def makeTypeWig(tranFN, wigDir, chrom, strand, species):
'''Using 14th column in transcripts for type info...might want to use something different?'''
coord_id = {}
f = open(tranFN, 'r')
for line in f:
ls = line.strip().split('\t')
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
if tChrom != chrom or tStrand != strand:
continue
tID = ls[0]
tType = ls[13]
tStart, tEnd = int(ls[3]), int(
ls[4]
) - 1 #0BASE CONVERSION !!! it might have to be 0BASE for making wig...?
tIDType = '%s:%s' % (tID, tType)
for i in xrange(tStart, tEnd + 1):
coord_id[i] = coord_id.get(i, '') + '%s ' % tIDType
#unique, string
for i, ids in coord_id.iteritems():
coord_id[i] = ','.join([x for x in set(ids.strip().split(' '))])
#write wig to file
writeWigDictToWig(coord_id, chrom, strand, species, 'tType', wigDir,
'None')
def makeTranscriptome(tranFN, outFN):
p = bioLibCG.cgPrint()
p.show = False
gf = GenomeFetch.GenomeFetch('hg19')
fOut = open(outFN, 'w')
f = open(tranFN, 'r')
for line in f:
ls = line.strip().split('\t')
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
exonStarts = [int(x) + 1 for x in ls[8][:-1].split(',')]
exonEnds = [int(x) for x in ls[9][:-1].split(',')]
exonPairs = zip(exonStarts, exonEnds)
tID = ls[0]
gID = ls[10]
seqList = []
for eStart, eEnd in exonPairs:
tcc = bioLibCG.makeTcc(tChrom, tStrand, eStart, eEnd)
seqList.append(gf.getSequence(tcc))
mRNA = ''.join(seqList)
#reverse direction if negative strand
if tStrand == '-1':
mRNA = mRNA[::-1]
fOut.write('> %s:%s:%s\n' % (tID, gID, len(mRNA)))
fOut.write(mRNA + '\n\n')
fOut.close()
f.close()
def makeTranscriptome(tranFN, outFN):
p = bioLibCG.cgPrint()
p.show = False
gf = GenomeFetch.GenomeFetch('hg19')
fOut = open(outFN, 'w')
f = open(tranFN, 'r')
for line in f:
ls = line.strip().split('\t')
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
exonStarts = [int(x) + 1 for x in ls[8][:-1].split(',')]
exonEnds = [int(x) for x in ls[9][:-1].split(',')]
exonPairs = zip(exonStarts, exonEnds)
tID = ls[0]
gID = ls[10]
seqList = []
for eStart, eEnd in exonPairs:
tcc = bioLibCG.makeTcc(tChrom, tStrand, eStart, eEnd)
seqList.append(gf.getSequence(tcc))
mRNA = ''.join(seqList)
#reverse direction if negative strand
if tStrand == '-1':
mRNA = mRNA[::-1]
fOut.write('> %s:%s:%s\n' % (tID, gID, len(mRNA)))
fOut.write(mRNA + '\n\n')
fOut.close()
f.close()
def convertEnsemblBiomart(fN, outFN):
fOut = open(outFN, 'w')
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
chrom = 'chr' + ls[1]
strand = bioLibCG.switchStrandFormat(ls[2])
ls[5], ls[6] = ls[4], ls[4]
numBlocks = 1
eStarts = ls[3] + ','
eEnds = ls[4] + ','
cs, ce = 'none', 'none'
tType = ls[8] + '_noncoding'
intType = 'None'
gType = 'noncoding_noncoding'
pString = [
ls[0], chrom, strand, ls[3], ls[4], ls[5], ls[6], numBlocks,
eStarts, eEnds, ls[7], cs, ce, tType, intType, gType
]
pString = [str(x) for x in pString]
pString = '\t'.join(pString)
fOut.write(pString + '\n')
f.close()
fOut.close()
def get3UTRFromTranscriptome(tranFN, outFN, wholeGene = False ):
fOut = open(outFN, 'w')
f = open(tranFN, 'r')
for i, line in enumerate(f):
ls = line.strip().split('\t')
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
tStart, tEnd = int(ls[3]), int(ls[4]) - 1
cStart, cEnd = int(ls[5]), int(ls[6]) - 1
if wholeGene:
utrTcc = bioLibCG.makeTcc(tChrom, tStrand, tStart, tEnd)
fOut.write('%s\n' % utrTcc)
continue
#5UTR
if tStrand == '1':
range5 = (tStart, cStart - 1)
else:
range5 = (cEnd + 1, tEnd)
#3UTR
if tStrand == '1':
range3 = (cEnd + 1, tEnd)
else:
range3 = (tStart, cStart - 1)
utrTcc = bioLibCG.makeTcc(tChrom, tStrand, range3[0], range3[1])
fOut.write('%s\n' % utrTcc)
f.close()
fOut.close()
def makeTypeWig(tranFN, wigDir, chrom, strand, species):
'''Using 14th column in transcripts for type info...might want to use something different?'''
