def transcriptSetOverlapTargets(aDir):
geneSetFN = '/home/chrisgre/dataSources/known/Human/geneSets/ensemblAllTranscripts.tsv'
allExons = cgGenes.createGeneSetFromFile(geneSetFN)
#get degradome TCCS
#note that you need to test the AS peaks, this is the location of the targetted transcript
aDC = cgDB.dataController(aDir, cgAlignment.cgAlignment)
id_alignment = aDC.load()
#create list of unique tccs.
uniqTccs = []
for alignment in id_alignment.values():
chrom, strand, start, end = cg.tccSplit(alignment.tTcc)
offset = alignment.tStart
sLen = alignment.sLength
if strand == '1':
start = start - 19 + offset
end = start + sLen
else:
end = end + 19 - offset
start = end - sLen
tcc = cg.makeTcc(chrom, strand, start, end)
if tcc not in uniqTccs: uniqTccs.append(tcc)
degTccs = [cg.convertToAS(x) for x in uniqTccs]
#find all overlapping exons/transcripts, then all results sequences that overlap exons
overlappingExons = allExons.transcriptOverlaps(degTccs)
overlappingExonTccs = [x.tcc for x in overlappingExons]
overlappingDegTccs = compare.compareTwoTcc(degTccs, overlappingExonTccs, 1)
#update
for obj in id_alignment.values():
chrom, strand, start, end = cg.tccSplit(alignment.tTcc)
offset = alignment.tStart
sLen = alignment.sLength
if strand == '1':
start = start - 19 + offset
end = start + sLen
else:
end = end + 19 - offset
start = end - sLen
tcc = cg.makeTcc(chrom, strand, start, end)
degTcc = cg.convertToAS(tcc)
if degTcc in overlappingDegTccs:
obj.transcriptOverlap = True
else:
obj.transcriptOverlap = False
aDC.commit(id_alignment)
def transcriptSetOverlapTargets(aDir):
geneSetFN = '/home/chrisgre/dataSources/known/Human/geneSets/ensemblAllTranscripts.tsv'
allExons = cgGenes.createGeneSetFromFile(geneSetFN)
#get degradome TCCS
#note that you need to test the AS peaks, this is the location of the targetted transcript
aDC = cgDB.dataController(aDir, cgAlignment.cgAlignment)
id_alignment = aDC.load()
#create list of unique tccs.
uniqTccs = []
for alignment in id_alignment.values():
chrom, strand, start, end = cg.tccSplit(alignment.tTcc)
offset = alignment.tStart
sLen = alignment.sLength
if strand == '1':
start = start - 19 + offset
end = start + sLen
else:
end = end + 19 - offset
start = end - sLen
tcc = cg.makeTcc(chrom, strand, start, end)
if tcc not in uniqTccs: uniqTccs.append(tcc)
degTccs = [cg.convertToAS(x) for x in uniqTccs]
#find all overlapping exons/transcripts, then all results sequences that overlap exons
overlappingExons = allExons.transcriptOverlaps(degTccs)
overlappingExonTccs = [x.tcc for x in overlappingExons]
overlappingDegTccs = compare.compareTwoTcc(degTccs, overlappingExonTccs, 1)
#update
for obj in id_alignment.values():
chrom, strand, start, end = cg.tccSplit(alignment.tTcc)
offset = alignment.tStart
sLen = alignment.sLength
if strand == '1':
start = start - 19 + offset
end = start + sLen
else:
end = end + 19 - offset
start = end - sLen
tcc = cg.makeTcc(chrom, strand, start, end)
degTcc = cg.convertToAS(tcc)
if degTcc in overlappingDegTccs:
obj.transcriptOverlap = True
else:
obj.transcriptOverlap = False
aDC.commit(id_alignment)
def transcriptSetOverlapDegFileHitmap(degFile, runningChrom, runningStrand):
geneSetFN = '/home/chrisgre/dataSources/known/Human/geneSets/ensemblAllTranscripts.tsv'
allExons = cgGenes.createGeneSetFromFile(geneSetFN)
transcriptTccs = []
for gene in allExons.set.values():
for transcript in gene.transcripts:
transcriptTccs.append(transcript.tcc)
#create hitmap
coordSet = set()
for tcc in transcriptTccs:
chrom, strand, start, end = cg.tccSplit(tcc)
if chrom != runningChrom:
continue
if strand != runningStrand:
continue
for i in range(start, end + 1):
coordSet.add(i)
#find overlapping degTccs
print 'done creating hitmap'
f = open(degFile, 'r')
newLines = []
for line in f:
ls = line.strip().split('\t')
degTcc = cg.convertToAS(ls[1])
chrom, strand, start, end = cg.tccSplit(degTcc)
if chrom != runningChrom:
continue
if strand != runningStrand:
continue
inTran = '0'
for i in xrange(start, end + 1):
if i in coordSet:
inTran = '1'
break
#update newLines
newLine = cg.appendToLine(line, inTran, 3)
newLines.append(newLine)
f.close()
f = open(degFile + '.%s.%s' % (runningChrom, runningStrand), 'w')
f.writelines(newLines)
f.close()
def transcriptSetOverlapDegFileHitmap(degFile, runningChrom, runningStrand):
geneSetFN = '/home/chrisgre/dataSources/known/Human/geneSets/ensemblAllTranscripts.tsv'
allExons = cgGenes.createGeneSetFromFile(geneSetFN)
transcriptTccs = []
for gene in allExons.set.values():
for transcript in gene.transcripts:
transcriptTccs.append(transcript.tcc)
#create hitmap
coordSet = set()
for tcc in transcriptTccs:
chrom, strand, start, end = cg.tccSplit(tcc)
if chrom != runningChrom:
continue
if strand != runningStrand:
continue
for i in range(start, end + 1):
coordSet.add(i)
#find overlapping degTccs
print 'done creating hitmap'
f = open(degFile, 'r')
newLines = []
for line in f:
ls = line.strip().split('\t')
degTcc = cg.convertToAS(ls[1])
chrom, strand, start, end = cg.tccSplit(degTcc)
if chrom != runningChrom:
continue
if strand != runningStrand:
continue
inTran = '0'
for i in xrange(start, end + 1):
if i in coordSet:
inTran = '1'
break
#update newLines
newLine = cg.appendToLine(line, inTran, 3)
