def getLongestRepeat(seq, minTimes, kmerLength):
#get all di seqs, uniquify
kSeqs = bioLibCG.returnFrames(seq, kmerLength)
kSeqs = set(kSeqs)
highestSLen = 0
for kmer in kSeqs:
#slides
for slide in range(0, kmerLength):
sLen = 0
for i in range(slide, len(seq), kmerLength):
try:
if seq[i:i + kmerLength] == kmer:
sLen += 1
else:
#if stretch is long enough, mask
if sLen >= minTimes:
if sLen > highestSLen: highestSLen = sLen
sLen = 0
else:
sLen = 0
except IndexError:
#check for masking
if sLen > minTimes:
if sLen > highestSLen: highestSLen = sLen
sLen = 0
print highestSLen
return highestSLen
def plotInfoNTFrameEnrichment(dir, frameWidth, outFN):
frameNum_seqs = {}
for fChrom in bioLibCG.humanChromosomes:
for fStrand in ('1', '-1'):
print fChrom, fStrand
fN = '%s/%s.%s.primeSeqs' % (dir, fChrom, fStrand)
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
seq = ls[0]
for i, frame in enumerate(bioLibCG.returnFrames(seq, frameWidth)):
frameNum_seqs.setdefault(i, []).append(frame)
f.close()
let_frameCounts = {}
for fNum in sorted(frameNum_seqs.keys()):
seqs = frameNum_seqs[fNum]
let_count = getSeqEnrichment(seqs)
for let, count in let_count.items():
let_frameCounts.setdefault(let, []).append(count)
fOut = open(outFN, 'w')
for let, fCounts in let_frameCounts.items():
fCounts = ','.join([str(x) for x in fCounts])
fOut.write('%s\t%s\n' % (let, fCounts))
fOut.close()
def getLongestRepeat(seq, minTimes, kmerLength):
#get all di seqs, uniquify
kSeqs = bioLibCG.returnFrames(seq, kmerLength)
kSeqs = set(kSeqs)
seqLength = len(seq)
highestSLen = 0
for kmer in kSeqs:
for slide in range(0, kmerLength):
sLen = 0
adjustedRange = seqLength - (kmerLength - 1) #takes into account excluding right side numbers
scanRange = range(slide, adjustedRange, kmerLength)
for i in scanRange:
#print seq[:i] + ' [' + seq[i:i + kmerLength] + ']', highestSLen, sLen, i, i + kmerLength
if (seq[i:i + kmerLength] == kmer):
sLen += 1
#take care of last nt
if i == scanRange[-1]:
if sLen >= minTimes:
if sLen > highestSLen: highestSLen = sLen
sLen = 0
else:
#if stretch is long enough, mask
if sLen >= minTimes:
if sLen > highestSLen: highestSLen = sLen
sLen = 0
return highestSLen
def kMask(seq, mask, minTimes, kmerLength):
if not mask:
mask = [False for x in seq]
#get all di seqs, uniquify
kSeqs = bioLibCG.returnFrames(seq, kmerLength)
kSeqs = set(kSeqs)
for kmer in kSeqs:
#slides
for slide in range(0, kmerLength):
sLen = 0
for i in range(slide, len(seq), kmerLength):
try:
if seq[i:i + kmerLength] == kmer:
sLen += 1
else:
#if stretch is long enough, mask
if sLen >= minTimes:
maskStart = i - (kmerLength * sLen)
for i in range(maskStart, i): mask[i] = True
sLen = 0
else:
sLen = 0
except IndexError:
#check for masking
if sLen > minTimes:
maskStart = i - (kmerLength * sLen)
for i in range(maskStart, i): mask[i] = True
sLen = 0
return mask
def plotInfoNTFrameEnrichment(dir, frameWidth, outFN):
frameNum_seqs = {}
for fChrom in bioLibCG.humanChromosomes:
for fStrand in ('1', '-1'):
print fChrom, fStrand
fN = '%s/%s.%s.primeSeqs' % (dir, fChrom, fStrand)
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
seq = ls[0]
for i, frame in enumerate(
bioLibCG.returnFrames(seq, frameWidth)):
frameNum_seqs.setdefault(i, []).append(frame)
