def test_write_vars_plus_flanks_to_fasta(self):
'''test _write_vars_plus_flanks_to_fasta'''
ref_seqs = {
# 12345678901234567890
'ref1': pyfastaq.sequences.Fasta('ref1', 'AGTGGATAGCTAGCTAGAGA'),
'ref2': pyfastaq.sequences.Fasta('ref2', 'AGGAGAGAGAGAGAGAA'),
'ref3': pyfastaq.sequences.Fasta('ref3', 'AGCTTCATAGAGAGGTTTA'),
}
vcf_records = {
'ref1': [
vcf_record.VcfRecord(
'ref1\t3\tid_1\tT\tC,AG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80'
),
vcf_record.VcfRecord(
'ref1\t10\tid_2\tCT\tA\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80'
),
],
'ref3': [
vcf_record.VcfRecord(
'ref3\t4\tid_3\tT\tG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80'
),
],
}
tmp_file = 'tmp.mapping_based_verifier.write_vars_plus_flanks_to_fasta.fa'
mapping_based_verifier.MappingBasedVerifier._write_vars_plus_flanks_to_fasta(
tmp_file, vcf_records, ref_seqs, 3)
expected_file = os.path.join(data_dir,
'write_vars_plus_flanks_to_fasta.fa')
self.assertTrue(filecmp.cmp(expected_file, tmp_file, shallow=False))
os.unlink(tmp_file)
def test_load_vcf_files(self):
'''test _load_vcf_files'''
vcf_file_1 = os.path.join(data_dir, 'load_vcf_files.1.vcf')
expected_headers = {
vcf_file_1: ['#file1 header1', '#file1 header2'],
}
expected_records = {
'ref.1': [
vcf_record.VcfRecord('ref.1\t5\tid3\tG\tA\tPASS\tSVTYPE=SNP\tGT\t1/1'),
vcf_record.VcfRecord('ref.1\t10\tid1\tA\tT\tPASS\tSVTYPE=SNP\tGT\t1/1'),
],
'ref.2': [vcf_record.VcfRecord('ref.2\t42\tid2\tG\tC\tPASS\tSVTYPE=SNP\tGT\t1/1')],
}
expected_sample = 'sample'
got_sample, got_headers, got_records = vcf_clusterer.VcfClusterer._load_vcf_files([vcf_file_1])
self.assertEqual(expected_sample, got_sample)
self.assertEqual(expected_headers, got_headers)
self.assertEqual(expected_records, got_records)
vcf_file_2 = os.path.join(data_dir, 'load_vcf_files.2.vcf')
expected_headers[vcf_file_2] = ['#file2 header', '#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO\tFORMAT\tsample_from_vcf_2']
expected_records['ref.3'] = [vcf_record.VcfRecord('ref.3\t8\tid5\tA\tG\tPASS\tSVTYPE=SNP\tGT\t1/1')]
expected_records['ref.1'].insert(1, vcf_record.VcfRecord('ref.1\t8\tid4\tC\tG\tPASS\tSVTYPE=SNP\tGT\t1/1'))
expected_sample = 'sample_from_vcf_2'
got_sample, got_headers, got_records = vcf_clusterer.VcfClusterer._load_vcf_files([vcf_file_1, vcf_file_2])
self.assertEqual(expected_sample, got_sample)
self.assertEqual(expected_headers, got_headers)
self.assertEqual(expected_records, got_records)
def test_update_vcf_record_using_gramtools_allele_depths_homozygous(self):
"""test update_using_gramtools_allele_depths homozygous"""
record = vcf_record.VcfRecord(
"ref\t4\t.\tT\tTC,G\t228\t.\tINDEL;IDV=54;IMF=0.885246;DP=61;VDB=7.33028e-19;SGB=-0.693147;MQSB=0.9725;MQ0F=0;AC=2;AN=2;DP4=0,0,23,31;MQ=57\tGT:PL\t1/1:255,163,0"
)
allele_combination_cov = {"1": 1, "2": 80}
allele_groups_dict = {"1": {0}, "2": {2}, "3": {1, 2}}
allele_per_base_cov = [[1], [0, 0], [80]]
expected = vcf_record.VcfRecord(
"ref\t4\t.\tT\tTC,G\t.\t.\t.\tGT:DP:COV:GT_CONF\t2/2:81:1,0,80:87.29"
)
mean_depth = 85
error_rate = 0.001
