def mutscan_signature_rsq():
for dir in mysetting.rsqMutscanDirL:
fileL = filter(lambda x: 'bak' not in x, map(lambda x: x.rstrip(), os.popen('find %s -name *dbsnp_flt' % dir).readlines()))
for file in fileL:
if 'splice_Z' in file:
sid = re.match('(.*)_splice[2]*.mutscan.dbsnp_flt', os.path.basename(file)).group(1)
else:
sid = re.match('(.*)_RSq_splice.mutscan.dbsnp_flt', os.path.basename(file)).group(1)
cntH = {}
total = 0
inFile = open(file, 'r')
for line in inFile:
colL = line.rstrip().split('\t')
chrom = colL[0]
ref = colL[2]
alt = colL[3]
if ref == 'N' or len(alt)>1:
continue
n_ref = int(colL[4])
n_alt = int(colL[5])
if n_alt >= MIN_MUT_N and (n_alt+n_ref) >= MIN_COV:
if ref not in ['C','T']:
ref = mybasic.rc(ref)
alt = mybasic.rc(alt)
if (ref,alt) in cntH:
cntH[(ref,alt)] += 1
else:
cntH[(ref,alt)] = 1
total += 1
##for line
for (r,a) in cntH:
sys.stdout.write('%s\t%s>%s\t%s\t%s\n' % (sid, r,a, cntH[(r,a)], total))
def mutation_signaturei_t(inDir, outName=''):
if outName == '':
outFile = sys.stdout
else:
outFile = open(outName, 'w')
outFile.write("samp_id\tmutation\tcontext\tfreq\tn_mut\tn_total\n")
mutFileNL = map(
lambda x: x.rstrip(),
os.popen('ls %s/*.mutect | grep -v union_pos' % inDir).readlines())
for mutFileN in mutFileNL:
sampN = mutFileN.split('/')[-1].split('.')[0].split('_')[0]
print sampN, mutFileN
mutFile = open(mutFileN, 'r')
mutFile.readline()
headerL = mutFile.readline().rstrip().split('\t')
idxH = {}
sigH = {}
cntH = {}
for i in range(len(headerL)):
idxH[headerL[i]] = i
total = 0
for line in mutFile:
colL = line.rstrip().split('\t')
chr = colL[idxH['contig']]
pos = colL[idxH['position']]
context = colL[idxH['context']]
ref = colL[idxH['ref_allele']]
alt = colL[idxH['alt_allele']]
status = colL[idxH['judgement']]
if status == 'REJECT' or chr == 'chrMT' or chr == 'chrM':
continue
total += 1
tri = context[2] + ref + context[4]
if ref == 'C' or ref == 'T':
nt_ch = ref + '>' + alt
else:
nt_ch = rc(ref) + '>' + rc(alt)
tri = rc(tri)
if (nt_ch, tri) in sigH:
sigH[(nt_ch, tri)] += 1
else:
sigH[(nt_ch, tri)] = 1
if (nt_ch) in cntH:
cntH[(nt_ch)] += 1
else:
cntH[(nt_ch)] = 1
mutFile.close()
for key in sigH:
(type, tri) = key
freq = sigH[key]
outFile.write('%s\t%s\t%s\t%s\t%s\t%s\n' %
(sampN, type, tri, freq, cntH[type], total))
outFile.flush()
outFile.close()
def mutation_signaturei_t(inDir, outName=''):
if outName == '':
outFile = sys.stdout
else:
outFile = open(outName, 'w')
outFile.write("samp_id\tmutation\tcontext\tfreq\tn_mut\tn_total\n")
mutFileNL = map(lambda x: x.rstrip(), os.popen('ls %s/*.mutect | grep -v union_pos' % inDir).readlines())
for mutFileN in mutFileNL:
sampN = mutFileN.split('/')[-1].split('.')[0].split('_')[0]
print sampN, mutFileN
mutFile = open(mutFileN, 'r')
mutFile.readline()
headerL = mutFile.readline().rstrip().split('\t')
idxH = {}
sigH = {}
cntH = {}
for i in range(len(headerL)):
idxH[headerL[i]] = i
total = 0
for line in mutFile:
colL = line.rstrip().split('\t')
chr = colL[idxH['contig']]
pos = colL[idxH['position']]
context = colL[idxH['context']]
ref = colL[idxH['ref_allele']]
alt = colL[idxH['alt_allele']]
status = colL[idxH['judgement']]
if status == 'REJECT' or chr == 'chrMT' or chr == 'chrM':
continue
total += 1
tri = context[2] + ref + context[4]
if ref == 'C' or ref == 'T':
nt_ch = ref + '>' + alt
else:
nt_ch = rc(ref) + '>' + rc(alt)
tri = rc(tri)
if (nt_ch,tri) in sigH:
sigH[(nt_ch,tri)] += 1
else:
sigH[(nt_ch,tri)] = 1
if (nt_ch) in cntH:
cntH[(nt_ch)] += 1
else:
cntH[(nt_ch)] = 1
mutFile.close()
for key in sigH:
(type, tri) = key
freq = sigH[key]
outFile.write('%s\t%s\t%s\t%s\t%s\t%s\n' % (sampN, type, tri, freq, cntH[type], total))
outFile.flush()
outFile.close()
