def loadExonH(): exnH = {} refFlatH = mygenome.loadRefFlatByChr() for chrom in refFlatH.keys(): if chrom not in exnH: exnH[chrom] = [] for tH in refFlatH[chrom]: for i in range(len(tH['exnList'])): exnH[chrom].append(tH['exnList'][i]) exnH[chrom] = list(set(exnH[chrom])) exnH[chrom].sort(lambda x,y: cmp(x[1],y[1])) exnH[chrom].sort(lambda x,y: cmp(x[0],y[0])) random.seed() tmpN = 'tmp_%d' % random.randint(1, 1000) tmpF = open(tmpN,'w') for chrom in exnH: if len(chrom) > 6: continue for i in range(len(exnH[chrom])): tmpF.write('%s\t%s\t%s\n' % (chrom, int(exnH[chrom][i][0])-1, int(exnH[chrom][i][1]))) tmpF.flush() tmpF.close() totalLen = os.popen('%s/NGS/quality/non-overlap.sh %s' % (mysetting.SRC_HOME, tmpN)).readlines()[0].rstrip() return exnH, int(totalLen)
def loadExonH(): exonH = {} refFlatH = mygenome.loadRefFlatByChr() for chrom in refFlatH.keys(): if chrom not in exonH: exonH[chrom] = [] for tH in refFlatH[chrom]: for i in range(len(tH['exnList'])): exonH[chrom].append(tH['exnList'][i]) kgH = mygenome.loadKgByChr() for chrom in kgH.keys(): if chrom not in exonH: exonH[chrom] = [] for tH in kgH[chrom]: for i in range(len(tH['exnList'])): exonH[chrom].append(tH['exnList'][i]) exonH[chrom] = list(set(exonH[chrom])) exonH[chrom].sort(lambda x, y: cmp(x[1], y[1])) exonH[chrom].sort(lambda x, y: cmp(x[0], y[0])) return exonH
def loadExonH(): exonH = {} refFlatH = mygenome.loadRefFlatByChr() for chrom in refFlatH.keys(): if chrom not in exonH: exonH[chrom] = [] for tH in refFlatH[chrom]: for i in range(len(tH["exnList"])): exonH[chrom].append(tH["exnList"][i]) kgH = mygenome.loadKgByChr() for chrom in kgH.keys(): if chrom not in exonH: exonH[chrom] = [] for tH in kgH[chrom]: for i in range(len(tH["exnList"])): exonH[chrom].append(tH["exnList"][i]) exonH[chrom] = list(set(exonH[chrom])) exonH[chrom].sort(lambda x, y: cmp(x[1], y[1])) exonH[chrom].sort(lambda x, y: cmp(x[0], y[0])) return exonH
def loadAnnot(geneL=[]): refFlatH = mygenome.loadRefFlatByChr() eiH = {} ei_keyH = {} juncInfoH = {} for chrom in refFlatH.keys(): eiH[chrom] = {} juncInfoH[chrom] = {} refFlatL = refFlatH[chrom] for tH in refFlatL: if geneL!=[] and tH['geneName'] not in geneL: continue for i in range(len(tH['exnList'])): if tH['strand'] == '+': pos = tH['exnList'][i][1] e_num = i+1 else: pos = tH['exnList'][i][0] e_num = len(tH['exnList'])-i mybasic.addHash(juncInfoH[chrom], pos, '%s%s:%s:%s/%s' % (tH['strand'], tH['geneName'], tH['refSeqId'], e_num, len(tH['exnList']))) eiH[chrom][pos] = 0 ei_keyH[chrom] = eiH[chrom].keys() ei_keyH[chrom].sort() ei_cntH = {} for chrom in juncInfoH.keys(): ei_cntH[chrom] = {} i = 0 for pos in sorted(juncInfoH[chrom].keys()): i += 1 ei_cntH[chrom][pos] = i return eiH,ei_keyH,juncInfoH,ei_cntH
def loadAnnot(geneL=[]): refFlatH = mygenome.loadRefFlatByChr() eiH = {} ei_keyH = {} juncInfoH = {} for chrom in refFlatH.keys(): eiH[chrom] = {} juncInfoH[chrom] = {} refFlatL = refFlatH[chrom] for tH in refFlatL: if geneL!=[] and tH['geneName'] not in geneL: continue for i in range(len(tH['exnList'])): if tH['strand'] == '+': pos = tH['exnList'][i][1] e_num = i+1 else: pos = tH['exnList'][i][0] e_num = len(tH['exnList'])-i mybasic.addHash(juncInfoH[chrom], pos, '%s%s:%s:%s/%s' % (tH['strand'], tH['geneName'], tH['refSeqId'], e_num, len(tH['exnList']))) eiH[chrom][pos] = 0 cursor.execute('replace into temp_table (chrom,pos) values ("%s",%s)' % (chrom,pos)) ei_keyH[chrom] = eiH[chrom].keys() ei_keyH[chrom].sort() return eiH,ei_keyH,juncInfoH