got_filtered = gramtools.update_vcf_record_using_gramtools_allele_depths(
record,
allele_combination_cov,
allele_per_base_cov,
allele_groups_dict,
mean_depth,
error_rate,
)
self.assertEqual(expected, record)
expected_filtered = vcf_record.VcfRecord(
"ref\t4\t.\tT\tG\t.\t.\t.\tGT:DP:COV:GT_CONF\t1/1:81:1,80:87.29")
self.assertEqual(expected_filtered, got_filtered)
def test_vcf_file_to_dict(self):
'''test vcf_file_to_dict'''
expected_header = ['# header1', '# header2']
lines = [
'ref_42\t11\tid_foo\tA\tG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80',
'ref_42\t12\tid_foo\tC\tG\t42.43\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,53:39.81',
'ref_43\t42\tid_foo\tT\tG\t43.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,54:39.82',
'ref_43\t43\tid_foo\tT\tG,*\t43.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,54:39.83',
'ref_43\t44\tid_foo\tT\t*\t43.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,54:39.84',
]
expected_records = {
'ref_42': [vcf_record.VcfRecord(lines[0]), vcf_record.VcfRecord(lines[1])],
'ref_43': [vcf_record.VcfRecord(lines[x]) for x in (2, 3, 4)],
}
infile = os.path.join(data_dir, 'vcf_file_to_dict.vcf')
got_header, got_records = vcf_file_read.vcf_file_to_dict(infile)
self.assertEqual(expected_records, got_records)
self.assertEqual(expected_header, got_header)
infile = os.path.join(data_dir, 'vcf_file_to_dict.vcf.gz')
got_header, got_records = vcf_file_read.vcf_file_to_dict(infile)
self.assertEqual(expected_records, got_records)
self.assertEqual(expected_header, got_header)
expected_records['ref_43'].pop()
expected_records['ref_43'][-1].remove_asterisk_alts()
infile = os.path.join(data_dir, 'vcf_file_to_dict.vcf')
got_header, got_records = vcf_file_read.vcf_file_to_dict(infile, remove_asterisk_alts=True)
self.assertEqual(expected_records, got_records)
self.assertEqual(expected_header, got_header)
def test_VcfRecord_constructor(self):
'''test VcfRecord constructor'''
line = 'ref_42\t11\tid_foo\tA\tG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80\n'
record = vcf_record.VcfRecord(line)
self.assertEqual(record.CHROM, 'ref_42')
self.assertEqual(record.POS, 10)
self.assertEqual(record.ID, 'id_foo')
self.assertEqual(record.REF, 'A')
self.assertEqual(record.ALT, ['G'])
self.assertEqual(record.QUAL, 42.42)
self.assertEqual(record.FILTER, 'PASS')
self.assertEqual(record.INFO, {
'KMER': '31',
'SVLEN': '0',
'SVTYPE': 'SNP'
})
self.assertEqual(record.FORMAT, {
'GT': '1/1',
'COV': '0,52',
'GT_CONF': '39.80'
})
line = 'ref_42\t11\tid_foo\tA\tG,TC\t.\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80\n'
record = vcf_record.VcfRecord(line)
self.assertEqual(record.QUAL, None)
self.assertEqual(record.ALT, ['G', 'TC'])
def test_make_all_variants_intervals(self):
"""test _make_all_variants_intervals"""
variants = {
"seq.1": [
vcf_record.VcfRecord(
"seq.1\t15\t.\tAGTTGTC\tA\t.\t.\tSVTYPE=DEL"),
vcf_record.VcfRecord("seq.1\t100\t.\tT\tA\t.\t.\tSVTYPE=SNP"),
],
"seq.2":
[vcf_record.VcfRecord("seq.1\t43\t.\tA\tACGTA\t.\t.\tSVTYPE=INS")],
}
big_variant_intervals = {
"seq.1": [
pyfastaq.intervals.Interval(9, 19),