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$\n' % (rL[0].seq(),mybasic.rc(rL[1].seq())),
for loc in locL:
outFile.write('%s\t%s\t%s\n' % (loc.chrom,loc.chrSta,loc.chrEnd))
def process_bp(inFileName,outFileName,coordH,regionL):
result = mygsnap.gsnapFile(inFileName,True)
outFile = open(outFileName, 'w')
outFile.write('browser full knownGene\n')
outFile.write('track name="%s" visibility=2\n' % inFileName)
for rL in result:
if not (rL[0].nLoci==1 and rL[1].nLoci==1 and rL[0].pairRel=='unpaired'):
raise Exception
locL = [mygenome.locus(rL[0].matchL()[0].segL[0][2]), mygenome.locus(rL[1].matchL()[0].segL[0][2])]
for loc in locL:
loc.chrSta += coordH[loc.chrom][1] -1
loc.chrEnd += coordH[loc.chrom][1] -1
loc.chrom = coordH[loc.chrom][0]
flag = False
for loc in locL:
for region in regionL:
if loc.overlap(region) > 0:
flag = True
if flag:
print '^%s.*%s$\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_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 mutect_weblogo_sub(sampN, inFileN, outFileN, pdfFileN):
inFile = open(inFileN, 'r')
inFile.readline() #comment line
headerL = inFile.readline().rstrip().split('\t')
idxH = {}
for i in range(len(headerL)):
idxH[headerL[i]] = i
outFile = open(outFileN,'w')
for line in inFile:
colL = line.rstrip().split('\t')
context = colL[idxH['context']]
ref = colL[idxH['ref_allele']]
alt = colL[idxH['alt_allele']]
status = colL[idxH['judgement']]
if status == 'REJECT':
continue
head = context[:3]
tail = context[-3:]
context = head + ref + tail
if ref not in ['C','T']:
context = mybasic.rc(context)
ref = mybasic.rc(ref)
alt = mybasic.rc(alt)
if ref == 'C' and alt == 'T':## TMZ context only
outFile.write('%s\n' % context)
outFile.flush()
outFile.close()
fin = open(outFileN,'r')
seqs = weblogolib.read_seq_data(fin)
data = weblogolib.LogoData.from_seqs(seqs)
options = weblogolib.LogoOptions()
options.show_fineprint = False
options.first_index = -3
options.logo_title = sampN
format = weblogolib.LogoFormat(data, options)
fout = open(pdfFileN, 'w')
weblogolib.pdf_formatter(data, format, fout)
def mutect_weblogo_sub(sampN, inFileN, outFileN, pdfFileN):
inFile = open(inFileN, 'r')
inFile.readline() #comment line
headerL = inFile.readline().rstrip().split('\t')
idxH = {}
for i in range(len(headerL)):
idxH[headerL[i]] = i
outFile = open(outFileN, 'w')
for line in inFile:
colL = line.rstrip().split('\t')
context = colL[idxH['context']]
ref = colL[idxH['ref_allele']]
alt = colL[idxH['alt_allele']]
status = colL[idxH['judgement']]
if status == 'REJECT':
continue
head = context[:3]
tail = context[-3:]
context = head + ref + tail
if ref not in ['C', 'T']:
context = mybasic.rc(context)
ref = mybasic.rc(ref)
alt = mybasic.rc(alt)
if ref == 'C' and alt == 'T': ## TMZ context only
outFile.write('%s\n' % context)
outFile.flush()
outFile.close()
fin = open(outFileN, 'r')
seqs = weblogolib.read_seq_data(fin)
data = weblogolib.LogoData.from_seqs(seqs)
options = weblogolib.LogoOptions()
options.show_fineprint = False
options.first_index = -3
options.logo_title = sampN
format = weblogolib.LogoFormat(data, options)
fout = open(pdfFileN, 'w')
weblogolib.pdf_formatter(data, format, fout)
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 mutscan_signature_rsq():
for dir in mysetting.rsqMutscanDirL:
fileL = filter(
lambda x: 'bak' not in x,
map(lambda x: x.rstrip(),
os.popen('find %s -name *dbsnp_flt' % dir).readlines()))
for file in fileL:
if 'splice_Z' in file:
sid = re.match('(.*)_splice[2]*.mutscan.dbsnp_flt',
os.path.basename(file)).group(1)
else:
sid = re.match('(.*)_RSq_splice.mutscan.dbsnp_flt',
os.path.basename(file)).group(1)
cntH = {}
total = 0
inFile = open(file, 'r')
for line in inFile:
colL = line.rstrip().split('\t')
chrom = colL[0]
ref = colL[2]
alt = colL[3]
