def test_get_sample_names(self):
with vcf.VcfFile(self.example_vcf, 'r') as vcffile:
vcffile.jump2variants()
sample_names = vcffile.get_sample_names()
self.assertEqual(len(sample_names), 3)
self.assertTrue('NA00001' in sample_names)
self.assertTrue('NA00002' in sample_names)
self.assertTrue('NA00003' in sample_names)
def test_is_variant_line(self):
with vcf.VcfFile(self.example_vcf, 'r') as vcffile:
self.assertFalse(vcffile.is_variant_line(vcffile.readline()))
self.assertFalse(vcffile.is_variant_line(vcffile.readline()))
vcffile.jump2variants()
self.assertTrue(vcffile.is_variant_line(vcffile.readline()))
self.assertTrue(
vcffile.is_variant_line(
'20 14370 rs6054257 G A 29 PASS NS=3;DP=14;AF=0.5;DB;H2 GT:GQ:DP:HQ 0|0:48:1:51,51 1|0:48:8:51,51 1/1:43:5:.,.\n'
))
def test_jump2variants(self):
with vcf.VcfFile(self.example_vcf, 'r') as vcffile:
vcffile.jump2variants()
# Test that column names are set
self.assertEqual(vcffile.column_names,
self.example_vcf_column_names)
# Test that the next line is the first variant
self.assertEqual(
'20 14370 rs6054257 G A 29 PASS NS=3;DP=14;AF=0.5;DB;H2 GT:GQ:DP:HQ 0|0:48:1:51,51 1|0:48:8:51,51 1/1:43:5:.,.\n',
vcffile.readline())
def test_set_column_names(self):
with vcf.VcfFile(self.example_vcf, 'r') as vcffile:
# Test that error is raised when trying to input non-header line
self.assertRaises(ValueError, vcffile.set_column_names,
(vcffile.readline()))
# Check that the header line is correctly parsed
vcffile.set_column_names(self.example_vcf_column_names_str)
self.assertEqual(vcffile.column_names[0], 'CHROM')
self.assertEqual(vcffile.column_names[1], 'POS')
self.assertEqual(vcffile.column_names[7], 'INFO')
def test_parse_line(self):
with vcf.VcfFile(self.example_vcf, 'r') as vcffile:
vcffile.jump2variants()
i = 0
for line in vcffile:
variant = vcffile.parse_line(line)
if i == 0:
self.assertEqual(variant['CHROM'], '20')
self.assertEqual(variant['ID'], 'rs6054257')
self.assertEqual(variant['NA00003'], '1/1:43:5:.,.')
elif i == 1:
self.assertEqual(variant['INFO'], 'NS=3;DP=11;AF=0.017')
self.assertEqual(variant['FORMAT'], 'GT:GQ:DP:HQ')
self.assertEqual(variant['NA00003'], '0/0:41:3')
break # exit loop after 2 iterations
i += 1
def test_read_variant(self):
with vcf.VcfFile(self.example_vcf, 'r') as vcffile:
# Check that error is raised when reading non-variant line
self.assertRaises(ValueError, vcffile.read_variant)
# Check that the variant row is read in correctly
vcffile.jump2variants()
variant = vcffile.read_variant()
self.assertEqual(variant['CHROM'], '20')
self.assertEqual(variant['ID'], 'rs6054257')
self.assertEqual(variant['NA00003'], '1/1:43:5:.,.')
variant = vcffile.read_variant()
self.assertEqual(variant['INFO'], 'NS=3;DP=11;AF=0.017')
self.assertEqual(variant['FORMAT'], 'GT:GQ:DP:HQ')
self.assertEqual(variant['NA00003'], '0/0:41:3')
def test_parse_info(self):
with vcf.VcfFile(self.example_vcf, 'r') as vcffile:
# Check that the variant row is read in correctly
vcffile.jump2variants()
variant = vcffile.read_variant()
info_map, info_single = vcffile.parse_info(variant)
self.assertEqual(info_map['AF'], '0.5')
self.assertEqual(info_map['DP'], '14')
self.assertEqual(len(info_single), 2)
self.assertTrue('DB' in info_single)
self.assertTrue('H2' in info_single)
variant = vcffile.read_variant()
info_map, info_single = vcffile.parse_info(variant)
self.assertEqual(info_map['NS'], '3')
self.assertEqual(info_map['AF'], '0.017')
def test_parse_samples(self):
with vcf.VcfFile(self.example_vcf, 'r') as vcffile:
vcffile.jump2variants()
variant = vcffile.read_variant()
sample2field2val = vcffile.parse_samples(variant)
self.assertEqual(len(sample2field2val), 3)
self.assertTrue('NA00001' in sample2field2val)
self.assertTrue('NA00002' in sample2field2val)
self.assertTrue('NA00003' in sample2field2val)
self.assertEqual(sample2field2val['NA00001']['GT'], '0|0')
self.assertEqual(sample2field2val['NA00002']['GQ'], '48')
self.assertEqual(sample2field2val['NA00003']['HQ'], '.,.')
