def test_unmapped_service_accessions(self):
expected_accessions = {
'array_express': {'A1', 'A2'},
'eva': {'EVA1', 'EVA2'}
}
submission = ExcelSubmission()
sequence1 = {
'sequence_array_express_accession': 'A1',
'sequence_eva_accession': 'EVA1',
}
sequence2 = {
'sequence_array_express_accession': 'A2',
'sequence_eva_accession': 'EVA2',
}
sequence_entity1 = ExcelLoader.map_row_entity(submission, 1,
'sequence', sequence1)
sequence_entity2 = ExcelLoader.map_row_entity(submission, 2,
'sequence', sequence2)
self.assertDictEqual(expected_accessions,
submission.get_all_accessions())
self.assertEqual('A1', sequence_entity1.get_accession('array_express'))
self.assertEqual('EVA1', sequence_entity1.get_accession('eva'))
self.assertEqual('A2', sequence_entity2.get_accession('array_express'))
self.assertEqual('EVA2', sequence_entity2.get_accession('eva'))
def test_mapped_accessions(self):
expected_accessions = {
'BioStudies': {'S-BSST1'},
'BioSamples': {'SAME1'},
'ENA_Project': {'PRJEB1'},
'ENA_Study': {'ERP1'},
'ENA_Sample': {'ERS1'},
'ENA_Experiment': {'ERX1', 'ERX2'},
'ENA_Run': {'ERR1', 'ERR2'}
}
submission = ExcelSubmission()
study = {
'study_biostudies_accession': 'S-BSST1',
'study_ena_project_accession': 'PRJEB1',
'study_ena_study_accession': 'ERP1'
}
sample = {
'sample_biosamples_accession': 'SAME1',
'sample_ena_sample_accession': 'ERS1'
}
run_experiment1 = {
'run_experiment_ena_experiment_accession': 'ERX1',
'run_experiment_ena_run_accession': 'ERR1',
}
run_experiment2 = {
'run_experiment_ena_experiment_accession': 'ERX2',
'run_experiment_ena_run_accession': 'ERR2',
}
study_entity1 = ExcelLoader.map_row_entity(submission, 1, 'study',
study)
sample_entity1 = ExcelLoader.map_row_entity(submission, 1, 'sample',
sample)
run_entity1 = ExcelLoader.map_row_entity(submission, 1,
'run_experiment',
run_experiment1)
run_entity2 = ExcelLoader.map_row_entity(submission, 2,
'run_experiment',
run_experiment2)
self.assertDictEqual(expected_accessions,
submission.get_all_accessions())
self.assertEqual('S-BSST1', study_entity1.get_accession('BioStudies'))
self.assertEqual('PRJEB1', study_entity1.get_accession('ENA_Project'))
self.assertEqual('ERP1', study_entity1.get_accession('ENA_Study'))
self.assertEqual('SAME1', sample_entity1.get_accession('BioSamples'))
self.assertEqual('ERS1', sample_entity1.get_accession('ENA_Sample'))
self.assertEqual('ERR1', run_entity1.get_accession('ENA_Run'))
self.assertEqual('ERX1', run_entity1.get_accession('ENA_Experiment'))
self.assertEqual('ERR2', run_entity2.get_accession('ENA_Run'))
self.assertEqual('ERX2', run_entity2.get_accession('ENA_Experiment'))
def test_loading_differing_index_equality(self):
# Note: I'm not sure if this is a feature or a bug so if this test fails it's okay,
# but it's a situation worth noting about our current implementation
s_1 = {
's_alias': 'STUD',
's_index': 'over-written index1',
's_name': 'over-written name1'
}
s_2 = {
's_index': 'STUD',
's_name': 'over-written name2'
}
s_3 = {
's_name': 'STUD'
}
expected_s = {
's_alias': 'STUD',
's_index': 'STUD',
's_name': 'STUD'
}
submission = ExcelSubmission()
s_entity1 = ExcelLoader.map_row_entity(submission, 1, 's', s_1)
s_entity2 = ExcelLoader.map_row_entity(submission, 2, 's', s_2)
s_entity3 = ExcelLoader.map_row_entity(submission, 3, 's', s_3)
self.assertEqual(s_entity1, s_entity2)
self.assertEqual(s_entity1, s_entity3)
self.assertDictEqual(expected_s, s_entity1.attributes)
def test_loading_without_index(self):
submission = ExcelSubmission()
entity1 = ExcelLoader.map_row_entity(submission, 1, 'lorem', {})
entity2 = ExcelLoader.map_row_entity(submission, 2, 'ipsum', {})
self.assertNotEqual(entity1, entity2)
self.assertEqual('lorem:1', entity1.identifier.index)
self.assertEqual('ipsum:2', entity2.identifier.index)
def test_loading_accession_equality(self):
submission = ExcelSubmission()
study_entity1 = ExcelLoader.map_row_entity(submission, 1, 'study', {'study_accession': 'STUD1'})
sample_entity1 = ExcelLoader.map_row_entity(submission, 1, 'sample', {'sample_accession': 'SAME1'})
