def test_combinatorial_metadata_parsing(self):
"""
Tests parsing of combinatorial assay metadata from an Experiment Description file, e.g.
as used in the Skyline workflow for proteomics.
"""
ed_file = factory.test_file_path(
"experiment_description/advanced.xlsx")
study = factory.StudyFactory()
# creating *AFTER* setup of testing database records
cache = ExperimentDescriptionContext()
importer = CombinatorialCreationImporter(study, self.testuser, cache)
parser = ExperimentDescFileParser(cache, importer)
parsed: List[CombinatorialDescriptionInput] = parser.parse_excel(
ed_file)
combo_input: CombinatorialDescriptionInput = parsed[0]
time_meta_type = models.MetadataType.system("Time")
self.assertDictEqual(
combo_input.protocol_to_combinatorial_meta_dict,
{
self.targeted_proteomics.pk: {
time_meta_type.pk: [8.0, 24.0]
},
self.metabolomics.pk: {
time_meta_type.pk: [4.0, 6.0]
},
},
)
def test_combinatorial_assay_creation(self):
"""
Tests combinatorial assay creation, e.g. from combinatorial time values in the Skyline
workflow.
"""
study = models.Study.objects.create(name="Test")
file = factory.test_file_path(
"experiment_description/combinatorial_assays.xlsx")
# creating *AFTER* setup of testing database records
cache = ExperimentDescriptionContext()
importer = CombinatorialCreationImporter(study, self.testuser, cache)
with open(file, "rb") as fh:
importer.do_import(
fh,
ExperimentDescriptionOptions(),
filename="Fake Excel file",
file_extension="xlsx",
)
# verify results
line = models.Line.objects.get(study_id=study.pk)
self.assertEqual(line.assay_set.count(), 2)
times = {
assay.metadata_get(cache.assay_time_mtype)
for assay in models.Assay.objects.filter(
line=line, protocol=self.targeted_proteomics)
}
self.assertEqual(times, {8, 24})
def test_advanced_experiment_description_xlsx(self):
strain, _ = models.Strain.objects.get_or_create(name="JW0111")
# creating *AFTER* setup of testing database records
cache = ExperimentDescriptionContext()
cache.strains_by_pk = {strain.pk: strain}
advanced_experiment_def_xlsx = factory.test_file_path(
"experiment_description/advanced.xlsx")
study = factory.StudyFactory()
importer = CombinatorialCreationImporter(study, self.testuser, cache)
parser = ExperimentDescFileParser(cache, importer)
parsed = parser.parse_excel(advanced_experiment_def_xlsx)
self.assertEqual(len(parsed), 1,
"Expected a single set of parsed input")
# TODO: following two calls should go away
self._map_input_strains(parsed, cache, {"JBx_002078": strain})
importer._query_related_object_context(parsed)
result = parsed[0].populate_study(study, importer.cache,
ExperimentDescriptionOptions())
self.assertFalse(importer.errors, "Import generated errors")
self.assertFalse(importer.warnings, "Import generated warnings")
for line in result.lines_created:
self.assertEqual(line.description, "Description blah blah")
self.assertEqual(line.assay_set.count(), 4)
def test_ed_file_parse_err_detection(self):
""""
Tests for Experiment Description file errors that can be caught during
parsing. Error detection includes catching column headers that don't
match any of:
1) Line attributes supported by the parser (e.g. line name)
2) Line metadata defined in the database
3) Protocol + assay metadata defined in the database
Also tests duplicate column header detection for each type of column
definition implied by 1-3 above.
"""
# parameters don't matter for test, but need to be there
importer = CombinatorialCreationImporter(None, None)
file_path = factory.test_file_path(
"experiment_description/parse_errors.xlsx")
# creating *AFTER* setup of testing database records
cache = ExperimentDescriptionContext()
parser = ExperimentDescFileParser(cache, importer)
parser.parse_excel(file_path)
# expect to find these errors
exp_errors = [
(
constants.BAD_FILE_CATEGORY,
constants.INVALID_COLUMN_HEADER,
'"T3mperature" (col B)',
),
(
constants.BAD_FILE_CATEGORY,
constants.DUPLICATE_LINE_METADATA,
'"Media" (col G)',
),
(
constants.BAD_FILE_CATEGORY,
constants.UNMATCHED_ASSAY_COL_HEADERS_KEY,
'"Tomperature" (col H)',
),
(
constants.BAD_FILE_CATEGORY,
constants.DUPLICATE_LINE_ATTR,
'"Line Name" (col I)',
),
(
constants.BAD_FILE_CATEGORY,
constants.DUPLICATE_LINE_ATTR,
'"Replicate Count" (col J)',
),
(
constants.BAD_FILE_CATEGORY,
constants.DUPLICATE_ASSAY_METADATA,
'"Targeted Proteomics Time" (col K)',
),
(
constants.BAD_FILE_CATEGORY,
constants.DUPLICATE_LINE_METADATA,
'"Strain(s)" (col L)',
),
(
constants.BAD_GENERIC_INPUT_CATEGORY,
constants.INVALID_REPLICATE_COUNT,
'"X" (D2)',
),
(
constants.INVALID_FILE_VALUE_CATEGORY,
constants.MISSING_REQUIRED_LINE_NAME,
"A3",
),
(
constants.INVALID_FILE_VALUE_CATEGORY,
constants.INCORRECT_TIME_FORMAT,
'"A" (F4)',
),
(
constants.INVALID_FILE_VALUE_CATEGORY,
constants.DUPLICATE_LINE_NAME_LITERAL,
'"181-aceF" (A2, A4)',
),
]
for category, heading, detail in exp_errors:
# make sure the category shows up
assert category in importer.errors
# make sure the heading shows up
category_group = importer.errors.get(category)
assert heading in category_group
# make sure the heading has the right details
assert detail in category_group.get(heading)._occurrence_details