def test_haplotypes(self):
# REF TGG TT G-
# 2093 2099 3200
# varset1 CCC GG
# CC GG
# v1 v2,v3 v4
v1 = factory.from_edit('chr24', 2093, 'TGG', 'CCC')
v2 = factory.from_edit('chr24', 2098, 'TT', 'GG')
v3 = factory.from_edit('chr24', 2098, 'TT', 'CC')
v4 = factory.from_edit('chr24', 3200, 'G', 'GG')
hap = variant.Haplotype.from_variants([v1, v2])
s1 = VariantSet.from_variants([hap, v3, v4])
s2 = VariantSet.from_variants(s1.iter_vrt())
vrt = list(s2.find_vrt('chr24', 2090, 2095, expand=True))
self.assertEqual(len(vrt), 2)
self.assertEqual(s2.nof_unit_vrt(), 8)
self.assertEqual(s1.diff(s2).nof_unit_vrt(), 0)
self.assertEqual(len(list(s2.diff_vrt(s1).iter_vrt())), 0)
h1 = variant.Haplotype.from_variants([v1, v2])
h2 = variant.Haplotype.from_variants([v3, v4])
s3 = VariantSet.from_variants([h1, h2])
self.assertEqual(s3.diff(s2).nof_unit_vrt(), 0)
self.assertEqual(s2.diff(s3).nof_unit_vrt(), 0)
def test_diff_vrt(self):
# REF G TTGG C
# 1206 2093 10044
# s1 C CCC T
# G CCC G
# C
# 2092 10044
# s2 TT T
# AC T
variants1 = [('chr24', 1206, 'G', 'C'), ('chr24', 2093, 'TG', 'CC'),
('chr24', 2095, 'G', 'C'), ('chr24', 10044, 'C', 'T'),
('chr24', 10044, 'C', 'G')]
variants2 = [('chr24', 2092, 'TT', 'AC'), ('chr24', 10044, 'C', 'T')]
s1 = VariantSet.from_variants(
[factory.from_edit(*v) for v in variants1])
s2 = VariantSet.from_variants(
[factory.from_edit(*v) for v in variants2])
diff = s1.diff(s2)
self.assertEqual(diff.nof_unit_vrt(), 4)
v1, v2, v3, v4 = list(diff.iter_vrt())
self.assertEqual([v1.start, v1.ref, v1.alt], [1206, 'G', 'C'])
self.assertEqual([v4.start, v4.ref, v4.alt], [10044, 'C', 'G'])
def test_VariantSet_cmp2(self):
vcf1 = pkg_file('genomvar.test', 'data/example3.vcf')
s1 = VariantSet.from_vcf(vcf1)
s2 = VariantSet.from_variants(reversed(list(s1.iter_vrt())))
diff = s1.diff(s2)
self.assertEqual(diff.nof_unit_vrt(), 0)
comm = s1.comm(s2)
self.assertEqual(comm.nof_unit_vrt(), s1.nof_unit_vrt())
self.assertEqual(s2.comm(s1).nof_unit_vrt(), comm.nof_unit_vrt())
def test_to_vcf_no_ref(self):
vs1 = VariantSet.from_variants(
[variant.Del("chr24", 23, 24),
variant.SNP("chr24", 1206, "C")])
buf = io.StringIO()
vs1.to_vcf(buf, reference=self.chr24)
buf.seek(0)
vs2 = VariantSet.from_vcf(buf)
self.assertEqual(vs1.comm(vs2).nof_unit_vrt(), 2)
def test_many_chroms_shuffled(self):
vs1 = VariantSet.from_vcf(
pkg_file('genomvar.test', 'data/example3.vcf'))
vrt = list(vs1.iter_vrt())
vrt2 = [vrt[i] for i in (0, 7, 1, 6, 2, 5, 3, 4)]
vs2 = VariantSet.from_variants(vrt2)
self.assertEqual(vs1.comm(vs2).nof_unit_vrt(), vs1.nof_unit_vrt())
self.assertEqual(vs1.diff(vs2).nof_unit_vrt(), 0)
self.assertEqual(vs2.comm(vs1).nof_unit_vrt(), vs2.nof_unit_vrt())
self.assertEqual(vs2.diff(vs1).nof_unit_vrt(), 0)
def test_diff_of_ambig_indel(self):
# 9945
# CTTTTTCAT
# s1 CTT--TCAT
# s2 C--TTTCAT
s1 = VariantSet.from_variants(
[factory.from_edit('chr24', 9947, 'TT', 'T')])
s2 = VariantSet.from_variants(
[factory.from_edit('chr24', 9945, 'CTT', 'C')])
# diff1 = s1.diff(s2, match_partial=False).iter_vrt()
# self.assertEqual(len(list(diff1)), 0) #TODO fix/reconsider
diff2 = s1.diff(s2).iter_vrt()
self.assertEqual(len(list(diff2)), 1) #TODO fix/reconsider
def test_ambig_difference_snp_in_locus(self):