def loadExonH(): exnH = {} refFlatH = mygenome.loadRefFlatByChr() for chrom in refFlatH.keys(): if chrom not in exnH: exnH[chrom] = [] for tH in refFlatH[chrom]: for i in range(len(tH['exnList'])): exnH[chrom].append(tH['exnList'][i]) exnH[chrom] = list(set(exnH[chrom])) exnH[chrom].sort(lambda x, y: cmp(x[1], y[1])) exnH[chrom].sort(lambda x, y: cmp(x[0], y[0])) random.seed() tmpN = 'tmp_%d' % random.randint(1, 1000) tmpF = open(tmpN, 'w') for chrom in exnH: if len(chrom) > 6: continue for i in range(len(exnH[chrom])): tmpF.write( '%s\t%s\t%s\n' % (chrom, int(exnH[chrom][i][0]) - 1, int(exnH[chrom][i][1]))) tmpF.flush() tmpF.close() totalLen = os.popen('%s/NGS/quality/non-overlap.sh %s' % (mysetting.SRC_HOME, tmpN)).readlines()[0].rstrip() return exnH, int(totalLen)
data = {} filePathPrefix = bedgraphFileN.split('.bedgraph')[0] split = filePathPrefix.split('/') sampN = split[len(split) - 1] #if 'D-' in sampN: # DorW='D' #elif 'SOLiD' in sampN: # DorW='W-SOLiD' #else: # DorW='W' bedgraph = open(bedgraphFileN, 'r') refFlat = mygenome.loadRefFlatByChr(refFlatFileN) for line in bedgraph: l = line.split('\t') chr_sample = l[0] if refFlat.has_key(chr_sample): s = int(l[1]) e = int(l[2]) d = int(l[3]) for gene in refFlat[chr_sample]: GeneN = gene['geneName'] if not data.has_key(GeneN): data.update({GeneN: {}}) SeqId = gene['refSeqId'] if not data[GeneN].has_key(SeqId): data[GeneN].update(
data={} filePathPrefix = bedgraphFileN.split('.bedgraph')[0] split=filePathPrefix.split('/') sampN = split[len(split)-1] #if 'D-' in sampN: # DorW='D' #elif 'SOLiD' in sampN: # DorW='W-SOLiD' #else: # DorW='W' bedgraph=open(bedgraphFileN,'r') refFlat=mygenome.loadRefFlatByChr(refFlatFileN) for line in bedgraph: l=line.split('\t') chr_sample=l[0] if refFlat.has_key(chr_sample): s=int(l[1]) e=int(l[2]) d=int(l[3]) for gene in refFlat[chr_sample]: GeneN=gene['geneName'] if not data.has_key(GeneN): data.update({GeneN:{}}) SeqId=gene['refSeqId'] if not data[GeneN].has_key(SeqId):
def exonSkip_proc(inGsnapFileName, outGsnapFileName, outReportFileName, sampN): geneNameH = mygenome.geneNameH() geneSetH = mygenome.geneSetH() geneInfoH = mygenome.geneInfoH(geneNameH, geneSetH) refFlatH = mygenome.loadRefFlatByChr() result = mygsnap.gsnapFile(inGsnapFileName, False) juncHH = {} for r in result: match = r.matchL()[0] if not '(transloc)' in r.pairRel: raise Exception if len(match.segL) != 2: raise Exception splice_type = re.search('splice_type:([^,\t]*)', match.segL[0][3]).group(1) direction = re.search('dir:([^,\t]*)', match.segL[0][3]).group(1) offset = int(re.search('\.\.