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 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()
import getHairpins
import cgGenes
import cgConfig as c
import bioLibCG as cg
mConf = c.getConfig('Main.conf')
geneSetFolder = mConf.conf['geneSetsHuman']
fN = '/home/chrisgre/projects/NoncodingHuman/results/NChuman-s3k8b17.results.sorted.introns.sorted'
cHairs = getHairpins.getHairpins(fN)
ensGenes = cgGenes.createGeneSetFromFile(geneSetFolder + '/ensemblAllTranscripts.tsv')
cDesc = {} #CID:gDesc
for CID in cHairs:
tcc = cHairs[CID]
cDesc[CID] = "NONE"
overlappingGenes = ensGenes.geneOverlaps([tcc])
if len(overlappingGenes) > 0:
print overlappingGenes[0].type
cDesc[CID] = overlappingGenes[0].type
f = open(fN, 'r')
newLines = []
for line in f:
CID = line.strip().split('\t')[7]
newLines.append(cg.appendToLine(line, cDesc[CID], 16))
f.close()
import cgGenes
import compareData as compare
import cgConfig as c
cName = 'mm9.conf'
mConf = c.getConfig('Main.conf')
conf = c.getConfig(cName)
organism = conf.conf['organism']
geneSetFolder = mConf.conf['geneSets%s' % organism]
genes = cgGenes.createGeneSetFromFile(geneSetFolder + '/allTransciptsType.tsv')
peakTccs = compare.tccFileToList('peakData.500.mm9', 0)
tOverlaps = genes.transcriptOverlaps(peakTccs)
typeDict = {}
for transcript in tOverlaps:
if transcript.type not in typeDict:
typeDict[transcript.type] = 1
else:
typeDict[transcript.type] += 1
#count the amounts of each type for each transcript
amount = {}
for gene in genes.genes:
for t in gene.transcripts:
if t.type in amount:
amount[t.type] += 1
else:
amount[t.type] = 1
print 'Total Peaks:', len(peakTccs)
import getHairpins
import cgGenes
import cgConfig as c
import bioLibCG as cg
mConf = c.getConfig('Main.conf')
geneSetFolder = mConf.conf['geneSetsHuman']
fN = '/home/chrisgre/projects/NoncodingHuman/results/NChuman-s3k8b17.results.sorted.introns.sorted'
cHairs = getHairpins.getHairpins(fN)
ensGenes = cgGenes.createGeneSetFromFile(geneSetFolder +
'/ensemblAllTranscripts.tsv')
cDesc = {} #CID:gDesc
for CID in cHairs:
tcc = cHairs[CID]
cDesc[CID] = "NONE"
overlappingGenes = ensGenes.geneOverlaps([tcc])
if len(overlappingGenes) > 0:
print overlappingGenes[0].type
cDesc[CID] = overlappingGenes[0].type
f = open(fN, 'r')
newLines = []
for line in f:
CID = line.strip().split('\t')[7]
newLines.append(cg.appendToLine(line, cDesc[CID], 16))
f.close()
def exonNoisy(cName = None):
#init
mConf = c.cgConfig('Main.conf')
conf = c.getConfig(cName)
cHairs = getHairpins.getHairpins(conf.conf['resultsExons']) #CID: HAIRPIN
organism = conf.conf['organism']
geneSetFolder = mConf.conf['geneSets%s' % organism]
#make prediction overlap hitmap
print 'Making prediction list'
predList = []
for CID in cHairs:
hPin = cHairs[CID]
predList.append(hPin)
if compare.checkIfOverlaps(predList):
predList = compare.collapseOverlaps(predList)
#make genes for Ensemble/make list of tccs for exons.
print 'Creating gene set'
ensGenes = cgGenes.createGeneSetFromFile(geneSetFolder + '/ensemblAllExons.tsv')
print ' loaded # genes:', len(ensGenes.set)
#collect levels for each haipin region
print '[Checking all levels]'
cidLevels = {}
for CID in cHairs:
print CID
hPin = cHairs[CID]
#for each hairpin, --> find overlapping transcripts in same gene
overlappingGenes = ensGenes.geneOverlaps([hPin])
if len(overlappingGenes) > 0:
gIDs = [gene.id for gene in overlappingGenes]
allTccs = ensGenes.getTccsFromGIDs(gIDs)
if compare.checkIfOverlaps:
print ' Overlaps...collapsing'
allTccs = compare.collapseOverlaps(allTccs)
else:
print 'NO GENE OVERLAPS!!!!!', CID, hPin
#filter out my predictions.
print ' Filtering out predictions'
checkList = compare.subtractTwoTccLists(allTccs, predList)
#Get Expression level for gene.
print ' Retrieving Expression levels:', cg.getTccListTotalLength(checkList)
levels = []
hPinLevels = stepVectorScan.scanVectorsHist(checkList, cName)
for hPin in hPinLevels:
levels.extend(hPinLevels[hPin])
cidLevels[CID] = levels
#output levels to file
print 'Outputting to file'
#find longest
longest = 0
for CID in cidLevels:
length = len(cidLevels[CID])
if length > longest:
longest = length
sortedKeys = cidLevels.keys()
sortedKeys.sort()
#print sortedKeys
newLines = []
for j in range(0, longest): #how many lines are there
newLine = []
for CID in sortedKeys:
if len(cidLevels[CID]) > j:# add it
newLine.append(str(cidLevels[CID][j]))
else:
newLine.append('NA')
newLines.append('\t'.join(newLine) + '\n')
outFileN = conf.conf['exonNoiseData']
outFile = open(outFileN, 'w')
outFile.write('\t'.join(sortedKeys) + '\n')
outFile.writelines(newLines)
outFile.close()