# 10043
# Ref TC-ACA--G
# v1 s1 CA
# v1 s2 G
# v2 s2 T
fac = variant.VariantFactory(reference=self.chr24, normindel=True)
s1 = VariantSet.from_variants(
[fac.from_edit('chr24', 10047, 'A', 'ACA')])
s2 = VariantSet.from_variants([
fac.from_edit('chr24', 10044, 'C', 'G'),
fac.from_edit('chr24', 10044, 'C', 'CT')
])
self.assertEqual(len(list(s1.comm(s2, match_ambig=True).iter_vrt())),
0)
def test_from_vcf_with_attr(self):
s = VariantSet.from_vcf(pkg_file('genomvar.test', 'data/example1.vcf'),
parse_info=True)
_vrt = list(s.find_vrt('chr24', 150, 160))
self.assertEqual(len(_vrt), 1)
vrt = _vrt[0]
self.assertEqual(vrt.attrib['info']['AF'], 1.0)
# Check multiallelic locus
_vrt = list(s.find_vrt('chr24', 20, 30))
self.assertEqual(len(_vrt), 2)
for vrt in _vrt:
if not vrt.is_variant_instance(variant.Null):
self.assertEqual(vrt.attrib['info']['AF'], 0.5)
# Check None/KeyError cases (".",field absent...)
_vrt = list(
filter(lambda o: not o.is_variant_instance(variant.Null),
s.find_vrt('chr24', 450, 460)))
self.assertEqual(len(_vrt), 1)
vrt = _vrt[0]
with self.assertRaises(ValueError):
vrt.attrib['info']['Randomfields']
_vrt = list(
filter(lambda o: not o.is_variant_instance(variant.Null),
s.find_vrt('chr24', 4750, 4760)))
self.assertEqual(len(_vrt), 1)
vrt = _vrt[0]
self.assertEqual(vrt.attrib['info']['STR'], True)
def test_match_with_haplotypes(self):
# REF TGG TT G|
# 2093 2099 3200
# varset1 CCC GG
# CC GG
# r1 r2,r3 r4
r1 = factory.from_edit('chr24', 2093, 'TGG', 'CCC')
r2 = factory.from_edit('chr24', 2098, 'TT', 'GG')
r3 = factory.from_edit('chr24', 2098, 'TT', 'CC')
r4 = factory.from_edit('chr24', 3200, 'GG', 'G')
hap1 = Haplotype.from_variants([r1, r2])
ccc1 = sorted(hap1.variants, key=lambda o: o.start)[0]
hap2 = Haplotype.from_variants([r3, r4])
s1 = VariantSet.from_variants([hap1, hap2])
hap = Haplotype.from_variants([r1, r4])
ccc2 = sorted(hap.variants, key=lambda o: o.start)[0]
match = s1.match(hap)
self.assertEqual(len(match), 1)
d1 = {k: [v2.base for v2 in v] for k, v in match.items()}
d2 = {ccc2.key: [ccc1]}
self.assertEqual(len(d1), len(d2))
self.assertEqual(list(d1.keys()), list(d2.keys()))
for k in d1:
self.assertTrue(
all([v1.edit_equal(v2) for v1, v2 in zip(d1[k], d2[k])]))
def test_variant_cluster2(self):
vs1 = VariantSet.from_variants([
variant.SNP('chr1', 13366968, 'A'),
variant.SNP('chr1', 13366969, 'T'),
variant.SNP('chr1', 13366970, 'G')
])
vs2 = VariantSet.from_variants([
variant.SNP('chr1', 13366968, 'A'),
variant.SNP('chr1', 13366969, 'T'),
variant.SNP('chr1', 13366970, 'G'),
variant.Ins('chr1', 13366970, 'TTT')
])
com = list(vs1.comm(vs2, match_partial=False).iter_vrt())
self.assertEqual(len(com), 3)
def test_asterisk_variant(self):
vset = VariantSet.from_vcf(pkg_file(
'genomvar.test', 'data/example_with_asterisk.vcf.gz'),
parse_info=True)
vrt = list(vset.find_vrt('chr1', 995507, 995515))
self.assertEqual(len(vrt), 3)
def test_from_variants_vcf(self):
vs0 = varset.VariantSet.from_vcf(pkg_file('genomvar.test',
'data/example1.vcf'),
parse_info=True)
variants1 = sorted(vs0.iter_vrt(), key=lambda v: v.key)
vs = VariantSet.from_variants(variants1)
_desc = 'Test for multinumber field'
