def transcriptSetOverlap(aDir, AS):
AS = bool(AS)
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
oRNA_DC = cgNexusFlat.dataController(aDir, cgOriginRNA.OriginRNA)
id_oRNA = oRNA_DC.load()
if AS == True:
degTccs = [cg.convertToAS(x.tcc) for x in id_oRNA.values()]
else:
degTccs = [x.tcc for x in id_oRNA.values()]
#find all overlapping exons/transcripts, then all results sequences that overlap exons
overlappingExons = allExons.transcriptOverlaps(degTccs)
#print len(overlappingExons), "num of overlapping exons"
overlappingExonTccs = [x.tcc for x in overlappingExons]
overlappingDegTccs = compare.compareTwoTcc(degTccs, overlappingExonTccs, 1)
#write new file
for obj in id_oRNA.values():
if AS:
degTcc = cg.convertToAS(obj.tcc)
else:
degTcc = obj.tcc
if degTcc in overlappingDegTccs:
obj.transcriptOverlap = True
else:
obj.transcriptOverlap = False
oRNA_DC.commit(id_oRNA)
def transcriptSetOverlap(aDir, AS):
AS = bool(AS)
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
oRNA_DC = cgDB.dataController(aDir, cgOriginRNA.OriginRNA)
id_oRNA = oRNA_DC.load()
if AS == True:
degTccs = [cg.convertToAS(x.tcc) for x in id_oRNA.values()]
else:
degTccs = [x.tcc for x in id_oRNA.values()]
#find all overlapping exons/transcripts, then all results sequences that overlap exons
overlappingExons = allExons.transcriptOverlaps(degTccs)
#print len(overlappingExons), "num of overlapping exons"
overlappingExonTccs = [x.tcc for x in overlappingExons]
overlappingDegTccs = compare.compareTwoTcc(degTccs, overlappingExonTccs, 1)
#write new file
for obj in id_oRNA.values():
if AS:
degTcc = cg.convertToAS(obj.tcc)
else:
degTcc = obj.tcc
if degTcc in overlappingDegTccs:
obj.transcriptOverlap = True
else:
obj.transcriptOverlap = False
oRNA_DC.commit(id_oRNA)
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 profileTargetsHistoAS(tccList, cName, name='boxplot'):
range = 50
histDict = {} # {coord: []}
histDictAS = {}
for tcc in tccList:
chrom, strand, start, end = cg.tccSplit(tcc)
#Get highest peak (sense)
tccStretch = cgPeaks.stretch(tcc, cName)
tccStretch.createPeaks(span=2)
highestCoord = tccStretch.getHighestPeak()
if highestCoord == None: continue
#AS
tccAS = cg.convertToAS(tcc)
tccStretch = cgPeaks.stretch(tccAS, cName)
tccStretch.createPeaks(span=2)
highestCoordAS = tccStretch.getHighestPeak()
if highestCoordAS == None: continue
#profile around point (Sense)
zPoint = cg.makeTcc(chrom, strand, highestCoord, end)
cProfile = svs.profileAroundPoint(zPoint, range, cName, ratio=True)
for coord in cProfile:
try:
histDict[coord].append(cProfile[coord])
except: #quicker way to initialize
histDict[coord] = [cProfile[coord]]
#profile around point (AS)
zPoint = cg.convertToAS(zPoint)
cProfile = svs.profileAroundPoint(zPoint,
range,
cName,
ratio=True,
ratioCoord=highestCoordAS)
for coord in cProfile:
try:
histDictAS[coord].append(cProfile[coord])
except: #quicker way to initialize
histDictAS[coord] = [cProfile[coord]]
plot.boxPlotHistoAS(histDict, histDictAS, name=name)
def profileTargetsHistoAS(tccList, cName, name = 'boxplot'):
range = 50
histDict = {} # {coord: []}
histDictAS = {}
for tcc in tccList:
chrom, strand, start, end = cg.tccSplit(tcc)
#Get highest peak (sense)
tccStretch = cgPeaks.stretch(tcc, cName)
tccStretch.createPeaks(span = 2)
highestCoord = tccStretch.getHighestPeak()
if highestCoord == None: continue
#AS
tccAS = cg.convertToAS(tcc)
tccStretch = cgPeaks.stretch(tccAS, cName)
tccStretch.createPeaks(span = 2)
highestCoordAS = tccStretch.getHighestPeak()
if highestCoordAS == None: continue
#profile around point (Sense)
zPoint = cg.makeTcc(chrom, strand, highestCoord, end)
cProfile = svs.profileAroundPoint(zPoint, range, cName, ratio = True)
for coord in cProfile:
try:
histDict[coord].append(cProfile[coord])
except: #quicker way to initialize
histDict[coord] = [cProfile[coord]]
#profile around point (AS)
zPoint = cg.convertToAS(zPoint)
cProfile = svs.profileAroundPoint(zPoint, range, cName, ratio = True, ratioCoord = highestCoordAS)
for coord in cProfile:
try:
histDictAS[coord].append(cProfile[coord])
except: #quicker way to initialize
histDictAS[coord] = [cProfile[coord]]
plot.boxPlotHistoAS(histDict, histDictAS, name = name)
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 transcriptSetOverlapDegFile(degFile):
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
degTccs = []
f = open(degFile, 'r')
for line in f:
ls = line.strip().split('\t')
degTccs.append(ls[1])
f.close()
degTccs = [cg.convertToAS(x) for x in degTccs]
#find all overlapping exons/transcripts, then all results sequences that overlap exons
overlappingExons = allExons.transcriptOverlaps(degTccs)
#print len(overlappingExons), "num of overlapping exons"
overlappingExonTccs = [x.tcc for x in overlappingExons]
overlappingDegTccs = compare.compareTwoTcc(degTccs, overlappingExonTccs, 1)
f = open(degFile, 'r')
newLines = []
for line in f:
degTcc = cg.convertToAS(ls[1])
inTran = '0'
if degTcc in overlappingDegTccs:
inTran = '1'
#update newLines
newLine = cg.appendToLine(line, inTran, 3)
f.close()
def transcriptSetOverlapDegFile(degFile):
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
degTccs = []
f = open(degFile, 'r')
for line in f:
ls = line.strip().split('\t')
degTccs.append(ls[1])
f.close()
degTccs = [cg.convertToAS(x) for x in degTccs]
#find all overlapping exons/transcripts, then all results sequences that overlap exons
overlappingExons = allExons.transcriptOverlaps(degTccs)
#print len(overlappingExons), "num of overlapping exons"
overlappingExonTccs = [x.tcc for x in overlappingExons]
overlappingDegTccs = compare.compareTwoTcc(degTccs, overlappingExonTccs, 1)
f = open(degFile, 'r')
newLines = []
for line in f:
degTcc = cg.convertToAS(ls[1])
inTran = '0'
if degTcc in overlappingDegTccs:
inTran = '1'
#update newLines
newLine = cg.appendToLine(line, inTran, 3)
f.close()