pyfastaq.intervals.Interval(50, 60),
],
"seq.3": [pyfastaq.intervals.Interval(42, 45)],
}
got = dnadiff.Dnadiff._make_all_variants_intervals(
variants, big_variant_intervals)
expected = {
"seq.1": [
pyfastaq.intervals.Interval(9, 20),
pyfastaq.intervals.Interval(50, 60),
pyfastaq.intervals.Interval(99, 99),
],
"seq.2": [pyfastaq.intervals.Interval(42, 42)],
"seq.3": [pyfastaq.intervals.Interval(42, 45)],
}
self.assertEqual(expected, got)
def test_update_vcf_record_using_gramtools_allele_depths_heterozygous(
self):
"""test update_using_gramtools_allele_depths heterozygous"""
record = vcf_record.VcfRecord(
"ref\t4\t.\tT\tA,G,TC\t228\t.\tINDEL;IDV=54;IMF=0.885246;DP=61;VDB=7.33028e-19;SGB=-0.693147;MQSB=0.9725;MQ0F=0;AC=2;AN=2;DP4=0,0,23,31;MQ=57\tGT:PL\t1/1:255,163,0"
)
allele_combination_cov = {"1": 9, "2": 7, "3": 1}
allele_groups_dict = {"1": {0}, "2": {2}, "3": {2, 3}}
allele_per_base_cov = [[0], [9], [7], [1, 0]]
expected = vcf_record.VcfRecord(
"ref\t4\t.\tT\tA,G,TC\t.\t.\t.\tGT:DP:COV:GT_CONF\t0/2:17:9,0,7,0:54.46"
)
mean_depth = 15
error_rate = 0.001
got_filtered = gramtools.update_vcf_record_using_gramtools_allele_depths(
record,
allele_combination_cov,
allele_per_base_cov,
allele_groups_dict,
mean_depth,
error_rate,
)
self.assertEqual(expected, record)
expected_filtered = vcf_record.VcfRecord(
"ref\t4\t.\tT\tG\t.\t.\t.\tGT:DP:COV:GT_CONF\t0/1:17:9,7:54.46")
self.assertEqual(expected_filtered, got_filtered)
def test_write_vars_plus_flanks_to_fasta(self):
"""test _write_vars_plus_flanks_to_fasta"""
ref_seqs = {
# 12345678901234567890
"ref1": pyfastaq.sequences.Fasta("ref1", "AGTGGATAGCTAGCTAGAGA"),
"ref2": pyfastaq.sequences.Fasta("ref2", "AGGAGAGAGAGAGAGAA"),
"ref3": pyfastaq.sequences.Fasta("ref3", "AGCTTCATAGAGAGGTTTA"),
}
vcf_records = {
"ref1": [
vcf_record.VcfRecord(
"ref1\t3\tid_1\tT\tC,AG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80"
),
vcf_record.VcfRecord(
"ref1\t10\tid_2\tCT\tA\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80"
),
],
"ref3": [
vcf_record.VcfRecord(
"ref3\t4\tid_3\tT\tG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80"
)
],
}
tmp_file = "tmp.mapping_based_verifier.write_vars_plus_flanks_to_fasta.fa"
mapping_based_verifier.MappingBasedVerifier._write_vars_plus_flanks_to_fasta(
tmp_file, vcf_records, ref_seqs, 3)
expected_file = os.path.join(data_dir,
"write_vars_plus_flanks_to_fasta.fa")
self.assertTrue(filecmp.cmp(expected_file, tmp_file, shallow=False))
os.unlink(tmp_file)
def test_VcfRecord_constructor(self):
"""test VcfRecord constructor"""
line = "ref_42\t11\tid_foo\tA\tG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80\n"
record = vcf_record.VcfRecord(line)
self.assertEqual(record.CHROM, "ref_42")
self.assertEqual(record.POS, 10)
self.assertEqual(record.ID, "id_foo")
self.assertEqual(record.REF, "A")
self.assertEqual(record.ALT, ["G"])
self.assertEqual(record.QUAL, 42.42)
self.assertEqual(record.FILTER, {"PASS"})
self.assertEqual(record.INFO, {
"KMER": "31",
"SVLEN": "0",
"SVTYPE": "SNP"
})
self.assertEqual(record.FORMAT, {
"GT": "1/1",
"COV": "0,52",