info_spec_tuples = [('DP4', 4, 'Integer', _desc),
('NSV', 1, 'Integer')]
info_spec_dict = vs0.dtype['info']
for info_spec in (info_spec_tuples, info_spec_dict):
tf = tempfile.NamedTemporaryFile(suffix='.vcf')
with open(tf.name, 'wt') as fh:
vs.to_vcf(fh, info_spec=info_spec)
with open(tf.name, 'rt') as fh:
self.assertIn(
'##INFO=<ID=DP4,Number=4,Type=Integer,Description="{}">'\
.format(_desc),
fh.read().splitlines())
fh.seek(0)
# print(fh.read())
variants2 = sorted(VCFReader(tf.name).iter_vrt(parse_info=True),
key=lambda v: v.key)
self.assertEqual(len(variants1), len(variants2))
cnt = 0
for v1, v2 in zip(variants1, variants2):
self.assertTrue(v1.edit_equal(v2))
self.assertEqual(v1.attrib['info']['NSV'],
v2.attrib['info']['NSV'])
def test_from_vcf_to_records(self):
vs = VariantSet.from_vcf(pkg_file('genomvar.test',
'data/example1.vcf'),
parse_info=True,
parse_samples=True)
self.assertEqual(vs._samples, ['SAMP1'])
# Test nested dtype
recs = vs.to_records(nested=True)
self.assertEqual(list(recs.dtype.fields), [
'chrom', 'start', 'end', 'ref', 'alt', 'vartype', 'phase_group',
'info', 'SAMPLES'
])
self.assertEqual(
list(recs['info'].dtype.fields),
['NSV', 'AF', 'DP4', 'ECNT', 'pl', 'mt', 'RECN', 'STR'])
self.assertEqual(list(recs['SAMPLES'].dtype.fields), ['SAMP1'])
self.assertEqual(list(recs['SAMPLES']['SAMP1'].dtype.fields), ['GT'])
# Test not nested
recs = vs.to_records(nested=False)
self.assertEqual(list(recs.dtype.fields), [
'chrom', 'start', 'end', 'ref', 'alt', 'vartype', 'phase_group',
'info_NSV', 'info_AF', 'info_DP4', 'info_ECNT', 'info_pl',
'info_mt', 'info_RECN', 'info_STR', 'SAMPLES_SAMP1_GT'
])
def test_from_variants_to_vcf_with_sampdata(self):
file = pkg_file('genomvar.test', 'data/example3.vcf')
variants1 = sorted(VCFReader(file).iter_vrt(parse_samples=True),
key=lambda v: v.key)
vs = VariantSet.from_variants(variants1)
tf = tempfile.NamedTemporaryFile(suffix='.vcf')
with open(tf.name, 'wt') as fh:
vs.to_vcf(
fh,
format_spec=[RESERVED_FORMAT.GT, ('AD', 'R', 'Integer', '')],
samples=['SAMP1'])
with open(tf.name, 'rt') as fh:
fh.seek(0)
self.assertIn(
'##FORMAT=<ID=AD,Number=R,Type=Integer,'\
+'Description="">',
fh.read().splitlines())
variants2 = sorted(VCFReader(tf.name).iter_vrt(parse_samples=True),
key=lambda v: v.key)
self.assertEqual(len(variants1), len(variants2))
cnt = 0
for v1, v2 in zip(variants1, variants2):
self.assertTrue(v1.edit_equal(v2))
self.assertEqual(v1.attrib['samples']['SAMP1']['AD'],
v2.attrib['samples']['SAMP1']['AD'])
def test_sort_chroms(self):
vs = VariantSet.from_vcf(
pkg_file('genomvar.test', 'data/example2.vcf.gz'))
vs.sort_chroms()
self.assertEqual(list(vs.get_chroms()), ['chr23', 'chr24'])
vs.sort_chroms(key=lambda c: 1 if c == 'chr24' else 2)
self.assertEqual(list(vs.get_chroms()), ['chr24', 'chr23'])
def test_empty_vcf(self):
buf = io.StringIO()
with open(pkg_file('genomvar.test', 'data/example1.vcf')) as fh:
for line in itertools.takewhile(lambda l: l.startswith('#'), fh):
buf.write(line)
buf.seek(0)
vs = VariantSet.from_vcf(buf)
self.assertEqual(vs.nof_unit_vrt(), 0)
def test_mnp_com_split(self):
# 23
# TTCACTTAGCATAATGTCTTCAAGATT
# v1 TT -single
# v2 TT -splitted
vset1 = VariantSet.from_variants(
[factory.from_edit('chr24', 22, 'AG', 'TT')])
