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 genKgCompositeModel(outTextFileName, outFaFileName):
kgH = mygenome.loadKgByChr()
outTextFile = open(outTextFileName, 'w')
outFaFile = open(outFaFileName, 'w')
for chrNum in range(1, 23) + ['X', 'Y', 'M']:
#for chrNum in [1]:
chrom = 'chr%s' % chrNum
txnLocusL_combined = []
for strand in ['+', '-']:
txnLocusL = [
mygenome.locus(
'%s:%s-%s%s' % (chrom, h['txnSta'], h['txnEnd'], strand),
h['kgId'])
for h in filter(lambda x: x['strand'] == strand, kgH[chrom])
]
n_before = len(txnLocusL)
txnLocusL = mygenome.mergeLoci(txnLocusL)
n_after = len(txnLocusL)
#print chrom, strand, n_before, n_after
txnLocusL_combined += txnLocusL
txnLocusL_combined.sort(lambda x, y: cmp(x.chrEnd, y.chrEnd))
txnLocusL_combined.sort(lambda x, y: cmp(x.chrSta, y.chrSta))
for txnLoc in txnLocusL_combined:
exnLocusL = []
for h in filter(lambda x: x['kgId'] in txnLoc.id, kgH[chrom]):
for (exnSta, exnEnd) in h['exnList']:
exnLocusL.append(
mygenome.locus('%s:%s-%s%s' %
(chrom, exnSta, exnEnd, h['strand'])))
exnLocusL.sort(lambda x, y: cmp(x.chrEnd, y.chrEnd))
exnLocusL.sort(lambda x, y: cmp(x.chrSta, y.chrSta))
exnLocusL = mygenome.mergeLoci(exnLocusL)
exnStaL = [str(exnLoc.chrSta) for exnLoc in exnLocusL]
exnEndL = [str(exnLoc.chrEnd) for exnLoc in exnLocusL]
outTextFile.write('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n' %
(txnLoc.id, txnLoc.chrom, txnLoc.strand,
txnLoc.chrSta, txnLoc.chrEnd, len(exnLocusL),
','.join(exnStaL), ','.join(exnEndL)))
outFaFile.write('>%s|%s|%s|%s|%s\n' %
(txnLoc.id, txnLoc.chrom, txnLoc.strand,
txnLoc.chrSta, txnLoc.chrEnd))
# for exnLoc in exnLocusL:
# outFaFile.write(exnLoc.nibFrag())
txnLocCopy = copy.deepcopy(
txnLoc) # print whole txn sequence in positive strand
txnLocCopy.strand = '+'
outFaFile.write(txnLocCopy.nibFrag())
outFaFile.write('\n')
outTextFile.close()
outFaFile.close()
def genCompositeModel(outTextFileName,outFaFileName,intronSize=100):
geneNameH = mygenome.geneNameH()
geneSetH = mygenome.geneSetH()
geneInfoH = mygenome.geneInfoH(geneNameH,geneSetH)
geneH = mygenome.loadKgByChr()
#geneH = mygenome.loadLincByChr(h=geneH)
outTextFile = open(outTextFileName, 'w')
outFaFile = open(outFaFileName, 'w')
for chrNum in range(1,23)+['X','Y','M']:
#for chrNum in [1]:
chrom = 'chr%s' % chrNum
geneH_byChr = filter(lambda x: mygenome.gene(x['geneId'],geneNameH,geneSetH,geneInfoH).geneName in mygenome.TK, geneH[chrom])
txnLocusL_combined = []
for strand in ['+','-']:
txnLocusL = [mygenome.locus('%s:%s-%s%s' % (chrom,h['txnSta'],h['txnEnd'],strand),h['geneId']) for h in filter(lambda x: x['strand']==strand, geneH_byChr)]
n_before = len(txnLocusL)
txnLocusL = mygenome.mergeLoci(txnLocusL)
n_after = len(txnLocusL)
#print chrom, strand, n_before, n_after
txnLocusL_combined += txnLocusL
txnLocusL_combined.sort(lambda x,y: cmp(x.chrEnd,y.chrEnd))
txnLocusL_combined.sort(lambda x,y: cmp(x.chrSta,y.chrSta))
for txnLoc in txnLocusL_combined:
exnLocusL = []
for h in filter(lambda x: x['geneId'] in txnLoc.id, geneH_byChr):
for (exnSta,exnEnd) in h['exnList']:
exnLocusL.append(mygenome.locus('%s:%s-%s%s' % (chrom, exnSta, exnEnd, h['strand'])))
exnLocusL.sort(lambda x,y: cmp(x.chrEnd,y.chrEnd))
exnLocusL.sort(lambda x,y: cmp(x.chrSta,y.chrSta))
exnLocusL = mygenome.mergeLoci(exnLocusL)
exnStaL = [str(exnLoc.chrSta) for exnLoc in exnLocusL]
exnEndL = [str(exnLoc.chrEnd) for exnLoc in exnLocusL]
outTextFile.write('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n' % (txnLoc.id,txnLoc.chrom,txnLoc.strand,txnLoc.chrSta,txnLoc.chrEnd,len(exnLocusL),','.join(exnStaL),','.join(exnEndL)))
outFaFile.write('>%s|%s|%s|%s|%s\n' % (txnLoc.id,txnLoc.chrom,txnLoc.strand,txnLoc.chrSta,txnLoc.chrEnd))
for i in range(len(exnLocusL)):
exnLocCopy = copy.deepcopy(exnLocusL[i])
exnLocCopy.strand = '+'
if i > 0:
exnLocCopy.chrSta -= min(intronSize, int((exnLocusL[i].chrSta - exnLocusL[i-1].chrEnd)/2))
if i < len(exnLocusL)-1:
exnLocCopy.chrEnd += min(intronSize, int((exnLocusL[i+1].chrSta - exnLocusL[i].chrEnd)/2))
outFaFile.write(exnLocCopy.nibFrag())
outFaFile.write('\n')
outTextFile.close()
outFaFile.close()
def genKgCompositeModel(outTextFileName,outFaFileName):
kgH = mygenome.loadKgByChr()
outTextFile = open(outTextFileName, 'w')
outFaFile = open(outFaFileName, 'w')
for chrNum in range(1,23)+['X','Y','M']:
#for chrNum in [1]:
chrom = 'chr%s' % chrNum
txnLocusL_combined = []
for strand in ['+','-']:
txnLocusL = [mygenome.locus('%s:%s-%s%s' % (chrom,h['txnSta'],h['txnEnd'],strand),h['kgId']) for h in filter(lambda x: x['strand']==strand, kgH[chrom])]
n_before = len(txnLocusL)
txnLocusL = mygenome.mergeLoci(txnLocusL)
n_after = len(txnLocusL)
#print chrom, strand, n_before, n_after
txnLocusL_combined += txnLocusL
txnLocusL_combined.sort(lambda x,y: cmp(x.chrEnd,y.chrEnd))
txnLocusL_combined.sort(lambda x,y: cmp(x.chrSta,y.chrSta))
for txnLoc in txnLocusL_combined:
exnLocusL = []
for h in filter(lambda x: x['kgId'] in txnLoc.id, kgH[chrom]):
for (exnSta,exnEnd) in h['exnList']:
exnLocusL.append(mygenome.locus('%s:%s-%s%s' % (chrom, exnSta, exnEnd, h['strand'])))
exnLocusL.sort(lambda x,y: cmp(x.chrEnd,y.chrEnd))
exnLocusL.sort(lambda x,y: cmp(x.chrSta,y.chrSta))
exnLocusL = mygenome.mergeLoci(exnLocusL)
exnStaL = [str(exnLoc.chrSta) for exnLoc in exnLocusL]
exnEndL = [str(exnLoc.chrEnd) for exnLoc in exnLocusL]
outTextFile.write('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n' % (txnLoc.id,txnLoc.chrom,txnLoc.strand,txnLoc.chrSta,txnLoc.chrEnd,len(exnLocusL),','.join(exnStaL),','.join(exnEndL)))
outFaFile.write('>%s|%s|%s|%s|%s\n' % (txnLoc.id,txnLoc.chrom,txnLoc.strand,txnLoc.chrSta,txnLoc.chrEnd))
# for exnLoc in exnLocusL:
# outFaFile.write(exnLoc.nibFrag())
txnLocCopy = copy.deepcopy(txnLoc) # print whole txn sequence in positive strand
txnLocCopy.strand = '+'
outFaFile.write(txnLocCopy.nibFrag())
outFaFile.write('\n')
outTextFile.close()
outFaFile.close()
