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 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 gsnap_process_junction(inReportFileName,outReportFileName):
geneNameH = mygenome.geneNameH()
geneSetH = mygenome.geneSetH()
geneInfoH = mygenome.geneInfoH(geneNameH,geneSetH)
outReportFile = open(outReportFileName,'w')
for line in open(inReportFileName):
(spliceType,sampN,bp1,bp2,t1,t2,nmatch,nseq,nreg) = line[:-1].split('\t')
gene1 = set()
if t1:
transcript1 = tuple(t1.split(';'))
for t in transcript1:
g = mygenome.gene(t,geneNameH,geneSetH,geneInfoH)
if g.geneName:
gene1.add(g.geneName)
else:
gene1 = ()
gene2 = set()
if t2:
transcript2 = tuple(t2.split(';'))
for t in transcript2:
g = mygenome.gene(t,geneNameH,geneSetH,geneInfoH)
if g.geneName:
gene2.add(g.geneName)
else:
gene2 = ()
bp_gene1 = set()
# transcript1 = tuple([x for x in bp1.split('|') if "uc" in x])
for t in tuple([x for x in bp1.split('|') if "uc" in x]):
g = mygenome.gene(t,geneNameH,geneSetH,geneInfoH)
if g.geneName:
bp_gene1.add(g.geneName)
bp_gene2 = set()
# transcript2 = tuple([x for x in bp2.split('|') if "uc" in x])
for t in tuple([x for x in bp2.split('|') if "uc" in x]):
g = mygenome.gene(t,geneNameH,geneSetH,geneInfoH)
if g.geneName:
bp_gene2.add(g.geneName)
# ch1 = tuple([x for x in id1.split('|') if not "uc" in x])
# ch2 = tuple([x for x in id2.split('|') if not "uc" in x])
if tuple([x for x in bp1.split('|') if "chr" in x])[0] == tuple([x for x in bp2.split('|') if "chr" in x])[0]:
type = 'intra'
else:
type = 'inter'
geneInfo1 = []
censusInfo1 = []
for geneName in gene1:
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 gene2:
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, spliceType, sampN, bp1, bp2, \
t1, t2, ';'.join(gene1), ';'.join(gene2), ';'.join(geneInfo1), ';'.join(geneInfo2), \
';'.join(censusInfo1), ';'.join(censusInfo2), ','.join(bp_gene1), ','.join(bp_gene2), \
nmatch ,nseq, nreg))
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')
def gsnap_process_junction(inReportFileName, outReportFileName):
geneNameH = mygenome.geneNameH()
geneSetH = mygenome.geneSetH()
geneInfoH = mygenome.geneInfoH(geneNameH, geneSetH)
outReportFile = open(outReportFileName, 'w')
for line in open(inReportFileName):
(spliceType, sampN, bp1, bp2, t1, t2, nmatch, nseq,
nreg) = line[:-1].split('\t')
gene1 = set()
if t1:
transcript1 = tuple(t1.split(';'))
for t in transcript1:
g = mygenome.gene(t, geneNameH, geneSetH, geneInfoH)
if g.geneName:
gene1.add(g.geneName)
else:
gene1 = ()
gene2 = set()
if t2:
transcript2 = tuple(t2.split(';'))
for t in transcript2:
g = mygenome.gene(t, geneNameH, geneSetH, geneInfoH)
if g.geneName:
gene2.add(g.geneName)
else:
gene2 = ()
bp_gene1 = set()
# transcript1 = tuple([x for x in bp1.split('|') if "uc" in x])
for t in tuple([x for x in bp1.split('|') if "uc" in x]):
g = mygenome.gene(t, geneNameH, geneSetH, geneInfoH)
if g.geneName:
bp_gene1.add(g.geneName)
bp_gene2 = set()
# transcript2 = tuple([x for x in bp2.split('|') if "uc" in x])
for t in tuple([x for x in bp2.split('|') if "uc" in x]):
g = mygenome.gene(t, geneNameH, geneSetH, geneInfoH)
if g.geneName:
bp_gene2.add(g.geneName)
# ch1 = tuple([x for x in id1.split('|') if not "uc" in x])
# ch2 = tuple([x for x in id2.split('|') if not "uc" in x])
if tuple([x for x in bp1.split('|') if "chr" in x
])[0] == tuple([x for x in bp2.split('|') if "chr" in x])[0]:
type = 'intra'
else:
type = 'inter'
geneInfo1 = []
censusInfo1 = []
for geneName in gene1:
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 gene2:
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, spliceType, sampN, bp1, bp2, \
t1, t2, ';'.join(gene1), ';'.join(gene2), ';'.join(geneInfo1), ';'.join(geneInfo2), \
';'.join(censusInfo1), ';'.join(censusInfo2), ','.join(bp_gene1), ','.join(bp_gene2), \
nmatch ,nseq, nreg))