if ref == 'N' or len(alt) > 1:
continue
n_ref = int(colL[4])
n_alt = int(colL[5])
if n_alt >= MIN_MUT_N and (n_alt + n_ref) >= MIN_COV:
if ref not in ['C', 'T']:
ref = mybasic.rc(ref)
alt = mybasic.rc(alt)
if (ref, alt) in cntH:
cntH[(ref, alt)] += 1
else:
cntH[(ref, alt)] = 1
total += 1
##for line
for (r, a) in cntH:
sys.stdout.write('%s\t%s>%s\t%s\t%s\n' %
(sid, r, a, cntH[(r, a)], total))
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 trim4x(inFqFileName, outFqFilePrefix, trimLen):
if inFqFileName == 'stdin':
inFqFile = sys.stdin
else:
inFqFile = open(inFqFileName)
outFqFile1 = open('%s.1.fastq' % outFqFilePrefix, 'w')
outFqFile2 = open('%s.2.fastq' % outFqFilePrefix, 'w')
while 1:
line = inFqFile.readline()
if not line:
break
seq = inFqFile.readline()[:-1]
if line[0] != '@':
raise Exception
seqN = line[1:].rstrip().split(' ')[0]
inFqFile.readline()
qual = inFqFile.readline()[:-1]
if 'N' in seq[:trimLen] or 'N' in seq[-trimLen:]:
continue
outFqFile1.write('@%s/1\n%s\n+\n%s\n' %
(seqN, seq[:trimLen], qual[:trimLen]))
outFqFile2.write(
'@%s/2\n%s\n+\n%s\n' %
(seqN, mybasic.rc(seq[-trimLen:]), mybasic.rev(qual[-trimLen:])))
outFqFile1.close()
outFqFile2.close()
def trim4x(inFqFileName, outFqFilePrefix, trimLen):
if inFqFileName == "stdin":
inFqFile = sys.stdin
else:
inFqFile = open(inFqFileName)
outFqFile1 = open("%s.1.fastq" % outFqFilePrefix, "w")
outFqFile2 = open("%s.2.fastq" % outFqFilePrefix, "w")
while 1:
line = inFqFile.readline()
if not line:
break
seq = inFqFile.readline()[:-1]
if line[0] != "@":
raise Exception
seqN = line[1:].rstrip().split(" ")[0]
inFqFile.readline()
qual = inFqFile.readline()[:-1]
if "N" in seq[:trimLen] or "N" in seq[-trimLen:]:
continue
outFqFile1.write("@%s/1\n%s\n+\n%s\n" % (seqN, seq[:trimLen], qual[:trimLen]))
outFqFile2.write("@%s/2\n%s\n+\n%s\n" % (seqN, mybasic.rc(seq[-trimLen:]), mybasic.rev(qual[-trimLen:])))
outFqFile1.close()
outFqFile2.close()
#!/usr/bin/python
import sys
import mybasic
motifL = ['TAAT', 'TAATT', 'TAATTG']
motifL_rc = [mybasic.rc(m, 'DNA') for m in motifL]
bed = open('/data1/IRCR/PKS/promoter_hg19.bed')
fa = open('/data1/IRCR/PKS/promoter_hg19.fa')
geneNameL = [x.split('\t')[3] for x in bed]
sys.stdout.write('geneName')
for i in range(len(motifL)):
sys.stdout.write('\tm%sf\tm%sr\tm%st' % (i + 1, i + 1, i + 1))
sys.stdout.write('\n')
idx = 0
while True:
h = fa.readline()[:-1]
s = fa.readline()[:-1].upper()
sys.stdout.write('%s' % (geneNameL[idx]))
countL = (s.count(motifL[i]), s.count(motifL_rc[i]))
def mutscan_signature(mode='WXS', outFileN=''):
if outFileN == '':
outFile = sys.stdout
else:
outFile = open(outFileN, 'w')
dirLH = {'WXS': mysetting.wxsMutscanDirL, 'RSQ': mysetting.rsqMutscanDirL}
contextH = {}
for dir in dirLH[mode]:
fileL = filter(lambda x: 'bak' not in x, map(lambda x: x.rstrip(), os.popen('find %s -name *dbsnp_flt' % dir).readlines()))
for file in fileL:
if mode=='RSQ':
if os.path.basename(file) == 'S647_splice.mutscan.dbsnp_flt': ## duplicated files
continue
if 'splice_Z' in file:
sid = re.match('(.*)_splice2.mutscan.dbsnp_flt', os.path.basename(file)).group(1)
else:
sid = re.match('(.*)_RSq_splice.mutscan.dbsnp_flt', os.path.basename(file)).group(1)
else:
sid = re.match('(.*).mutscan.dbsnp_flt', os.path.basename(file)).group(1)
sigH = {}
cntH = {}
total = 0
inFile = open(file, 'r')
for line in inFile:
colL = line.rstrip().split('\t')
chrom = colL[0]
pos = int(colL[1])
ref = colL[2]
alt = colL[3]
if ref == 'N' or len(alt) > 1:
continue
n_ref = int(colL[4])
n_alt = int(colL[5])
if n_alt >= MIN_MUT_N and (n_alt+n_ref) >= MIN_COV:
start = pos - 1
end = pos + 1
if chrom in contextH and pos in contextH[chrom]:
context = contextH[chrom][pos]
else:
resL = os.popen('samtools faidx /data1/Sequence/ucsc_hg19/hg19.fasta %s:%s-%s' % (chrom,start,end)).readlines()
context = resL[1].rstrip().upper()
if chrom not in contextH:
contextH[chrom] = {}
contextH[chrom][pos] = context
if ref not in ['C','T']:
ref = mybasic.rc(ref)
alt = mybasic.rc(alt)
context = mybasic.rc(context)
ch = ref + '>' + alt
if ch in cntH:
cntH[ch] += 1
else:
cntH[ch] = 1
if (ch,context) in sigH:
sigH[(ch,context)] += 1
else:
sigH[(ch,context)] = 1
total += 1
#if
##for line
for (type,context) in sigH:
outFile.write('%s\t%s\t%s\t%s\t%s\t%s\n' % (sid, type, context, sigH[(type,context)], cntH[type], total))
##for file
#for dir
outFile.flush()
outFile.close()
def mutation_signature_ttt(inDirN, outDirN):
sampN = os.path.basename(inDirN)
outName = '%s/%s.mutation_signature.txt' % (outDirN, sampN)
# outFile = open(outName, 'w')
#
# outFile.write("samp_id\tmutation\tcontext\tfreq\tn_mut\tn_total\n")
mutFileNL = map(lambda x: x.rstrip(), os.popen('ls %s/*mutect' % inDirN).readlines())
if mutFileNL == []:
mutFileNL = map(lambda x: x.rstrip(), os.popen('ls %s/*rerun' % inDirN).readlines())
pdfFileN = '%s/%s.mutation_signature.pdf' % (outDirN, sampN)
os.system('Rscript %s/NGS/mutation/mutect_mutation_signature_plot.R %s %s' % (mysetting.SRC_HOME, outName, pdfFileN))
return()
for mutFileN in mutFileNL:
(id, postfix) = re.search('(.*)_([A-Z0-9]{1,})_[TKNCS]{2}', sampN).groups()
if postfix != 'T':
sid = '%s_%s' % (id, postfix)
else:
sid = id
mutFile = open(mutFileN, 'r')
mutFile.readline()
headerL = mutFile.readline().rstrip().split('\t')
idxH = {}
sigH = {}
cntH = {}
for i in range(len(headerL)):
idxH[headerL[i]] = i
total = 0
for line in mutFile:
colL = line.rstrip().split('\t')
chr = colL[idxH['contig']]
pos = colL[idxH['position']]
context = colL[idxH['context']]
ref = colL[idxH['ref_allele']]
alt = colL[idxH['alt_allele']]
status = colL[idxH['judgement']]
if status == 'REJECT' or chr == 'chrMT' or chr == 'chrM':
continue
total += 1
tri = context[2] + ref + context[4]
if ref == 'C' or ref == 'T':
nt_ch = ref + '>' + alt
else:
nt_ch = rc(ref) + '>' + rc(alt)
tri = rc(tri)
if (nt_ch,tri) in sigH:
sigH[(nt_ch,tri)] += 1
else:
sigH[(nt_ch,tri)] = 1
if (nt_ch) in cntH:
cntH[(nt_ch)] += 1
else:
cntH[(nt_ch)] = 1
#for line
mutFile.close()
for key in sigH:
(type, tri) = key
freq = sigH[key]
outFile.write('%s\t%s\t%s\t%s\t%s\t%s\n' % (sid, type, tri, freq, cntH[type], total))
#for mutFile
outFile.flush()
outFile.close()
def mutation_signature(inDirN, outDirN, outName=''):
sampN = os.path.basename(inDirN)
if outName == '':
outName = '%s/%s.mutation_signature.txt' % (outDirN, sampN)
outFile = open(outName, 'w')
outFile.write('samp_id\tmutation\tcontext\tfreq\tn_mut\tn_total\n') # header
mutFileNL = map(lambda x: x.rstrip(), os.popen('ls %s/*mutect*filter.vcf' % inDirN).readlines())
if mutFileNL != []:
mutH = {}
for mutFileN in mutFileNL:
(id, postfix) = re.search('(.*)_([A-Z0-9]{1,})_[TKNCS]{2}', sampN).groups()
if postfix != 'T':
sid = '%s_%s' % (id, postfix)
else:
sid = id
mutFile = open(mutFileN, 'r')
for line in mutFile:
if line[0] == '#':
continue
colL = line.rstrip().split('\t')
chr = colL[0]
pos = colL[1]
ref = colL[3]
alt = colL[4]
mutH[(chr,pos,ref,alt)] = 1
#for line
#for mutFileN
#if there's mutation vcf
mutFileNL = map(lambda x: x.rstrip(), os.popen('ls %s/*mutect' % inDirN).readlines())
if mutFileNL == []:
mutFileNL = map(lambda x: x.rstrip(), os.popen('ls %s/*mutect_rerun' % inDirN).readlines())
if mutFileNL != []:
sigH = {}
cntH = {}
total = 0
for mutFileN in mutFileNL:
mutFile = open(mutFileN, 'r')
mutFile.readline()
headerL = mutFile.readline().rstrip().split('\t')
idxH = {}
for i in range(len(headerL)):
idxH[headerL[i]] = i
for line in mutFile:
colL = line.rstrip().split('\t')
