def test_should_return_true_if_no_GL_or_PL_present(self):
schema = Schema()
schema.set_sample_data('GT', '1', 'String', '')
schema.samples = ['foo']
records = list(
generate_records(
schema,
['chrZ', '200', '.', 'C', 'A', '.', 'PASS', '.', 'GT', '0/1']))
self.assertTrue(records[0].sample_info.has_no_likelihoods())
def test_should_return_false_if_one_sample_okay_for_GL(self):
schema = Schema()
schema.set_sample_data('GT', '1', 'String', '')
schema.set_sample_data('GL', 'G', 'Float', '')
schema.samples = ['foo']
records = list(
generate_records(schema, [
'chrZ', '200', '.', 'C', 'A', '.', 'PASS', '.', 'GT:GL',
'0/1:90,1,120', '0/1:.,.,.'
]))
self.assertFalse(records[0].sample_info.has_no_likelihoods())
def test_should_return_true_if_all_likelihoods_are_none_for_PL(self):
schema = Schema()
schema.set_sample_data('GT', '1', 'String', '')
schema.set_sample_data('PL', 'G', 'Float', '')
schema.samples = ['foo']
records = list(
generate_records(schema, [
'chrZ', '200', '.', 'C', 'A', '.', 'PASS', '.', 'GT:PL',
'0/1:.,.,.'
]))
self.assertTrue(records[0].sample_info.has_no_likelihoods())
def test_should_parse_column_headers_with_complex_sample_names(self):
lines = [
'##fileformat=VCFv4.2\n',
'#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO\tFORMAT\tOWEN_TOBY-RHYS.JONES\n',
]
reader = VCFReader(iter(lines))
header = reader.read_header()
expected = Schema()
expected.samples = ['OWEN_TOBY-RHYS.JONES']
self.assertEqual(expected, header)
def test_should_warn_when_GT_is_not_present(self, log):
schema = Schema()
schema.set_sample_data('GL', 'G', 'Float', '')
schema.samples = ['foo']
records = list(
generate_records(schema, [
'chrZ', '200', '.', 'C', 'A,T', '.', 'PASS', '.', 'GL', '1,2,3'
]))
for index, record in enumerate(records):
self.assertEqual(
(index, ['1', '2', '3']),
(index, record.sample_info.get_field('foo', 'GL')))
log.check(('wecall.vcfutils.fieldmetadata', 'WARNING',
'Unknown ploidy when parsing genotype likelihood'), )
def test_should_drop_genotype_likelihood_with_mismatch_ploidy(self):
schema = Schema()
schema.set_sample_data('GT', '1', 'String', '')
schema.set_sample_data('GL', 'G', 'Float', '')
schema.samples = ['foo']
records = list(
generate_records(schema, [
'chrZ', '200', '.', 'C', 'A,T', '.', 'PASS', '.', 'GT:GL',
'0/1:1,2,3,4'
]))
self.assertEqual(GenotypeCall("0/1"),
records[0].sample_info.get_field('foo', 'GT'))
self.assertEqual([None, None, None],
records[0].sample_info.get_field('foo', 'GL'))
self.assertEqual(GenotypeCall("0/0"),
records[1].sample_info.get_field('foo', 'GT'))
self.assertEqual([None, None, None],
records[1].sample_info.get_field('foo', 'GL'))
def test_should_split_genotype_likelihood_properly(self):
schema = Schema()
schema.set_sample_data('GT', '1', 'String', '')
schema.set_sample_data('GL', 'G', 'Float', '')
schema.samples = ['foo']
records = list(
generate_records(schema, [
'chrZ', '200', '.', 'C', 'A,T', '.', 'PASS', '.', 'GT:GL',
'0/1:1,2,3,4,5,6'
]))
self.assertEqual(GenotypeCall("0/1"),
records[0].sample_info.get_field('foo', 'GT'))
self.assertEqual([1.0, 2.0, 3.0],
records[0].sample_info.get_field('foo', 'GL'))
self.assertEqual(GenotypeCall("0/0"),
records[1].sample_info.get_field('foo', 'GT'))
self.assertEqual([1.0, 4.0, 6.0],
records[1].sample_info.get_field('foo', 'GL'))
