def test_fastq_to_tabular_msa(self):
class CustomSequence(GrammaredSequence):
@classproperty
@overrides(GrammaredSequence)
def gap_chars(cls):
return set('-.')
@classproperty
@overrides(GrammaredSequence)
def default_gap_char(cls):
return '-'
@classproperty
@overrides(GrammaredSequence)
def nondegenerate_chars(cls):
return set(string.ascii_letters)
@classproperty
@overrides(GrammaredSequence)
def degenerate_map(cls):
return {}
for valid_files, kwargs, components in self.valid_configurations:
for valid in valid_files:
for observed_kwargs in kwargs:
_drop_kwargs(observed_kwargs, 'seq_num')
if 'constructor' not in observed_kwargs:
observed_kwargs['constructor'] = CustomSequence
constructor = observed_kwargs['constructor']
expected_kwargs = {}
expected_kwargs['lowercase'] = 'introns'
observed_kwargs['lowercase'] = 'introns'
expected = TabularMSA([
constructor(c[2],
metadata={
'id': c[0],
'description': c[1]
},
positional_metadata={
'quality': np.array(c[3],
dtype=np.uint8)
},
**expected_kwargs) for c in components
])
observed = _fastq_to_tabular_msa(valid, **observed_kwargs)
self.assertEqual(observed, expected)
def test_fastq_to_tabular_msa(self):
class CustomSequence(GrammaredSequence):
@classproperty
@overrides(GrammaredSequence)
def gap_chars(cls):
return set('-.')
@classproperty
@overrides(GrammaredSequence)
def default_gap_char(cls):
return '-'
@classproperty
@overrides(GrammaredSequence)
def definite_chars(cls):
return set(string.ascii_letters)
@classproperty
@overrides(GrammaredSequence)
def degenerate_map(cls):
return {}
for valid_files, kwargs, components in self.valid_configurations:
for valid in valid_files:
for observed_kwargs in kwargs:
_drop_kwargs(observed_kwargs, 'seq_num')
if 'constructor' not in observed_kwargs:
observed_kwargs['constructor'] = CustomSequence
constructor = observed_kwargs['constructor']
expected_kwargs = {}
expected_kwargs['lowercase'] = 'introns'
observed_kwargs['lowercase'] = 'introns'
expected = TabularMSA(
[constructor(
c[2], metadata={'id': c[0],
'description': c[1]},
positional_metadata={'quality': np.array(c[3],
dtype=np.uint8)},
**expected_kwargs)
for c in components])
observed = _fastq_to_tabular_msa(valid, **observed_kwargs)
self.assertEqual(observed, expected)
def test_fastq_to_tabular_msa_no_constructor(self):
with self.assertRaisesRegex(ValueError, '`constructor`'):
_fastq_to_tabular_msa(get_data_path('fastq_multi_seq_sanger'))
