def process_bp(inFileName,outFileName,coordH,regionL):
result = mygsnap.gsnapFile(inFileName,True)
outFile = open(outFileName, 'w')
outFile.write('browser full knownGene\n')
outFile.write('track name="%s" visibility=2\n' % inFileName)
for rL in result:
if not (rL[0].nLoci==1 and rL[1].nLoci==1 and rL[0].pairRel=='unpaired'):
raise Exception
locL = [mygenome.locus(rL[0].matchL()[0].segL[0][2]), mygenome.locus(rL[1].matchL()[0].segL[0][2])]
for loc in locL:
loc.chrSta += coordH[loc.chrom][1] -1
loc.chrEnd += coordH[loc.chrom][1] -1
loc.chrom = coordH[loc.chrom][0]
flag = False
for loc in locL:
for region in regionL:
if loc.overlap(region) > 0:
flag = True
if flag:
print '^%s.*%s$\n' % (rL[0].seq(),mybasic.rc(rL[1].seq())),
for loc in locL:
outFile.write('%s\t%s\t%s\n' % (loc.chrom,loc.chrSta,loc.chrEnd))
def process_bed(inFileName, outFileName, coordH):
result = mygsnap.gsnapFile(inFileName, True)
outFile = open(outFileName, 'w')
count_all = 0
count_strand = 0
outFile.write('browser full knownGene\n')
outFile.write('track name="targeted" visibility=2\n')
for rL in result:
if not (rL[0].nLoci == 1 and rL[1].nLoci == 1
and rL[0].pairRel == 'unpaired'):
raise Exception
locL = [
mygenome.locus(rL[0].matchL()[0].segL[0][2]),
mygenome.locus(rL[1].matchL()[0].segL[0][2])
]
for loc in locL:
loc.chrSta += coordH[loc.chrom][1] - 1
loc.chrEnd += coordH[loc.chrom][1] - 1
loc.chrom = coordH[loc.chrom][0]
for loc in locL:
outFile.write('%s\t%s\t%s\n' % (loc.chrom, loc.chrSta, loc.chrEnd))
def filter_transloc(inFileName, outFileName):
result = mygsnap.gsnapFile(inFileName, False)
outFile = open(outFileName, 'w')
count_all = 0
count_transloc = 0
for r in result:
count_all += 1
if not '(transloc)' in r.pairRel:
continue
match = r.matchL()[0]
segObjL = match.getSegInfo()
skip = False
for segObj in segObjL:
if segObj.span - segObj.numMatch > 2 or segObj.percMatch < 90 or segObj.span < 5:
skip = True
break
if skip:
continue
outFile.write(r.rawText() + '\n')
count_transloc += 1
print 'Results:', count_transloc, count_all
def filter_transloc(inFileName,outFileName): result = mygsnap.gsnapFile(inFileName,False) outFile = open(outFileName, 'w') count_all = 0 count_transloc = 0 for r in result: count_all += 1 if not '(transloc)' in r.pairRel: continue match = r.matchL()[0] segObjL = match.getSegInfo() skip = False for segObj in segObjL: if segObj.span - segObj.numMatch > 2 or segObj.percMatch < 90 or segObj.span < 5: skip = True break if skip: continue outFile.write(r.rawText()+'\n') count_transloc += 1 print 'Results:',count_transloc,count_all
def filter_strand(inFileName,outFileName):
result = mygsnap.gsnapFile(inFileName,True)
outFile = open(outFileName, 'w')
count_all = 0
count_strand = 0
for rL in result:
if not (rL[0].nLoci==1 and rL[1].nLoci==1 and rL[0].pairRel=='unpaired'):
raise Exception
chrom0 = rL[0].matchL()[0].segL[0][2].split(':')[0]
chrom1 = rL[1].matchL()[0].segL[0][2].split(':')[0]
if ((chrom0[0]==chrom0[-1] and chrom1[0]!=chrom1[-1]) or (chrom0[0]!=chrom0[-1] and chrom1[0]==chrom1[-1])) and chrom0[1:-1]!=chrom1[1:-1]:
for i in (0,1):
outFile.write(rL[i].rawText()+'\n')
count_strand += 1
else:
for i in (0,1):
print rL[i].rawText()
count_all += 1
print count_strand, count_all
def process_bp(inFileName,outFileName,regionL):
result = mygsnap.gsnapFile(inFileName,True)
outFile = open(outFileName, 'w')
outFile.write('browser full knownGene\n')
outFile.write('track name="%s" visibility=2\n' % inFileName)
for rL in result:
if not (rL[0].nLoci==1 and rL[1].nLoci==1 and rL[0].pairRel=='concordant'):
raise Exception
locL = [mygenome.locus(rL[0].matchL()[0].segL[0][2]), mygenome.locus(rL[1].matchL()[0].segL[0][2])]
flag = False
for loc in locL:
for region in regionL:
if loc.overlap(region) > 0:
flag = True
if flag:
print '^%s.*%s$\n' % (rL[0].seq(),mybasic.rc(rL[1].seq())),
for loc in locL:
outFile.write('%s\t%s\t%s\n' % (loc.chrom,loc.chrSta,loc.chrEnd))
def filter_strand(inFileName, outFileName):
result = mygsnap.gsnapFile(inFileName, True)
outFile = open(outFileName, 'w')
count_all = 0
count_strand = 0
for rL in result:
if not (rL[0].nLoci == 1 and rL[1].nLoci == 1
and rL[0].pairRel == 'unpaired'):
raise Exception
chrom0 = rL[0].matchL()[0].segL[0][2].split(':')[0]
chrom1 = rL[1].matchL()[0].segL[0][2].split(':')[0]
if ((chrom0[0] == chrom0[-1] and chrom1[0] != chrom1[-1]) or
(chrom0[0] != chrom0[-1]
and chrom1[0] == chrom1[-1])) and chrom0[1:-1] != chrom1[1:-1]:
for i in (0, 1):
outFile.write(rL[i].rawText() + '\n')
count_strand += 1
else:
for i in (0, 1):
print rL[i].rawText()
count_all += 1
print count_strand, count_all
def process_bed(inFileName,outFileName,coordH):
result = mygsnap.gsnapFile(inFileName,True)
outFile = open(outFileName, 'w')
count_all = 0
count_strand = 0
outFile.write('browser full knownGene\n')
