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 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 updateELevel(oFN, wigDir, rn=None, tn=None):
oNX = cgNexusFlat.Nexus(oFN, degPeak.degPeak)
oNX.load(['tcc', 'eLevel'], [rn, tn])
wigDict = cgWig.loadWigDict(wigDir)
for oID in oNX.eLevel:
expandTcc = bioLibCG.expandTcc(oNX.tcc[oID], 3)
coord_value = cgWig.getExpressionProfile(expandTcc, wigDict)
oNX.eLevel[oID] = max(coord_value.values())
oNX.save()
def updateELevel2(dFN, dForm, wigDir):
'''Dont need to do it by chromosome because it is small enough'''
'''Also dont need to flip the strand because the wig is opposite as well'''
NX = Nexus(dFN, dForm)
NX.load(['tcc', 'eLevel'])
wigDict = cgWig.loadWigDictFloat(wigDir)
while NX.nextID():
coord_value = cgWig.getExpressionProfile(NX.tcc, wigDict)
NX.eLevel = max(coord_value.values())
NX.save()
def updateELevel(oFN, wigDir, rn = None, tn = None):
'''Dont need to do it by chromosome because it is small enough'''
'''Also dont need to flip the strand because the wig is opposite as well'''
oNX = cgNexusFlat.Nexus(oFN, cgOriginRNAFlat.OriginRNA)
oNX.load(['tcc', 'eLevel'], [rn, tn])
wigDict = cgWig.loadWigDict(wigDir)
for oID in oNX.eLevel:
coord_value = cgWig.getExpressionProfile(oNX.tcc[oID], wigDict)
oNX.eLevel[oID] = max(coord_value.values())
oNX.save()
def updateELevel2(dFN, dForm, wigDir):
'''Dont need to do it by chromosome because it is small enough'''
'''Also dont need to flip the strand because the wig is opposite as well'''
NX = Nexus(dFN, dForm)
NX.load(['tcc', 'eLevel'])
wigDict = cgWig.loadWigDictFloat(wigDir)
while NX.nextID():
coord_value = cgWig.getExpressionProfile(NX.tcc, wigDict)
NX.eLevel = max(coord_value.values())
NX.save()
def updateELevel(oFN, wigDir, rn = None, tn = None):
'''Dont need to do it by chromosome because it is small enough'''
'''Also dont need to flip the strand because the wig is opposite as well'''
oNX = cgNexusFlat.Nexus(oFN, cgOriginRNAFlat.OriginRNA)
oNX.load(['tcc', 'eLevel'], [rn, tn])
wigDict = cgWig.loadWigDict(wigDir)
for oID in oNX.eLevel:
coord_value = cgWig.getExpressionProfile(oNX.tcc[oID], wigDict)
oNX.eLevel[oID] = max(coord_value.values())
oNX.save()
def updateELevel(oFN, wigDir, rn = None, tn = None):
oNX = cgNexusFlat.Nexus(oFN, degPeak.degPeak)
oNX.load(['tcc', 'eLevel'], [rn, tn])
wigDict = cgWig.loadWigDict(wigDir)
for oID in oNX.eLevel:
expandTcc = bioLibCG.expandTcc(oNX.tcc[oID], 3)
coord_value = cgWig.getExpressionProfile(expandTcc, wigDict)
oNX.eLevel[oID] = max(coord_value.values())
oNX.save()
def getPairInfo2(oFN, aFN, oWigDir, aWigDir, outFN, assembly):
oNX = cgNexusFlat.Nexus(oFN, cgOriginRNAFlat.OriginRNA)
oNX.load(['tcc', 'filteredTargets'])
aNX = cgNexusFlat.Nexus(aFN, cgAlignmentFlat.cgAlignment)
aNX.load(['tTcc', 'tStart', 'tEnd', 'tLength', 'mismatchPositions'])
#get expression Dicts for alignments and small RNAs
