def store_bio_searchio_hit(self, searchio_hit: Hit, target: str) -> None:
"""Store bio searchio hit."""
organism_obj, created = Organism.objects.get_or_create(
abbreviation="multispecies",
genus="multispecies",
species="multispecies",
common_name="multispecies",
)
if not hasattr(searchio_hit, "accession"):
searchio_hit.accession = None
# if interproscan-xml parsing, get db name from Hit.attributes.
if target == "InterPro":
db_name = searchio_hit.attributes["Target"].upper()
# prevents the creation of multiple databases for SIGNALP
if db_name.startswith("SIGNALP"):
db_name = "SIGNALP"
db, created = Db.objects.get_or_create(name=db_name)
# if blast-xml parsing, db name is self.db ("BLAST_source")
else:
db = self.db
dbxref, created = Dbxref.objects.get_or_create(
db=db, accession=searchio_hit.id)
feature, created = Feature.objects.get_or_create(
organism=organism_obj,
uniquename=searchio_hit.id,
type_id=self.so_term_protein_match.cvterm_id,
name=searchio_hit.accession,
dbxref=dbxref,
defaults={
"is_analysis": False,
"is_obsolete": False,
"timeaccessioned": datetime.now(timezone.utc),
"timelastmodified": datetime.now(timezone.utc),
},
)
for aux_dbxref in searchio_hit.dbxrefs:
aux_db, aux_term = aux_dbxref.split(":", 1)
if aux_db == "GO":
try:
term_db = Db.objects.get(name=aux_db.upper())
dbxref = Dbxref.objects.get(db=term_db, accession=aux_term)
cvterm = Cvterm.objects.get(dbxref=dbxref)
FeatureCvterm.objects.get_or_create(
feature=feature,
cvterm=cvterm,
pub=self.pub,
is_not=False,
rank=0,
)
except ObjectDoesNotExist:
self.ignored_goterms.add(aux_dbxref)
else:
term_db, created = Db.objects.get_or_create(
name=aux_db.upper())
dbxref, created = Dbxref.objects.get_or_create(
db=term_db, accession=aux_term)
FeatureDbxref.objects.get_or_create(feature=feature,
dbxref=dbxref,
is_current=1)
return None
def test_store_bio_searchio_hit(self):
"""Tests - store bio searchio hit."""
# create RO term: contained in
test_db = Db.objects.create(name="RO")
test_dbxref = Dbxref.objects.create(accession="00002", db=test_db)
test_cv = Cv.objects.create(name="relationship")
Cvterm.objects.create(
name="contained in",
cv=test_cv,
dbxref=test_dbxref,
is_obsolete=0,
is_relationshiptype=0,
)
# create SO terms: protein_match
test_cv = Cv.objects.create(name="sequence")
test_db = Db.objects.create(name="SO")
test_dbxref = Dbxref.objects.create(accession="00001", db=test_db)
Cvterm.objects.create(
name="protein_match",
cv=test_cv,
dbxref=test_dbxref,
is_obsolete=0,
is_relationshiptype=0,
)
test_dbxref = Dbxref.objects.create(accession="00002", db=test_db)
Cvterm.objects.create(
name="polypeptide",
cv=test_cv,
dbxref=test_dbxref,
is_obsolete=0,
is_relationshiptype=0,
)
# create GO term
test_db = Db.objects.create(name="GO")
test_dbxref = Dbxref.objects.create(accession="1234", db=test_db)
test_cv = Cv.objects.create(name="biological_process")
Cvterm.objects.create(
name="GO:1234",
cv=test_cv,
dbxref=test_dbxref,
is_obsolete=0,
is_relationshiptype=0,
)
# create a bio searchio hit
test_searchio_hit = Hit()
test_searchio_hit.id = "PF1234"
test_searchio_hit.accession = "PFAM mock domain"
test_searchio_hit.attributes["Target"] = "PFAM"
test_searchio_hit.dbxrefs = [
"GO:1234", "IPR:IPR012345", "Reactome:R-HSA-12345"
]
Organism.objects.create(genus="test", species="organism")
# instantiate the loader
test_feature_file = FeatureLoader(filename="file.name",
source="InterproScan_source")
# store the bio searchio hit
# From interproscan
target = "InterPro"
test_feature_file.store_bio_searchio_hit(test_searchio_hit, target)
test_feature = Feature.objects.get(uniquename="PF1234")
self.assertEqual("PFAM mock domain", test_feature.name)
test_dbxref = Dbxref.objects.get(accession="IPR012345")
test_feature_dbxref = FeatureDbxref.objects.get(feature=test_feature,
dbxref=test_dbxref)
self.assertEqual(True, test_feature_dbxref.is_current)
test_cvterm = Cvterm.objects.get(name="GO:1234")
test_feature_cvterm = FeatureCvterm.objects.get(feature=test_feature,
cvterm=test_cvterm)
self.assertEqual(0, test_feature_cvterm.rank)