newLines.append(newLine)
f.close()
f = open(degFile + '.%s.%s' % (runningChrom, runningStrand), 'w')
f.writelines(newLines)
f.close()
def countWithBinsSet(dFN, binDir, type = 'INTRON'):
dNX = cgNexusFlat.Nexus(dFN, cgDegPeak.Peak)
dNX.load(['tcc'])
numBins = 1
c_s_bin_set = {}
for chrom in bioLibCG.humanChromosomes:
for strand in ('1', '-1'):
#initialize data structure
for i in range(0, numBins):
c_s_bin_set.setdefault(chrom, {}).setdefault(strand, {})[i] = set()
f = open(binDir + '/%s.%s.%s.bins' % (type, chrom, strand), 'r')
for line in f:
ls = line.strip().split('\t')
tccs = ls[1:numBins + 1]
for i in range(0,numBins):
ch, st, sta, end = bioLibCG.tccSplit(tccs[i])
for j in range(sta, end + 1):
c_s_bin_set[chrom][strand][i].add(j)
#collect dTtcs in list
dTccs = []
for dID in dNX.tcc:
tcc = dNX.tcc[dID]
c, s, st, en = bioLibCG.tccSplit(tcc)
if s == '1':
s = '-1'
en = st
else:
s = '1'
st = en
dTccs.append(bioLibCG.makeTcc(c,s,st,en))
#make bCounts
binCounts = [0] * numBins
#count for each bin
for i in range(0, numBins):
for dTcc in dTccs:
c, s, st, en = bioLibCG.tccSplit(dTcc)
for j in range(st, en + 1):
if j in c_s_bin_set[c][s][i]:
binCounts[i] += 1
print '%s\t%s' % (i, binCounts[i])
def countWithBinsSet(dFN, binDir, type='INTRON'):
dNX = cgNexusFlat.Nexus(dFN, cgDegPeak.Peak)
dNX.load(['tcc'])
numBins = 1
c_s_bin_set = {}
for chrom in bioLibCG.humanChromosomes:
for strand in ('1', '-1'):
#initialize data structure
for i in range(0, numBins):
c_s_bin_set.setdefault(chrom, {}).setdefault(strand,
{})[i] = set()
f = open(binDir + '/%s.%s.%s.bins' % (type, chrom, strand), 'r')
for line in f:
ls = line.strip().split('\t')
tccs = ls[1:numBins + 1]
for i in range(0, numBins):
ch, st, sta, end = bioLibCG.tccSplit(tccs[i])
for j in range(sta, end + 1):
c_s_bin_set[chrom][strand][i].add(j)
#collect dTtcs in list
dTccs = []
for dID in dNX.tcc:
tcc = dNX.tcc[dID]
c, s, st, en = bioLibCG.tccSplit(tcc)
if s == '1':
s = '-1'
en = st
else:
s = '1'
st = en
dTccs.append(bioLibCG.makeTcc(c, s, st, en))
#make bCounts
binCounts = [0] * numBins
#count for each bin
for i in range(0, numBins):
for dTcc in dTccs:
c, s, st, en = bioLibCG.tccSplit(dTcc)
for j in range(st, en + 1):
if j in c_s_bin_set[c][s][i]:
binCounts[i] += 1
print '%s\t%s' % (i, binCounts[i])
def testOverlaps(dataFN, oFF):
dataNX = Nexus(dataFN, oFF)
dataNX.load(['tcc'])
#check for overlaps
overlappingIDs = set()
chrom_strand_range = {}
while dataNX.nextID():
chrom, strand, start, end = bioLibCG.tccSplit(dataNX.tcc)
#check if overlap
chrom_strand_range.setdefault(chrom, {}).setdefault(strand, set())
overlap = False
for i in range(start, end + 1):
if i in chrom_strand_range[chrom][strand]:
overlap = True
break
#tag or add these coordinates
if overlap:
overlappingIDs.add(dataNX.id)
else:
for i in range(start, end + 1):
chrom_strand_range[chrom][strand].add(i)
print "THESE OVERLAP", overlappingIDs
def updateContext(oFN, wigDir, chrom, strand, switchStrand = False):
oNX = cgNexusFlat.Nexus(oFN, cgDegPeak.Peak)
oNX.load(['tcc', 'context'])
if switchStrand:
strand = str(-int(strand))
else:
strand = str(strand)
print 'loading wig'
coord_contexts = cgWig.loadSingleWigContext(wigDir, chrom, strand, 'context')
print 'done loading'
ds = bioLibCG.dominantSpotter(['C_EXON', 'C_3UTR', 'C_5UTR', 'NC_EXON', 'NC_3UTR', 'NC_5UTR', 'C_INTRON', 'NC_INTRON', 'INTER'])
for oID in oNX.tcc:
oChrom, oStrand, start, end = bioLibCG.tccSplit(oNX.tcc[oID])
#deg wigs is AS to actual clipping site
if switchStrand:
oStrand = str(-int(strand))
else:
oStrand = str(oStrand)
if oChrom == chrom and oStrand == strand:
contexts = coord_contexts.get(start, 'INTER').split(',')
oNX.context[oID] = ds.spotItem(contexts)
oNX.save()
def updateContext(oFN, wigDir, chrom, strand, rn=None, tn=None):
oNX = cgNexusFlat.Nexus(oFN, degPeak.degPeak)
oNX.load(['context', 'tcc'], [rn, tn])
print 'loading wig'
coord_contexts = cgWig.loadSingleWigContext(wigDir, chrom, strand,
'context')
print 'done loading'
ds = bioLibCG.dominantSpotter([
'C_EXON', 'C_3UTR', 'C_5UTR', 'NC_EXON', 'NC_3UTR', 'NC_5UTR',
'C_INTRON', 'NC_INTRON', 'INTER'
])
for oID in oNX.tcc:
oChrom, oStrand, start, end = bioLibCG.tccSplit(oNX.tcc[oID])
#deg wigs is AS to actual clipping site
if oStrand == '1':
oStrand = '-1'
else:
oStrand = '1'
if oChrom == chrom and oStrand == strand:
contexts = coord_contexts.get(start, 'INTER').split(',')
oNX.context[oID] = ds.spotItem(contexts)
oNX.save()
def newCoords(fN):
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
chrom, strand, start, end = bioLibCG.tccSplit(ls[0])
print '%s:%s-%s' % (chrom, start, end)
def updateIContext(oFN, wigDir, chrom, strand, rn = None, tn = None):
oNX = cgNexusFlat.Nexus(oFN, cgDegPeak.Peak)
oNX.load(['tcc', 'iContexts'], [rn, tn])
if strand == '1':
strand = '-1'
else:
strand = '1'
print 'loading wig'
coord_contexts = cgWig.loadSingleWigContext(wigDir, chrom, strand, 'iContext')
print 'done loading'
for oID in oNX.tcc:
oChrom, oStrand, start, end = bioLibCG.tccSplit(oNX.tcc[oID])
#deg wigs is AS to actual clipping site
if oStrand == '1':
oStrand = '-1'
else:
oStrand = '1'
if oChrom == chrom and oStrand == strand:
contexts = coord_contexts.get(start, '-1').split(',')
contexts = [int(x) for x in contexts]
