def _apply_metadata_to_experiment(experiment: Experiment, metadata: dict):
experiment.source_url = ENA_URL_TEMPLATE.format(
experiment.accession_code)
experiment.source_database = "SRA"
experiment.technology = "RNA-SEQ"
# We don't get this value from the API, unfortunately.
# experiment.platform_accession_code = experiment["platform_accession_code"]
if not experiment.description:
experiment.description = "No description."
if "study_title" in metadata:
experiment.title = metadata["study_title"]
if "study_abstract" in metadata:
experiment.description = metadata["study_abstract"]
if "lab_name" in metadata:
experiment.submitter_institution = metadata["lab_name"]
if "experiment_design_description" in metadata:
experiment.protocol_description = metadata[
"experiment_design_description"]
if "pubmed_id" in metadata:
experiment.pubmed_id = metadata["pubmed_id"]
experiment.has_publication = True
if "study_ena_first_public" in metadata:
experiment.source_first_published = parse_date(
metadata["study_ena_first_public"])
if "study_ena_last_update" in metadata:
experiment.source_last_modified = parse_date(
metadata["study_ena_last_update"])
# We only want GEO alternate accessions for SRA samples
if re.match(r"^GSE\d{2,6}", metadata.get("external_id",
"")) is not None:
experiment.alternate_accession_code = metadata["external_id"]
# Rare, but it happens.
if not experiment.protocol_description:
# metadata.get() doesn't work here because sometimes the
# key is present but its value is None, in which case None
# is returned, causing our database constraint to be
# violated.
if ("library_construction_protocol" in metadata
and metadata["library_construction_protocol"]):
experiment.protocol_description = metadata[
"library_construction_protocol"]
else:
experiment.protocol_description = "Protocol was never provided."
# Scrape publication title and authorship from Pubmed
if experiment.pubmed_id:
pubmed_metadata = utils.get_title_and_authors_for_pubmed_id(
experiment.pubmed_id)
experiment.publication_title = pubmed_metadata[0]
experiment.publication_authors = pubmed_metadata[1]
def setUp(self):
experiment = Experiment()
experiment.accession_code = "GSE000"
experiment.alternate_accession_code = "E-GEOD-000"
experiment.title = "NONONONO"
experiment.description = "Boooooourns. Wasabi."
experiment.technology = "RNA-SEQ"
experiment.save()
self.experiment = experiment
# Create some samples to attach keywords to
sample = Sample()
sample.accession_code = "SRR123"
sample.technology = "RNA-SEQ"
sample.source_database = "SRA"
sample.title = "Not important"
sample.save()
experiment_sample_association = ExperimentSampleAssociation()
experiment_sample_association.sample = sample
experiment_sample_association.experiment = experiment
experiment_sample_association.save()
sample2 = Sample()
sample2.accession_code = "SRR456"
sample2.technology = "RNA-SEQ"
sample2.source_database = "SRA"
sample2.title = "Not important"
sample2.save()
experiment_sample_association = ExperimentSampleAssociation()
experiment_sample_association.sample = sample2
experiment_sample_association.experiment = experiment
experiment_sample_association.save()
# Create the ontology terms I'm using in the tests
name = OntologyTerm()
name.ontology_term = "PATO:0000122"
name.human_readable_name = "length"
name.save()
unit = OntologyTerm()
unit.ontology_term = "UO:0010012"
unit.human_readable_name = "thou"
unit.save()
contribution = Contribution()
contribution.source_name = "refinebio_tests"
contribution.methods_url = "ccdatalab.org"
contribution.save()
self.contribution = contribution
def setUp(self):
# Saving this for if we have protected endpoints
# self.superuser = User.objects.create_superuser('john', '*****@*****.**', 'johnpassword')
# self.client.login(username='******', password='******')
# self.user = User.objects.create(username="******")
experiment = Experiment()
experiment.accession_code = "GSE000"
experiment.alternate_accession_code = "E-GEOD-000"
experiment.title = "NONONONO"
experiment.description = "Boooooourns. Wasabi."
experiment.technology = "RNA-SEQ"
experiment.save()
experiment = Experiment()
experiment.accession_code = "GSE123"
experiment.title = "Hey Ho Let's Go"
experiment.description = (
"This is a very exciting test experiment. Faygo soda. Blah blah blah."