([0-9]*)', match.segL[0][1]).group(1)) transcript1 = re.search('label_[12]:([^,\t]*)', match.segL[0][3]) gene1 = set() if transcript1: transcript1 = tuple( [x.split('.exon')[0] for x in transcript1.group(1).split('|')]) for t in transcript1: g = mygenome.gene(t, geneNameH, geneSetH, geneInfoH) if g.geneName: gene1.add(g.geneName) else: transcript1 = () transcript2 = re.search('label_[12]:([^,\t]*)', match.segL[1][3]) gene2 = set() if transcript2: transcript2 = tuple( [x.split('.exon')[0] for x in transcript2.group(1).split('|')]) for t in transcript2: g = mygenome.gene(t, geneNameH, geneSetH, geneInfoH) if g.geneName: gene2.add(g.geneName) else: transcript2 = () s1 = match.segL[0][2] s2 = match.segL[1][2] bp1 = re.match('([+-])([^:]+):[0-9]+..([0-9]+)', s1) bp2 = re.match('([+-])([^:]+):([0-9]+)..[0-9]+', s2) if (bp1.group(1), direction) in (('+', 'sense'), ('-', 'antisense')): trans_strand1 = '+' elif (bp1.group(1), direction) in (('+', 'antisense'), ('-', 'sense')): trans_strand1 = '-' else: raise Exception if (bp2.group(1), direction) in (('+', 'sense'), ('-', 'antisense')): trans_strand2 = '+' elif (bp2.group(1), direction) in (('+', 'antisense'), ('-', 'sense')): trans_strand2 = '-' else: raise Exception bp_gene1 = mygenome.locus( '%s:%s-%s%s' % (bp1.group(2), int(bp1.group(3)) - 1, bp1.group(3), trans_strand1)).overlappingGeneL( refFlatH=refFlatH, strand_sensitive=True) bp_gene2 = mygenome.locus( '%s:%s-%s%s' % (bp2.group(2), int(bp2.group(3)) - 1, bp2.group(3), trans_strand2)).overlappingGeneL( refFlatH=refFlatH, strand_sensitive=True) if direction == 'sense': key = (bp1.groups()[1:], bp2.groups()[1:]) transcript = (transcript1, transcript2) gene = (tuple(gene1), tuple(gene2)) bp_gene = (bp_gene1, bp_gene2) elif direction == 'antisense': key = (bp2.groups()[1:], bp1.groups()[1:]) transcript = (transcript2, transcript1) gene = (tuple(gene2), tuple(gene1)) bp_gene = (bp_gene2, bp_gene1) else: raise Exception if key in juncHH: juncHH[key]['match'].append(r) juncHH[key]['seq'].append(r.seq()) juncHH[key]['reg'].append((direction, offset)) else: juncHH[key] = { 'match': [r], 'splice_type': splice_type, 'seq': [r.seq()], 'reg': [(direction, offset)], 'transcript': transcript, 'gene': gene, 'bp_gene': bp_gene } juncKH = juncHH.items() juncKH.sort(lambda x, y: cmp(len(set(y[1]['reg'])), len(set(x[1]['reg'])))) outGsnapFile = open(outGsnapFileName, 'w') outReportFile = open(outReportFileName, 'w') for (key, juncH) in juncKH: if key[0][0] == key[1][0]: type = 'intra' else: type = 'inter' geneInfo1 = [] censusInfo1 = [] for geneName in juncH['gene'][0]: gene = mygenome.gene(geneName, geneNameH, geneSetH, geneInfoH) geneInfo1.append( '%s:%s:%s' % (geneName, gene.getAttr('desc'), gene.getAttr('summary'))) censusInfo1.append('%s:%s:%s:%s' % (gene.getAttr('census_somatic'), gene.getAttr('census_germline'), gene.getAttr('census_mutType'), gene.getAttr('census_translocPartners'))) geneInfo2 = [] censusInfo2 = [] for geneName in juncH['gene'][1]: gene = mygenome.gene(geneName, geneNameH, geneSetH, geneInfoH) geneInfo2.append( '%s:%s:%s' % (geneName, gene.getAttr('desc'), gene.getAttr('summary'))) censusInfo2.append('%s:%s:%s:%s' % (gene.getAttr('census_somatic'), gene.getAttr('census_germline'), gene.getAttr('census_mutType'), gene.getAttr('census_translocPartners'))) outReportFile.write('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n' % \ (type, juncH['splice_type'], sampN, ':'.join(key[0]), ':'.join(key[1]), \ ';'.join(juncH['transcript'][0]), ';'.join(juncH['transcript'][1]), ';'.join(juncH['gene'][0]), ';'.join(juncH['gene'][1]), ';'.join(geneInfo1), ';'.join(geneInfo2), \ ';'.join(censusInfo1), ';'.join(censusInfo2), ','.join(juncH['bp_gene'][0]), ','.join(juncH['bp_gene'][1]), \ len(juncH['match']) ,len(set(juncH['seq'])), len(set(juncH['reg'])))) for m in juncH['match']: outGsnapFile.write(m.rawText() + '\n')
def gsnap_process_junction(inGsnapFileName,outGsnapFileName,outReportFileName,sampN): geneNameH = mygenome.geneNameH() geneSetH = mygenome.geneSetH() geneInfoH = mygenome.geneInfoH(geneNameH,geneSetH) refFlatH = mygenome.loadRefFlatByChr() result = mygsnap.gsnapFile(inGsnapFileName,False) juncHH = {} for r in result: match = r.matchL()[0] if not '(transloc)' in r.pairRel: raise Exception if len(match.segL) != 2: raise Exception splice_type = re.search('splice_type:([^,\t]*)', match.segL[0][3]).group(1) direction = re.search('dir:([^,\t]*)', match.segL[0][3]).group(1) offset = int(re.search('\.\.([0-9]*)', match.segL[0][1]).group(1)) rm = re.search('label_[12]:([^,\t]*)', match.segL[0][3]) gene1 = set() if rm: trans_exon1 = rm.group(1).split('|') for t in trans_exon1: g = mygenome.gene(t.split('.exon')[0],geneNameH,geneSetH,geneInfoH) if g.geneName: gene1.add(g.geneName) else: trans_exon1 = () rm = re.search('label_[12]:([^,\t]*)', match.segL[0][3]) gene2 = set() if rm: trans_exon2 = rm.group(1).split('|') for t in trans_exon2: g = mygenome.gene(t.split('.exon')[0],geneNameH,geneSetH,geneInfoH) if g.geneName: gene2.add(g.geneName) else: trans_exon2 = () s1 = match.segL[0][2] s2 = match.segL[1][2] bp1 = re.match('([+-])([^:]+):[0-9]+..([0-9]+)',s1) bp2 = re.match('([+-])([^:]+):([0-9]+)..[0-9]+',s2) if (bp1.group(1),direction) in (('+','sense'),('-','antisense')): trans_strand1 = '+' elif (bp1.group(1),direction) in (('+','antisense'),('-','sense')): trans_strand1 = '-' else: raise Exception if (bp2.group(1),direction) in (('+','sense'),('-','antisense')): trans_strand2 = '+' elif (bp2.group(1),direction) in (('+','antisense'),('-','sense')): trans_strand2 = '-' else: raise Exception locus1 = mygenome.locus('%s:%s-%s%s' % (bp1.group(2),int(bp1.group(3))-1,bp1.group(3),trans_strand1)) bp_gene1 = list(set(locus1.overlappingGeneL(refFlatH=refFlatH,strand_sensitive=True)).difference(gene1)) locus2 = mygenome.locus('%s:%s-%s%s' % (bp2.group(2),int(bp2.group(3))-2,bp2.group(3),trans_strand2)) bp_gene2 = list(set(locus2.overlappingGeneL(refFlatH=refFlatH,strand_sensitive=True)).difference(gene2)) if direction=='sense': key = (bp1.groups()[1:],bp2.groups()[1:]) trans_exon = (trans_exon1,trans_exon2) gene = (list(gene1),list(gene2)) bp_gene = (bp_gene1,bp_gene2) elif direction=='antisense': key = (bp2.groups()[1:],bp1.groups()[1:]) trans_exon = (trans_exon2,trans_exon1) gene = (list(gene2),list(gene1)) bp_gene = (bp_gene2,bp_gene1) else: raise Exception if key in juncHH: juncHH[key]['match'].append(r) juncHH[key]['seq'].append(r.seq()) juncHH[key]['reg'].append((direction,offset)) else: juncHH[key] = {'match':[r], 'splice_type':splice_type, 'seq':[r.seq()], 'reg':[(direction,offset)], 'trans_exon':trans_exon, 'gene':gene, 'bp_gene':bp_gene} juncKH = juncHH.items() juncKH.sort(lambda x,y: cmp(len(set(y[1]['reg'])),len(set(x[1]['reg'])))) outGsnapFile = open(outGsnapFileName,'w') outReportFile = open(outReportFileName,'w') for (key, juncH) in juncKH: if key[0][0] == key[1][0]: type = 'intra' else: type = 'inter' geneInfo1 = [] censusInfo1 = [] for geneName in juncH['gene'][0]+juncH['bp_gene'][0]: gene = mygenome.gene(geneName,geneNameH,geneSetH,geneInfoH) geneInfo1.append('%s:%s:%s' % (geneName,gene.getAttr('desc'),gene.getAttr('summary'))) censusInfo1.append('%s:%s:%s:%s' % (gene.getAttr('census_somatic'),gene.getAttr('census_germline'),gene.getAttr('census_mutType'),gene.getAttr('census_translocPartners'))) geneInfo2 = [] censusInfo2 = [] for geneName in juncH['gene'][1]+juncH['bp_gene'][1]: gene = mygenome.gene(geneName,geneNameH,geneSetH,geneInfoH) geneInfo2.append('%s:%s:%s' % (geneName,gene.getAttr('desc'),gene.getAttr('summary'))) censusInfo2.append('%s:%s:%s:%s' % (gene.getAttr('census_somatic'),gene.getAttr('census_germline'),gene.getAttr('census_mutType'),gene.getAttr('census_translocPartners'))) outReportFile.write('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s;%s\t%s;%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n' % \ (type, juncH['splice_type'], sampN, ':'.join(key[0]), ':'.join(key[1]), \ ','.join(juncH['trans_exon'][0]), ','.join(juncH['trans_exon'][1]), \ ','.join(juncH['gene'][0]), ','.join(juncH['bp_gene'][0]), ','.join(juncH['gene'][1]), ','.join(juncH['bp_gene'][1]), \ ';'.join(geneInfo1), ';'.join(geneInfo2), ';'.join(censusInfo1), ';'.join(censusInfo2), \ len(juncH['match']) ,len(set(juncH['seq'])), len(set(juncH['reg'])))) for m in juncH['match']: outGsnapFile.write(m.rawText()+'\n')
inFileName = sys.argv[1] outFileName = sys.argv[2] else: # this is default file, analysis specific files should be indicated as commandline arguments inFileName = 'GH_ft.txt' outFileName = 'GH_ft_process.txt' refFlatFileName = '/home/gye_hyeon/RNASeq/code_jk/refFlat_hg19.txt' resolveGene = False result = mygsnap.gsnapFile(inFileName) outFile = open(outFileName, 'w') if resolveGene: refFlatH = mygenome.loadRefFlatByChr(refFlatFileName) for rL in result: for i in (0,1): outFile.write(','.join([x.toString() for x in rL[i].matchL()[0].mergedLocusL()]) + '\t') if resolveGene: overlappingGeneL = [] for loc in rL[i].locusMergedL: overlappingGeneL += loc.overlappingGeneL(refFlatH=refFlatH) outFile.write(','.join(overlappingGeneL))