oNX.iContexts[oID] = contexts
oNX.save()
def getPlotData(aSites, wigDir, outFN):
'''get box plot data from sites in degradome'''
#load and init
spreadRange = range(-200, 201) #200 +/- ... might want to check distance each AAUAAA is from each other
relCoord_degVals = dict( (i, []) for i in spreadRange )
for chrom in bioLibCG.humanChromosomes:
for strand in ('1', '-1'):
print chrom, strand
coord_value = cgWig.loadSingleWig(wigDir, chrom, strand, 'ALL')
f = open(aSites, 'r')
for line in f:
ls = line.strip().split('\t')
ichrom, istrand, start, end = bioLibCG.tccSplit(ls[0])
if ichrom != chrom or istrand != strand: continue
for i in spreadRange:
degVal = coord_value.get(end + i, 0)
relCoord_degVals[i].append(degVal)
f.close()
#output box data
#each row is a histogram of spread position (e.g., first row is -200)
f = open(outFN, 'w')
outLines = []
for i in spreadRange:
l = [str(x) for x in relCoord_degVals[i]]
outLines.append('\t'.join(l) + '\n')
f.writelines(outLines)
f.close()
def returnContBlocks(profile, tcc, minLevel = 5):
chrom, strand, start, end = cg.tccSplit(tcc)
pCoords = profile.keys()
pCoords.sort()
inBlock = False
bStart = None
blocks = []
for pCoord in pCoords:
if int(profile[pCoord]) > minLevel: #expression's high enough
if inBlock:
continue
else: #block Start
bStart = pCoord
inBlock = True
else: #not high enough
if inBlock: #end the block
blocks.append('%s:%s:%s:%s' % (chrom, strand, bStart, pCoord - 1))
inBlock = False
else:
continue
return blocks
def testOverlaps(dataFN, oFF):
dataNX = Nexus(dataFN, oFF)
dataNX.load(['tcc'])
#check for overlaps
overlappingIDs = set()
chrom_strand_range = {}
while dataNX.nextID():
chrom, strand, start, end = bioLibCG.tccSplit(dataNX.tcc)
#check if overlap
chrom_strand_range.setdefault(chrom, {}).setdefault(strand, set())
overlap = False
for i in range(start, end + 1):
if i in chrom_strand_range[chrom][strand]:
overlap = True
break
#tag or add these coordinates
if overlap:
overlappingIDs.add(dataNX.id)
else:
for i in range(start, end + 1):
chrom_strand_range[chrom][strand].add(i)
print "THESE OVERLAP", overlappingIDs
def markCenterExpression(aFN, wigDir, rn = None, tn = None):
extend = 25
timer = bioLibCG.cgTimer()
timer.start()
aNX = cgNexusFlat.Nexus(aFN, cgAlignmentFlat.cgAlignment)
aNX.load(['centerExpression', 'tTcc', 'tStart', 'sLength', 'tELevel'], [rn, tn])
#load expression of degradome
wigDict = cgWig.loadWigDict(wigDir)
for aID in aNX.centerExpression:
aNX.centerExpression[aID] = [0.0, 0.0, 0.0]
chrom, strand, start, end = bioLibCG.tccSplit(aNX.tTcc[aID])
offset = aNX.tStart[aID]
sLen = aNX.sLength[aID]
if strand == '1':
start = start - extend + offset
end = start + sLen
else:
end = end + extend - offset
start = end - sLen
scanRange = bioLibCG.makeTcc(chrom, strand, start, end)
stretch = cgWig.getExpressionProfile(scanRange, wigDict)
#make sure peak is in the small range
peakLevel = aNX.tELevel[aID]
peakInRange = (peakLevel in stretch.values())
expressionSum = sum(stretch.values())
sortedKeys = stretch.keys()
sortedKeys.sort()
if strand == '-1':
sortedKeys.reverse()
if expressionSum != 0 and peakInRange:
sumE = 0.0
for key in sortedKeys[8:12]:
sumE += stretch[key]
aNX.centerExpression[aID][0] = sumE/expressionSum
sumE = 0.0
for key in sortedKeys[7:13]:
sumE += stretch[key]
aNX.centerExpression[aID][1] = sumE/expressionSum
sumE = 0.0
for key in sortedKeys[6:14]:
sumE += stretch[key]
aNX.centerExpression[aID][2] = sumE/expressionSum
aNX.save()
def collectIntronSeqs(fN, outFN, assembly, amount=100, prime3=True):
myG = gf.GenomeFetch(assembly)
fOut = open(outFN, 'w')
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
if not 'C_INTRON' in ls[1]: continue
chrom, strand, start, end = bioLibCG.tccSplit(ls[2])
if end - start < amount: continue
if strand == '1':
if prime3:
tcc = bioLibCG.makeTcc(chrom, strand, end - amount, end)
seq = myG.getSequence(tcc)
fOut.write('%s\n' % seq[::-1])
else:
if prime3:
tcc = bioLibCG.makeTcc(chrom, strand, start, start + amount)
seq = myG.getSequence(tcc)
fOut.write('%s\n' % seq[::-1])
f.close()
fOut.close()
def updateHitMap(chrom_strand_coord, tcc):
chrom, strand, start, end = bioLibCG.tccSplit(tcc)
for i in xrange(start, end + 1):
chrom_strand_coord.setdefault(chrom, {}).setdefault(strand,
set()).add(i)
def probe(tcc, conf = None):
if not conf:
mConf = c.cgConfig('Main.conf')
smallPath = mConf.conf['smallPath']
chrom, strand, start, end = cg.tccSplit(tcc)
total = 0
for lib in cg.recurseDir(smallPath, end = 'mapped.%s.wig' % strand):
try:
eLevels = stepVectorScan.scanVectorsFile(lib, [tcc])
except:
print lib, 'index failed'
continue
#find highest expression level
highest = 0
for coord in eLevels:
if eLevels[coord] > highest:
highest = eLevels[coord]
if highest > 0:
print lib, highest
total += highest
#print eLevels
print total
def getNumberContextPosition(iContextDir, outFN, cType = 'INTRON', prime5 = True, ):
#load all context NXs
print 'loading NXs'
chrom_strand_NX = {}
for chrom in bioLibCG.humanChromosomes:
for strand in ('1', '-1'):
iFN = iContextDir + '/%s.%s.ids' % (chrom, strand)
NX = cgNexusFlat.Nexus(iFN, cgIContext.IContext)
NX.load(['type', 'tcc'])
chrom_strand_NX.setdefault(chrom, {})[strand] = NX
print 'done...'