outFile.write('track name="targeted" visibility=2\n')
for rL in result:
if not (rL[0].nLoci==1 and rL[1].nLoci==1 and rL[0].pairRel=='unpaired'):
raise Exception
locL = [mygenome.locus(rL[0].matchL()[0].segL[0][2]), mygenome.locus(rL[1].matchL()[0].segL[0][2])]
for loc in locL:
loc.chrSta += coordH[loc.chrom][1] -1
loc.chrEnd += coordH[loc.chrom][1] -1
loc.chrom = coordH[loc.chrom][0]
for loc in locL:
outFile.write('%s\t%s\t%s\n' % (loc.chrom,loc.chrSta,loc.chrEnd))
def process_bp(inGsnapFileName, outBpFileName):
result = mygsnap.gsnapFile(inGsnapFileName, False)
outBpFile = open(outBpFileName, 'w')
seqH = {}
for r in result:
match = r.matchL()[0]
if not '(transloc)' in r.pairRel:
raise Exception
if len(match.segL) != 2:
raise Exception
s1 = match.segL[0][2]
s2 = match.segL[1][2]
if s1[0] != s2[0]:
raise Exception
strand = s1[0]
s1T = re.match('[+-]([^:]+):[0-9]+..([0-9]+)', s1).groups()
s2T = re.match('[+-]([^:]+):([0-9]+)..[0-9]+', s2).groups()
if strand == '+':
seq = r.seq()
offset = int(match.segL[0][1].split('..')[1])
junction = (s1T, s2T)
else:
seq = mybasic.rc(r.seq(), 'DNA')
offset = len(seq) - int(match.segL[0][1].split('..')[1])
junction = (s2T, s1T)
mybasic.addHash(seqH, junction, (offset, seq))
for ((j1, j2), vL) in seqH.items():
vL.sort(lambda x, y: cmp(x[0], y[0]))
vL_mod = []
for (offset, seq) in vL:
offset = blockSize - offset + 1
vL_mod.append('%s:%s' % (offset, seq))
outBpFile.write('%s:%s-%s,%s:%s-%s,%s\n' %
(j1[0].split('_')[0], int(j1[1]) - blockSize, j1[1],
j1[0].split('_')[0], j1[1], int(j1[1]) + blockSize,
'|'.join(vL_mod)))
def process_bp(inGsnapFileName):
result = mygsnap.gsnapFile(inGsnapFileName, False)
#outBpFile = open(outBpFileName, 'w')
seqH = {}
for r in result:
match = r.matchL()[0]
if not '(transloc)' in r.pairRel:
raise Exception
if len(match.segL) != 2:
raise Exception
s1 = match.segL[0][2]
s2 = match.segL[1][2]
direction = re.search('dir:([^,\t]*)', match.segL[0][3]).group(1)
bp1 = re.match('[+-]([^:]+):[0-9]+..([0-9]+)', s1).groups()
bp2 = re.match('[+-]([^:]+):([0-9]+)..[0-9]+', s2).groups()
# if bp1[0] == bp2[0]:
# continue
if direction == 'sense':
seq = r.seq()
offset = int(match.segL[0][1].split('..')[1])
bp12 = (bp1, bp2)
else:
seq = mybasic.rc(r.seq(), 'DNA')
offset = len(seq) - int(match.segL[0][1].split('..')[1])
bp12 = (bp2, bp1)
mybasic.addHash(seqH, bp12, (offset, seq))
seqL = seqH.items()
seqL.sort(lambda x, y: cmp(len(y[1]), len(x[1])))
for ((bp1, bp2), vL) in seqL:
vL.sort(lambda x, y: cmp(y[0], x[0]))
maxOffset = vL[0][0]
print '\n', bp1, bp2, len(vL), '\n'
for (offset, seq) in vL:
print '%s%s %s' % (' ' * (maxOffset - offset), seq[:offset],
seq[offset:])
def main(inGsnapFileName, outReportFileName, sampN, geneNL=[], overlap=10):
eiH, ei_keyH, juncInfoH = loadAnnot(geneNL)
print 'Finished loading refFlat'
result = mygsnap.gsnapFile(inGsnapFileName, False)
count = 0
for r in result:
if r.nLoci != 1:
continue
match = r.matchL()[0]
for seg in match.segL:
loc = seg[2]
rm = re.match('([+-])([^:]+):([0-9,]+)..([0-9,]+)', loc)
strand = rm.group(1)
chrom = rm.group(2)
chrPosL = [int(rm.group(3)), int(rm.group(4))]
chrSta = min(chrPosL) - 1
chrEnd = max(chrPosL)
for pos in ei_keyH[chrom]:
if chrSta + overlap <= pos <= chrEnd - overlap:
eiH[chrom][pos] += 1
elif chrEnd - overlap < pos:
break
# count += 1
#
# if count % 10000 == 0:
# print count
outReportFile = open(outReportFileName, 'w')
for chrom in ei_keyH.keys():
for e in ei_keyH[chrom]:
if eiH[chrom][e] == []:
continue
outReportFile.write(
'%s\t%s\t%s\t%s\n' %
(sampN, '%s:%s' %
(chrom, e), ','.join(juncInfoH[chrom][e]), eiH[chrom][e]))
def process_bp(inGsnapFileName,outBpFileName):
result = mygsnap.gsnapFile(inGsnapFileName,False)
outBpFile = open(outBpFileName, 'w')
seqH = {}
for r in result:
match = r.matchL()[0]
if not '(transloc)' in r.pairRel:
raise Exception
if len(match.segL) != 2:
raise Exception
s1 = match.segL[0][2]
s2 = match.segL[1][2]
if s1[0] != s2[0]:
raise Exception
strand = s1[0]
s1T = re.match('[+-]([^:]+):[0-9]+..([0-9]+)',s1).groups()
s2T = re.match('[+-]([^:]+):([0-9]+)..[0-9]+',s2).groups()
if strand == '+':
seq = r.seq()
offset = int(match.segL[0][1].split('..')[1])
junction = (s1T, s2T)
else:
seq = mybasic.rc(r.seq(),'DNA')
offset = len(seq)-int(match.segL[0][1].split('..')[1])
junction = (s2T, s1T)
mybasic.addHash(seqH,junction,(offset,seq))
for ((k1,k2), v) in seqH.items():
v.sort(lambda x,y: cmp(y[0],x[0]))
k1T = re.match()
k2T = re.match()
k1_pos =
k2_pos =
k1_seq =
k2_seq =
outBpFile.write('%s,%s,%s\n' % (':'.join(k1),':'.join(k2),'|'.join(['%s:%s' % (offset,seq) for (offset,seq) in v])))
def process_bp(inGsnapFileName):
result = mygsnap.gsnapFile(inGsnapFileName,False)
#outBpFile = open(outBpFileName, 'w')
seqH = {}
for r in result:
match = r.matchL()[0]
if not '(transloc)' in r.pairRel:
raise Exception
if len(match.segL) != 2:
raise Exception
s1 = match.segL[0][2]
s2 = match.segL[1][2]
direction = re.search('dir:([^,\t]*)', match.segL[0][3]).group(1)
bp1 = re.match('[+-]([^:]+):[0-9]+..([0-9]+)',s1).groups()
bp2 = re.match('[+-]([^:]+):([0-9]+)..[0-9]+',s2).groups()
# if bp1[0] == bp2[0]:
# continue
if direction == 'sense':
seq = r.seq()
offset = int(match.segL[0][1].split('..')[1])
bp12 = (bp1, bp2)
else:
seq = mybasic.rc(r.seq(),'DNA')
offset = len(seq)-int(match.segL[0][1].split('..')[1])
bp12 = (bp2, bp1)
mybasic.addHash(seqH,bp12,(offset,seq))
seqL = seqH.items()
seqL.sort(lambda x,y: cmp(len(y[1]),len(x[1])))
for ((bp1,bp2), vL) in seqL:
vL.sort(lambda x,y: cmp(y[0],x[0]))
maxOffset = vL[0][0]
print '\n',bp1,bp2,len(vL),'\n'
for (offset,seq) in vL:
print '%s%s %s' % (' ' * (maxOffset-offset),seq[:offset],seq[offset:])
def process_bp(inGsnapFileName,outBpFileName):
result = mygsnap.gsnapFile(inGsnapFileName,False)
outBpFile = open(outBpFileName, 'w')
seqH = {}
for r in result:
match = r.matchL()[0]
if not '(transloc)' in r.pairRel:
raise Exception
if len(match.segL) != 2:
raise Exception
s1 = match.segL[0][2]
s2 = match.segL[1][2]
if s1[0] != s2[0]:
raise Exception
strand = s1[0]
s1T = re.match('[+-]([^:]+):[0-9]+..([0-9]+)',s1).groups()
s2T = re.match('[+-]([^:]+):([0-9]+)..[0-9]+',s2).groups()
if strand == '+':
seq = r.seq()
offset = int(match.segL[0][1].split('..')[1])
junction = (s1T, s2T)
else:
seq = mybasic.rc(r.seq(),'DNA')
offset = len(seq)-int(match.segL[0][1].split('..')[1])
junction = (s2T, s1T)
mybasic.addHash(seqH,junction,(offset,seq))
for ((j1,j2), vL) in seqH.items():
vL.sort(lambda x,y: cmp(x[0],y[0]))
vL_mod = []
for (offset,seq) in vL:
offset = blockSize-offset+1
vL_mod.append('%s:%s' % (offset,seq))
outBpFile.write('%s:%s-%s,%s:%s-%s,%s\n' % (j1[0].split('_')[0],int(j1[1])-blockSize,j1[1], j1[0].split('_')[0],j1[1],int(j1[1])+blockSize, '|'.join(vL_mod)))
def main(inGsnapFileName,outReportFileName,sampN,geneNL=[],overlap=10):
eiH, ei_keyH, juncInfoH = loadAnnot(geneNL)
print 'Finished loading refFlat'
result = mygsnap.gsnapFile(inGsnapFileName,False)
count = 0
for r in result:
if r.nLoci != 1:
continue
match = r.matchL()[0]
for seg in match.segL:
loc = seg[2]
rm = re.match('([+-])([^:]+):([0-9,]+)..([0-9,]+)',loc)
strand = rm.group(1)
chrom = rm.group(2)
chrPosL = [int(rm.group(3)), int(rm.group(4))]
chrSta = min(chrPosL) - 1
chrEnd = max(chrPosL)
for pos in ei_keyH[chrom]:
if chrSta+overlap <= pos <= chrEnd-overlap:
eiH[chrom][pos] += 1
elif chrEnd-overlap < pos:
break
# count += 1
#
# if count % 10000 == 0:
# print count
outReportFile = open(outReportFileName,'w')
for chrom in ei_keyH.keys():
for e in ei_keyH[chrom]:
if eiH[chrom][e]==[]:
continue
outReportFile.write('%s\t%s\t%s\t%s\n' % (sampN, '%s:%s' % (chrom,e), ','.join(juncInfoH[chrom][e]), eiH[chrom][e]))
def main(inGsnapFileName,outReportFileName,sampN,geneNL=[],overlap=10):
eiH, ei_keyH, juncInfoH, ei_cntH = loadAnnot(geneNL)
print 'Finished loading refFlat'
result = mygsnap.gsnapFile(inGsnapFileName,False)
count = 0
for r in result:
if r.nLoci != 1:
continue
match = r.matchL()[0]
for seg in match.segL:
loc = mygenome.locus(seg[2])
if loc.chrSta + overlap > loc.chrEnd - overlap:
continue
cnt_s = findCut(ei_cntH[loc.chrom], ei_keyH[loc.chrom], loc.chrSta + overlap - 1)
cnt_e = findCut(ei_cntH[loc.chrom], ei_keyH[loc.chrom], loc.chrEnd - overlap)
if cnt_e < 1: ## no junction overlaps
continue
elif cnt_s != cnt_e: # overlapping junction exists
pos_min = bisect.bisect_right(ei_keyH[loc.chrom], loc.chrSta + overlap - 1) - 1
pos_max = bisect.bisect_right(ei_keyH[loc.chrom], loc.chrEnd - overlap)
for pos in range(pos_min, pos_max):
if loc.chrSta+overlap <= ei_keyH[loc.chrom][pos] <= loc.chrEnd-overlap:
eiH[loc.chrom][ei_keyH[loc.chrom][pos]] += 1
# count += 1
#
# if count % 10000 == 0:
# print count
outReportFile = open(outReportFileName,'w')
for chrom in ei_keyH.keys():
for e in ei_keyH[chrom]:
if eiH[chrom][e]==[]:
continue
outReportFile.write('%s\t%s\t%s\t%s\n' % (sampN, '%s:%s' % (chrom,e), ','.join(juncInfoH[chrom][e]), eiH[chrom][e]))
def filter_annot1(inFileName, outFileName):
result = mygsnap.gsnapFile(inFileName, False)
outFile = open(outFileName, 'w')
count_all = 0
count_include = 0
for r in result:
if not '(transloc)' in r.pairRel:
raise Exception
match = r.matchL()[0]
geneSetL = []
for i in range(len(match.segL)):
rm = re.search('label_[12]:([^,\t]*)', match.segL[i][3])
if rm:
geneSetL.append(
set([x.split('.exon')[0] for x in rm.group(1).split('|')]))
else:
geneSetL.append(set())
geneSetCommon = geneSetL[0]
for s in geneSetL[1:]:
geneSetCommon = geneSetCommon.intersection(s)
if len(geneSetCommon) == 0:
outFile.write(r.rawText() + '\n')
count_include += 1
# else:
# print r.rawText()
count_all += 1