def fusion_proc_annot(inReportFileName,outReportFileName,inCnaGctFileName=None): geneDB = mygenome.getGeneDB() frameInfoH = mygenome.getFrameInfoH() refFlatH = mygenome.loadRefFlatByChr() if inCnaGctFileName: cnaDB = mygenome.tcgaCnaDB(inCnaGctFileName) else: cnaDB = None outReportFile = open(outReportFileName,'w') for line in open(inReportFileName): (splice_type,sampN,bp1,bp2,teStr1,teStr2,nmatch,nseq,nreg) = line[:-1].split('\t') if inCnaGctFileName: indivId = re.match('.*(TCGA-[0-9]{2}-[0-9]{4}).*',sampN).group(1) geneStatL = [] for (bp,teStr) in ((bp1,teStr1),(bp2,teStr2)): geneS = set() teL = [] for te in teStr.split(','): rm = re.match('(.*)\.exon([0-9]*)/[0-9]*',te) if rm: t = rm.group(1) e = int(rm.group(2)) g = mygenome.gene(t,geneDB=geneDB) if g.geneName: geneS.add(g.geneName) teL.append((t,e)) rm = re.match('([+-])(chr[^:]*):([0-9]*)',bp) bp_geneS = set(mygenome.locus('%s:%s-%s%s' % (rm.group(2),int(rm.group(3))-1,rm.group(3),rm.group(1))).overlappingGeneL(refFlatH=refFlatH,strand_sensitive=True)) bp_geneS = bp_geneS.difference(geneS) cnaInfo = [] geneInfo = [] censusInfo = [] goInfoS = set() keggInfoS = set() biocartaInfoS = set() for geneName in list(geneS) + list(bp_geneS): gene = mygenome.gene(geneName,geneDB=geneDB) if cnaDB: cnaInfo.append('%s:%s' % (geneName,cnaDB.query(indivId,geneName))) geneInfo.append('%s:%s:%s' % (geneName,gene.getAttr('desc'),gene.getAttr('summary'))) censusInfo.append('%s:%s:%s:%s' % (gene.getAttr('census_somatic'),gene.getAttr('census_germline'),gene.getAttr('census_mutType'),gene.getAttr('census_translocPartners'))) goInfoS = goInfoS.union(set(gene.getAttr('go'))) keggInfoS = keggInfoS.union(set(gene.getAttr('kegg'))) biocartaInfoS = biocartaInfoS.union(set(gene.getAttr('biocarta'))) geneStatL.append((bp1.split(':')[0],bp,teStr,teL,geneS,bp_geneS,cnaInfo,geneInfo,censusInfo,goInfoS,keggInfoS,biocartaInfoS)) (chrom1,bp1,teStr1,teL1,geneS1,bp_geneS1,cnaInfo1,geneInfo1,censusInfo1,goInfoS1,keggInfoS1,biocartaInfoS1) = geneStatL[0] (chrom2,bp2,teStr2,teL2,geneS2,bp_geneS2,cnaInfo2,geneInfo2,censusInfo2,goInfoS2,keggInfoS2,biocartaInfoS2) = geneStatL[1] if chrom1 == chrom2: type = 'intra' else: type = 'inter' frameL = [] for (t1,e1) in teL1: for (t2,e2) in teL2: cons = mygenome.frameCons(t1,e1, t2,e2, frameInfoH) if cons=='Y': frameL.append('%s.%s-%s.%s:%s' % (t1,e1,t2,e2,cons)) outReportFile.write('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n' % \ (sampN, splice_type, type, bp1,bp2, teStr1,teStr2, ','.join(frameL), ','.join(cnaInfo1), ','.join(cnaInfo2), \ '%s;%s' % (','.join(geneS1),','.join(bp_geneS1)), ';'.join(geneInfo1), ';'.join(censusInfo1), \ ';'.join(map(str,goInfoS1)), ';'.join(map(str,keggInfoS1)), ';'.join(map(str,biocartaInfoS1)), '%s;%s' % (','.join(geneS2),','.join(bp_geneS2)), ';'.join(geneInfo2), ';'.join(censusInfo2), \ ';'.join(map(str,goInfoS2)), ';'.join(map(str,keggInfoS2)), ';'.join(map(str,biocartaInfoS2)), nmatch,nseq,nreg))
if len(sys.argv) >= 3: inFileName = sys.argv[1] outFileName = sys.argv[2] else: # this is default file, analysis specific files should be indicated as commandline arguments inFileName = 'GH_ft.txt' outFileName = 'GH_ft_process.txt' refFlatFileName = '/home/gye_hyeon/RNASeq/code_jk/refFlat_hg19.txt' resolveGene = False result = mygsnap.gsnapFile(inFileName) outFile = open(outFileName, 'w') if resolveGene: refFlatH = mygenome.loadRefFlatByChr(refFlatFileName) for rL in result: for i in (0, 1): outFile.write( ','.join([x.toString() for x in rL[i].matchL()[0].mergedLocusL()]) + '\t') if resolveGene: overlappingGeneL = [] for loc in rL[i].locusMergedL: overlappingGeneL += loc.overlappingGeneL(refFlatH=refFlatH)