vset2 = VariantSet.from_variants([
factory.from_edit('chr24', 22, 'A', 'T'),
factory.from_edit('chr24', 23, 'G', 'T')
])
com = vset1.comm(vset2)
v = list(com.iter_vrt())[0]
self.assertEqual([v.start, v.ref], [22, 'AG'])
self.assertFalse(list(vset1.diff(vset2).iter_vrt()))
def test_strip_order_dependent_Ambig(self):
# 10043
# R T--CA--CAG
# v1 TCACA--CAG
# v2 T--CACACAG
factory = variant.VariantFactory(reference=pkg_file(
'genomvar.test', 'data/chr24.fna'),
normindel=True)
v1 = factory.from_edit('chr24', 10043, 'T', 'TCA')
v2 = factory.from_edit('chr24', 10045, 'A', 'ACA')
s1 = VariantSet.from_variants([v1])
s2 = VariantSet.from_variants([v2])
diff = s1.diff(s2, match_ambig=True)
self.assertEqual(len(list(diff.iter_vrt())), 0)
diff = s1.diff(s2, match_ambig=False)
self.assertEqual(len(list(diff.iter_vrt())), 1)
def test_sv_types(self):
with warnings.catch_warnings(record=True) as wrn:
vs = VariantSet.from_vcf(
pkg_file('genomvar.test', 'data/example4.vcf.gz'))
warnings.simplefilter('always')
self.assertEqual(vs.nof_unit_vrt(), 100)
self.assertGreater(len(wrn), 1)
self.assertIn('Structural', str(wrn[-1].message))
def test_from_vcf_with_samples(self):
vset = VariantSet.from_vcf(pkg_file('genomvar.test',
'data/example1.vcf'),
parse_samples=True)
vrt = list(vset.find_vrt('chr24', 1200, 1210))
self.assertEqual(len(vrt), 2)
v1, v2 = vrt
self.assertEqual(v1.attrib['samples']['SAMP1']['GT'], (1, 0))
def test_ambig_difference_different_ambig(self):
# 10043
# Ref T--CA--CA--G
# v1 s1 T--CA--CACAG ins CA right
# v2 s1 T------CA--G del CA left
# v1 s2 TCACA--CA--G ins CA left
# v2 s2 T--CA------G del CA right
fac = variant.VariantFactory(reference=self.chr24, normindel=True)
s1 = VariantSet.from_variants([
fac.from_edit('chr24', 10047, 'A', 'ACA'),
fac.from_edit('chr24', 10043, 'TCA', 'T')
])
s2 = VariantSet.from_variants([
fac.from_edit('chr24', 10043, 'T', 'TCA'),
fac.from_edit('chr24', 10045, 'ACA', 'A')
])
self.assertEqual(len(list(s1.diff(s2, match_ambig=True).iter_vrt())),
0)
def test_from_variants(self):
vfset = VariantSetFromFile(
pkg_file('genomvar.test', 'data/example1.vcf'))
vset = VariantSet.from_variants(list(vfset.iter_vrt()))
vrt = list(vset.find_vrt('chr24', 1200, 1210))
self.assertEqual(len(vrt), 2)
# Test error on out of reference bounds
with self.assertRaises(ValueError):
VariantSet.from_variants(list(vfset.iter_vrt()) +
[variant.SNP('chr24', 10000000, 'C')],
reference=self.chr24)
# Test error on chromosome not in reference
with self.assertRaises(ValueError):
vs = VariantSet.from_variants(list(vfset.iter_vrt()) +
[variant.SNP('chr2', 10, 'C')],
reference=self.chr24)
def test_from_string_buffer(self):
buf = io.StringIO()
with open(pkg_file('genomvar.test', 'data/example1.vcf'), 'rt') as fh:
for line in fh:
buf.write(line)
buf.seek(0)
vs = VariantSet.from_vcf(buf)
self.assertEqual(len(list(vs.find_vrt('chr24', 150, 160))), 1)
self.assertEqual(len(list(vs.find_vrt('chr24', 20, 30))), 2)
def test_in_the_middle(self):
# 2
# TTCACTTAGCAT
# v1 GGG
# v2 G
s1 = VariantSet.from_variants(
[factory.from_edit('chr24', 2, 'CAC', 'GGG')])
s2 = VariantSet.from_variants(
[factory.from_edit('chr24', 3, 'A', 'G')])
diff = s1.diff(s2)