print 'getting context occupancy'
#get the amount of introns/etc that occupy X
pos_numIntrons = {}
for chrom in chrom_strand_NX:
for strand in chrom_strand_NX[chrom]:
NX = chrom_strand_NX[chrom][strand]
for id in NX.ids:
if cType not in NX.type[id]: continue
c, s, st, en = bioLibCG.tccSplit(NX.tcc[id])
iLen = en - st
for i in xrange(0, iLen):
pos_numIntrons[i] = pos_numIntrons.get(i, 0) + 1
fOut = open(outFN, 'w')
for pos, numIntrons in pos_numIntrons.iteritems():
fOut.write('%s\t%s\n' % (pos, numIntrons) )
fOut.close()
def collectIntronSeqs(fN, outFN, assembly, amount = 100, prime3 = True):
myG = gf.GenomeFetch(assembly)
fOut = open(outFN, 'w')
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
if not 'C_INTRON' in ls[1]: continue
chrom, strand, start, end = bioLibCG.tccSplit(ls[2])
if end - start < amount: continue
if strand == '1':
if prime3:
tcc = bioLibCG.makeTcc(chrom, strand, end - amount, end)
seq = myG.getSequence(tcc)
fOut.write('%s\n' % seq[::-1])
else:
if prime3:
tcc = bioLibCG.makeTcc(chrom, strand, start, start + amount)
seq = myG.getSequence(tcc)
fOut.write('%s\n' % seq[::-1])
f.close()
fOut.close()
def newCoords(fN):
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
chrom, strand, start, end = bioLibCG.tccSplit(ls[0])
print '%s:%s-%s' % (chrom, start, end)
def updateTranscriptOverlap(oFN, wigDir, chrom, strand, rn=None, tn=None):
oNX = cgNexusFlat.Nexus(oFN, cgDegPeak.Peak)
oNX.load(['tOverlap', 'tcc'], [rn, tn])
#load the AS wig file for this degradome strand
if strand == '1':
strand = '-1'
else:
strand = '1'
coord_transcripts = cgWig.loadSingleWigTranscript(wigDir, chrom, strand,
'transcript')
for oID in oNX.tOverlap:
tChrom, tStrand, start, end = bioLibCG.tccSplit(oNX.tcc[oID])
if tStrand == '1':
tStrand = '-1'
else:
tStrand = '1'
if tChrom != chrom or tStrand != strand: continue
oNX.tOverlap[oID] = False
for i in xrange(start, end + 1):
if i in coord_transcripts:
oNX.tOverlap[oID] = True
break
oNX.save()
def probe(tcc, conf=None):
if not conf:
mConf = c.cgConfig('Main.conf')
smallPath = mConf.conf['smallPath']
chrom, strand, start, end = cg.tccSplit(tcc)
total = 0
for lib in cg.recurseDir(smallPath, end='mapped.%s.wig' % strand):
try:
eLevels = stepVectorScan.scanVectorsFile(lib, [tcc])
except:
print lib, 'index failed'
continue
#find highest expression level
highest = 0
for coord in eLevels:
if eLevels[coord] > highest:
highest = eLevels[coord]
if highest > 0:
print lib, highest
total += highest
#print eLevels
print total
def updateGeneName(dFN,
fFN,
wigDir,
chrom,
strand,
prefix,
switchStrand=False):
NX = Nexus(dFN, fFN)
NX.load(['geneNames', 'tcc'])
if switchStrand:
strand = -strand
strand = str(strand)
coord_gName = cgWig.loadSingleWigTranscript(wigDir, chrom, strand, prefix)
while NX.nextID():
chrom, strand, start, end = bioLibCG.tccSplit(NX.tcc)
overlappingGenes = coord_gName.get(start, ".")
if overlappingGenes == "NONE":
NX.geneNames = []
else:
NX.geneNames = overlappingGenes.split(',')
NX.save()
def updateTranscriptOverlap(oFN, wigDir, chrom, strand, rn = None, tn = None):
oNX = cgNexusFlat.Nexus(oFN, cgDegPeak.Peak)
oNX.load(['tOverlap', 'tcc'], [rn, tn])
#load the AS wig file for this degradome strand
if strand == '1':
strand = '-1'
else:
strand = '1'
coord_transcripts = cgWig.loadSingleWigTranscript(wigDir, chrom, strand, 'transcript')
for oID in oNX.tOverlap:
tChrom, tStrand, start, end = bioLibCG.tccSplit(oNX.tcc[oID])
if tStrand == '1':
tStrand = '-1'
else:
tStrand = '1'
if tChrom != chrom or tStrand != strand: continue
oNX.tOverlap[oID] = False
for i in xrange(start, end + 1):
if i in coord_transcripts:
oNX.tOverlap[oID] = True
break
oNX.save()
def updateContext(oFN, wigDir, chrom, strand, rn = None, tn = None):
oNX = cgNexusFlat.Nexus(oFN, degPeak.degPeak)
oNX.load(['context', 'tcc'], [rn, tn])
print 'loading wig'
coord_contexts = cgWig.loadSingleWigContext(wigDir, chrom, strand, 'context')
print 'done loading'
ds = bioLibCG.dominantSpotter(['C_EXON', 'C_3UTR', 'C_5UTR', 'NC_EXON', 'NC_3UTR', 'NC_5UTR', 'C_INTRON', 'NC_INTRON', 'INTER'])
for oID in oNX.tcc:
oChrom, oStrand, start, end = bioLibCG.tccSplit(oNX.tcc[oID])
#deg wigs is AS to actual clipping site
if oStrand == '1':
oStrand = '-1'
else:
oStrand = '1'
if oChrom == chrom and oStrand == strand:
contexts = coord_contexts.get(start, 'INTER').split(',')
oNX.context[oID] = ds.spotItem(contexts)
oNX.save()
def scanVectorsFile(fN, tccList):
'''Given tcc list --> scan wig files and return coord:value...