chr = colL[idxH['contig']]
pos = colL[idxH['position']]
context = colL[idxH['context']]
ref = colL[idxH['ref_allele']]
alt = colL[idxH['alt_allele']]
if (chr,pos,ref,alt) in mutH:
total += 1
tri = context[2] + ref + context[4]
if ref == 'C' or ref == 'T':
nt_ch = ref + '>' + alt
else:
nt_ch = rc(ref) + '>' + rc(alt)
tri = rc(tri)
if (nt_ch, tri) in sigH:
sigH[(nt_ch, tri)] += 1
else:
sigH[(nt_ch, tri)] = 1
if nt_ch in cntH:
cntH[nt_ch] += 1
else:
cntH[nt_ch] = 1
# if not filtered out
#for line
#for mutFileN
for key in sigH:
(type, tri) = key
freq = sigH[key]
outFile.write('%s\t%s\t%s\t%s\t%s\t%s\n' % (sid, type, tri, freq, cntH[type], total))
def main(inFileName,geneList=[]):
dataH = {}
# nameL = ('Mutation GRCh37 genome position', 'Mutation GRCh37 strand','Gene name','ID_sample','ID_tumour','Primary site', \
# 'Site subtype','Primary histology','Histology subtype','Genome-wide screen','Mutation ID','Mutation CDS','Mutation AA', \
# 'Mutation Description','Mutation zygosity','Mutation somatic status','Pubmed_PMID','Sample source','Tumor origin','Comments')
nameL = ('Gene name','Mutation CDS','Mutation AA','Mutation Description','Mutation GRCh37 genome position','Mutation GRCh37 strand','Mutation somatic status')
inFile = open(inFileName)
headerL = inFile.readline()[:-1].split('\t')
idxH = dict([(x, headerL.index(x)) for x in nameL])
for line in inFile:
valueL = line[:-1].split('\t')
geneN = valueL[idxH['Gene name']]
if '_ENST' in geneN:
geneN = geneN.split('_ENST')[0]
if len(geneList)>0 and geneN not in geneList:
continue
coord = valueL[idxH['Mutation GRCh37 genome position']]
if not coord:
continue
somatic = valueL[idxH['Mutation somatic status']]
if not 'somatic' in somatic:
continue
(chrNum,chrSta,chrEnd) = re.search('([^:-]+):([^:-]+)-([^:-]+)', coord).groups()
cds = valueL[idxH['Mutation CDS']]
aa = valueL[idxH['Mutation AA']]
desc = valueL[idxH['Mutation Description']]
strand = valueL[idxH['Mutation GRCh37 strand']]
rm = re.match('c\.[\+\-_0-9]+([atgcATGC]*)(>|ins|del)([atgcATGC]*)',cds)
if rm:
(ref,vtype,alt) = rm.groups()
else:
ref,alt = '',''
if strand == '-':
ref = mybasic.rc(ref)
alt = mybasic.rc(alt)
chr = chrNum
if chr == '23':
chr = 'X'
chrNum = 'X'
elif chr == '24':
chr = 'Y'
chrNum = 'Y'
elif chr == '25':
chr = 'M'
chrNum = 'M'
# if vtype == 'del':
# rm = re.search('([ACGT]+)', alt.upper())
# ## if deleted bases are specified
# if alt != '' and rm:
# ## check if deleted bases are the same as reference sequences at the location
# new_ref = os.popen('samtools faidx /data1/Sequence/ucsc_hg19/hg19.fa chr%s:%s-%s' % (chr,chrSta,chrEnd)).readlines()[1:]
# new_ref = "".join(map(lambda x: x.rstrip().upper(), new_ref))
# if new_ref == alt.upper():
# chrSta = str(int(chrSta) - 1)
# ref = os.popen('samtools faidx /data1/Sequence/ucsc_hg19/hg19.fa chr%s:%s-%s' % (chr,chrSta,chrEnd)).readlines()[1:]
# ref = "".join(map(lambda x: x.rstrip().upper(), ref))
# alt = ref[0]
key = (chrNum,chrSta,chrEnd,strand,ref,alt)
if key in dataH:
mybasic.pushHash(dataH[key],'geneN',geneN)
mybasic.pushHash(dataH[key],'cds',cds)
mybasic.pushHash(dataH[key],'aa',aa)
mybasic.pushHash(dataH[key],'desc',desc)
else:
dataH[key] = {'geneN':set([geneN]), 'cds':set([cds]), 'aa':set([aa]), 'desc':set([desc])}
for ((chrNum,chrSta,chrEnd,strand,ref,alt),infoH) in dataH.iteritems():
sys.stdout.write('chr%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n' % (chrNum,chrSta,chrEnd,strand, ref,alt,\
','.join(filter(lambda x: not x.startswith('ENSG'), list(infoH['geneN']))), ','.join(infoH['cds']), ','.join(infoH['aa']), ','.join(infoH['desc'])))
def mutation_signature_ttt(inDirN, outDirN):
sampN = os.path.basename(inDirN)
outName = '%s/%s.mutation_signature.txt' % (outDirN, sampN)
