def main():
genes = fo.getRecords(sys.argv[2])
annotatee = fo.getRecords(sys.argv[1])
withinGeneType = sys.argv[3]
withoutGeneType = sys.argv[4]
fraction = float(sys.argv[5])
ao,_ = fo.getNearFeatures(annotatee, genes,fraction,False,False)
f = open(sys.argv[1])
out = open(sys.argv[1]+'.annotation.txt','w')
for r in f:
if r.startswith('#'):
out.write(r.strip() +"\ttype\tgene" + "\n")
else:
tokens = r.strip().split('\t')
tempAnn = ao[tokens[3]]
if len(tempAnn) > 0:
annStr = withinGeneType + "\t" + ','.join(list(tempAnn))
else:
annStr = withoutGeneType
tokens.append(annStr)
out.write('\t'.join(tokens))
out.write('\n')
out.close()
def main():
genes = fo.getRecords(sys.argv[1])
fraction = float(sys.argv[-1])
for prefix in sys.argv[2:-1]:
out = open(prefix+'.all.bed','w')
out.write("track name=\"%s\" visibility=2 itemRgb=\"On\"\n"%(prefix+'_all',))
#novel = fo.getRecords(prefix+".novel.bed")
#anns,_ = fo.getNearFeatures(novel,genes,fraction,False,False)
f = open(prefix+'.novel.bed')
for r in f:
if r.startswith('#'):
out.write(r.strip()+"\tthickStart\tthickEnd\titemRgb\n")
else:
tokens = r.strip().split('\t')
tempAnn = anns[tokens[3]]
tokens.append(tokens[1])
tokens.append(tokens[2])
cIndex = getColorId(int(tokens[4]))
if len(tempAnn) > 0:
tokens.append(INTRONIC_COLORS[cIndex])
else:
tokens.append(UNKNOWN_COLORS[cIndex])
out.write('\t'.join(tokens))
out.write('\n')
f.close()
#exonic = fo.getRecords(prefix+".exonic.bed")
#anns,_ = fo.getNearFeatures(exonic,genes,fraction,False,False)
f = open(prefix+".exonic.bed")
for r in f:
if r.startswith('#'):
continue
else:
tokens = r.strip().split('\t')
tempAnn = anns[tokens[3]]
tokens.append(tokens[1])
tokens.append(tokens[2])
cIndex = getColorId(int(tokens[4]))
tokens.append(EXONIC_COLORS[cIndex])
tokens[3] = 'e'+tokens[3]
out.write('\t'.join(tokens))
out.write('\n')
f.close()
out.close()
for r in records:
if not re.match("^chr(\d+|X|Y|M)$",r['chrom']):
continue
count = 0
for read in sam.fetch(r['chrom'],int(r['chromStart']),int(r['chromEnd'])-1):
count += 1
maxCov = 0
for c in sam.pileup(r['chrom'],int(r['chromStart']), int(r['chromEnd']) -1):
if c.n > maxCov:
maxCov = c.n
length = int(r['chromEnd']) - int(r['chromStart'])
rpkm = gc.calRPKM(count,length,numMapped)
outBed.write('\t'.join([r['chrom'],r['chromStart'],r['chromEnd'],str(index),str(rpkm),r['strand']]))
outBed.write('\n')
outCov.write('\t'.join([str(index),str(count),str(maxCov),str(rpkm)]))
outCov.write('\n')
index += 1
if __name__=="__main__":
nameMap = {"exonic":"cds", "intronic":"noncds"}
mapped = c2r.getMappedReads(sys.argv[1])
for prefix in sys.argv[2:]:
for k in nameMap:
records = fo.getRecords("../genome/hg19/hg19."+nameMap[k]+".bed")
removeDuplicate(records)
sam = pysam.Samfile(prefix+".npcrd.unique."+k+".bam","rb")
outBed = open(prefix+"."+k+".nonCluster.bed",'w')
outCov = open(prefix + "."+k+".nonCluster.cov.txt",'w')
calculateCov(records, sam, outBed, outCov, long(mapped[prefix.split('/')[-1].lower()]))
