def test_collate_samples(self):
"""Ensure collate_samples task works."""
sample1_data = {'name': 'Sample01', KRAKEN_NAME: create_kraken()}
sample2_data = {'name': 'Sample02', KRAKEN_NAME: create_kraken()}
sample1 = Sample(**sample1_data).save()
sample2 = Sample(**sample2_data).save()
sample_group = add_sample_group(name='SampleGroup01')
sample_group.samples = [sample1, sample2]
db.session.commit()
samples = [sample.fetch_safe() for sample in [sample1, sample2]]
result = collate_samples.delay(KRAKEN_NAME, ['taxa'], samples).get()
self.assertIn('Sample01', result)
self.assertIn('Sample02', result)
self.assertIn('taxa', result['Sample01'])
self.assertIn('taxa', result['Sample02'])
def test_run_taxa_tree_sample(self): # pylint: disable=invalid-name
"""Ensure TaxaTree run_sample produces correct results."""
kwargs = {
KrakenResultModule.name(): create_kraken(),
KrakenHLLResultModule.name(): create_krakenhll(),
Metaphlan2ResultModule.name(): create_metaphlan2(),
}
self.generic_run_sample_test(kwargs, TaxaTreeDisplayModule)
def test_tsne_returns_data(self):
"""
Ensure run_tsne method returns array of the correct size.
tSNE is non-deterministic so that is as close as we can get to a real test.
"""
sample_dict = {f'SMPL_{i}': dict(create_kraken().taxa) for i in range(3)}
tsne_output = run_tsne(sample_dict)
self.assertEqual((3, 2), tsne_output.shape)
def create_sample(i):
"""Create unique sample for index i."""
args = {
'name': f'Sample{i}',
'metadata': {'foobar': f'baz{i}'},
CARDAMRResultModule.name(): create_card_amr(),
KrakenResultModule.name(): create_kraken(),
Metaphlan2ResultModule.name(): create_metaphlan2(),
}
return Sample(**args).save()
def create_sample(i):
"""Create unique sample for index i."""
metadata = {'foobar': f'baz{i}'}
sample_data = {
'name': f'Sample{i}',
'metadata': metadata,
KRAKEN_NAME: create_kraken(),
KRAKENHLL_NAME: create_krakenhll(),
METAPHLAN2_NAME: create_metaphlan2(),
}
return Sample(**sample_data).save()
def create_sample(i):
"""Create unique sample for index i."""
return Sample(
**{
'name': f'Sample{i}',
'metadata': {
'foobar': f'baz{i}'
},
KrakenResultModule.name(): create_kraken(),
KrakenHLLResultModule.name(): create_krakenhll(),
Metaphlan2ResultModule.name(): create_metaphlan2(),
}).save()
def test_clean_zeroed_samples(self):
"""Ensure get_clean_samples method removes features below threshold."""
sample_dict = {f'SMPL_{i}': dict(create_kraken().taxa) for i in range(3)}
sample_dict['SMPL_1']['somebadkingdom'] = 0.0000001
all_feature_ids = set([])
for feature_set in sample_dict.values():
all_feature_ids |= set(feature_set.keys())
result = get_clean_samples(sample_dict)
for feature_set in result.values():
self.assertNotIn('somebadkingdom', feature_set)
def test_clean_samples(self):
"""Ensure get_clean_samples method adds missing features to all samples."""
sample_dict = {f'SMPL_{i}': create_kraken().taxa for i in range(3)}
all_feature_ids = set([])
for feature_set in sample_dict.values():
all_feature_ids |= set(feature_set.keys())
result = get_clean_samples(sample_dict, no_zero_features=False)
for feature_set in result.values():
for feature_id in all_feature_ids:
self.assertIn(feature_id, feature_set)
def create_sample(i):
"""Create unique sample for index."""
sample_data = {'name': f'SMPL_{i}', KRAKEN_NAME: create_kraken()}
return Sample(**sample_data)
def test_tool_result_names(self):
"""Ensure tool_result_names property works as expected."""
sample_data = {'name': 'SMPL_01', KRAKEN_NAME: create_kraken()}
sample = Sample(**sample_data).save()
self.assertEqual(len(sample.tool_result_names), 1)
self.assertIn(KRAKEN_NAME, sample.tool_result_names)