# outFile = open(outName, 'w')
#
# outFile.write("samp_id\tmutation\tcontext\tfreq\tn_mut\tn_total\n")
mutFileNL = map(lambda x: x.rstrip(),
os.popen('ls %s/*mutect' % inDirN).readlines())
if mutFileNL == []:
mutFileNL = map(lambda x: x.rstrip(),
os.popen('ls %s/*rerun' % inDirN).readlines())
pdfFileN = '%s/%s.mutation_signature.pdf' % (outDirN, sampN)
os.system(
'Rscript %s/NGS/mutation/mutect_mutation_signature_plot.R %s %s' %
(mysetting.SRC_HOME, outName, pdfFileN))
return ()
for mutFileN in mutFileNL:
(id, postfix) = re.search('(.*)_([A-Z0-9]{1,})_[TKNCS]{2}',
sampN).groups()
if postfix != 'T':
sid = '%s_%s' % (id, postfix)
else:
sid = id
mutFile = open(mutFileN, 'r')
mutFile.readline()
headerL = mutFile.readline().rstrip().split('\t')
idxH = {}
sigH = {}
cntH = {}
for i in range(len(headerL)):
idxH[headerL[i]] = i
total = 0
for line in mutFile:
colL = line.rstrip().split('\t')
chr = colL[idxH['contig']]
pos = colL[idxH['position']]
context = colL[idxH['context']]
ref = colL[idxH['ref_allele']]
alt = colL[idxH['alt_allele']]
status = colL[idxH['judgement']]
if status == 'REJECT' or chr == 'chrMT' or chr == 'chrM':
continue
total += 1
tri = context[2] + ref + context[4]
if ref == 'C' or ref == 'T':
nt_ch = ref + '>' + alt
else:
nt_ch = rc(ref) + '>' + rc(alt)
tri = rc(tri)
if (nt_ch, tri) in sigH:
sigH[(nt_ch, tri)] += 1
else:
sigH[(nt_ch, tri)] = 1
if (nt_ch) in cntH:
cntH[(nt_ch)] += 1
else:
cntH[(nt_ch)] = 1
#for line
mutFile.close()
for key in sigH:
(type, tri) = key
freq = sigH[key]
outFile.write('%s\t%s\t%s\t%s\t%s\t%s\n' %
(sid, type, tri, freq, cntH[type], total))
#for mutFile
outFile.flush()
outFile.close()
def mutation_signature(inDirN, outDirN, outName=''):
sampN = os.path.basename(inDirN)
if outName == '':
outName = '%s/%s.mutation_signature.txt' % (outDirN, sampN)
outFile = open(outName, 'w')
outFile.write(
'samp_id\tmutation\tcontext\tfreq\tn_mut\tn_total\n') # header
mutFileNL = map(lambda x: x.rstrip(),
os.popen('ls %s/*mutect*filter.vcf' % inDirN).readlines())
if mutFileNL != []:
mutH = {}
for mutFileN in mutFileNL:
(id, postfix) = re.search('(.*)_([A-Z0-9]{1,})_[TKNCS]{2}',
sampN).groups()
if postfix != 'T':
sid = '%s_%s' % (id, postfix)
else:
sid = id
mutFile = open(mutFileN, 'r')
for line in mutFile:
if line[0] == '#':
continue
colL = line.rstrip().split('\t')
chr = colL[0]
pos = colL[1]
ref = colL[3]
alt = colL[4]
mutH[(chr, pos, ref, alt)] = 1
#for line
#for mutFileN
#if there's mutation vcf
mutFileNL = map(lambda x: x.rstrip(),
os.popen('ls %s/*mutect' % inDirN).readlines())
if mutFileNL == []:
mutFileNL = map(lambda x: x.rstrip(),
os.popen('ls %s/*mutect_rerun' % inDirN).readlines())
if mutFileNL != []:
sigH = {}
cntH = {}
total = 0
for mutFileN in mutFileNL:
mutFile = open(mutFileN, 'r')
mutFile.readline()
headerL = mutFile.readline().rstrip().split('\t')
idxH = {}
for i in range(len(headerL)):
idxH[headerL[i]] = i
for line in mutFile:
colL = line.rstrip().split('\t')
chr = colL[idxH['contig']]
pos = colL[idxH['position']]
context = colL[idxH['context']]
ref = colL[idxH['ref_allele']]
alt = colL[idxH['alt_allele']]
if (chr, pos, ref, alt) in mutH:
total += 1
tri = context[2] + ref + context[4]
if ref == 'C' or ref == 'T':
nt_ch = ref + '>' + alt
else:
nt_ch = rc(ref) + '>' + rc(alt)
tri = rc(tri)
if (nt_ch, tri) in sigH:
sigH[(nt_ch, tri)] += 1
else:
sigH[(nt_ch, tri)] = 1
if nt_ch in cntH:
cntH[nt_ch] += 1
else:
cntH[nt_ch] = 1
# if not filtered out
#for line
#for mutFileN
for key in sigH:
(type, tri) = key
freq = sigH[key]
outFile.write('%s\t%s\t%s\t%s\t%s\t%s\n' %
(sid, type, tri, freq, cntH[type], total))
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 main(inFileName, geneList=[]):
dataH = {}