"GT_CONF": "39.80"
})
line = "ref_42\t11\tid_foo\tA\tG,TC\t.\tFilter1;Filter2\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80\n"
record = vcf_record.VcfRecord(line)
self.assertEqual(record.QUAL, None)
self.assertEqual(record.ALT, ["G", "TC"])
self.assertEqual(record.FILTER, {"Filter1", "Filter2"})
line = "ref_42\t11\tid_foo\tA\tG,TC\t.\t.\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80\n"
record = vcf_record.VcfRecord(line)
self.assertEqual(record.FILTER, set())
def test_add_vcf_record_and_len(self):
'''test add_vcf_record and len'''
record1 = vcf_record.VcfRecord(
'ref_42\t11\tid_1\tA\tG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80'
)
record2 = vcf_record.VcfRecord(
'ref_42\t12\tid_2\tC\tG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80'
)
record3 = vcf_record.VcfRecord(
'ref_42\t15\tid_2\tC\tG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80'
)
record4 = vcf_record.VcfRecord(
'ref_42\t19\tid_2\tCCCCC\tG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80'
)
record5 = vcf_record.VcfRecord(
'ref_42\t23\tid_2\tC\tG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80'
)
cluster = vcf_record_cluster.VcfRecordCluster(
max_distance_between_variants=3)
self.assertEqual(0, len(cluster))
self.assertTrue(cluster.add_vcf_record(record1))
self.assertEqual(1, len(cluster))
self.assertTrue(cluster.add_vcf_record(record2))
self.assertEqual(2, len(cluster))
self.assertTrue(cluster.add_vcf_record(record3))
self.assertEqual(3, len(cluster))
self.assertFalse(cluster.add_vcf_record(record4))
self.assertEqual(3, len(cluster))
cluster.max_distance_between_variants = 5
self.assertTrue(cluster.add_vcf_record(record4))
self.assertEqual(4, len(cluster))
self.assertTrue(cluster.add_vcf_record(record5))
self.assertEqual(5, len(cluster))
def test_vcf_records_make_same_allele_combination():
ref_seqs = {"ref1": "GCTGT"}
record1 = vcf_record.VcfRecord("ref1\t1\t.\tGCT\tGC,GCGT\t.\t.\t.")
record2 = vcf_record.VcfRecord("ref1\t5\t.\tT\tTGG,G\t.\t.\t.")
record3 = vcf_record.VcfRecord("ref2\t5\t.\tT\tTGG,G\t.\t.\t.")
assert variant_tracking.vcf_records_make_same_allele_combination(
record1, record2, ref_seqs)
assert not variant_tracking.vcf_records_make_same_allele_combination(
record1, record3, ref_seqs)
def test_total_coverage(self):
"""test total_coverage"""
record = vcf_record.VcfRecord(
"ref\t3\tid_foo\tC\tA\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:GT_CONF\t1/1:39.80\n"
)
self.assertEqual(None, record.total_coverage())
record = vcf_record.VcfRecord(
"ref\t3\tid_foo\tC\tA\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:1,2,39:39.80\n"
)
self.assertEqual(42, record.total_coverage())
def test_to_record_per_alt(self):
"""test to_record_per_alt"""
vcf = vcf_record.VcfRecord("ref\t42\t.\tA\tC\t.\tPASS\tSVTYPE=SNP\n")
self.assertEqual([vcf], vcf.to_record_per_alt())
vcf = vcf_record.VcfRecord(
"ref\t42\t.\tA\tC,TC\t.\tPASS\tSVTYPE=SNP\n")
vcf_c = vcf_record.VcfRecord("ref\t42\t.\tA\tC\t.\tPASS\tSVTYPE=SNP\n")
vcf_tc = vcf_record.VcfRecord(
"ref\t42\t.\tA\tTC\t.\tPASS\tSVTYPE=SNP\n")