'''
timer = cg.cgTimer()
timer.start()
coordDict = {} # tcc: [list values]
for tcc in tccList:
chrom, strand, tccStart, tccEnd = cg.tccSplit(tcc)
#goto correct line in index
fIndex = cgIndex.lineIndex(fN, header = True) #!!!there actually is a header...have to deal with this...
fIndex.passCheckFunction(cgIndex.wigCheckFunction)
fIndex.binarySearch(tcc) #places file pointer at beginning of tcc as beginning
stop = False
for line in fIndex.file:
#print 'Line:', line.strip()
lBeg = int(cg.ss(line)[1])
lEnd = int(cg.ss(line)[2])
lValue = int(cg.ss(line)[3].split('.')[0])
if tccStart > lBeg:
lBeg = tccStart
if tccEnd < lEnd:
lEnd = tccEnd
stop = True
#print timer.split()
for i in range(lBeg, lEnd):
coordDict[i] = lValue
if stop: break
#fIndex.close()
return coordDict
def makeWig(fN, assembly, format=None, name=None):
'''format assumes bowtie
suitible for medium mapped files.
takes longer.'''
#assume bowtie
if not format: format = 'Bowtie'
parserFunction = returnParserFunction(format)
if not name: name = cg.getBaseFileName(fN, naked=True)
lDict = cg.returnChromLengthDict(assembly)
for chrom in lDict:
if not chrom in cg.acceptableChroms: continue
for strand in ['1', '-1']:
f = open(fN, 'r')
#create hitmap of chrom and strand
print chrom, strand, 'hitmap'
hitDict = {}
for line in f:
lChrom, lStrand, start, end = cg.tccSplit(parserFunction(line))
lStrand = str(lStrand)
start = int(start)
end = int(end)
if chrom == lChrom and strand == lStrand:
for i in range(start, end + 1):
try:
hitDict[i] += 1
except KeyError:
hitDict[i] = 1
#write results to wig file
writeWigFromHitDict(hitDict, assembly)
def updateContext(fN, fF, wigDir, chrom, strand, switchStrand = False):
NX = Nexus(fN, fF)
NX.load(['tcc', 'context'])
if switchStrand:
strand = str(-int(strand))
else:
strand = str(strand)
print 'loading wig'
coord_contexts = cgWig.loadSingleWigContext(wigDir, chrom, strand, 'context')
print 'done loading'
ds = bioLibCG.dominantSpotter(['C_EXON', 'C_3UTR', 'C_5UTR', 'NC_EXON', 'NC_3UTR', 'NC_5UTR', 'C_INTRON', 'NC_INTRON', 'INTER'])
while NX.nextID():
oChrom, oStrand, start, end = bioLibCG.tccSplit(NX.tcc)
#deg wigs is AS to actual clipping site
if switchStrand:
oStrand = str(-int(strand))
else:
oStrand = str(oStrand)
if oChrom == chrom and oStrand == strand:
contexts = coord_contexts.get(start, 'INTER').split(',')
NX.context = ds.spotItem(contexts)
NX.save()
def profileAroundPoint(zeroPoint, span, cName, ratio=False, ratioCoord=None):
'''span is +/- that number... if you put in 30, you'll get
-29 to 29 around 0
zeroPoint must be in tcc format with point at start
ratio will return all points around zero point as a ratio of zero point
so zeroPoint = 1.0 and the rest will be fractions of that...'''
chrom, strand, zPoint, end = cg.tccSplit(zeroPoint)
rStart = zPoint - span
rEnd = zPoint + span
rTcc = cg.makeTcc(chrom, strand, rStart, rEnd)
scanDict = svCoord([rTcc], cName)
#reorient so that zero is at zero
#find position of zero
returnDict = {}
if not ratio:
for i in range(1 - span, span):
returnDict[i] = scanDict[zPoint + i]
else:
if ratioCoord:
zeroVal = scanDict[ratioCoord]
else:
zeroVal = scanDict[zPoint]
if zeroVal == 0: zeroVal = 1
for i in range(1 - span, span):
r = float(scanDict[zPoint + i]) / float(zeroVal)
returnDict[i] = r
return returnDict
def scanVectorsFile(fN, tccList):
'''Given tcc list --> scan wig files and return coord:value...
'''
timer = cg.cgTimer()
timer.start()
coordDict = {} # tcc: [list values]
for tcc in tccList:
chrom, strand, tccStart, tccEnd = cg.tccSplit(tcc)
#goto correct line in index
fIndex = cgIndex.lineIndex(fN, header = True) #!!!there actually is a header...have to deal with this...
fIndex.passCheckFunction(cgIndex.wigCheckFunction)
fIndex.binarySearch(tcc) #places file pointer at beginning of tcc as beginning
stop = False
for line in fIndex.file:
#print 'Line:', line.strip()
lBeg = int(cg.ss(line)[1])
lEnd = int(cg.ss(line)[2])
lValue = int(cg.ss(line)[3].split('.')[0])
if tccStart > lBeg:
lBeg = tccStart
if tccEnd < lEnd:
lEnd = tccEnd
stop = True
#print timer.split()
for i in range(lBeg, lEnd):
coordDict[i] = lValue
if stop: break
#fIndex.close()
return coordDict
def profileAroundPoint(zeroPoint, span, cName, ratio = False, ratioCoord = None):
'''span is +/- that number... if you put in 30, you'll get
-29 to 29 around 0
zeroPoint must be in tcc format with point at start
ratio will return all points around zero point as a ratio of zero point
so zeroPoint = 1.0 and the rest will be fractions of that...'''
chrom, strand, zPoint, end = cg.tccSplit(zeroPoint)
rStart = zPoint - span
rEnd = zPoint + span
rTcc = cg.makeTcc(chrom, strand, rStart, rEnd)
scanDict = svCoord([rTcc], cName)
#reorient so that zero is at zero
#find position of zero
returnDict = {}
if not ratio:
for i in range(1-span, span):
returnDict[i] = scanDict[zPoint + i]
else:
if ratioCoord:
zeroVal = scanDict[ratioCoord]
else:
zeroVal = scanDict[zPoint]
if zeroVal == 0: zeroVal = 1
for i in range(1-span, span):
r = float(scanDict[zPoint + i])/float(zeroVal)
returnDict[i] = r
return returnDict
def returnContBlocks(profile, tcc, minLevel=5):
chrom, strand, start, end = cg.tccSplit(tcc)
pCoords = profile.keys()
pCoords.sort()
inBlock = False
bStart = None
blocks = []
for pCoord in pCoords:
if int(profile[pCoord]) > minLevel: #expression's high enough
if inBlock:
continue
else: #block Start
bStart = pCoord
inBlock = True
else: #not high enough
if inBlock: #end the block
blocks.append('%s:%s:%s:%s' %
(chrom, strand, bStart, pCoord - 1))
inBlock = False
else:
continue
return blocks
def makeBins5(fN, fOut, typeFilter):
fOut = open(fOut, 'w')
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
id = ls[0]
type, tcc = ls[1:3]
chrom, strand, st, en = bioLibCG.tccSplit(tcc)
#only take seqs that are long enough
if en - st < 100: continue
if not typeFilter in type: continue
tccBins = [] #0 is the first nt from the 3' end
if strand == '1':
for i in range(0, 100):
s, e = st + i, st + i
tccBins.append(bioLibCG.makeTcc(chrom,strand,s,e))
elif strand == '-1':
for i in range(0, 100):
s, e = en - i, en - i
tccBins.append(bioLibCG.makeTcc(chrom,strand,s,e))
pString = [id] + tccBins
fOut.write('\t'.join([str(x) for x in pString]) + '\n')
def makeWig(fN, assembly, format = None, name = None):
'''format assumes bowtie
suitible for medium mapped files.