)
experiment.technology = "MICROARRAY"
experiment.save()
self.experiment = experiment
experiment_annotation = ExperimentAnnotation()
experiment_annotation.data = {"hello": "world", "123": 456}
experiment_annotation.experiment = experiment
experiment_annotation.save()
# Create 26 test organisms numbered 0-25 for pagination test, so there should be 29 organisms total (with the 3 others below)
for i in range(26):
Organism(name=("TEST_ORGANISM_{}".format(i)),
taxonomy_id=(1234 + i)).save()
ailuropoda = Organism(name="AILUROPODA_MELANOLEUCA",
taxonomy_id=9646,
is_scientific_name=True)
ailuropoda.save()
self.homo_sapiens = Organism(name="HOMO_SAPIENS",
taxonomy_id=9606,
is_scientific_name=True)
self.homo_sapiens.save()
self.danio_rerio = Organism(name="DANIO_RERIO",
taxonomy_id=1337,
is_scientific_name=True)
self.danio_rerio.save()
sample = Sample()
sample.title = "123"
sample.accession_code = "123"
sample.is_processed = True
sample.organism = ailuropoda
sample.save()
sample = Sample()
sample.title = "789"
sample.accession_code = "789"
sample.is_processed = True
sample.organism = ailuropoda
sample.save()
self.sample = sample
# add qn target for sample organism
result = ComputationalResult()
result.commands.append("create_qn_target.py")
result.is_ccdl = True
result.is_public = True
result.processor = None
result.save()
cra = ComputationalResultAnnotation()
cra.result = result
cra.data = {"organism_id": ailuropoda.id, "is_qn": True}
cra.save()
ailuropoda.qn_target = result
ailuropoda.save()
sample_annotation = SampleAnnotation()
sample_annotation.data = {"goodbye": "world", "789": 123}
sample_annotation.sample = sample
sample_annotation.save()
original_file = OriginalFile()
original_file.save()
original_file_sample_association = OriginalFileSampleAssociation()
original_file_sample_association.sample = sample
original_file_sample_association.original_file = original_file
original_file_sample_association.save()
downloader_job = DownloaderJob()
downloader_job.save()
download_assoc = DownloaderJobOriginalFileAssociation()
download_assoc.original_file = original_file
download_assoc.downloader_job = downloader_job
download_assoc.save()
processor_job = ProcessorJob()
processor_job.save()
processor_assoc = ProcessorJobOriginalFileAssociation()
processor_assoc.original_file = original_file
processor_assoc.processor_job = processor_job
processor_assoc.save()
experiment_sample_association = ExperimentSampleAssociation()
experiment_sample_association.sample = sample
experiment_sample_association.experiment = experiment
experiment_sample_association.save()
experiment.num_total_samples = 1
experiment.num_processed_samples = 1
experiment.save()
result = ComputationalResult()
result.save()
sra = SampleResultAssociation()
sra.sample = sample
sra.result = result
sra.save()
result = ComputationalResult()
result.save()
sra = SampleResultAssociation()
sra.sample = sample
sra.result = result
sra.save()
processor = Processor()
processor.name = "Salmon Quant"
processor.version = "v9.9.9"
processor.docker_image = "dr_salmon"
processor.environment = '{"some": "environment"}'
processor.save()
computational_result_short = ComputationalResult(processor=processor)
computational_result_short.save()
organism_index = OrganismIndex()
organism_index.index_type = "TRANSCRIPTOME_SHORT"
organism_index.organism = self.danio_rerio
organism_index.result = computational_result_short
organism_index.absolute_directory_path = (
"/home/user/data_store/salmon_tests/TRANSCRIPTOME_INDEX/SHORT")
organism_index.is_public = True
organism_index.s3_url = "not_blank"
organism_index.save()
return