def makeGene(self,root):
gene=Gene()
root["object"]=gene
gene.ID=root["extra"]["ID"]
children=root.get("children",None)
if(children is None): return gene
for child in children:
obj=self.labelStructure(child)
if(obj is None): continue
if(type(obj)==Transcript):
gene.addTranscript(obj)
obj.gene=gene
obj.geneId=gene.getId()
extra=root["extra"]
gene.extraFields=""
for key in extra:
gene.extraFields+=key+"="+extra[key]+";"
return gene
def makeGene(self, root):
gene = Gene()
root["object"] = gene
gene.ID = root["extra"]["ID"]
children = root.get("children", None)
if (children is None): return gene
for child in children:
obj = self.labelStructure(child)
if (obj is None): continue
if (type(obj) == Transcript):
gene.addTranscript(obj)
obj.gene = gene
obj.geneId = gene.getId()
extra = root["extra"]
gene.extraFields = ""
for key in extra:
gene.extraFields += key + "=" + extra[key] + ";"
return gene
def assembleTranscriptome(self):
"""
Loop over uniqueGeneSet, in which the ENSG-IDs are saved,
and assemble all the transcripts and exons for this gene and save it as a Gene object.
This gene obeject ist then added to the geneList of this Genome object
"""
# transcriptsByType = defaultdict(list)
# construct Genes
for uniqGene in self.uniqGeneSet:
geneId, chromosome, strand = uniqGene
geneNames = list(self.uniqGene_to_names[uniqGene])
geneType = self.uniqGene_to_source[uniqGene]
geneExons = set()
geneCds = set()
# get exons from all transcripts
for transcriptId in self.uniqGene_to_transcriptIds[uniqGene]:
geneExons |= self.transcriptId_to_exons[
transcriptId] # add new exon tuples to geneExons set
geneCds |= self.transcriptId_to_cds[transcriptId]
# usually gtf Files are sorted, but this can't be assumed
geneExons = sorted(geneExons, reverse=not strand)
geneCds = sorted(geneCds, reverse=not strand)
gene = Gene(geneId, chromosome, strand, geneType, geneNames,
geneExons, geneCds)
geneExons = dict(zip(geneExons, xrange(1000000)))
geneCds = dict(zip(geneCds, xrange(1000000)))
self.geneList.append(gene)
# construct transcripts
for transcriptId in self.uniqGene_to_transcriptIds[uniqGene]:
transcriptNames = self.transcriptId_to_names[transcriptId]
protId = self.transcriptId_to_protId[
transcriptId] if transcriptId in self.transcriptId_to_protId else None
exons = sorted(self.transcriptId_to_exons[transcriptId])
codingExons = sorted(self.transcriptId_to_cds[transcriptId])
exonIndices = array('H', [geneExons[e] for e in exons])
codingExonIndices = array('H',
[geneCds[e] for e in codingExons])
codingFrames = array('H', [
int(frame) for exonNumber, frame in sorted(
self.transcriptId_to_codingFrames[transcriptId])
])
startCodon = tuple([
interval for exonNumber, interval in sorted(
self.transcriptId_to_startCodons[transcriptId])
])
stopCodon = tuple([
interval for exonNumber, interval in sorted(
self.transcriptId_to_stopCodons[transcriptId])
])
if len(codingExons) != len(codingFrames):
raise Exception(
"Number of coding Exons and Frames differ for %s %s" %
geneId, transcriptId)
transcript = Transcript(gene, transcriptId,
list(transcriptNames), protId,
exonIndices, codingExonIndices,
codingFrames, startCodon, stopCodon)
gene.addTranscript(transcript)