takes longer.'''
#assume bowtie
if not format: format = 'Bowtie'
parserFunction = returnParserFunction(format)
if not name: name = cg.getBaseFileName(fN, naked = True)
lDict = cg.returnChromLengthDict(assembly)
for chrom in lDict:
if not chrom in cg.acceptableChroms: continue
for strand in ['1', '-1']:
f = open(fN, 'r')
#create hitmap of chrom and strand
print chrom, strand, 'hitmap'
hitDict = {}
for line in f:
lChrom, lStrand, start, end = cg.tccSplit(parserFunction(line))
lStrand = str(lStrand)
start = int(start)
end = int(end)
if chrom == lChrom and strand == lStrand:
for i in range(start, end + 1):
try:
hitDict[i] += 1
except KeyError:
hitDict[i] = 1
#write results to wig file
writeWigFromHitDict(hitDict, assembly)
def truncate(fN):
tCount = 0
shortCount = 0
fOut = open(fN + '.trun', 'w')
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
type, tcc = ls[1], ls[2]
tCount += 1
c, s, st, en = bioLibCG.tccSplit(tcc)
cLen = en - st
if cLen < 50:
shortCount += 1
continue
if s == '1':
en = en - 50
elif s == '-1':
st = st + 50
else:
print 'error'
return 1
ls[2] = bioLibCG.makeTcc(c, s, st, en)
line = '\t'.join(str(x) for x in [ls[0], ls[1], ls[2]]) + '\n'
fOut.write(line)
print shortCount, tCount
def truncate5(fN):
tCount = 0
shortCount = 0
fOut = open(fN + '.trun5', 'w')
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
type, tcc = ls[1], ls[2]
tCount += 1
c, s, st, en = bioLibCG.tccSplit(tcc)
cLen = en - st
if cLen < 50:
shortCount += 1
continue
if s == '1':
st = st + 50
elif s == '-1':
en = en - 50
else:
print 'error'
return 1
ls[2] = bioLibCG.makeTcc(c, s, st, en)
line = '\t'.join(str(x) for x in [ls[0], ls[1], ls[2]]) + '\n'
fOut.write(line)
print shortCount, tCount
def markCenterExpression(aFN, wigDir, rn=None, tn=None):
extend = 25
timer = bioLibCG.cgTimer()
timer.start()
aNX = cgNexusFlat.Nexus(aFN, cgAlignmentFlat.cgAlignment)
aNX.load(['centerExpression', 'tTcc', 'tStart', 'sLength', 'tELevel'],
[rn, tn])
#load expression of degradome
wigDict = cgWig.loadWigDict(wigDir)
for aID in aNX.centerExpression:
aNX.centerExpression[aID] = [0.0, 0.0, 0.0]
chrom, strand, start, end = bioLibCG.tccSplit(aNX.tTcc[aID])
offset = aNX.tStart[aID]
sLen = aNX.sLength[aID]
if strand == '1':
start = start - extend + offset
end = start + sLen
else:
end = end + extend - offset
start = end - sLen
scanRange = bioLibCG.makeTcc(chrom, strand, start, end)
stretch = cgWig.getExpressionProfile(scanRange, wigDict)
#make sure peak is in the small range
peakLevel = aNX.tELevel[aID]
peakInRange = (peakLevel in stretch.values())
expressionSum = sum(stretch.values())
sortedKeys = stretch.keys()
sortedKeys.sort()
if strand == '-1':
sortedKeys.reverse()
if expressionSum != 0 and peakInRange:
sumE = 0.0
for key in sortedKeys[8:12]:
sumE += stretch[key]
aNX.centerExpression[aID][0] = sumE / expressionSum
sumE = 0.0
for key in sortedKeys[7:13]:
sumE += stretch[key]
aNX.centerExpression[aID][1] = sumE / expressionSum
sumE = 0.0
for key in sortedKeys[6:14]:
sumE += stretch[key]
aNX.centerExpression[aID][2] = sumE / expressionSum
aNX.save()
def testPeaks(degFN, dForm, allGeneInfo, gForm, switchStrand = False):
#load/configure gene Info
gNX = Nexus(allGeneInfo, gForm)
gNX.load(['geneName', 'numReads', 'numSpots'])
gName_numReads = {}
gName_numSpots = {}
while gNX.nextID():
gName_numReads[gNX.geneName] = gNX.numReads
gName_numSpots[gNX.geneName] = gNX.numSpots
#load degFN info
dNX = Nexus(degFN, dForm)
dNX.load(['tcc', 'eLevel', 'geneNames', 'pValBin'])
while dNX.nextID():
gNames, readsForPeak = dNX.geneNames, dNX.eLevel
chrom, strand, start, end = bioLibCG.tccSplit(dNX.tcc)
if switchStrand:
strand = -int(strand)
pVals = []
for gName in gNames:
#may have to change gene name cuz of multiple spans
try:
totGeneReads = gName_numReads[gName]
numSpotsForGene = gName_numSpots[gName]
except KeyError:
try:
gName = gName + '_RE_%s_%s' % (chrom, strand)
totGeneReads = gName_numReads[gName]
numSpotsForGene = gName_numSpots[gName]
except KeyError:
print "FIX THIS GENE NAME", gName
continue
#add psuedocount
totGeneReads += 1
numSpotsForGene += 1 # not sure whether to do this yet...
