def convertTranscriptEffectVEP(self, annStr, hgvsG):
"""
Takes the ANN string of a VEP generated VCF, splits it
and returns a populated GA4GH transcript effect object.
:param annStr: String
:param hgvsG: String
:return: effect protocol.TranscriptEffect
"""
effect = self._createGaTranscriptEffect()
(alt, effects, impact, symbol, geneName, featureType,
featureId, trBiotype, exon, intron, hgvsC, hgvsP,
cdnaPos, cdsPos, protPos, aminos, codons,
existingVar, distance, strand, symbolSource,
hgncId, hgvsOffset) = annStr.split('|')
effect.alternateBases = alt
effect.effects = self.convertSeqOntology(effects)
effect.featureId = featureId
effect.hgvsAnnotation = protocol.HGVSAnnotation()
effect.hgvsAnnotation.genomic = hgvsG
effect.hgvsAnnotation.transcript = hgvsC
effect.hgvsAnnotation.protein = hgvsP
self.addLocations(effect, protPos, cdnaPos)
effect.id = self.getTranscriptEffectId(effect)
effect.analysisResults = []
return effect
def _addTranscriptEffectLocations(self, effect, ann):
# TODO Make these valid HGVS values
effect.hgvsAnnotation = protocol.HGVSAnnotation()
effect.hgvsAnnotation.genomic = str(ann.start)
effect.hgvsAnnotation.transcript = str(ann.start)
effect.hgvsAnnotation.protein = str(ann.start)
effect.proteinLocation = self._createGaAlleleLocation()
effect.proteinLocation.start = ann.start
effect.CDSLocation = self._createGaAlleleLocation()
effect.CDSLocation.start = ann.start
effect.cDNALocation = self._createGaAlleleLocation()
effect.cDNALocation.start = ann.start
return effect
def _createCsqTranscriptEffect(
self, alt, term, protPos, cdnaPos, featureId):
effect = self._createGaTranscriptEffect()
effect.alternateBases = alt
effect.effects = self.convertSeqOntology(term)
effect.featureId = featureId
effect.hgvsAnnotation = protocol.HGVSAnnotation()
# These are not present in the data
effect.hgvsAnnotation.genomic = None
effect.hgvsAnnotation.transcript = None
effect.hgvsAnnotation.protein = None
self.addLocations(effect, protPos, cdnaPos)
effect.id = self.getTranscriptEffectId(effect)
effect.analysisResults = []
return effect
def testAddLocations(self):
effect = protocol.TranscriptEffect()
effect.hgvsAnnotation = protocol.HGVSAnnotation()
effect.hgvsAnnotation.protein = "NM_001005484.1:p.Ile144Asn"
effect.hgvsAnnotation.transcript = "NM_001005484.1:c.431T>A"
effect.proteinLocation = protocol.AlleleLocation()
effect.cDNALocation = protocol.AlleleLocation()
effect.CDSLocation = protocol.AlleleLocation()
effect.proteinLocation.alternateSequence = "Asn"
effect.proteinLocation.referenceSequence = "Ile"
effect.proteinLocation.start = 143
effect.CDSLocation.alternateSequence = "A"
effect.CDSLocation.referenceSequence = "T"
effect.CDSLocation.start = 430
effect.cDNALocation.start = 430
protPos = "144/305"
cdnaPos = "431/918"
testEffect = self._variantAnnotationSet.addLocations(
effect, protPos, cdnaPos)
self.assertEqual(testEffect, effect)
def convertTranscriptEffectSnpEff(self, annStr, hgvsG):
"""
Takes the ANN string of a SnpEff generated VCF, splits it
and returns a populated GA4GH transcript effect object.
:param annStr: String
:param hgvsG: String
:return: effect protocol.TranscriptEffect()
"""
effect = self._createGaTranscriptEffect()
# SnpEff and VEP don't agree on this :)
(alt, effects, impact, geneName, geneId, featureType,
featureId, trBiotype, rank, hgvsC, hgvsP, cdnaPos,
cdsPos, protPos, distance, errsWarns) = annStr.split('|')
effect.alternateBases = alt
effect.effects = self.convertSeqOntology(effects)
effect.featureId = featureId
effect.hgvsAnnotation = protocol.HGVSAnnotation()
effect.hgvsAnnotation.genomic = hgvsG
effect.hgvsAnnotation.transcript = hgvsC
effect.hgvsAnnotation.protein = hgvsP
self.addLocations(effect, protPos, cdnaPos)
effect.id = self.getTranscriptEffectId(effect)
effect.analysisResults = []
return effect