vv = list(diff.iter_vrt())
self.assertEqual(len(vv), 2)
v1, v2 = vv
self.assertEqual([v1.start, v1.ref, v1.alt], [2, 'C', 'G'])
self.assertEqual([v2.start, v2.ref, v2.alt], [4, 'C', 'G'])
self.assertEqual(
len(list(s1.diff(s2, match_partial=False).iter_vrt())), 1)
def test_from_vcf_with_info(self):
vset = VariantSet.from_vcf(pkg_file('genomvar.test',
'data/example1.vcf'),
parse_info=True)
vrt = list(vset.find_vrt('chr24', 1200, 1210))
self.assertEqual(len(vrt), 2)
v1, v2 = vrt
self.assertEqual(v1.alt, 'C')
self.assertEqual(v1.attrib['info']['NSV'], 1)
def test_from_vcf_problematic(self):
vset = VariantSet.from_vcf(pkg_file('genomvar.test',
'data/example5.vcf.gz'),
parse_info=True)
v = list(vset.find_vrt('1', 160109406, 160109409))
self.assertEqual(len(v), 1)
v = v[0]
self.assertEqual(v.attrib['info']['PH'], ('.', ))
def test_VariantSet_cmp(self):
vcf1 = pkg_file('genomvar.test', 'data/example1.vcf')
vcf2 = pkg_file('genomvar.test', 'data/example2.vcf.gz')
s1 = VariantSet.from_vcf(vcf1, parse_info=True, parse_samples=True)
s2 = VariantSet.from_vcf(vcf2)
diff = s1.diff(s2)
self.assertEqual(diff.nof_unit_vrt(), 14)
# Now same diff but without loading in memory
N = 0
for vrt in s1.diff_vrt(s2).iter_vrt():
N += vrt.nof_unit_vrt()
self.assertEqual(N, 14)
comm = s1.comm(s2)
self.assertEqual(len(list(comm.iter_vrt())), 4)
v1, v2 = sorted(comm.iter_vrt(), key=lambda v: v.key)[:2]
self.assertEqual(v1.attrib['info']['NSV'], 1)
self.assertEqual(v1.attrib['samples']['SAMP1']['GT'], (0, 1))
self.assertEqual(s2.comm(s1).nof_unit_vrt(), comm.nof_unit_vrt())
def test_ovlp(self):
# 23
# TTCACTTAGCATAATGTCTTCAAG|ATT
# G
# C <- not interfering
ins = factory.from_edit(chrom='chr24', start=23, ref='G', alt='GG')
s1 = VariantSet.from_variants([ins], )
vb = factory.from_edit(chrom='chr24', start=24, ref='A', alt='C')
self.assertEqual(len(s1.ovlp(vb)), 0)
vb2 = factory.from_edit(chrom='chr24', start=23, ref='G', alt='GG')
self.assertEqual(len(s1.ovlp(vb2, match_ambig=False)), 1)
def test_match2(self):
# 1
# R TCACAG
# del1 T--CAG
# del2 TCA--G
del1 = variant.Del('chrom', 1, 3)
adel1 = variant.AmbigDel('chrom', (1, 1), (5, 3), 'CA', '')
del2 = variant.Del('chrom', 2, 4)
adel2 = variant.AmbigDel('chrom', (1, 2), (5, 4), 'AC', '')
vs = VariantSet.from_variants([del1])
self.assertEqual(len(vs.match(adel1)), 1)
self.assertEqual(len(vs.match(adel2)), 0)
self.assertEqual(len(vs.match(adel2, match_ambig=True)), 1)
vs2 = VariantSet.from_variants([adel1])
self.assertEqual(len(vs2.match(adel1)), 1)
self.assertEqual(len(vs2.match(del1)), 1)
self.assertEqual(len(vs2.match(adel2)), 0)
self.assertEqual(len(vs2.match(adel2, match_ambig=True)), 1)
self.assertEqual(len(vs2.match(del2, match_ambig=True)), 1)
self.assertEqual(len(vs2.match(del2, match_ambig=False)), 0)
def test_find_vrt(self):
ivfs = VariantSetFromFile(pkg_file('genomvar.test',
'data/example2.vcf.gz'),
index=True)
vs = VariantSet.from_vcf(
pkg_file('genomvar.test', 'data/example2.vcf.gz'))
self.assertEqual(
sum([v.nof_unit_vrt() for v in ivfs.find_vrt('chr24')]),
sum([v.nof_unit_vrt() for v in vs.find_vrt('chr24')]))
self.assertEqual(sum([v.nof_unit_vrt() for v in ivfs.iter_vrt()]),
sum([v.nof_unit_vrt() for v in vs.iter_vrt()]))