self.assertEqual([vcf_c, vcf_tc], vcf.to_record_per_alt())
def test_to_record_per_alt(self):
'''test to_record_per_alt'''
vcf = vcf_record.VcfRecord('ref\t42\t.\tA\tC\t.\tPASS\tSVTYPE=SNP\n')
self.assertEqual([vcf], vcf.to_record_per_alt())
vcf = vcf_record.VcfRecord(
'ref\t42\t.\tA\tC,TC\t.\tPASS\tSVTYPE=SNP\n')
vcf_c = vcf_record.VcfRecord('ref\t42\t.\tA\tC\t.\tPASS\tSVTYPE=SNP\n')
vcf_tc = vcf_record.VcfRecord(
'ref\t42\t.\tA\tTC\t.\tPASS\tSVTYPE=SNP\n')
self.assertEqual([vcf_c, vcf_tc], vcf.to_record_per_alt())
def test_vcf_file_to_dict():
vcf_file = os.path.join(data_dir, "vcf_file_to_dict.vcf")
expect = {
"ref1":
[vcf_record.VcfRecord("ref1\t42\t1\tT\tA\t.\tPASS\t.\tGT\t1/1")],
"ref2": [
vcf_record.VcfRecord("ref2\t43\t3\tT\tA,C\t.\tPASS\t.\tGT\t2/2"),
vcf_record.VcfRecord("ref2\t44\t2\tT\tA\t.\tPASS\t.\tGT\t1/1"),
],
}
got = recall._vcf_file_to_dict(vcf_file)
assert got == expect
def test_load_vcf_files(self):
"""test _load_vcf_files"""
vcf_file_1 = os.path.join(data_dir, "load_vcf_files.1.vcf")
expected_headers = {vcf_file_1: ["#file1 header1", "#file1 header2"]}
expected_records = {
"ref.1": [
vcf_record.VcfRecord(
"ref.1\t5\tid3\tG\tA\tPASS\tSVTYPE=SNP\tGT\t1/1"),
vcf_record.VcfRecord(
"ref.1\t10\tid1\tA\tT\tPASS\tSVTYPE=SNP\tGT\t1/1"),
],
"ref.2": [
vcf_record.VcfRecord(
"ref.2\t42\tid2\tG\tC\tPASS\tSVTYPE=SNP\tGT\t1/1")
],
}
expected_sample = "sample"
(
got_sample,
got_headers,
got_records,
) = vcf_clusterer.VcfClusterer._load_vcf_files([vcf_file_1], None)
self.assertEqual(expected_sample, got_sample)
self.assertEqual(expected_headers, got_headers)
self.assertEqual(expected_records, got_records)
vcf_file_2 = os.path.join(data_dir, "load_vcf_files.2.vcf")
expected_headers[vcf_file_2] = [
"#file2 header",
"#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO\tFORMAT\tsample_from_vcf_2",
]
expected_records["ref.3"] = [
vcf_record.VcfRecord(
"ref.3\t8\tid5\tA\tG\tPASS\tSVTYPE=SNP\tGT\t1/1")
]
expected_records["ref.1"].insert(
1,
vcf_record.VcfRecord(
"ref.1\t8\tid4\tC\tG\tPASS\tSVTYPE=SNP\tGT\t1/1"))
expected_sample = "sample_from_vcf_2"
(
got_sample,
got_headers,
got_records,
) = vcf_clusterer.VcfClusterer._load_vcf_files(
[vcf_file_1, vcf_file_2], None)
self.assertEqual(expected_sample, got_sample)
self.assertEqual(expected_headers, got_headers)
self.assertEqual(expected_records, got_records)
def test_record_with_zero_pos_valueerror_raised(self):
ref_seq = 'AGCTATCTGCGTATTCGATC'
record_1 = vcf_record.VcfRecord(
'ref\t0\t.\tC\tCG\t42.42\tPASS\tSVTPYPE=INDEL\tGT\t1/1')
record_2 = vcf_record.VcfRecord(
'ref\t1\t.\tT\tA\t42.42\tPASS\tSVTPYPE=SNP\tGT\t1/1')
cluster = vcf_record_cluster.VcfRecordCluster(
vcf_record=record_1, max_distance_between_variants=1)
cluster.add_vcf_record(record_2)
with self.assertRaises(ValueError):
cluster.make_one_merged_vcf_record_for_gramtools(ref_seq)
def test_str(self):
"""test __str__"""
line = "ref_42\t11\tid_foo\tA\tG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80"
record = vcf_record.VcfRecord(line)
self.assertEqual(line, str(record))
line = "ref_42\t11\tid_foo\tA\tG\t42.42\tPASS\t."