# nameL = ('Mutation GRCh37 genome position', 'Mutation GRCh37 strand','Gene name','ID_sample','ID_tumour','Primary site', \
# 'Site subtype','Primary histology','Histology subtype','Genome-wide screen','Mutation ID','Mutation CDS','Mutation AA', \
# 'Mutation Description','Mutation zygosity','Mutation somatic status','Pubmed_PMID','Sample source','Tumor origin','Comments')
nameL = ('Gene name', 'Mutation CDS', 'Mutation AA',
'Mutation Description', 'Mutation GRCh37 genome position',
'Mutation GRCh37 strand', 'Mutation somatic status')
inFile = open(inFileName)
headerL = inFile.readline()[:-1].split('\t')
idxH = dict([(x, headerL.index(x)) for x in nameL])
for line in inFile:
valueL = line[:-1].split('\t')
geneN = valueL[idxH['Gene name']]
if '_ENST' in geneN:
geneN = geneN.split('_ENST')[0]
if len(geneList) > 0 and geneN not in geneList:
continue
coord = valueL[idxH['Mutation GRCh37 genome position']]
if not coord:
continue
somatic = valueL[idxH['Mutation somatic status']]
if not 'somatic' in somatic:
continue
(chrNum, chrSta, chrEnd) = re.search('([^:-]+):([^:-]+)-([^:-]+)',
coord).groups()
cds = valueL[idxH['Mutation CDS']]
aa = valueL[idxH['Mutation AA']]
desc = valueL[idxH['Mutation Description']]
strand = valueL[idxH['Mutation GRCh37 strand']]
rm = re.match('c\.[\+\-_0-9]+([atgcATGC]*)(>|ins|del)([atgcATGC]*)',
cds)
if rm:
(ref, vtype, alt) = rm.groups()
else:
ref, alt = '', ''
if strand == '-':
ref = mybasic.rc(ref)
alt = mybasic.rc(alt)
chr = chrNum
if chr == '23':
chr = 'X'
chrNum = 'X'
elif chr == '24':
chr = 'Y'
chrNum = 'Y'
elif chr == '25':
chr = 'M'
chrNum = 'M'
# if vtype == 'del':
# rm = re.search('([ACGT]+)', alt.upper())
# ## if deleted bases are specified
# if alt != '' and rm:
# ## check if deleted bases are the same as reference sequences at the location
# new_ref = os.popen('samtools faidx /data1/Sequence/ucsc_hg19/hg19.fa chr%s:%s-%s' % (chr,chrSta,chrEnd)).readlines()[1:]
# new_ref = "".join(map(lambda x: x.rstrip().upper(), new_ref))
# if new_ref == alt.upper():
# chrSta = str(int(chrSta) - 1)
# ref = os.popen('samtools faidx /data1/Sequence/ucsc_hg19/hg19.fa chr%s:%s-%s' % (chr,chrSta,chrEnd)).readlines()[1:]
# ref = "".join(map(lambda x: x.rstrip().upper(), ref))
# alt = ref[0]
key = (chrNum, chrSta, chrEnd, strand, ref, alt)
if key in dataH:
mybasic.pushHash(dataH[key], 'geneN', geneN)
mybasic.pushHash(dataH[key], 'cds', cds)
mybasic.pushHash(dataH[key], 'aa', aa)
mybasic.pushHash(dataH[key], 'desc', desc)
else:
dataH[key] = {
'geneN': set([geneN]),
'cds': set([cds]),
'aa': set([aa]),
'desc': set([desc])
}
for ((chrNum, chrSta, chrEnd, strand, ref, alt),
infoH) in dataH.iteritems():
sys.stdout.write('chr%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n' % (chrNum,chrSta,chrEnd,strand, ref,alt,\
','.join(filter(lambda x: not x.startswith('ENSG'), list(infoH['geneN']))), ','.join(infoH['cds']), ','.join(infoH['aa']), ','.join(infoH['desc'])))
#!/usr/bin/python
import sys
import mybasic
motifL = ["TAAT", "TAATT", "TAATTG"]
motifL_rc = [mybasic.rc(m, "DNA") for m in motifL]
bed = open("/data1/IRCR/PKS/promoter_hg19.bed")
fa = open("/data1/IRCR/PKS/promoter_hg19.fa")
geneNameL = [x.split("\t")[3] for x in bed]
sys.stdout.write("geneName")
for i in range(len(motifL)):
sys.stdout.write("\tm%sf\tm%sr\tm%st" % (i + 1, i + 1, i + 1))
sys.stdout.write("\n")
idx = 0
while True:
h = fa.readline()[:-1]
s = fa.readline()[:-1].upper()
sys.stdout.write("%s" % (geneNameL[idx]))
countL = (s.count(motifL[i]), s.count(motifL_rc[i]))
def mutation_signature(inDirN, outDirN, outName=""):
sampN = os.path.basename(inDirN)
if outName == "":
outName = "%s/%s.mutation_signature.txt" % (outDirN, sampN)
outFile = open(outName, "w")