coord_id = {}
f = open(tranFN, 'r')
for line in f:
ls = line.strip().split('\t')
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
if tChrom != chrom or tStrand != strand:
continue
tID = ls[0]
tType = ls[13]
tStart, tEnd = int(ls[3]), int(ls[4]) - 1 #0BASE CONVERSION !!! it might have to be 0BASE for making wig...?
tIDType = '%s:%s' % (tID, tType)
for i in xrange(tStart, tEnd + 1):
coord_id[i] = coord_id.get(i, '') + '%s ' % tIDType
#unique, string
for i, ids in coord_id.iteritems():
coord_id[i] = ','.join([x for x in set(ids.strip().split(' '))])
#write wig to file
writeWigDictToWig(coord_id, chrom, strand, species, 'tType', wigDir, 'None')
def makeGeneWig(tranFN, wigDir, chrom, strand):
coord_id = {}
f = open(tranFN, 'r')
for line in f:
ls = line.strip().split('\t')
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
if tChrom != chrom or tStrand != strand:
continue
gID = ls[10]
#gID = gID.replace(" ", "_")
tStart, tEnd = int(ls[3]), int(ls[4]) - 1
for i in xrange(tStart, tEnd + 1):
coord_id[i] = coord_id.get(
i, '') + '%s$' % gID #$ is used because of spaces
#unique, string
for i, ids in coord_id.iteritems():
coord_id[i] = ','.join(
[x for x in set(ids.strip().split('$')) if x != ''])
#write wig to file
writeWigDictToWig(coord_id, chrom, strand, 'hg19', 'ALL', wigDir, 'None')
def convertEnsemblBiomart(fN, outFN):
fOut = open(outFN, 'w')
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
chrom = 'chr' + ls[1]
strand = bioLibCG.switchStrandFormat(ls[2])
ls[5], ls[6] = ls[4], ls[4]
numBlocks = 1
eStarts = ls[3] + ','
eEnds = ls[4] + ','
cs, ce = 'none', 'none'
tType = ls[8] + '_noncoding'
intType = 'None'
gType = 'noncoding_noncoding'
pString = [ls[0], chrom, strand, ls[3], ls[4], ls[5], ls[6], numBlocks, eStarts, eEnds, ls[7], cs, ce, tType, intType, gType]
pString = [str(x) for x in pString]
pString = '\t'.join(pString)
fOut.write(pString + '\n')
f.close()
fOut.close()
def getTccs(fN):
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
chrom, strand, start, end = ls[0], bioLibCG.switchStrandFormat(ls[5]), ls[1], ls[2]
print bioLibCG.makeTcc(chrom, strand, start, end)
def getTccs(fN):
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
chrom, strand, start, end = ls[0], bioLibCG.switchStrandFormat(
ls[5]), ls[1], ls[2]
print bioLibCG.makeTcc(chrom, strand, start, end)
def collectIDs2(fN, fN2, fN3):
'''Used for getting the # repeat reads on target results'''
idSet = set()
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
for id in ls[9].split(','):
idSet.add(id)
idSetDeg = set()
f = open(fN2, 'r')
for line in f:
ls = line.strip().split('\t')
if ls[18] == "F": continue
if ls[0] in idSet:
idSetDeg.add(ls[11])
chrom_strand_coord = getHitMap(list(idSetDeg))
#print 'target Tccs', len(idSetDeg)
#print chrom_strand_coord['chr1']['1']
eSet = set()
eDict = {}
readNames = set()
f = open(fN3, 'r')
for line in f:
ls = line.strip().split('\t')
chrom, strand, start = ls[2], ls[1], int(ls[3])
strand = bioLibCG.switchStrandFormat(strand)
for i in range(start - 3, start + 3):
try:
if i in chrom_strand_coord[chrom][strand]:
readNames.add(ls[0])
break
except KeyError:
continue
f.close()
#now go back through and count the times the read appears
readName_count = {}
f = open(fN3, 'r')
for line in f:
ls = line.strip().split('\t')
if ls[0] in readNames:
readName_count[ls[0]] = readName_count.get(ls[0], 0) + 1
for read, count in readName_count.iteritems():
print '%s\t%s' % (read, count)
def collectIDs2(fN, fN2, fN3):
'''Used for getting the # repeat reads on target results'''
idSet = set()