oWigDict = cgWig.loadWigDict(oWigDir)
aWigDict = cgWig.loadWigDict(aWigDir)
myG = gf.GenomeFetch(assembly)
fOut = open(outFN, 'w')
for oID in oNX.tcc:
for aID in oNX.filteredTargets[oID]:
#expand the oTcc to fit the target
oChrom, oStrand, oStart, oEnd = bioLibCG.tccSplit(oNX.tcc[oID])
if oStrand == '1':
oStart -= aNX.tStart[aID] #5'
oEnd += (aNX.tLength[aID] - 1) - aNX.tEnd[aID] #3'
oTcc = bioLibCG.makeTcc(oChrom, oStrand, oStart, oEnd)
elif oStrand == '-1':
oStart -= (aNX.tLength[aID] - 1) - aNX.tEnd[aID] #3'
oEnd += aNX.tStart[aID] #5'
oTcc = bioLibCG.makeTcc(oChrom, oStrand, oStart, oEnd)
#expand the peak tcc to be the target tcc (peak is only 1-4nt long)
aChrom, aStrand, aStart, aEnd = bioLibCG.tccSplit(aNX.tTcc[aID])
aStart -= 25
aEnd += 25
aTcc = bioLibCG.makeTcc(aChrom, aStrand, aStart, aEnd)
#get expression
oCoord_value = cgWig.getExpressionProfile(oTcc, oWigDict)
aCoord_value = cgWig.getExpressionProfile(aTcc, aWigDict)
oValues = [oCoord_value[x] for x in sorted(oCoord_value.keys())]
aValues = [aCoord_value[x] for x in sorted(aCoord_value.keys())]
if oStrand == '-1':
oValues.reverse()
if aStrand == '-1':
aValues.reverse()
#get sequences
aSeq = myG.getSequence(aTcc)
oSeq = myG.getSequence(oTcc)
pString = [str(oID),
str(aID),
','.join([str(x) for x in oValues]),
','.join([str(x) for x in aValues]),
str(aNX.tStart[aID]),
str(aNX.tEnd[aID]),
oNX.tcc[oID],
aNX.tTcc[aID],
oSeq,
aSeq,
','.join([str(x) for x in aNX.mismatchPositions[aID]])]
fOut.write('\t'.join(pString) + '\n')
fOut.close()
def getPairInfo(oFN, aFN, oWigDir, aWigDir, outFN, assembly):
oNX = cgNexusFlat.Nexus(oFN, cgOriginRNAFlat.OriginRNA)
oNX.load(['tcc', 'filteredTargets', 'sequence'])
aNX = cgNexusFlat.Nexus(aFN, cgAlignmentFlat.cgAlignment)
aNX.load(['tTcc', 'tStart', 'tEnd', 'tLength', 'targetSequence'])
#get expression Dicts for alignments and small RNAs
oWigDict = cgWig.loadWigDict(oWigDir)
aWigDict = cgWig.loadWigDict(aWigDir)
myG = gf.GenomeFetch(assembly)
fOut = open(outFN, 'w')
for oID in oNX.tcc:
for aID in oNX.filteredTargets[oID]:
#expand the oTcc to fit the target
oChrom, oStrand, oStart, oEnd = bioLibCG.tccSplit(oNX.tcc[oID])
if oStrand == '-1':
oStart -= (aNX.tLength[aID] - 1) - aNX.tEnd[aID]
oEnd += aNX.tStart[aID]
oTcc = bioLibCG.makeTcc(oChrom, oStrand, oStart, oEnd)
elif oStrand == '1':
oStart -= aNX.tStart[aID]
oEnd += (aNX.tLength[aID] - 1) - aNX.tEnd[aID]
oTcc = bioLibCG.makeTcc(oChrom, oStrand, oStart, oEnd)
#expand the peak tcc to be the target tcc (peak is only 1-4nt long)
aChrom, aStrand, aStart, aEnd = bioLibCG.tccSplit(aNX.tTcc[aID])
aStart -= 25
aEnd += 25
aTcc = bioLibCG.makeTcc(aChrom, aStrand, aStart, aEnd)
print oNX.tcc[oID], oTcc, aNX.tTcc[aID], aTcc
#get expression
oCoord_value = cgWig.getExpressionProfile(oTcc, oWigDict)
aCoord_value = cgWig.getExpressionProfile(aTcc, aWigDict)
