1.query_name == a2.query_name
if a1.has_tag('HP'):
assert a2.has_tag('HP')
assert a1.get_tag('HP') == a2.get_tag('HP')
def test_haplotag2():
with TemporaryDirectory() as tempdir:
outbam = tempdir + '/output.bam'
run_haplotag(variant_file='tests/data/haplotag_2.vcf.gz', alignment_file='tests/data/haplotag.bam', output=outbam)
ps_count = 0
for alignment in pysam.AlignmentFile(outbam):
if alignment.has_tag('PS'):
ps_count += 1
if alignment.has_tag('HP'):
# simulated bam, we know from which haplotype each read originated (given in read name)
true_ht = int(alignment.query_name[-1])
assert true_ht == alignment.get_tag('HP')
assert ps_count > 0
def test_haplotag():
with TemporaryDirectory() as tempdir:
outbam1 = tempdir + '/output1.bam'
outbam2 = tempdir + '/output2.bam'
# run haplotag with two vcfs containing opposite phasings (i.e. 1|0 - 0|1 ..)
run_haplotag(variant_file='tests/data/haplotag_1.vcf.gz', alignment_file='tests/data/haplotag.bam', output=outbam1)
run_haplotag(variant_file='tests/data/haplotag_2.vcf.gz', alignment_file='tests/data/haplotag.bam', output=outbam2)
for a1, a2 in zip(pysam.AlignmentFile(outbam1), pysam.AlignmentFile(outbam2)):
assert a1.query_name == a2.query_name
if a1.has_tag('HP'):
assert a2.has_tag('HP')
assert a1.get_tag('HP') != a2.get_tag('HP')
def test_haplotag_10X():
with TemporaryDirectory() as tempdir:
outbam = tempdir + '/output.bam'
run_haplotag(variant_file='tests/data/haplotag.10X.vcf.gz', alignment_file='tests/data/haplotag.10X.bam', output=outbam)
# map BX tag --> readlist
BX_tag_to_readlist = defaultdict(list)
for alignment in pysam.AlignmentFile(outbam):
if alignment.has_tag('BX') and alignment.has_tag('HP'):
BX_tag_to_readlist[alignment.get_tag('BX')].append(alignment)
# reads having same BX tag need to be assigned to same haplotype
for tag in BX_tag_to_readlist.keys():
haplotype = BX_tag_to_readlist[tag][0].get_tag('HP')
for read in BX_tag_to_readlist[tag]:
assert haplotype == read.get_tag('HP')
def test_haplotag_no_readgroups2(): with raises(SystemExit): # vcf contains multiple samples, there should be an error run_haplotag(alignment_file='tests/data/haplotag_noRG.bam', variant_file='tests/data/haplotag_noRG.vcf.gz', output='/dev/null', ignore_read_groups=True)
