def test_infer_annotation_pipeline(self):
anno_fields = ['CSQ']
header = self._get_sample_header_fields(with_annotation=True)
variant1 = self._get_sample_variant_1()
variant1.info['CSQ'] = [
'A|1|100|1.2', 'A|2|101|1.3', 'A|12|start|0', 'TT|13|end|7'
]
variant2 = self._get_sample_variant_1()
variant2.info['CSQ'] = [
'A|1|100|', 'A|2|101|', 'A|1.2|102|0', 'TT|1.3|103|7'
]
desc = 'Inferred type field for annotation {}.'
expected = vcf_header_io.VcfHeader(
infos={
'CSQ_Gene_TYPE':
createInfo('CSQ_Gene_TYPE', 1, 'Float', desc.format('Gene')),
'CSQ_Position_TYPE':
createInfo('CSQ_Position_TYPE', 1, 'String',
desc.format('Position')),
'CSQ_Score_TYPE':
createInfo('CSQ_Score_TYPE', 1, 'Float', desc.format('Score'))
})
with TestPipeline() as p:
inferred_headers = (
p
| Create([variant1, variant2])
| 'InferAnnotationTypes' >> infer_headers.InferHeaderFields(
defined_headers=header,
infer_headers=False,
annotation_fields_to_infer=anno_fields))
assert_that(inferred_headers,
asserts.header_fields_equal_ignore_order([expected]))
p.run()
def test_header_fields_inferred_from_two_variants(self):
with TestPipeline() as p:
variant_1 = self._get_sample_variant_1()
variant_2 = self._get_sample_variant_2()
inferred_headers = (
p
| Create([variant_1, variant_2])
| 'InferUndefinedHeaderFields' >>
infer_undefined_headers.InferUndefinedHeaderFields(
defined_headers=None))
expected_infos = {'IS': Info('IS', 1, 'String', '', '', ''),
'ISI': Info('ISI', 1, 'Integer', '', '', ''),
'ISF': Info('ISF', 1, 'Float', '', '', ''),
'IF': Info('IF', 1, 'Float', '', '', ''),
'IB': Info('IB', 0, 'Flag', '', '', ''),
'IA': Info('IA', None, 'Float', '', '', ''),
'IS_2': Info('IS_2', 1, 'String', '', '', '')}
expected_formats = {'FI': Format('FI', 1, 'Integer', ''),
'FU': Format('FU', None, 'Float', ''),
'FI_2': Format('FI_2', 1, 'Integer', '')}
expected = vcf_header_io.VcfHeader(
infos=expected_infos, formats=expected_formats)
assert_that(inferred_headers,
asserts.header_fields_equal_ignore_order([expected]))
p.run()
def test_pipeline(self):
infos = {'IS': Info('IS', 1, 'String', '', '', ''),
'ISI': Info('ISI', 1, 'Integer', '', '', ''),
'ISF': Info('ISF', 1, 'Float', '', '', ''),
'IB': Info('IB', 0, 'Flag', '', '', ''),
'IA': Info('IA', -1, 'Integer', '', '', '')}
formats = OrderedDict([
('FS', Format('FS', 1, 'String', 'desc')),
('FI', Format('FI', 2, 'Integer', 'desc')),
('GT', Format('GT', 2, 'Integer', 'Special GT key')),
('PS', Format('PS', 1, 'Integer', 'Special PS key'))])
with TestPipeline() as p:
variant_1 = self._get_sample_variant_info_ia_cardinality_mismatch()
variant_2 = self._get_sample_variant_format_fi_float_value()
inferred_headers = (
p
| Create([variant_1, variant_2])
| 'InferHeaderFields' >>
infer_headers.InferHeaderFields(
defined_headers=vcf_header_io.VcfHeader(infos=infos,
formats=formats),
allow_incompatible_records=True))
expected_infos = {'IA': Info('IA', None, 'Float', '', '', ''),
'IF': Info('IF', 1, 'Float', '', '', '')}
expected_formats = {'FI': Format('FI', 2, 'Float', 'desc'),
'FU': Format('FU', None, 'Float', '')}
expected = vcf_header_io.VcfHeader(infos=expected_infos,
formats=expected_formats)
assert_that(inferred_headers,
asserts.header_fields_equal_ignore_order([expected]))
p.run()
def test_defined_fields_filtered_two_variants(self):
# Only INFO and FORMAT in the first variants are already defined in the
# header section of the VCF files.
with TestPipeline() as p:
vcf_headers = self._get_sample_header_fields()
vcf_headers_side_input = p | 'vcf_header' >> Create([vcf_headers])
variant_1 = self._get_sample_variant_1()
variant_2 = self._get_sample_variant_2()
inferred_headers = (
p
| Create([variant_1, variant_2])
| 'InferUndefinedHeaderFields' >>
infer_undefined_headers.InferUndefinedHeaderFields(
pvalue.AsSingleton(vcf_headers_side_input)))
expected_infos = {'IS_2': Info('IS_2', 1, 'String', '', '', '')}
expected_formats = {'FI_2': Format('FI_2', 1, 'Integer', '')}
expected = vcf_header_io.VcfHeader(
infos=expected_infos, formats=expected_formats)
assert_that(inferred_headers,
asserts.header_fields_equal_ignore_order([expected]))
p.run()