for aCoord in sorted(aCoord_value.keys()):
print aCoord, aCoord_value[aCoord]
oValues = [oCoord_value[x] for x in sorted(oCoord_value.keys())]
aValues = [aCoord_value[x] for x in sorted(aCoord_value.keys())]
maxO = max(oValues[aNX.tStart[aID]:aNX.tEnd[aID]]) # might want to change to max only within the small range
maxA = max(aValues)
oValues = [(float(x)/maxO) * 20 for x in oValues]
aValues = [(float(x)/maxA) * 50 for x in aValues]
#get sequence
aSeq = myG.getSequence(aTcc)
oSeq = myG.getSequence(oTcc)
letter_value = zip(aSeq, aValues)
for letter, value in letter_value:
print letter, value
pString = [str(oID), str(aID), ','.join([str(x) for x in oValues]), ','.join([str(x) for x in aValues]), str(maxO), str(maxA), str(aNX.tStart[aID]), str(aNX.tEnd[aID]), oNX.tcc[oID], aNX.tTcc[aID], oSeq, aSeq]
fOut.write('\t'.join(pString) + '\n')
fOut.close()
def getPairInfo2(oFN, aFN, oWigDir, aWigDir, outFN, assembly):
oNX = cgNexusFlat.Nexus(oFN, cgOriginRNAFlat.OriginRNA)
oNX.load(['snrSS', 'tcc', 'filteredTargets', 'context'])
aNX = cgNexusFlat.Nexus(aFN, cgAlignmentFlat.cgAlignment)
aNX.load(
['tTcc', 'tStart', 'tEnd', 'tLength', 'mismatchPositions', 'context'])
#get expression Dicts for alignments and small RNAs
oWigDict = cgWig.loadWigDict(oWigDir)
aWigDict = cgWig.loadWigDict(aWigDir)
myG = gf.GenomeFetch(assembly)
fOut = open(outFN, 'w')
for oID in oNX.tcc:
for aID in oNX.filteredTargets[oID]:
#expand the oTcc to fit the target
oChrom, oStrand, oStart, oEnd = bioLibCG.tccSplit(oNX.tcc[oID])
if oStrand == '1':
oStart -= aNX.tStart[aID] #5'
oEnd += (aNX.tLength[aID] - 1) - aNX.tEnd[aID] #3'
oTcc = bioLibCG.makeTcc(oChrom, oStrand, oStart, oEnd)
elif oStrand == '-1':
oStart -= (aNX.tLength[aID] - 1) - aNX.tEnd[aID] #3'
oEnd += aNX.tStart[aID] #5'
oTcc = bioLibCG.makeTcc(oChrom, oStrand, oStart, oEnd)
#expand the peak tcc to be the target tcc (peak is only 1-4nt long)
aChrom, aStrand, aStart, aEnd = bioLibCG.tccSplit(aNX.tTcc[aID])
aStart -= 25
aEnd += 25
aTcc = bioLibCG.makeTcc(aChrom, aStrand, aStart, aEnd)
#get expression
oCoord_value = cgWig.getExpressionProfile(oTcc, oWigDict)
aCoord_value = cgWig.getExpressionProfile(aTcc, aWigDict)
oValues = [oCoord_value[x] for x in sorted(oCoord_value.keys())]
aValues = [aCoord_value[x] for x in sorted(aCoord_value.keys())]
if oStrand == '-1':
oValues.reverse()
if aStrand == '-1':
aValues.reverse()
#get sequences
aSeq = myG.getSequence(aTcc)
oSeq = myG.getSequence(oTcc)
oSNR = oNX.snrSS[oID]
oContext = oNX.context[oID]
aContext = aNX.context[aID]
pString = [
str(oID),
str(aID), ','.join([str(x) for x in oValues]),
','.join([str(x) for x in aValues]),
str(aNX.tStart[aID]),
str(aNX.tEnd[aID]), oNX.tcc[oID], aNX.tTcc[aID], oSeq, aSeq,
','.join([str(x) for x in aNX.mismatchPositions[aID]]),
str(oSNR), oContext, aContext
]
fOut.write('\t'.join(pString) + '\n')
fOut.close()