#check for hidden intron gene overlap
try:
q = 1.0/numSpotsForGene
except ZeroDivisionError:
continue #intron gene
#add p val
pVals.append(binom.sf(readsForPeak, totGeneReads, q))
dNX.pValBin = max(pVals) if pVals else -1.0
dNX.save()
def extendPeakTest(tcc, pRange, minVal, maxAvgNoise, minPeakLength, maxPeakLength, cName):
chrom, strand, peakPosition, end = cg.tccSplit(tcc)
cProfile = stepVectorScan.profileAroundPoint(tcc, pRange, cName, ratio = True)
#extend this peak left and right
leftRange = range(1-pRange, 0)
rightRange = range(1, pRange)
leftRange.reverse() #going from the middle outward
#left
startFinal = leftRange[-1]
for i in leftRange:
if cProfile[i] > minVal:
print ' extending stretch'
else:
print ' end of stretch L'
startFinal = i + 1
break
#right
endFinal = rightRange[-1]
for i in rightRange:
if cProfile[i] > minVal:
print ' extending stretch'
else:
print ' end of stretch R'
endFinal = i - 1
break
peakLength = endFinal - startFinal + 1
#avg expression around peak check...
#get total expression before peak
low = startFinal
high = endFinal
noiseExpression = 0
lowRange = range(1 - pRange, low)
highRange = range(high + 1, pRange)
totalLength = len(lowRange) + len(highRange)
print totalLength, pRange, low, high, lowRange, highRange
for i in lowRange:
noiseExpression += cProfile[i]
for i in highRange:
noiseExpression += cProfile[i]
try:
avgNoise = noiseExpression/float(totalLength)
except:
return False
#filter out peaks that look a certain way.
if (minPeakLength < peakLength < maxPeakLength) and (avgNoise < maxAvgNoise):
goodTcc = cg.makeTcc(chrom, strand, peakPosition + startFinal, peakPosition + endFinal)
print '*KEEPER'
return goodTcc
else:
return False
def updateTypeAlignment(oFN, wigDir, chrom, strand, rn = None, tn = None):
'''This is for ALIGNMENTS...NOT DEG PEAKS!'''
oNX = cgNexusFlat.Nexus(oFN, cgAlignmentFlat.cgAlignment)
oNX.load(['tTcc', 'type'], [rn, tn])
if strand == '1':
strand = '-1'
else:
strand = '1'
print 'loading wig'
coord_types = cgWig.loadSingleWigContext(wigDir, chrom, strand, 'tType')
print 'done loading'
domOrder = ['microRNA_noncoding',
'lincRNA_noncoding',
'longNC_noncoding',
'miRNA_pseudogene_noncoding',
'Mt_rRNA_noncoding',
'Mt_tRNA_noncoding',
'Mt_tRNA_pseudogene_noncoding',
'rRNA_noncoding',
'rRNA_pseudogene_noncoding',
'scRNA_pseudogene_noncoding',
'snoRNA_noncoding',
'snoRNA_pseudogene_noncoding',
'snRNA_noncoding',
'snRNA_pseudogene_noncoding',
'tRNA_pseudogene_noncoding',
'pseudogene_noncoding',
'protein_coding',
'None']
ds = bioLibCG.dominantSpotter(domOrder)
for oID in oNX.tTcc:
oChrom, oStrand, start, end = bioLibCG.tccSplit(oNX.tTcc[oID])
#deg wigs is AS to actual clipping site
if oStrand == '1':
oStrand = '-1'
else:
oStrand = '1'
if oChrom == chrom and oStrand == strand:
tranTypes = coord_types.get(start, 'None').split(',')
types = [x.split(':')[1] if x != 'None' else 'None' for x in tranTypes]
types = list(set(types))
oNX.type[oID] = ds.spotItem(types)
oNX.save()
def countWithBinsSetReads(readFN, binDir, type='INTRON'):
numBins = 100
#create bin sets
c_s_bin_set = {}
for chrom in bioLibCG.humanChromosomes:
for strand in ('1', '-1'):
#initialize data structure
for i in range(0, numBins):
c_s_bin_set.setdefault(chrom, {}).setdefault(strand,
{})[i] = set()
f = open(binDir + '/%s.%s.%s.bins' % (type, chrom, strand), 'r')
for line in f:
ls = line.strip().split('\t')
tccs = ls[1:numBins + 1]
for i in range(0, numBins):
ch, st, sta, end = bioLibCG.tccSplit(tccs[i])
for j in range(sta, end + 1):
c_s_bin_set[chrom][strand][i].add(j)
print 'creating read sets'
#creat read set
c_s_set = {}
for chrom in bioLibCG.humanChromosomes:
c_s_set[chrom] = {}
for strand in ('1', '-1'):
c_s_set[chrom][strand] = set()
f = open(readFN, 'r')
for line in f:
ls = line.strip().split('\t')
strand, chrom, start = ls[1:4]
if chrom not in bioLibCG.humanChromosomes: continue
start = int(start)
if strand == '+':
strand = '-1'
start += 20
else:
strand = '1'
c_s_set[chrom][strand].add(start)
print 'counting'
#make bCounts
binCounts = [0] * numBins
#count for each bin
for i in range(0, numBins):
for chrom in bioLibCG.humanChromosomes:
for strand in ('1', '-1'):
for j in c_s_set[chrom][strand]:
if j in c_s_bin_set[chrom][strand][i]:
binCounts[i] += 1
print '%s\t%s' % (i, binCounts[i])
def countWithBinsSetReads(readFN, binDir, type = 'INTRON'):
numBins = 100
#create bin sets
c_s_bin_set = {}
for chrom in bioLibCG.humanChromosomes:
for strand in ('1', '-1'):
#initialize data structure
for i in range(0, numBins):
c_s_bin_set.setdefault(chrom, {}).setdefault(strand, {})[i] = set()
f = open(binDir + '/%s.%s.%s.bins' % (type, chrom, strand), 'r')
for line in f:
ls = line.strip().split('\t')
tccs = ls[1:numBins + 1]
for i in range(0,numBins):
ch, st, sta, end = bioLibCG.tccSplit(tccs[i])
for j in range(sta, end + 1):
c_s_bin_set[chrom][strand][i].add(j)
print 'creating read sets'
#creat read set
c_s_set = {}
for chrom in bioLibCG.humanChromosomes:
c_s_set[chrom] = {}
for strand in ('1', '-1'):
c_s_set[chrom][strand] = set()
f = open(readFN, 'r')
for line in f:
ls = line.strip().split('\t')
strand, chrom, start = ls[1:4]
if chrom not in bioLibCG.humanChromosomes: continue
start = int(start)
if strand == '+':
strand = '-1'
start += 20
else:
strand = '1'
c_s_set[chrom][strand].add(start)