record = vcf_record.VcfRecord(line)
self.assertEqual(line, str(record))
line = "ref_42\t11\tid_foo\tA\tG\t42.42\tPASS\tFOO;KMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80"
record = vcf_record.VcfRecord(line)
self.assertEqual(line, str(record))
def test_GT_always_printed_first_when_present(self):
"""test GT always printed first when present"""
line1 = "ref\t11\tid\tA\tG\t42.0\tPASS\t.\tGT:COV:GT_CONF\t1/1:0,52:39.80"
line2 = "ref\t11\tid\tA\tG\t42.0\tPASS\t.\tCOV:GT:GT_CONF\t0,52:1/1:39.80"
vcf = vcf_record.VcfRecord(line1)
self.assertEqual(line1, str(vcf))
vcf = vcf_record.VcfRecord(line2)
self.assertEqual(line1, str(vcf))
line3 = "ref\t11\tid\tA\tG\t42.0\tPASS\t.\tCOV:GT_CONF\t0,52:39.80"
vcf = vcf_record.VcfRecord(line3)
self.assertEqual(line3, str(vcf))
vcf.set_format_key_value("GT", "1/1")
self.assertEqual(line1, str(vcf))
def test_remove_asterisk_alts(self):
'''test remove_asterisk_alts'''
record = vcf_record.VcfRecord(
'ref.3\t8\tid5\tA\tG\tPASS\tSVTYPE=SNP\tGT\t1/1')
record.remove_asterisk_alts()
self.assertEqual(['G'], record.ALT)
record = vcf_record.VcfRecord(
'ref.3\t8\tid5\tA\tG,*\tPASS\tSVTYPE=SNP\tGT\t1/1')
record.remove_asterisk_alts()
self.assertEqual(['G'], record.ALT)
record = vcf_record.VcfRecord(
'ref.3\t8\tid5\tA\t*\tPASS\tSVTYPE=SNP\tGT\t1/1')
record.remove_asterisk_alts()
self.assertEqual([], record.ALT)
def test_merge(self):
"""test merge"""
ref_seq = pyfastaq.sequences.Fasta("ref", "AGCTAGGTCAG")
record1 = vcf_record.VcfRecord("wrong_ref\t3\t.\tC\tA\t228\t.\t.\t.")
record2 = vcf_record.VcfRecord(
"ref\t1\t.\tAG\tGAA\t228\t.\tINDEL;IDV=54;IMF=0.885246;DP=61;VDB=7.33028e-19;SGB=-0.693147;MQSB=0.9725;MQ0F=0;AC=2;AN=2;DP4=0,0,23,31;MQ=57\tGT:PL\t1/1:255,163,0"
)
record3 = vcf_record.VcfRecord(
"ref\t2\t.\tG\tAA\t228\t.\tINDEL;IDV=54;IMF=0.885246;DP=61;VDB=7.33028e-19;SGB=-0.693147;MQSB=0.9725;MQ0F=0;AC=2;AN=2;DP4=0,0,23,31;MQ=57\tGT:PL\t1/1:255,163,0"
)
record4 = vcf_record.VcfRecord(
"ref\t3\t.\tC\tCAT\t21.4018\t.\tINDEL;IDV=2;IMF=0.0338983;DP=59;VDB=0.18;SGB=-0.453602;MQ0F=0;AC=2;AN=2;DP4=0,0,0,2;MQ=60\tGT:PL\t1/1:48,6,0"
)
record5 = vcf_record.VcfRecord("ref\t7\t.\tG\tC\t21.4018\t.\t.\t.\t.")
self.assertIsNone(record1.merge(record2, ref_seq))
self.assertIsNone(record2.merge(record1, ref_seq))
self.assertIsNone(record1.merge(record3, ref_seq))
self.assertIsNone(record3.merge(record1, ref_seq))
got = record3.merge(record4, ref_seq)
expected = vcf_record.VcfRecord(
"ref\t2\t.\tGC\tAACAT\t.\t.\tSVTYPE=MERGED\tGT\t1/1")
self.assertEqual(expected, got)
got = record4.merge(record3, ref_seq)
expected = vcf_record.VcfRecord(
"ref\t2\t.\tGC\tAACAT\t.\t.\tSVTYPE=MERGED\tGT\t1/1")
self.assertEqual(expected, got)
got = record4.merge(record5, ref_seq)
expected = vcf_record.VcfRecord(
"ref\t3\t.\tCTAGG\tCATTAGC\t.\t.\tSVTYPE=MERGED\tGT\t1/1")
self.assertEqual(expected, got)
def test_intersects(self):
"""test intersects"""
record1 = vcf_record.VcfRecord("ref_42\t11\t.\tA\tG\t42.42\tPASS\t.")
record2 = vcf_record.VcfRecord("ref_42\t12\t.\tC\tT\t42.42\tPASS\t.")
record3 = vcf_record.VcfRecord("ref_43\t12\t.\tC\tT\t42.42\tPASS\t.")
record4 = vcf_record.VcfRecord("ref_42\t11\t.\tCT\tT\t42.42\tPASS\t.")
