def _gaFeatureForFeatureDbRecord(self, feature):
"""
:param feature: The DB Row representing a feature
:return: the corresponding GA4GH protocol.Feature object
"""
gaFeature = protocol.Feature()
gaFeature.id = self.getCompoundIdForFeatureId(feature['id'])
if feature.get('parent_id'):
gaFeature.parent_id = self.getCompoundIdForFeatureId(
feature['parent_id'])
else:
gaFeature.parent_id = ""
gaFeature.feature_set_id = self.getId()
gaFeature.reference_name = pb.string(feature.get('reference_name'))
gaFeature.start = pb.int(feature.get('start'))
gaFeature.end = pb.int(feature.get('end'))
gaFeature.name = pb.string(feature.get('name'))
if feature.get('strand', '') == '-':
gaFeature.strand = protocol.NEG_STRAND
else:
# default to positive strand
gaFeature.strand = protocol.POS_STRAND
gaFeature.child_ids.extend(map(
self.getCompoundIdForFeatureId,
json.loads(feature['child_ids'])))
gaFeature.feature_type.CopyFrom(
self._ontology.getGaTermByName(feature['type']))
attributes = json.loads(feature['attributes'])
# TODO: Identify which values are ExternalIdentifiers and OntologyTerms
for key in attributes:
for v in attributes[key]:
gaFeature.attributes.vals[key].values.add().string_value = v
if 'gene_name' in attributes and len(attributes['gene_name']) > 0:
gaFeature.gene_symbol = pb.string(attributes['gene_name'][0])
return gaFeature
def _gaFeatureForFeatureDbRecord(self, feature):
"""
:param feature: The DB Row representing a feature
:return: the corresponding GA4GH protocol.Feature object
"""
gaFeature = protocol.Feature()
gaFeature.id = self.getCompoundIdForFeatureId(feature['id'])
if feature.get('parent_id'):
gaFeature.parentId = self.getCompoundIdForFeatureId(
feature['parent_id'])
else:
gaFeature.parentId = ""
gaFeature.featureSetId = self.getId()
gaFeature.referenceName = feature['reference_name']
gaFeature.start = int(feature['start'])
gaFeature.end = int(feature['end'])
if feature.get('strand', '') == '-':
gaFeature.strand = protocol.Strand.NEG_STRAND
else:
# default to positive strand
gaFeature.strand = protocol.Strand.POS_STRAND
gaFeature.childIds = map(
self.getCompoundIdForFeatureId,
json.loads(feature['child_ids']))
gaFeature.featureType = \
self._sequenceOntology.getGaTermByName(feature['type'])
gaFeature.attributes = protocol.Attributes()
gaFeature.attributes.vals = json.loads(feature['attributes'])
return gaFeature
def _generateSimulatedFeature(self, randomNumberGenerator):
feature = protocol.Feature()
feature.featureSetId = self.getId()
feature.start = randomNumberGenerator.randint(1000, 2000)
feature.end = feature.start + randomNumberGenerator.randint(1, 100)
feature.featureType = self._getRandomfeatureType(
randomNumberGenerator)
references = ["chr1", "chr2", "chrX"]
feature.referenceName = randomNumberGenerator.choice(references)
strands = [protocol.Strand.POS_STRAND, protocol.Strand.NEG_STRAND]
feature.strand = randomNumberGenerator.choice(strands)
feature.attributes = protocol.Attributes()
feature.attributes.vals = {
"gene_name": ["Frances", ],
"gene_type": ["mRNA", ],
"gene_status": ["UNKNOWN", ]}
return feature
def _getFeatureById(self, featureId):
"""
find a feature and return ga4gh representation, use 'native' id as
featureId
"""
featureRef = rdflib.URIRef(featureId)
featureDetails = self._detailTuples([featureRef])
feature = {}
for detail in featureDetails:
feature[detail['predicate']] = []
for detail in featureDetails:
feature[detail['predicate']].append(detail['object'])
pbFeature = protocol.Feature()
term = protocol.OntologyTerm()
# Schema for feature only supports one type of `type`
# here we default to first OBO defined
for featureType in sorted(feature[TYPE]):
if "obolibrary" in featureType:
term.term = self._featureTypeLabel(featureType)
term.id = featureType
pbFeature.feature_type.MergeFrom(term)
break
pbFeature.id = featureId
# Schema for feature only supports one type of `name` `symbol`
# here we default to shortest for symbol and longest for name
feature[LABEL].sort(key=len)
pbFeature.gene_symbol = feature[LABEL][0]
pbFeature.name = feature[LABEL][-1]
pbFeature.attributes.MergeFrom(protocol.Attributes())
for key in feature:
for val in sorted(feature[key]):
pbFeature.attributes.vals[key].values.add().string_value = val
if featureId in self._locationMap:
location = self._locationMap[featureId]
pbFeature.reference_name = location["chromosome"]
pbFeature.start = location["begin"]
pbFeature.end = location["end"]
return pbFeature
def _generateSimulatedFeature(self, randomNumberGenerator):
feature = protocol.Feature()
feature.feature_set_id = self.getId()
feature.start = randomNumberGenerator.randint(1000, 2000)
feature.end = feature.start + randomNumberGenerator.randint(1, 100)
feature.feature_type.CopyFrom(self._getRandomfeatureType(
randomNumberGenerator))
references = ["chr1", "chr2", "chrX"]
feature.reference_name = randomNumberGenerator.choice(references)
strands = [protocol.POS_STRAND, protocol.NEG_STRAND]
feature.strand = randomNumberGenerator.choice(strands)
attributes = {
"gene_name": "Frances",
"gene_type": "mRNA",
"gene_status": "UNKNOWN"}
for key, value in attributes.items():
feature.attributes.vals[key].values.add().string_value = value
return feature