print 'counting'
#make bCounts
binCounts = [0] * numBins
#count for each bin
for i in range(0, numBins):
for chrom in bioLibCG.humanChromosomes:
for strand in ('1', '-1'):
for j in c_s_set[chrom][strand]:
if j in c_s_bin_set[chrom][strand][i]:
binCounts[i] += 1
print '%s\t%s' % (i, binCounts[i])
def test(tcc, wigDir):
chrom, strand, start, end = bioLibCG.tccSplit(tcc)
print chrom, strand
coord_eLevel = cgWig.loadSingleWig(wigDir, chrom, strand, 'ALL')
sKeys = sorted(coord_eLevel.keys())
for i in range(start, end + 1):
print i, coord_eLevel.get(i, 0)
def getExpressionProfile(tcc, wigDict):
'''assume 1 based'''
chrom, strand, start, end = bioLibCG.tccSplit(tcc)
coord_value = {}
for i in range(start, end + 1):
coord_value[i] = wigDict[chrom][strand].get(i, 0)
return coord_value
def addOne(fN):
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
chrom, strand, start, end = bioLibCG.tccSplit(ls[0])
start += 1
end += 1
print bioLibCG.makeTcc(chrom,strand,start,end)
def getExpressionProfile(tcc, wigDict):
'''assume 1 based'''
chrom, strand, start, end = bioLibCG.tccSplit(tcc)
coord_value = {}
for i in range(start, end + 1):
coord_value[i] = wigDict[chrom][strand].get(i, 0)
return coord_value
def mapStartRangeCheckFunction(val, line):
lineStart = int(line.strip().split('\t')[3])
lineEnd = lineStart + len(line.strip().split('\t')[4])
chrom, strand, start, end = cg.tccSplit(val)
start = int(start)
end = int(end)
if cg.simpleOverlap(start, end, lineStart, lineEnd):
return 0
else:
return -1
def mapStartCheckFunction(val, line):
lineStart = int(line.strip().split('\t')[3])
chrom, strand, start, end = cg.tccSplit(val)
start = int(start)
if start < lineStart:
return -1
elif start > lineStart:
return 1
elif start == lineStart:
return 0
def makeWigMem(fN, assembly, format=None, name=None, directory=None):
'''format assumes bowtie
suitible for small mapped files.'''
if not name: name = cg.getBaseFileName(fN, naked=True)
if not format: format = 'Bowtie'
parserFunction = returnParserFunction(format)
lDict = cg.returnChromLengthDict(assembly)
f = open(fN, 'r')
f.readline() #header...file might not have one but its one read...
#create hitmap of chrom and strand
hitDict = {} #format = chr: { strand : { coord : value
for line in f:
try:
lChrom, lStrand, start, end = cg.tccSplit(parserFunction(line))
except AttributeError:
continue
lStrand = str(lStrand)
start = int(start)
end = int(end)
if lChrom in cg.acceptableChroms:
#wig for degradome
if lStrand == '1':
i = start + 20
else:
i = start
try:
hitDict[lChrom][lStrand][i] += 1
except KeyError:
if lChrom not in hitDict:
hitDict[lChrom] = {}
if lStrand not in hitDict[lChrom]:
hitDict[lChrom][lStrand] = {}
hitDict[lChrom][lStrand][i] = 1
'''
for i in range(start, end):
try:
hitDict[lChrom][lStrand][i] += 1
except KeyError:
if lChrom not in hitDict:
hitDict[lChrom] = {}
if lStrand not in hitDict[lChrom]:
hitDict[lChrom][lStrand] = {}
hitDict[lChrom][lStrand][i] = 1
'''
f.close()
#write results to wig file
writeWigFromHitDict(hitDict, assembly, name, directory)
def mapStartCheckFunction(val, line):
lineStart = int(line.strip().split('\t')[3])
chrom, strand, start, end = cg.tccSplit(val)
start = int(start)
if start < lineStart:
return -1
elif start > lineStart:
return 1
elif start == lineStart:
return 0
def mapStartRangeCheckFunction(val, line):
lineStart = int(line.strip().split('\t')[3])
lineEnd = lineStart + len(line.strip().split('\t')[4])
chrom, strand, start, end = cg.tccSplit(val)
start = int(start)
end = int(end)
if cg.simpleOverlap(start, end, lineStart, lineEnd):
return 0
else:
return -1
def svCoord(tccList, config = None):
'''Given tcc list --> scan Organism wig files and coord:value...
'''
#init
config = c.getConfig(config)
org = config.conf['organism']
wigDir = config.conf['wigSetDir']
wigSetName = config.conf['wigSetName']
splitIntoChroms = config.conf['wigChromSplit']
if splitIntoChroms == 'True':
splitIntoChroms = True
else:
splitIntoChroms = False
coordDict = {} # tcc: [list values]
for tcc in tccList:
chrom, strand, tccStart, tccEnd = cg.tccSplit(tcc)
if splitIntoChroms:
fN = wigDir + '/%s.%s.%s.wig' % (wigSetName, chrom, strand)
else:
fN = wigDir + '/Merge.%s.%s.wig' % (org.lower(), strand)
fIndex = cgIndex.lineIndex(fN, header = True)
fIndex.passCheckFunction(cgIndex.wigCheckFunction)
fIndex.binarySearch(tcc) #places file pointer at beginning of tcc as beginning
stop = False
for line in fIndex.file:
#print 'Line:', line.strip()
lBeg = int(cg.ss(line)[1]) + 1
#print 'lBeg', lBeg
lEnd = int(cg.ss(line)[2])
#print 'lEnd', lEnd
#print '--'
lValue = int(cg.ss(line)[3].split('.')[0])
if tccStart > lBeg:
lBeg = tccStart
if tccEnd < lEnd:
lEnd = tccEnd
stop = True
#print timer.split()
for i in range(lBeg, lEnd + 1):
coordDict[i] = lValue
if stop: break
fIndex.close() #close the file and the index after use...
return coordDict
def peakToSeq(peakFN, extend, outFN):
#extend is +25 for degradome and -6/-4 for oRNA
extend = int(extend)
gf = GenomeFetch.GenomeFetch('hg19')
outF = open(outFN, 'w')
f = open(peakFN, 'r')
for line in f:
ls = line.strip().split('\t')
chrom, strand, start, end = bioLibCG.tccSplit(ls[0])
start, end = start - extend, end + extend
newTcc = bioLibCG.makeTcc(chrom, strand, start, end)
outF.write(gf.getSequence(newTcc) + '\n')