print 'Results:', count_include, count_all
def filter_annot2(inFileName, outFileName):
result = mygsnap.gsnapFile(inFileName, False)
outFile = open(outFileName, "w")
count_all = 0
count_include = 0
for r in result:
if not "(transloc)" in r.pairRel:
raise Exception
match = r.matchL()[0]
geneSetL = []
isThereEmptySet = False
for i in range(len(match.segL)):
rm = re.search("label_[12]:([^,\t]*)", match.segL[i][3])
if rm:
geneSetL.append(set([x.split(".exon")[0] for x in rm.group(1).split("|")]))
else:
geneSetL.append(set())
isThereEmptySet = True
geneSetCommon = geneSetL[0]
for s in geneSetL[1:]:
geneSetCommon = geneSetCommon.intersection(s)
if len(geneSetCommon) == 0 and isThereEmptySet == False:
outFile.write(r.rawText() + "\n")
count_include += 1
else:
print r.rawText()
count_all += 1
print count_include, count_all
def filter_crossMap(inFileName,outFileName): result = mygsnap.gsnapFile(inFileName,True) outFile = open(outFileName, 'w') count_all = 0 count_crossMap = 0 for rL in result: if rL[0].nLoci==1 and rL[1].nLoci==1 and rL[0].pairRel=='unpaired': for i in (0,1): outFile.write(rL[i].rawText()+'\n') count_crossMap += 1 count_all += 1 print count_crossMap,count_all
def main(inGsnapFileName,outReportFileName,sampN,geneNL=[],overlap=10):
eiH, ei_keyH, juncInfoH = loadAnnot(geneNL)
print 'Finished loading refFlat'
result = mygsnap.gsnapFile(inGsnapFileName,False)
count = 0
for r in result:
if r.nLoci != 1:
continue
match = r.matchL()[0]
for seg in match.segL:
loc = mygenome.locus(seg[2])
cursor.execute('select 1 from temp_table where chrom="%s" and pos>=%s and pos<=%s' % (loc.chrom,loc.chrSta+overlap,loc.chrEnd-overlap))
if cursor.fetchone():
eiH[loc.chrom][pos] += 1
count += 1
if count % 10000 == 0:
print count
outReportFile = open(outReportFileName,'w')
for chrom in ei_keyH.keys():
for e in ei_keyH[chrom]:
if eiH[chrom][e]==[]:
continue
outReportFile.write('%s\t%s\t%s\t%s\n' % (sampN, '%s:%s' % (chrom,e), ','.join(juncInfoH[chrom][e]), eiH[chrom][e]))
def filter_crossMap(inFileName, outFileName):
result = mygsnap.gsnapFile(inFileName, True)
outFile = open(outFileName, 'w')
count_all = 0
count_crossMap = 0
for rL in result:
if rL[0].nLoci == 1 and rL[1].nLoci == 1 and rL[
0].pairRel == 'unpaired':
for i in (0, 1):
outFile.write(rL[i].rawText() + '\n')
count_crossMap += 1
count_all += 1
print count_crossMap, count_all
def bp_filter(inFileName,outFileName,expSize): ''' filters gsnap records with (no-mismatch, no-indel, no-splicing) and (insert_length > N-nt) ''' result = mygsnap.gsnapFile(inFileName,True) outFile = open(outFileName, 'w') count_all = 0 count_include = 0 for rL in result: count_all += 1 if rL[0].nLoci != 1 or rL[1].nLoci != 1: continue skip = False for i in (0,1): match = rL[i].matchL()[0] seg = match.getSegInfo()[0] if len(match.segL) > 1 or match.pairInfo()[0] < 1000 or match.pairInfo()[0] > 10000 or (seg.len - seg.numMatch) > 0: skip = True break if skip: continue for i in (0,1): print rL[i].rawText() print rL[i].matchL()[0].getSegInfo()[0].len, rL[i].matchL()[0].getSegInfo()[0].numMatch outFile.write(rL[i].rawText()+'\n') count_include += 1 print count_include, count_all
def filter_annot1(inFileName,outFileName):
result = mygsnap.gsnapFile(inFileName,False)
outFile = open(outFileName, 'w')
count_all = 0
count_include = 0
for r in result:
if not '(transloc)' in r.pairRel:
raise Exception
match = r.matchL()[0]
geneSetL = []
for i in range(len(match.segL)):
rm = re.search('label_[12]:([^,\t]*)',match.segL[i][3])
if rm:
geneSetL.append(set([x.split('.exon')[0] for x in rm.group(1).split('|')]))
else:
geneSetL.append(set())
geneSetCommon = geneSetL[0]
for s in geneSetL[1:]:
geneSetCommon = geneSetCommon.intersection(s)
if len(geneSetCommon) == 0:
outFile.write(r.rawText()+'\n')
count_include += 1
# else:
# print r.rawText()
count_all += 1
print 'Results:',count_include,count_all
def bp_filter(inFileName, outFileName, expSize):
'''
filters gsnap records with (no-mismatch, no-indel, no-splicing) and (insert_length > N-nt)
'''
result = mygsnap.gsnapFile(inFileName, True)
outFile = open(outFileName, 'w')
count_all = 0
count_include = 0
for rL in result:
count_all += 1
if rL[0].nLoci != 1 or rL[1].nLoci != 1 or rL[0].pairRel not in (
'concordant', 'paired'):
continue
skip = False
for i in (0, 1):
match = rL[i].matchL()[0]
if len(match.segL) > 1 or match.pairInfo()[0] == int(
expSize) or match.getSegInfo()[0].numMismatch > 0:
skip = True
break
if skip:
continue
for i in (0, 1):
outFile.write(rL[i].rawText() + '\n')
count_include += 1
print count_include, count_all
def bp_filter(inFileName,outFileName,expSize):
'''
filters gsnap records with (no-mismatch, no-indel, no-splicing) and (insert_length > N-nt)
'''
result = mygsnap.gsnapFile(inFileName,True)
outFile = open(outFileName, 'w')
count_all = 0
count_include = 0