study_entity2 = ExcelLoader.map_row_entity(submission, 2, 'study', {'study_accession': 'STUD1'})
sample_entity2 = ExcelLoader.map_row_entity(submission, 2, 'sample', {'sample_accession': 'SAME2'})
self.assertEqual(study_entity1, study_entity2)
self.assertNotEqual(sample_entity1, sample_entity2)
def test_default_accessions(self):
expected_accessions = {
'BioStudies': {'S-BSST1'},
'BioSamples': {'SAME1'},
'ENA_Run': {'ERR1', 'ERR2'}
}
submission = ExcelSubmission()
study = {
'study_accession': 'S-BSST1',
}
sample = {
'sample_accession': 'SAME1',
}
run_experiment1 = {
'run_experiment_accession': 'ERR1',
}
run_experiment2 = {
'run_experiment_accession': 'ERR2',
}
study_entity1 = ExcelLoader.map_row_entity(submission, 1, 'study',
study)
sample_entity1 = ExcelLoader.map_row_entity(submission, 1, 'sample',
sample)
run_entity1 = ExcelLoader.map_row_entity(submission, 1,
'run_experiment',
run_experiment1)
run_entity2 = ExcelLoader.map_row_entity(submission, 2,
'run_experiment',
run_experiment2)
self.assertDictEqual(expected_accessions,
submission.get_all_accessions())
self.assertEqual('S-BSST1', study_entity1.identifier.index)
self.assertEqual('S-BSST1', study_entity1.get_accession('BioStudies'))
self.assertEqual('SAME1', sample_entity1.identifier.index)
self.assertEqual('SAME1', sample_entity1.get_accession('BioSamples'))
self.assertEqual('ERR1', run_entity1.identifier.index)
self.assertEqual('ERR1', run_entity1.get_accession('ENA_Run'))
self.assertEqual('ERR2', run_entity2.identifier.index)
self.assertEqual('ERR2', run_entity2.get_accession('ENA_Run'))
def test_umnapped_no_service_accession(self):
# In the case that {object}_accession is used in the excel for an object that isn't configured with a default service
# Use the accession as an index but do not add the accession to the entity.
attributes = {'lorem_accession': 'ipsum'}
submission = ExcelSubmission()
lorem_entity = ExcelLoader.map_row_entity(submission, 1, 'lorem',
attributes)
self.assertEqual('ipsum', lorem_entity.identifier.index)
self.assertFalse(lorem_entity.get_accessions())
def test_loading_index_equality(self):
submission = ExcelSubmission()
study_entity1 = ExcelLoader.map_row_entity(submission, 1, 'study', {'study_alias': 'STUD1'})
sample_entity1 = ExcelLoader.map_row_entity(submission, 1, 'sample', {'sample_index': 'SAME1'})
run_entity1 = ExcelLoader.map_row_entity(submission, 1, 'run', {'run_name': 'John'})
study_entity2 = ExcelLoader.map_row_entity(submission, 2, 'study', {'study_alias': 'STUD1'})
sample_entity2 = ExcelLoader.map_row_entity(submission, 2, 'sample', {'sample_index': 'SAME1'})
run_entity2 = ExcelLoader.map_row_entity(submission, 2, 'run', {'run_name': 'Jill'})
self.assertEqual(study_entity1, study_entity2)
self.assertEqual(sample_entity1, sample_entity2)
self.assertNotEqual(run_entity1, run_entity2)
def test_links(self):
submission = ExcelSubmission()
study_entity1 = ExcelLoader.map_row_entity(submission, 1, 'study', {'study_accession': 'STUD1'})
sample_entity1 = ExcelLoader.map_row_entity(submission, 1, 'sample', {'sample_accession': 'SAME1'})
run_entity1 = ExcelLoader.map_row_entity(submission, 1, 'run_experiment', {'run_experiment_accession': 'RUN1'})
ExcelLoader.map_row_entity(submission, 2, 'study', {'study_accession': 'STUD1'})
ExcelLoader.map_row_entity(submission, 2, 'sample', {'sample_accession': 'SAME1'})
run_entity2 = ExcelLoader.map_row_entity(submission, 2, 'run_experiment', {'run_experiment_accession': 'RUN2'})
self.assertSetEqual({'SAME1'}, study_entity1.get_linked_indexes('sample'))
self.assertSetEqual({'RUN1', 'RUN2'}, study_entity1.get_linked_indexes('run_experiment'))
self.assertSetEqual({'STUD1'}, sample_entity1.get_linked_indexes('study'))
self.assertSetEqual({'RUN1', 'RUN2'}, sample_entity1.get_linked_indexes('run_experiment'))
self.assertSetEqual({'STUD1'}, run_entity1.get_linked_indexes('study'))
self.assertSetEqual({'SAME1'}, run_entity1.get_linked_indexes('sample'))
self.assertSetEqual({'STUD1'}, run_entity2.get_linked_indexes('study'))
self.assertSetEqual({'SAME1'}, run_entity2.get_linked_indexes('sample'))