self.assertTrue(record1.intersects(record1))
self.assertTrue(record2.intersects(record2))
self.assertFalse(record1.intersects(record2))
self.assertFalse(record2.intersects(record1))
self.assertFalse(record3.intersects(record2))
self.assertFalse(record2.intersects(record3))
self.assertTrue(record2.intersects(record4))
self.assertTrue(record4.intersects(record2))
def test_remove_asterisk_alts(self):
"""test remove_asterisk_alts"""
record = vcf_record.VcfRecord(
"ref.3\t8\tid5\tA\tG\tPASS\tSVTYPE=SNP\tGT\t1/1")
record.remove_asterisk_alts()
self.assertEqual(["G"], record.ALT)
record = vcf_record.VcfRecord(
"ref.3\t8\tid5\tA\tG,*\tPASS\tSVTYPE=SNP\tGT\t1/1")
record.remove_asterisk_alts()
self.assertEqual(["G"], record.ALT)
record = vcf_record.VcfRecord(
"ref.3\t8\tid5\tA\t*\tPASS\tSVTYPE=SNP\tGT\t1/1")
record.remove_asterisk_alts()
self.assertEqual([], record.ALT)
def test_ref_string_matches_dict_of_ref_sequences(self):
"""test ref_string_matches_dict_of_ref_sequences"""
record = vcf_record.VcfRecord(
"ref1\t3\t.\tA\tG\tPASS\tSVTYPE=SNP\tGT\t1/1")
ref_seqs = {"ref1": "GTACG", "ref2": "TTTTT"}
self.assertTrue(
record.ref_string_matches_dict_of_ref_sequences(ref_seqs))
record = vcf_record.VcfRecord(
"ref2\t3\t.\tA\tG\tPASS\tSVTYPE=SNP\tGT\t1/1")
self.assertFalse(
record.ref_string_matches_dict_of_ref_sequences(ref_seqs))
record = vcf_record.VcfRecord(
"ref3\t3\t.\tA\tG\tPASS\tSVTYPE=SNP\tGT\t1/1")
self.assertFalse(
record.ref_string_matches_dict_of_ref_sequences(ref_seqs))
def test_start_and_end(self):
'''test start_and_end'''
cluster = vcf_record_cluster.VcfRecordCluster(
max_distance_between_variants=3)
self.assertEqual((None, None), cluster.start_and_end())
record1 = vcf_record.VcfRecord(
'ref_42\t11\tid_1\tA\tG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80'
)
self.assertTrue(cluster.add_vcf_record(record1))
self.assertEqual((10, 10), cluster.start_and_end())
record2 = vcf_record.VcfRecord(
'ref_42\t12\tid_2\tC\tG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80'
)
self.assertTrue(cluster.add_vcf_record(record2))
self.assertEqual((10, 11), cluster.start_and_end())
def test_get_total_length_of_expected_regions_called(self):
'''test _get_total_length_of_expected_regions_called'''
expected_regions = {
'ref.1': [
pyfastaq.intervals.Interval(101,
200), # 100 long, 92 get called
pyfastaq.intervals.Interval(251,
260), # 10 long, none get called
],
'ref.2': [
pyfastaq.intervals.Interval(42, 43), # 2 long, none get called
],
}
called_vcf_records = {
'ref.1': [
vcf_record.VcfRecord(
'ref.1\t100\t.\tACGTACTGTA\tA,G\t42.0\t.\tDP4=42\tGT\t2/2'
),
],
}
got_all, got_called = mapping_based_verifier.MappingBasedVerifier._get_total_length_of_expected_regions_called(
expected_regions, called_vcf_records)
self.assertEqual(112, got_all)
self.assertEqual(8, got_called)
def test_ref_string_matches_ref_sequence(self):
"""test ref_string_matches_ref_sequence"""
record = vcf_record.VcfRecord(
"ref_name\t1\t.\tAGT\tG\tPASS\tSVTYPE=SNP\tGT\t1/1")
self.assertFalse(record.ref_string_matches_ref_sequence("AG"))
record = vcf_record.VcfRecord(
"ref_name\t3\t.\tA\tG\tPASS\tSVTYPE=SNP\tGT\t1/1")
self.assertTrue(record.ref_string_matches_ref_sequence("GCATG"))
self.assertFalse(record.ref_string_matches_ref_sequence("GCxTG"))
record = vcf_record.VcfRecord(
"ref_name\t3\t.\tAGT\tG\tPASS\tSVTYPE=SNP\tGT\t1/1")
self.assertTrue(record.ref_string_matches_ref_sequence("GCAGT"))
self.assertFalse(record.ref_string_matches_ref_sequence("GCAGC"))