for rL in result:
count_all += 1
if rL[0].nLoci != 1 or rL[1].nLoci != 1 or rL[0].pairRel not in ('concordant','paired'):
continue
skip = False
for i in (0,1):
match = rL[i].matchL()[0]
if len(match.segL) > 1 or match.pairInfo()[0] == int(expSize) or match.getSegInfo()[0].numMismatch > 0:
skip = True
break
if skip:
continue
for i in (0,1):
outFile.write(rL[i].rawText()+'\n')
count_include += 1
print count_include, count_all
def process_bp(inFileName,outFileName,regionL):
result = mygsnap.gsnapFile(inFileName,True)
outFile = open(outFileName, 'w')
outFile.write('browser full knownGene\n')
outFile.write('track name="%s" visibility=2\n' % inFileName)
for rL in result:
if not (rL[0].nLoci==1 and rL[1].nLoci==1 and rL[0].pairRel=='concordant'):
raise Exception
if int(rL[0].pairInfo()[0]) < 76:
continue
locL = [mygenome.locus(rL[0].matchL()[0].segL[0][2]), mygenome.locus(rL[1].matchL()[0].segL[0][2])]
flag = False
for loc in locL:
for region in regionL:
if loc.overlap(region) > 0:
flag = True
if flag:
# print '^%s.*%s$\n' % (rL[0].seq(),mybasic.rc(rL[1].seq())),
# for loc in locL:
# outFile.write('%s\t%s\t%s\n' % (loc.chrom,loc.chrSta,loc.chrEnd))
if locL[0].chrEnd < locL[1].chrSta:
outFile.write('%s\t%s\t%s\n' % (loc.chrom,locL[0].chrEnd,locL[1].chrSta))
else:
outFile.write('%s\t%s\t%s\n' % (loc.chrom,locL[1].chrEnd,locL[0].chrSta))
def fusion_proc_sort(inGsnapFileName,outGsnapFileName,outReportFileName,sampN):
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))
transcriptL = []
for i in range(2):
rm = re.search('label_[12]:([^,\t]*)', match.segL[i][3])
if rm:
transcriptL.append(rm.group(1).replace('|',','))
else:
transcriptL.append('')
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
if direction=='sense':
key = ((trans_strand1,)+bp1.groups()[1:],(trans_strand2,)+bp2.groups()[1:])
elif direction=='antisense':
key = ((trans_strand2,)+bp2.groups()[1:],(trans_strand1,)+bp1.groups()[1:])
transcriptL = transcriptL[::-1]
else:
raise Exception
if key in juncHH:
juncHH[key]['match'].append(r)
juncHH[key]['seq'].append(r.seq())
juncHH[key]['pos'].append((direction,offset))
else:
juncHH[key] = {'match':[r], 'splice_type':splice_type, 'seq':[r.seq()], 'pos':[(direction,offset)], 'transcript':transcriptL}
juncKH = juncHH.items()
juncKH.sort(lambda x,y: cmp(len(set(y[1]['pos'])),len(set(x[1]['pos']))))
outGsnapFile = open(outGsnapFileName,'w')
outReportFile = open(outReportFileName,'w')
for (key, juncH) in juncKH:
outReportFile.write('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n' % \
(juncH['splice_type'], sampN, key[0][0]+':'.join(key[0][1:]), key[1][0]+':'.join(key[1][1:]), \
juncH['transcript'][0], juncH['transcript'][1], \
len(juncH['match']), len(set(juncH['seq'])), len(set(juncH['pos']))))
for m in juncH['match']:
outGsnapFile.write(m.rawText()+'\n')
def gsnap_process_junction(inGsnapFileName,outGsnapFileName,outReportFileName,sampN):
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])
if transcript1:
transcript1 = tuple([x.split('.exon')[0] for x in transcript1.group(1).split('|')])
else:
transcript1 = ()
transcript2 = re.search('label_[12]:([^,\t]*)', match.segL[1][3])
if transcript2:
transcript2 = tuple([x.split('.exon')[0] for x in transcript2.group(1).split('|')])
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
if direction=='sense':
key = (bp1.groups()[1:],bp2.groups()[1:])
transcript = (transcript1,transcript2)
elif direction=='antisense':
key = (bp2.groups()[1:],bp1.groups()[1:])
transcript = (transcript2,transcript1)
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}
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:
outReportFile.write('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n' % \
(juncH['splice_type'], sampN,':'.join(key[0]), ':'.join(key[1]),\
';'.join(juncH['transcript'][0]), ';'.join(juncH['transcript'][1]),\
len(juncH['match']) ,len(set(juncH['seq'])), len(set(juncH['reg']))))
for m in juncH['match']:
outGsnapFile.write(m.rawText()+'\n')
def make_samse(ifileN, ofileN):
headerL = make_header('/data1/Sequence/ucsc_hg19/hg19.chrom.sizes')
# ofile = open(ofileN, 'w')
for header in headerL:
print header
# ofile.write('%s\n' % header)
result = mygsnap.gsnapFile(
'/pipeline/test_ini_gsnap2sam/S022_single.gsnap', False)
#result = mygsnap.gsnapFile('/pipeline/test_ini_gsnap2sam/test.gsnap',False)
#result = mygsnap.gsnapFile('/pipeline/test_ini_gsnap2sam/S022_pair.gsnap',True)
## for unpaired
for r in result:
qname = r.rid()
flag = 0x0
rname = '*'
pos = 0
mapq = 0
cigar = ''
rnext = '*' ## assume --npath=1 (maximum 1 alignment per read)
pnext = 0 ## assume --npath=1 (maximum 1 alignment per read)
tlen = 0 ## assume --npath=1 (maximum 1 alignment per read)
seq = r.seq()
qual = r.qual()
extra = 'NH:i:1\tHI:i:1' ## assume --npath=1 (maximum 1 alignment per read)
if r.nLoci > 1:
flag = flag | 0x4
cigar = '*'
new_cigar = '*'
extra = ''
print('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s' %
(qname, flag, rname, pos, mapq, new_cigar, rnext, pnext,
tlen, seq, qual, extra))
else:
if r.pairRel == '(transloc)':
match = r.matchL()[0]
segL = match.getSegInfo()
mapq = segL[0].mapq
seq = r.seq()
qual = r.qual()
for seg in segL:
flag = 0x0
(strand, rname, pos1,
pos2) = re.search('([\+\-])(.*):([0-9]+)\.\.([0-9]+)',
seg.seg[2]).groups()
pos = min(int(pos1), int(pos2))
(cigar, clip) = seg.toCIGAR_trans()
if clip < 0: ## first half
seq2 = seq[:clip]
qual2 = qual[:clip]
else: ## second half
seq2 = seq[clip:]
qual2 = qual[clip:]
if strand == '-':
flag = flag | 0x10
seq2 = mybasic.rc(seq2)
qual2 = mybasic.rev(qual2)
# print qname,seg.toCIGAR_trans()
print('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s' %
(qname, flag, rname, pos, mapq, cigar, rnext, pnext,
tlen, seq2, qual2, extra))
else:
match = r.matchL()[
0] ## assume --npath=1 (maximum 1 alignment per read)
segL = match.getSegInfo()
(strand, rname, pos1,
pos2) = re.search('([\+\-])(.*):([0-9]+)\.\.([0-9]+)',
segL[0].seg[2]).groups()
pos = min(int(pos1), int(pos2))
mapq = segL[0].mapq
seg_nm = segL[0].numSub
cigar2 = match.toCIGAR()
# print qname, match.toCIGAR()
if segL[0].start != '' and segL[0].start != '0':
cigar = str(segL[0].start) + 'S'
if strand == '-':
cigar = str(segL[0].numMatch +
segL[0].numSub) + 'M' + cigar
if segL[0].ins != '' and segL[0].ins != '0':
cigar = str(segL[0].ins) + 'I' + cigar
else:
cigar = cigar + str(segL[0].numMatch +
segL[0].numSub) + 'M'
if segL[0].ins != '' and segL[0].ins != '0':
cigar = cigar + str(segL[0].ins) + 'I'
if len(segL) == 1:
new_cigar = segL[0].toCIGAR(True)
else:
new_cigar = segL[0].toCIGAR()
prev_cigar = new_cigar
index = 0
for seg in segL[1:]:
index = index + 1
if index == (len(segL) - 1):
final = True
else:
final = False
rm = re.search('([\+\-])(.*):([0-9]+)\.\.([0-9]+)',
seg.seg[2]).groups()
match = str(seg.numMatch + seg.numSub) + 'M'
if seg.ins != '' and seg.ins != '0':
ins = str(seg.ins) + 'I'
else:
ins = ''
if pos == 0 or pos > min(int(rm[2]), int(rm[3])):
pos = min(int(rm[2]), int(rm[3]))
if strand == '-':
dist = int(pos2) - int(rm[2]) - 1
if dist > 0:
cigar = match + ins + str(dist) + 'N' + cigar
else:
cigar = match + ins + cigar
else:
dist = int(rm[2]) - int(pos2) - 1
if dist > 0:
cigar = cigar + str(dist) + 'N' + match + ins
else:
cigar = cigar + match + ins
seg_nm = seg_nm + seg.numSub
pos1 = rm[2]
pos2 = rm[3]
cur_cigar = seg.toCIGAR(final)
if strand == '-':
if dist > 0 and 'D' not in prev_cigar and 'I' not in prev_cigar:
new_cigar = cur_cigar + str(dist) + 'N' + new_cigar
else:
new_cigar = cur_cigar + new_cigar
else:
if dist > 0 and 'D' not in prev_cigar and 'I' not in prev_cigar:
new_cigar = new_cigar + str(dist) + 'N' + cur_cigar
else:
new_cigar = new_cigar + cur_cigar
prev_cigar = cur_cigar
if segL[-1].end != '' and segL[-1].end != '0': ## last segment
if strand == '-':
cigar = str(segL[-1].end) + 'S' + cigar
else:
cigar = cigar + str(segL[-1].end) + 'S'
extra = extra + ('\tNM:i:%s' % seg_nm)
if strand == '-':
flag = flag | 0x10
seq = mybasic.rc(seq)
qual = mybasic.rev(qual)
## print ('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s' % (qname, flag, rname, pos, mapq, cigar, rnext, pnext, tlen, seq, qual, extra))
print('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s' %
(qname, flag, rname, pos, mapq, cigar2, rnext, pnext,
tlen, seq, qual, extra))
def exonSkip_filter(inFileName,outFileName):
'''
filters-in exon-skipping candidates in splice-mapped gsnap
'''
result = mygsnap.gsnapFile(inFileName, False)
outFile = open(outFileName, 'w')
count_all = 0
count_include = 0
for r in result:
if r.nLoci != 1:
continue
match = r.matchL()[0]
if len(match.segL) != 2:
continue
segObjL = match.getSegInfo()
jncH = {}
skip = False
for segObj in segObjL:
if segObj.span - segObj.numMatch > 2 or segObj.percMatch < 90 or segObj.span < 5:
skip = True
break
if segObj.label == '':
break
for b in segObj.label.split('|'):
rm2 = re.match('(.*)\.exon([0-9]+)\/[0-9]+',b)
transId = rm2.group(1)
exonNum = int(rm2.group(2))
mybasic.addHash(jncH,transId,exonNum)
if skip:
continue
jncL = jncH.items()
if len(jncL)>0 and max([len(j[1]) for j in jncL])>1:
minDist = 100
for i in range(len(jncL)):
if len(jncL[i][1]) == 2 and abs(jncL[i][1][0]-jncL[i][1][1]) < minDist:
minDist = abs(jncL[i][1][0]-jncL[i][1][1])
if minDist > 1:
outFile.write(r.rawText()+'\n')
count_include += 1
count_all += 1
print 'Results:',count_include, count_all
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 exonSkip_proc_sort(inGsnapFileName,outGsnapFileName,outReportFileName,sampN):
result = mygsnap.gsnapFile(inGsnapFileName,False)
juncHH = {}
for r in result:
if r.nLoci != 1:
raise Exception
match = r.matchL()[0]