def fusion_proc_annot(inReportFileName, outReportFileName, inCnaGctFileName=None): geneDB = mygenome.getGeneDB() frameInfoH = mygenome.getFrameInfoH() refFlatH = mygenome.loadRefFlatByChr() if inCnaGctFileName: cnaDB = mygenome.tcgaCnaDB(inCnaGctFileName) else: cnaDB = None outReportFile = open(outReportFileName, 'w') for line in open(inReportFileName): (splice_type, sampN, bp1, bp2, teStr1, teStr2, nmatch, nseq, nreg) = line[:-1].split('\t') if inCnaGctFileName: indivId = re.match('.*(TCGA-[0-9]{2}-[0-9]{4}).*', sampN).group(1) geneStatL = [] for (bp, teStr) in ((bp1, teStr1), (bp2, teStr2)): geneS = set() teL = [] for te in teStr.split(','): rm = re.match('(.*)\.exon([0-9]*)/[0-9]*', te) if rm: t = rm.group(1) e = int(rm.group(2)) g = mygenome.gene(t, geneDB=geneDB) if g.geneName: geneS.add(g.geneName) teL.append((t, e)) rm = re.match('([+-])(chr[^:]*):([0-9]*)', bp) bp_geneS = set( mygenome.locus('%s:%s-%s%s' % (rm.group(2), int(rm.group(3)) - 1, rm.group(3), rm.group(1))).overlappingGeneL( refFlatH=refFlatH, strand_sensitive=True)) bp_geneS = bp_geneS.difference(geneS) cnaInfo = [] geneInfo = [] censusInfo = [] goInfoS = set() keggInfoS = set() biocartaInfoS = set() for geneName in list(geneS) + list(bp_geneS): gene = mygenome.gene(geneName, geneDB=geneDB) if cnaDB: cnaInfo.append('%s:%s' % (geneName, cnaDB.query(indivId, geneName))) geneInfo.append( '%s:%s:%s' % (geneName, gene.getAttr('desc'), gene.getAttr('summary'))) censusInfo.append('%s:%s:%s:%s' % (gene.getAttr('census_somatic'), gene.getAttr('census_germline'), gene.getAttr('census_mutType'), gene.getAttr('census_translocPartners'))) goInfoS = goInfoS.union(set(gene.getAttr('go'))) keggInfoS = keggInfoS.union(set(gene.getAttr('kegg'))) biocartaInfoS = biocartaInfoS.union( set(gene.getAttr('biocarta'))) geneStatL.append( (bp1.split(':')[0], bp, teStr, teL, geneS, bp_geneS, cnaInfo, geneInfo, censusInfo, goInfoS, keggInfoS, biocartaInfoS)) (chrom1, bp1, teStr1, teL1, geneS1, bp_geneS1, cnaInfo1, geneInfo1, censusInfo1, goInfoS1, keggInfoS1, biocartaInfoS1) = geneStatL[0] (chrom2, bp2, teStr2, teL2, geneS2, bp_geneS2, cnaInfo2, geneInfo2, censusInfo2, goInfoS2, keggInfoS2, biocartaInfoS2) = geneStatL[1] if chrom1 == chrom2: type = 'intra' else: type = 'inter' frameL = [] for (t1, e1) in teL1: for (t2, e2) in teL2: cons = mygenome.frameCons(t1, e1, t2, e2, frameInfoH) if cons == 'Y': frameL.append('%s.%s-%s.%s:%s' % (t1, e1, t2, e2, cons)) outReportFile.write('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n' % \ (sampN, splice_type, type, bp1,bp2, teStr1,teStr2, ','.join(frameL), ','.join(cnaInfo1), ','.join(cnaInfo2), \ '%s;%s' % (','.join(geneS1),','.join(bp_geneS1)), ';'.join(geneInfo1), ';'.join(censusInfo1), \ ';'.join(map(str,goInfoS1)), ';'.join(map(str,keggInfoS1)), ';'.join(map(str,biocartaInfoS1)), '%s;%s' % (','.join(geneS2),','.join(bp_geneS2)), ';'.join(geneInfo2), ';'.join(censusInfo2), \ ';'.join(map(str,goInfoS2)), ';'.join(map(str,keggInfoS2)), ';'.join(map(str,biocartaInfoS2)), nmatch,nseq,nreg))