self.assertFalse(record.ref_string_matches_ref_sequence("GCAG"))
self.assertFalse(record.ref_string_matches_ref_sequence("GCA"))
self.assertFalse(record.ref_string_matches_ref_sequence("GA"))
def make_separate_indels_and_one_alt_with_all_snps_no_combinations(self, ref_seq):
'''Returns a VCF record, where each indel from this
cluster is in a separate ALT. Then all the remaining SNPs are
applied to make one ALT. If >1 SNP in same place, either one
might be used'''
final_start_position = min([x.POS for x in self.vcf_records])
final_end_position = max([x.ref_end_pos() for x in self.vcf_records])
snps = []
new_vcf_records = []
for record in self.vcf_records:
if record.is_snp():
snps.append(copy.copy(record))
else:
new_record = copy.copy(record)
new_record.add_flanking_seqs(ref_seq, final_start_position, final_end_position)
new_vcf_records.append(new_record)
if len(snps):
new_record = copy.copy(snps[0])
for snp in snps[1:]:
merged = new_record.merge(snp, ref_seq)
if merged is not None:
new_record = merged
new_record.add_flanking_seqs(ref_seq, final_start_position, final_end_position)
new_vcf_records.append(new_record)
alts = ','.join([x.ALT[0] for x in new_vcf_records])
new_record = vcf_record.VcfRecord('\t'.join([self.vcf_records[0].CHROM, str(final_start_position + 1), '.', new_vcf_records[0].REF, alts, '.', 'PASS', '.']))
return new_record
def _variant_cluster_to_vcf_line(self, variants, variant_ids, max_alleles=None):
if len(variants) == 0:
return None
ref_seq = self.ref_seqs[self.ref_seq_names[variants[0].seq_id]]
if logging.getLogger().level <= logging.DEBUG:
logging.debug(f"Clustering variants:")
for v in variants:
logging.debug(f" {ref_seq.id}\t{v.pos}\t{v.ref}\t{v.alt}")
start, end, alts = allele_combinations.var_cluster_to_coords_and_alts(
variants, ref_seq, max_alleles=max_alleles
)
info_field = "."
if alts is None:
alts = set()
var_id_to_var = dict(zip(variant_ids, variants))
var_patterns = var_patterns_from_block_slices(
self.var_block_tabixes, var_id_to_var, variants[0].seq_id, start, end
)
for var_pattern in var_patterns:
alt_alleles = var_pattern_to_alleles(
var_id_to_var,
var_pattern,
ref_seq,
start,
end,
max_alleles=max_alleles,
)
if alt_alleles is None:
logging.warning("Conflicting allele combination:")
for var_id in sorted(list(var_pattern)):
var = var_id_to_var[var_id]
logging.warning(
f" {ref_seq.id} {var.pos+1} {var.ref} {var.alt}"
)
else:
alts.update(alt_alleles)
info_field = "High_variability"
if len(alts) == 0:
logging.warning("Could not make VCF record from these variants:")
for variant in variants:
logging.warning(" " + str(variant))
return None
else:
return vcf_record.VcfRecord(
"\t".join(
[
ref_seq.id,
str(start + 1),
".",
ref_seq[start : end + 1],
",".join(sorted(list(alts))),
".",
"PASS",
info_field,
]
)
)
def test_ref_end_pos(self):
"""test ref_end_pos"""
line = "ref_42\t11\tid_foo\tA\tG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80\n"
record = vcf_record.VcfRecord(line)
self.assertEqual(10, record.ref_end_pos())
line = "ref_42\t11\tid_foo\tAA\tG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80\n"
record = vcf_record.VcfRecord(line)
self.assertEqual(11, record.ref_end_pos())
line = "ref_42\t11\tid_foo\tAAG\tG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80\n"
record = vcf_record.VcfRecord(line)
self.assertEqual(12, record.ref_end_pos())
line = "ref_42\t11\tid_foo\t.\tG\t42.42\tPASS\tKMER=31;SVLEN=0;SVTYPE=SNP\tGT:COV:GT_CONF\t1/1:0,52:39.80\n"
record = vcf_record.VcfRecord(line)
self.assertEqual(10, record.ref_end_pos())