outFile.write("samp_id\tmutation\tcontext\tfreq\tn_mut\tn_total\n") # header
mutFileNL = map(lambda x: x.rstrip(), os.popen("ls %s/*mutect*filter.vcf" % inDirN).readlines())
if mutFileNL != []:
mutH = {}
for mutFileN in mutFileNL:
(id, postfix) = re.search("(.*)_([A-Z0-9]{1,})_[TKNCS]{2}", sampN).groups()
if postfix != "T":
sid = "%s_%s" % (id, postfix)
else:
sid = id
mutFile = open(mutFileN, "r")
for line in mutFile:
if line[0] == "#":
continue
colL = line.rstrip().split("\t")
chr = colL[0]
pos = colL[1]
ref = colL[3]
alt = colL[4]
mutH[(chr, pos, ref, alt)] = 1
# for line
# for mutFileN
# if there's mutation vcf
mutFileNL = map(lambda x: x.rstrip(), os.popen("ls %s/*mutect" % inDirN).readlines())
if mutFileNL == []:
mutFileNL = map(lambda x: x.rstrip(), os.popen("ls %s/*mutect_rerun" % inDirN).readlines())
if mutFileNL != []:
sigH = {}
cntH = {}
total = 0
for mutFileN in mutFileNL:
mutFile = open(mutFileN, "r")
mutFile.readline()
headerL = mutFile.readline().rstrip().split("\t")
idxH = {}
for i in range(len(headerL)):
idxH[headerL[i]] = i
for line in mutFile:
colL = line.rstrip().split("\t")
chr = colL[idxH["contig"]]
pos = colL[idxH["position"]]
context = colL[idxH["context"]]
ref = colL[idxH["ref_allele"]]
alt = colL[idxH["alt_allele"]]
if (chr, pos, ref, alt) in mutH:
total += 1
tri = context[2] + ref + context[4]
if ref == "C" or ref == "T":
nt_ch = ref + ">" + alt
else:
nt_ch = rc(ref) + ">" + rc(alt)
tri = rc(tri)
if (nt_ch, tri) in sigH:
sigH[(nt_ch, tri)] += 1
else:
sigH[(nt_ch, tri)] = 1
if nt_ch in cntH:
cntH[nt_ch] += 1
else:
cntH[nt_ch] = 1
# if not filtered out
# for line
# for mutFileN
for key in sigH:
(type, tri) = key
freq = sigH[key]
outFile.write("%s\t%s\t%s\t%s\t%s\t%s\n" % (sid, type, tri, freq, cntH[type], total))
# if raw mutect call
outFile.flush()
outFile.close()
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 mutscan_signature(mode='WXS', outFileN=''):
if outFileN == '':
outFile = sys.stdout
else:
outFile = open(outFileN, 'w')
dirLH = {'WXS': mysetting.wxsMutscanDirL, 'RSQ': mysetting.rsqMutscanDirL}
contextH = {}
for dir in dirLH[mode]:
fileL = filter(
lambda x: 'bak' not in x,
map(lambda x: x.rstrip(),
os.popen('find %s -name *dbsnp_flt' % dir).readlines()))
for file in fileL:
if mode == 'RSQ':
if os.path.basename(
file
) == 'S647_splice.mutscan.dbsnp_flt': ## duplicated files
continue
if 'splice_Z' in file:
sid = re.match('(.*)_splice2.mutscan.dbsnp_flt',
os.path.basename(file)).group(1)
else:
sid = re.match('(.*)_RSq_splice.mutscan.dbsnp_flt',
os.path.basename(file)).group(1)
else:
sid = re.match('(.*).mutscan.dbsnp_flt',
os.path.basename(file)).group(1)
sigH = {}
cntH = {}
total = 0
inFile = open(file, 'r')
for line in inFile:
colL = line.rstrip().split('\t')
chrom = colL[0]
pos = int(colL[1])
ref = colL[2]
alt = colL[3]
if ref == 'N' or len(alt) > 1:
continue
n_ref = int(colL[4])
n_alt = int(colL[5])
if n_alt >= MIN_MUT_N and (n_alt + n_ref) >= MIN_COV:
start = pos - 1
end = pos + 1
if chrom in contextH and pos in contextH[chrom]:
context = contextH[chrom][pos]
else:
resL = os.popen(
'samtools faidx /data1/Sequence/ucsc_hg19/hg19.fasta %s:%s-%s'
% (chrom, start, end)).readlines()
context = resL[1].rstrip().upper()
if chrom not in contextH:
contextH[chrom] = {}
contextH[chrom][pos] = context
if ref not in ['C', 'T']:
ref = mybasic.rc(ref)
alt = mybasic.rc(alt)
context = mybasic.rc(context)
ch = ref + '>' + alt
if ch in cntH:
cntH[ch] += 1
else:
cntH[ch] = 1
if (ch, context) in sigH:
sigH[(ch, context)] += 1
else:
sigH[(ch, context)] = 1
total += 1
#if
##for line
for (type, context) in sigH:
outFile.write('%s\t%s\t%s\t%s\t%s\t%s\n' %
(sid, type, context, sigH[(type, context)],
cntH[type], total))
##for file
#for dir
outFile.flush()
outFile.close()