def allTLSeqCounter():
for i in range(1, numofruns+1):
x = TSeqCounter(filenames[i])
y = LSeqCounter(filenames[i])
b = barcodechecker(filenames[i])
TSeqNums.update({i:x})
LSeqNums.update({i:y})
Barcodevalues.update({i:b})
def listinit(mink, maxk):
l = lvalues[1]
for i in range(0,numofruns+1):
poslist.append({})
rposlist.append({})
try:
avg = barcodechecker(filenames[i])
except:
continue
for k in range(mink, maxk+1):
poslist[i].update({k:{}})
rposlist[i].update({k:{}})
for x in range(0,(l+1-k-(2*avg))):
poslist[i][k].update({x:[]})
rposlist[i][k].update({x:[]})
def listinit():
for i in range(0,numofruns+1):
poslist.append({})
rposlist.append({})
consensuslist.append({})
cooccurencelist.append({})
cooccurenceindexlist.append({})
gaplist.append({})
gapstdev.append({})
for j in range(1, numofruns+1):
if extype == "SELEX":
l = lvalues[j]
if knownbarcode == False:
if nobarcode == True:
avg5 = 0
avg3 = 0
if nobarcode == False:
avg5 = barcodechecker(filenames[j])
avg3 = avg5
if knownbarcode == True:
avg5 = len(barcodeprimers53[j][0])
avg3 = len(barcodeprimers53[j][1])
for k in range(mink, maxk+1):
poslist[j][k] = {}
rposlist[j][k] = {}
consensuslist[j][k] = {}
cooccurencelist[j][k] = {}
cooccurenceindexlist[j][k] = {}
gaplist[j][k] = {}
gapstdev[j][k] = {}
for n in range(1, nmotifs+1):
poslist[j][k][n] = {}
rposlist[j][k][n] = {}
consensuslist[j][k][n] = {}
cooccurencelist[j][k][n] = {}
cooccurenceindexlist[j][k][n] = {}
gaplist[j][k][n] = []
gapstdev[j][k][n] = {}
gapstdev[j][k][n]["mean"] = 0
gapstdev[j][k][n]["stdev"] = 0
if extype == "SELEX":
for x in range(0,(l+1-k-(avg5+avg3))):
poslist[j][k][n].update({x:0})
rposlist[j][k][n].update({x:0})
else:
poslist[j][k][n] = []
rposlist[j][k][n] = []
def seqwtfbsfinder(FileName, k):
TFBSSeqNum = 0
l = lvalues[(ufilenames[FileName])]
fastaFileName = open(FileName, "r")
avg = barcodechecker(FileName)
for line in fastaFileName:
c = 0
line = line.strip()
if line.startswith(">"):
continue
if len(line) == l and "N" not in line:
for x in range(0,((len(line)+1)-k)-(2*avg)):
kmers = str(line[x+avg:x+k+avg])
if len(kmers) > 0 and line.count(kmers) == 1:
hkmer = kmer2hash(kmers)
if hkmer in hamminglist2[(ufilenames[FileName])][k]:
if c == 0:
TFBSSeqNum += 1
c += 1
else:
continue
if revcompwanted == True:
if len(line) == l and "N" not in line:
for x in range(0,((len(line)+1)-k)-(2*avg)):
rkmers = revComp(line[x+avg:x+k+avg])
if len(rkmers) > 0 and (line.count(kmers) + line.count(rkmers)) == 1:
hkmer = kmer2hash(rkmers)
if hkmer in hamminglist2[(ufilenames[FileName])][k]:
if c == 0:
TFBSSeqNum += 1
c += 1
else:
continue
return TFBSSeqNum
def CreatePosList(FileName, k, runnum):
fastaFileName = open(FileName, "r")
avg = barcodechecker(FileName)
for line in fastaFileName:
line = line.strip()
if line.startswith(">"):
continue
if len(line) == lvalues[runnum] and "N" not in line:
for x in range(0,((len(line)+1)-k)-(2*avg)):
kmers = str(line[x+avg:x+k+avg])
if len(kmers) > 0 and line.count(kmers) == 1:
hkmers = kmer2hash(kmers)
if hkmers in hamminglist2[runnum][k]:
poslist[runnum][k][x].append(hkmers)
if revcompwanted == True:
if len(line) == lvalues[runnum] and "N" not in line:
for x in range(0,((len(line)+1)-k)-(2*avg)):
kmers = str(line[x+avg:x+k+avg])
rkmers = revComp(line[x+avg:x+k+avg])
if len(rkmers) > 0 and (line.count(rkmers) + line.count(kmers)) == 1:
hrkmers = kmer2hash(rkmers)
if hrkmers in hamminglist2[runnum][k]:
rposlist[runnum][k][x].append(hrkmers)
def CreatePosListNORMAL(FileName, k, runnum):
fastaFileName = open(FileName, "r")
if knownbarcode == False:
if nobarcode == True:
avg5 = 0
avg3 = 0
if nobarcode == False:
avg5 = barcodechecker(FileName)
avg3 = avg5
if knownbarcode == True:
avg5 = len(barcodeprimers53[runnum][0])
avg3 = len(barcodeprimers53[runnum][1])
firstline = fastaFileName.readline()
firstline = firstline.strip()
if firstline.startswith(">"):
filetype = "fasta"
if firstline.startswith("@"):
filetype = "fastq"
for n in range(1, nmotifs+1):
seq1 = kmer2hash(consensuslist[runnum][k][n])
seq2 = kmer2hash(revComp(consensuslist[runnum][k][n]))
for sequence in SeqIO.parse(FileName, filetype):
line = str(sequence.seq)
lenline = len(line)
c = 0
LSeqNums[runnum] += 1
TSeqNums[runnum] += 1
done = set()
noisefilter = {"F": [],"R": []}
for x in range(0,(lenline+1-k)-(avg5+avg3)):
try:
kmers = kmer2hash(str(line[x+avg5:x+k+avg5]))
#if kmers not in done:
if kmers in hamminglist2[runnum][k][n]:
poslist[runnum][k][n].append(x/(lenline-k+1-(avg5+avg3)))
done.add(kmers)
if c == 0:
numoftfbsseq[runnum][k] += 1
c += 1
if kmers in rhamminglist2[runnum][k][n]:
rposlist[runnum][k][n].append((x)/(lenline-k+1-(avg5+avg3)))
done.add(kmers)
if c == 0:
numoftfbsseq[runnum][k] += 1
c += 1
if kmers == seq1:
noisefilter["F"].append(x)
if kmers == seq2:
noisefilter["R"].append(x)
except:
continue
if x == (len(line)-k)-(avg5+avg3):
if seq1 & seq2 in done:
numf = len(noisefilter["F"])
numr = len(noisefilter["R"])
a = 0
b = 0
mindiff = 1000
while (a < numf and b < numr):
if (k < abs(noisefilter["F"][a] - noisefilter["R"][b]) < mindiff):
mindiff = abs(noisefilter["F"][a] - noisefilter["R"][b])
if (noisefilter["F"][a] < noisefilter["R"][b]):
a += 1
else:
b += 1
if k < mindiff <= (k+10):
cooccurencelist[runnum][k][n]["fr"] += 1
gaplist[runnum][k][n].append(mindiff-k)
elif seq1 in done and seq2 not in done:
cooccurencelist[runnum][k][n]["f"] += 1
elif seq2 in done and seq1 not in done:
cooccurencelist[runnum][k][n]["r"] += 1