variant = vcffile.read_variant()
sample2field2val = vcffile.parse_samples(variant)
self.assertEqual(sample2field2val['NA00001']['HQ'], '58,50')
self.assertEqual(sample2field2val['NA00002']['GT'], '0|1')
self.assertEqual(sample2field2val['NA00003']['GQ'], '41')
def test_get_sample_gt(self):
with vcf.VcfFile(self.example_vcf, 'r') as vcffile:
vcffile.jump2variants()
variant = vcffile.read_variant()
gt = vcffile.get_sample_gt(variant, 'NA00001', phased=True)
self.assertEqual(gt, 'G|G')
variant = vcffile.read_variant()
gt = vcffile.get_sample_gt(variant, 'NA00002', phased=True)
self.assertEqual(gt, 'T|A')
variant = vcffile.read_variant()
gt = vcffile.get_sample_gt(variant, 'NA00003', phased=False)
self.assertEqual(gt, 'T/T')
variant = vcffile.read_variant()
variant = vcffile.read_variant()
gt = vcffile.get_sample_gt(variant, 'NA00001', phased=False)
self.assertEqual(gt, 'G/GTC')
def parse_vcf(vcf_in, fout):
'''
Read through the vcf file, and parse it.
Output results in tsv format
'''
with vcf.VcfFile(vcf_in, 'r') as vcffile:
# Skip to the variants section of the vcf file
vcffile.jump2variants()
sample_names = vcffile.get_sample_names()
sample_names_gt = [sn + '_gt' for sn in sample_names]
sample_names_dp = [sn + '_dp' for sn in sample_names]
# Output header line
fout.write('%s\n' % '\t'.join(
['Chrom', 'Position', 'Ref', 'Alt', 'Type', 'AF', 'NoCall'] +
sample_names_gt + sample_names_dp))
# Read in the variant lines
for line in vcffile:
# Get parsed variant data
variant = vcffile.parse_line(line)
# Record columns
chrom = variant['CHROM']
pos = variant['POS']
ref = variant['REF']
alt = variant['ALT']
# Parse the info column
info_map, info_single = vcffile.parse_info(variant)
af = info_map['AF']
# Variant type
variant_type = 'snp'
len_ref = len(ref)
len_alt = len(alt)
if len_ref > 1 or len_alt > 1:
if len_ref > len_alt:
variant_type = 'del'
elif len_ref < len_alt:
variant_type = 'ins'
else: # len_ref == len_alt
if len_ref == 2:
variant_type = 'dnp'
elif len_ref == 3:
variant_type = 'tnp'
else:
variant_type = 'onp'
# Sample genotypes
samples2field2val = vcffile.parse_samples(variant)
samples_gts = []
samples_dps = []
num_nocall = 0
for sample in sample_names:
sample_gt = vcffile.get_sample_gt(variant, sample)
if sample_gt == 'N/N':
num_nocall += 1
if 'DP' not in samples2field2val[sample]:
sample_dp = 'NA'
else:
sample_dp = samples2field2val[sample]['DP']
samples_gts.append(sample_gt)
samples_dps.append(sample_dp)
# Output to standard output
fout.write('%s\n' % '\t'.join(
[chrom, pos, ref, alt, variant_type, af,
str(num_nocall)] + samples_gts + samples_dps))
def test_is_header(self):
with vcf.VcfFile(self.example_vcf, 'r') as vcffile:
self.assertTrue(
vcffile.is_header(self.example_vcf_column_names_str))
self.assertFalse(
vcffile.is_header('#' + self.example_vcf_column_names_str))
def test_is_meta(self):
with vcf.VcfFile(self.example_vcf, 'r') as vcffile:
self.assertTrue(vcffile.is_meta('##fileDate=20120818'))
self.assertFalse(vcffile.is_meta('# blahblah'))
self.assertFalse(vcffile.is_meta('foo'))