if len(match.segL) != 2:
raise Exception
direction = re.search('dir:([^,\t]*)', match.segL[0][3]).group(1)
offset = int(re.search('\.\.([0-9]*)', match.segL[0][1]).group(1))
exonLP = []
for i in range(len(match.segL)):
rm = re.search('label_[12]:([^,\t]*)',match.segL[i][3])
if not rm:
raise Exception
exonLP.append(rm.group(1).replace('|',','))
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
if direction=='sense':
key = ((trans_strand1,)+bp1.groups()[1:],(trans_strand2,)+bp2.groups()[1:])
elif direction=='antisense':
key = ((trans_strand2,)+bp2.groups()[1:],(trans_strand1,)+bp1.groups()[1:])
exonLP = exonLP[::-1]
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], 'seq':[r.seq()], 'reg':[(direction,offset)], 'exonLP':exonLP}
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:
outReportFile.write('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n' % \
(sampN, key[0][0]+':'.join(key[0][1:]), key[1][0]+':'.join(key[1][1:]),\
juncH['exonLP'][0], juncH['exonLP'][1],\
len(juncH['match']) ,len(set(juncH['seq'])), len(set(juncH['reg']))))
for m in juncH['match']:
outGsnapFile.write(m.rawText()+'\n')
#!/usr/bin/python
import sys, re, numpy
import mygsnap
if len(sys.argv) >= 4:
inFileName = sys.argv[1]
outFileName_matches = sys.argv[2]
outFileName_mmPos = sys.argv[3]
else:
inFileName = 'GH.txt'
outFileName_matches = 'GH_matches.dst'
outFileName_mmPos = 'GH_mmPos.txt'
result = mygsnap.gsnapFile(inFileName)
out_matches = open(outFileName_matches, 'w')
out_mmPos = open(outFileName_mmPos, 'w')
matches_count = {'unpaired': [], 'concordant': []}
matches_score = {'unpaired': [], 'concordant': []}
totalPairs = {'unpaired': 0, 'concordant': 0}
mmPos = {'unpaired': {0: None, 1: None}, 'concordant': {0: None, 1: None}}
for rL in result:
if rL[0].nLoci == 1 and rL[1].nLoci == 1 and not '(transloc)' in rL[
0].pairRel: # unique, no-within-read-splicing
mL = [rL[0].matchL()[0], rL[1].matchL()[0]]
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 fusion_proc_sort(inGsnapFileName, outGsnapFileName, outReportFileName,
sampN):
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))
transcriptL = []
for i in range(2):
rm = re.search('label_[12]:([^,\t]*)', match.segL[i][3])
if rm:
transcriptL.append(rm.group(1).replace('|', ','))
else:
transcriptL.append('')
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
if direction == 'sense':
key = ((trans_strand1, ) + bp1.groups()[1:],
(trans_strand2, ) + bp2.groups()[1:])
elif direction == 'antisense':
key = ((trans_strand2, ) + bp2.groups()[1:],
(trans_strand1, ) + bp1.groups()[1:])
transcriptL = transcriptL[::-1]
else:
raise Exception
if key in juncHH:
juncHH[key]['match'].append(r)
juncHH[key]['seq'].append(r.seq())
juncHH[key]['pos'].append((direction, offset))
else:
juncHH[key] = {
'match': [r],
'splice_type': splice_type,
'seq': [r.seq()],
'pos': [(direction, offset)],
'transcript': transcriptL
}
juncKH = juncHH.items()
juncKH.sort(lambda x, y: cmp(len(set(y[1]['pos'])), len(set(x[1]['pos']))))
outGsnapFile = open(outGsnapFileName, 'w')
outReportFile = open(outReportFileName, 'w')
for (key, juncH) in juncKH:
outReportFile.write('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n' % \
(juncH['splice_type'], sampN, key[0][0]+':'.join(key[0][1:]), key[1][0]+':'.join(key[1][1:]), \
juncH['transcript'][0], juncH['transcript'][1], \
len(juncH['match']), len(set(juncH['seq'])), len(set(juncH['pos']))))
for m in juncH['match']:
outGsnapFile.write(m.rawText() + '\n')
def make_samse(ifileN, ofileN):
headerL = make_header('/data1/Sequence/ucsc_hg19/hg19.chrom.sizes')
# ofile = open(ofileN, 'w')
for header in headerL:
print header
# ofile.write('%s\n' % header)
result = mygsnap.gsnapFile('/pipeline/test_ini_gsnap2sam/S022_single.gsnap',False)
#result = mygsnap.gsnapFile('/pipeline/test_ini_gsnap2sam/test.gsnap',False)
#result = mygsnap.gsnapFile('/pipeline/test_ini_gsnap2sam/S022_pair.gsnap',True)
## for unpaired
for r in result:
qname = r.rid()
flag = 0x0
rname = '*'
pos = 0
mapq = 0
cigar = ''
rnext = '*' ## assume --npath=1 (maximum 1 alignment per read)
pnext = 0 ## assume --npath=1 (maximum 1 alignment per read)
tlen = 0 ## assume --npath=1 (maximum 1 alignment per read)
seq = r.seq()
qual = r.qual()
extra = 'NH:i:1\tHI:i:1' ## assume --npath=1 (maximum 1 alignment per read)
if r.nLoci > 1:
flag = flag | 0x4
cigar = '*'
new_cigar = '*'
extra = ''
print ('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s' % (qname, flag, rname, pos, mapq, new_cigar, rnext, pnext, tlen, seq, qual, extra))
else:
if r.pairRel == '(transloc)':
match = r.matchL()[0]
segL = match.getSegInfo()
mapq = segL[0].mapq
seq = r.seq()
qual = r.qual()
for seg in segL:
flag = 0x0
(strand, rname, pos1, pos2) = re.search('([\+\-])(.*):([0-9]+)\.\.([0-9]+)', seg.seg[2]).groups()
pos = min(int(pos1), int(pos2))
