def test_split_empty_sample_data_string(self):
schema = self.__get_example_schema("vcf_example.vcf")
cols = """1\t11082325\tRS1\tG\tC,A\t.\t.\tPP=.;DP=.;DPR=.;DPF=.;VC=.;VCR=.;VCF=.;ABPV=.;SBPV=.;MQ=.;BR=.;QD=.\tGT:PL:GQ\t1|0:3000,0,3000:1000\t1|1:2000,0,1000:3""".split(
) # noqa
first_record = next(record.generate_records(schema, cols))
self.assertEqual(first_record.alt, 'C')
self.assertTrue(first_record.from_multi_alt)
second_record = next(record.generate_records(schema, cols))
self.assertEqual(first_record.info, second_record.info)
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_warn_about_unrecognised_key_in_monoallelic_line(self, log):
records = list(generate_records(Schema(), [
'chrZ', '200', '.', 'C', 'T', '.', 'PASS', 'NEW_KEY=value'
]))
for index, record in enumerate(records):
self.assertEqual(
(index, ['value']), (index, record.info['NEW_KEY']))
log.check(
('root',
'WARNING',
'info field {!r} not defined in schema'.format('NEW_KEY')),
)
def test_should_warn_about_too_many_alts_in_field_of_allelic_cardinality(
self,
log):
schema = Schema()
schema.set_info_data('key', 'A', 'String', '')
records = list(generate_records(schema, [
'chrZ', '200', '.', 'C', 'A,T', '.', 'PASS', 'key=a,b,c'
]))
expected = [['a'], ['b']]
for index, record in enumerate(records):
self.assertEqual(expected[index], record.info['key'])
log.check(('wecall.vcfutils.fieldmetadata', 'WARNING',
'expected 2 items in {!r}'.format([['a'], ['b'], ['c']])), )
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'))
def test_should_add_default_parsing_rule_for_unknown_key_in_multiallelic_line(self):
schema = Schema()
records = list(generate_records(schema, [
'chrZ', '200', '.', 'C', 'A,T', '.', 'PASS', 'NEW_KEY=value'
]))
self.assertEqual(0, len(list(schema.iter_info_data())))
for index, record in enumerate(records):
self.assertEqual(
(index, ['value']), (index, record.info['NEW_KEY']))
self.assertEqual(1, len(list(schema.iter_info_data())))
info_metadata = schema.get_info_data('NEW_KEY')
self.assertEqual('.', info_metadata.number)
self.assertEqual('String', info_metadata.data_type)
self.assertEqual(
'Inferred from file content during parsing',
info_metadata.description)
self.assertEqual('vcfutils', info_metadata.source)
self.assertEqual('undefined', info_metadata.version)