f.close()
let_frameCounts = {}
for fNum in sorted(frameNum_seqs.keys()):
seqs = frameNum_seqs[fNum]
let_count = getSeqEnrichment(seqs)
for let, count in let_count.items():
let_frameCounts.setdefault(let, []).append(count)
fOut = open(outFN, 'w')
for let, fCounts in let_frameCounts.items():
fCounts = ','.join([str(x) for x in fCounts])
fOut.write('%s\t%s\n' % (let, fCounts))
fOut.close()
def getLongestRepeat(seq, minTimes, kmerLength):
#get all di seqs, uniquify
kSeqs = bioLibCG.returnFrames(seq, kmerLength)
kSeqs = set(kSeqs)
seqLength = len(seq)
highestSLen = 0
for kmer in kSeqs:
for slide in range(0, kmerLength):
sLen = 0
adjustedRange = seqLength - (
kmerLength - 1
) #takes into account excluding right side numbers
scanRange = range(slide, adjustedRange, kmerLength)
for i in scanRange:
#print seq[:i] + ' [' + seq[i:i + kmerLength] + ']', highestSLen, sLen, i, i + kmerLength
if (seq[i:i + kmerLength] == kmer):
sLen += 1
#take care of last nt
if i == scanRange[-1]:
if sLen >= minTimes:
if sLen > highestSLen: highestSLen = sLen
sLen = 0
else:
#if stretch is long enough, mask
if sLen >= minTimes:
if sLen > highestSLen: highestSLen = sLen
sLen = 0
return highestSLen
def kMask(seq, mask, minTimes, kmerLength):
if not mask:
mask = [False for x in seq]
#get all di seqs, uniquify
kSeqs = bioLibCG.returnFrames(seq, kmerLength)
kSeqs = set(kSeqs)
for kmer in kSeqs:
#slides
for slide in range(0, kmerLength):
sLen = 0
for i in range(slide, len(seq), kmerLength):
try:
if seq[i:i + kmerLength] == kmer:
sLen += 1
else:
#if stretch is long enough, mask
if sLen >= minTimes:
maskStart = i - (kmerLength * sLen)
for i in range(maskStart, i): mask[i] = True
sLen = 0
else:
sLen = 0
except IndexError:
#check for masking
if sLen > minTimes:
maskStart = i - (kmerLength * sLen)
for i in range(maskStart, i): mask[i] = True
sLen = 0
return mask
def createWordDatabase(seqList, wordSize):
#targetList should be a list of cgSeqs
wordDict = {} # word: [id, word start, word end]
for tSeq in seqList:
for i,word in enumerate(bioLibCG.returnFrames(tSeq.sequence, wordSize)):
try:
wordDict[word].append([tSeq.id, i, i + wordSize - 1])
except KeyError:
wordDict[word] = [[tSeq.id, i, i + wordSize - 1]]
return wordDict
def createWordDatabase(seqList, wordSize):
#targetList should be a list of cgSeqs
wordDict = {} # word: [id, word start, word end]
for tSeq in seqList:
for i, word in enumerate(bioLibCG.returnFrames(tSeq.sequence,
wordSize)):
try:
wordDict[word].append([tSeq.id, i, i + wordSize - 1])
except KeyError:
wordDict[word] = [[tSeq.id, i, i + wordSize - 1]]
return wordDict
def returnDiFreq(seqs):
#collect di count
di_freq = {}
totalFrames = 0.0
for seq in seqs:
dis = bioLibCG.returnFrames(seq, 2)
totalFrames += len(dis)
for di in dis:
di_freq[di] = di_freq.get(di, 0) + 1.0
#convert to frequencey
di_freq = dict( (x, di_freq[x]/totalFrames) for x in di_freq)
return di_freq
def alignQuery(qSeq, wordDatabase, sequenceDatabase, wordSize, maxNumMismatches, fOut):
#wordDataBase : [id, wordStart, wordEnd]F
#find all words in sequence --> align if word in database
for i,word in enumerate(bioLibCG.returnFrames(qSeq.sequence, wordSize)):
if word in wordDatabase:
for aInfo in wordDatabase[word]:
qWordPos = [i, i + wordSize - 1]
tSeq = cgSeq(aInfo[0], sequenceDatabase[aInfo[0]])
tWordPos = [aInfo[1], aInfo[2]]
newAlignment = alignSeqs(qSeq, tSeq, qWordPos, tWordPos, maxNumMismatches)
if newAlignment:
fOut.write(newAlignment.alignmentOutput() + '\n')
def returnDiFreq(seqs):
#collect di count
di_freq = {}
totalFrames = 0.0
for seq in seqs:
dis = bioLibCG.returnFrames(seq, 2)
totalFrames += len(dis)
for di in dis:
di_freq[di] = di_freq.get(di, 0) + 1.0
#convert to frequencey
di_freq = dict((x, di_freq[x] / totalFrames) for x in di_freq)
return di_freq
def alignQuery(qSeq, wordDatabase, sequenceDatabase, wordSize,
maxNumMismatches, fOut):
#wordDataBase : [id, wordStart, wordEnd]F
#find all words in sequence --> align if word in database
for i, word in enumerate(bioLibCG.returnFrames(qSeq.sequence, wordSize)):
if word in wordDatabase:
for aInfo in wordDatabase[word]:
qWordPos = [i, i + wordSize - 1]
tSeq = cgSeq(aInfo[0], sequenceDatabase[aInfo[0]])
tWordPos = [aInfo[1], aInfo[2]]
newAlignment = alignSeqs(qSeq, tSeq, qWordPos, tWordPos,
maxNumMismatches)
if newAlignment:
fOut.write(newAlignment.alignmentOutput() + '\n')
def locateASignals(dataFN, outFN, rn = None, tn = None):
#load data
NX = cgNexusFlat.Nexus(dataFN, ASite)
NX.load(['coord', 'sequence'], [rn, tn])
f = open(outFN, 'w')
for id in NX.ids:
chrom, strand, start, end = bioLibCG.tccSplit(NX.coord[id])
if len(NX.sequence[id]) < 10: continue
print NX.sequence[id], '\n'
checkFrames = bioLibCG.returnFrames(NX.sequence[id], 6)
for i, frame in enumerate(checkFrames):
if frame == 'AATAAA':
#assume 0-based...?
siteStart, siteEnd = start + i, start + i + 5
f.write('%s\n' % bioLibCG.makeTcc(chrom, strand, siteStart, siteEnd))
f.close()
def get20mers(aluFN):
gf = GenomeFetch.GenomeFetch('hg19')
seq_count = {}
f = open(aluFN, 'r')
for line in f:
ls = line.strip().split('\t')
coord = ls[0]
chrom, start, end = coord.split(':')[0], coord.split(':')[1].split(
'-')[0], coord.split(':')[1].split('-')[1]
strand = bioLibCG.switchStrandFormat(ls[2])
tcc = bioLibCG.makeTcc(chrom, strand, start, end)
seq = gf.getSequence(tcc)
frames = bioLibCG.returnFrames(seq, 20)
if frames == 1:
continue
for smallSeq in frames:
count = seq_count.get(smallSeq, 0)
seq_count[smallSeq] = count + 1
for seq, count in seq_count.items():
print '%s\t%s' % (seq, count)
def get20mers(aluFN):
gf = GenomeFetch.GenomeFetch('hg19')
seq_count = {}
f = open(aluFN, 'r')
for line in f:
ls = line.strip().split('\t')
coord = ls[0]
chrom, start, end = coord.split(':')[0], coord.split(':')[1].split('-')[0], coord.split(':')[1].split('-')[1]
strand = bioLibCG.switchStrandFormat(ls[2])
tcc = bioLibCG.makeTcc(chrom, strand, start, end)
seq = gf.getSequence(tcc)
frames = bioLibCG.returnFrames(seq, 20)
if frames == 1:
continue
for smallSeq in frames:
count = seq_count.get(smallSeq, 0)
seq_count[smallSeq] = count + 1
for seq, count in seq_count.items():
print '%s\t%s' % (seq, count)
def createSubSequences(oID_sequence, frameLength):
return dict( (oID, set(bioLibCG.returnFrames(oID_sequence[oID], frameLength))) for oID in oID_sequence)
def createSubSequences(oID_sequence, frameLength):
return dict(
(oID, set(bioLibCG.returnFrames(oID_sequence[oID], frameLength)))
for oID in oID_sequence)