(cigar,clip) = seg.toCIGAR_trans()
if clip < 0: ## first half
seq2 = seq[:clip]
qual2 = qual[:clip]
else: ## second half
seq2 = seq[clip:]
qual2 = qual[clip:]
if strand == '-':
flag = flag | 0x10
seq2 = mybasic.rc(seq2)
qual2 = mybasic.rev(qual2)
# print qname,seg.toCIGAR_trans()
print ('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s' % (qname, flag, rname, pos, mapq, cigar, rnext, pnext, tlen, seq2, qual2, extra))
else:
match = r.matchL()[0] ## assume --npath=1 (maximum 1 alignment per read)
segL = match.getSegInfo()
(strand, rname, pos1, pos2) = re.search('([\+\-])(.*):([0-9]+)\.\.([0-9]+)', segL[0].seg[2]).groups()
pos = min(int(pos1), int(pos2))
mapq = segL[0].mapq
seg_nm = segL[0].numSub
cigar2 = match.toCIGAR()
# print qname, match.toCIGAR()
if segL[0].start != '' and segL[0].start != '0':
cigar = str(segL[0].start) + 'S'
if strand == '-':
cigar = str(segL[0].numMatch + segL[0].numSub) + 'M' + cigar
if segL[0].ins != '' and segL[0].ins != '0':
cigar = str(segL[0].ins) + 'I' + cigar
else:
cigar = cigar + str(segL[0].numMatch + segL[0].numSub) + 'M'
if segL[0].ins != '' and segL[0].ins != '0':
cigar = cigar + str(segL[0].ins) + 'I'
if len(segL) == 1:
new_cigar = segL[0].toCIGAR(True)
else:
new_cigar = segL[0].toCIGAR()
prev_cigar = new_cigar
index = 0
for seg in segL[1:]:
index = index + 1
if index == (len(segL) - 1):
final = True
else:
final = False
rm = re.search('([\+\-])(.*):([0-9]+)\.\.([0-9]+)', seg.seg[2]).groups()
match = str(seg.numMatch + seg.numSub) + 'M'
if seg.ins != '' and seg.ins != '0':
ins = str(seg.ins) + 'I'
else:
ins = ''
if pos == 0 or pos > min(int(rm[2]), int(rm[3])):
pos = min(int(rm[2]), int(rm[3]))
if strand == '-':
dist = int(pos2) - int(rm[2]) - 1
if dist > 0:
cigar = match + ins + str(dist) + 'N' + cigar
else:
cigar = match + ins + cigar
else:
dist = int(rm[2]) - int(pos2) - 1
if dist > 0:
cigar = cigar + str(dist) + 'N' + match + ins
else:
cigar = cigar + match + ins
seg_nm = seg_nm + seg.numSub
pos1 = rm[2]
pos2 = rm[3]
cur_cigar = seg.toCIGAR(final)
if strand == '-':
if dist > 0 and 'D' not in prev_cigar and 'I' not in prev_cigar:
new_cigar = cur_cigar + str(dist) + 'N' + new_cigar
else:
new_cigar = cur_cigar + new_cigar
else:
if dist > 0 and 'D' not in prev_cigar and 'I' not in prev_cigar:
new_cigar = new_cigar + str(dist) + 'N' + cur_cigar
else:
new_cigar = new_cigar + cur_cigar
prev_cigar = cur_cigar
if segL[-1].end != '' and segL[-1].end != '0': ## last segment
if strand == '-':
cigar = str(segL[-1].end) + 'S' + cigar
else:
cigar = cigar + str(segL[-1].end) + 'S'
extra = extra + ('\tNM:i:%s' % seg_nm)
if strand == '-':
flag = flag | 0x10
seq = mybasic.rc(seq)
qual = mybasic.rev(qual)
## print ('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s' % (qname, flag, rname, pos, mapq, cigar, rnext, pnext, tlen, seq, qual, extra))
print ('%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s' % (qname, flag, rname, pos, mapq, cigar2, rnext, pnext, tlen, seq, qual, extra))
def exonSkip_filter(inFileName, outFileName):
'''
filters-in exon-skipping candidates in splice-mapped gsnap
'''
result = mygsnap.gsnapFile(inFileName, False)
if outFileName[-3:] == '.gz':
outFile = gzip.open(outFileName, 'wb')
else:
outFile = open(outFileName, 'w')
count_all = 0
count_include = 0
for r in result:
if r.nLoci != 1:
continue
match = r.matchL()[0]
if len(match.segL) != 2:
continue
segObjL = match.getSegInfo()
jncH = {}
skip = False
for segObj in segObjL:
if segObj.span - segObj.numMatch > 2 or segObj.percMatch < 90 or segObj.span < 5:
skip = True
break
if segObj.label == '':
break
for b in segObj.label.split('|'):
rm2 = re.match('(.*)\.exon([0-9]+)\/[0-9]+', b)
transId = rm2.group(1)
exonNum = int(rm2.group(2))
mybasic.addHash(jncH, transId, exonNum)
if skip:
continue
jncL = jncH.items()
if len(jncL) > 0 and max([len(j[1]) for j in jncL]) > 1:
minDist = 100
for i in range(len(jncL)):
if len(jncL[i][1]) == 2 and abs(jncL[i][1][0] -
jncL[i][1][1]) < minDist:
minDist = abs(jncL[i][1][0] - jncL[i][1][1])
if minDist == 1: # only difference
outFile.write(r.rawText() + '\n')
count_include += 1
count_all += 1
print 'Results:', count_include, count_all
import sys import mygsnap if len(sys.argv) >= 3: inFileName = sys.argv[1] outFileName = sys.argv[2] else: inFileName = '/Data2/RNASeq_SMC1_S02_result.txt' outFileName = 'GH_S02_matchfilter2.txt' matchCutOff = 90 result = mygsnap.gsnapFile(inFileName) outFile = open(outFileName, 'w') for rL in result: if not (rL[0].nLoci==1 and rL[1].nLoci==1 and rL[0].pairRel=='unpaired') or '(transloc)' in rL[0].pairRel: continue if not (len(rL[0].matchL()[0].mergedLocusL())==1 and len(rL[1].matchL()[0].mergedLocusL())==1): continue if not (rL[0].matchL()[0].numMatch()>=matchCutOff and rL[1].matchL()[0].numMatch()>=matchCutOff): continue for i in (0,1):