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
for id in ls[9].split(','):
idSet.add(id)
idSetDeg = set()
f = open(fN2, 'r')
for line in f:
ls = line.strip().split('\t')
if ls[18] == "F": continue
if ls[0] in idSet:
idSetDeg.add(ls[11])
chrom_strand_coord = getHitMap(list(idSetDeg))
#print 'target Tccs', len(idSetDeg)
#print chrom_strand_coord['chr1']['1']
eSet = set()
eDict = {}
readNames = set()
f = open(fN3, 'r')
for line in f:
ls = line.strip().split('\t')
chrom, strand, start = ls[2], ls[1], int(ls[3])
strand = bioLibCG.switchStrandFormat(strand)
for i in range(start - 3, start + 3):
try:
if i in chrom_strand_coord[chrom][strand]:
readNames.add(ls[0])
break
except KeyError:
continue
f.close()
#now go back through and count the times the read appears
readName_count = {}
f = open(fN3, 'r')
for line in f:
ls = line.strip().split('\t')
if ls[0] in readNames:
readName_count[ls[0]] = readName_count.get(ls[0], 0) + 1
for read, count in readName_count.iteritems():
print '%s\t%s' % (read, count)
def getCoords(fN, rn = None, tn = None):
oNX = cgNexusFlat.Nexus(fN, cgOriginRNAFlat.OriginRNA)
oNX.load(['tcc'], [rn, tn])
for id in oNX.tcc:
chrom, strand, start, end = bioLibCG.tccSplit(oNX.tcc[id])
name = 'None'
score = '0'
strand = bioLibCG.switchStrandFormat(strand)
thickStart = start
thickEnd = end
pString = [str(x) for x in [chrom, start, end, name, score, strand, thickStart, thickEnd]]
print '\t'.join(pString)
def getCoords(fN, rn=None, tn=None):
oNX = cgNexusFlat.Nexus(fN, cgOriginRNAFlat.OriginRNA)
oNX.load(['tcc'], [rn, tn])
for id in oNX.tcc:
chrom, strand, start, end = bioLibCG.tccSplit(oNX.tcc[id])
name = 'None'
score = '0'
strand = bioLibCG.switchStrandFormat(strand)
thickStart = start
thickEnd = end
pString = [
str(x) for x in
[chrom, start, end, name, score, strand, thickStart, thickEnd]
]
print '\t'.join(pString)
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 makeTranscriptWig(tranFN, wigDir, chrom, strand, species = 'hg19'):
coord_id = {}
f = open(tranFN, 'r')
for line in f:
ls = line.strip().split('\t')
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
if tChrom != chrom or tStrand != strand:
continue
tID = ls[0]
tStart, tEnd = int(ls[3]), int(ls[4]) - 1
for i in xrange(tStart, tEnd + 1):
coord_id[i] = coord_id.get(i, '') + '%s ' % tID
#unique, string
for i, ids in coord_id.iteritems():
coord_id[i] = ','.join([x for x in set(ids.strip().split(' '))])
#write wig to file
writeWigDictToWig(coord_id, chrom, strand, species, 'transcript', wigDir, 'None')
def updateMultipleTccs(oDir, mappedFN):
#parse bowtie
f = open(mappedFN, 'r')
oID_tccs = {}
for line in f:
ls = line.strip().split('\t')
oID = int(ls[0])
strand, chrom, firstCoord = bioLibCG.switchStrandFormat(ls[1]), ls[2], int(ls[3])
secondCoord = firstCoord + len(ls[4]) - 1
tcc = bioLibCG.makeTcc(chrom, strand, firstCoord, secondCoord)
oID_tccs.setdefault(oID, []).append(tcc)
oDC = cgDB.dataController(oDir, cgOriginRNA.OriginRNA)
id_oRNA = oDC.load()
for id, oRNA in id_oRNA.items():
oRNA.tccs = oID_tccs[id]
oDC.commit(id_oRNA)
def updateMultipleTccs(oDir, mappedFN):
#parse bowtie
f = open(mappedFN, 'r')
oID_tccs = {}
for line in f:
ls = line.strip().split('\t')
oID = int(ls[0])
strand, chrom, firstCoord = bioLibCG.switchStrandFormat(
ls[1]), ls[2], int(ls[3])
secondCoord = firstCoord + len(ls[4]) - 1
tcc = bioLibCG.makeTcc(chrom, strand, firstCoord, secondCoord)
oID_tccs.setdefault(oID, []).append(tcc)
oDC = cgDB.dataController(oDir, cgOriginRNA.OriginRNA)
id_oRNA = oDC.load()
for id, oRNA in id_oRNA.items():
oRNA.tccs = oID_tccs[id]
oDC.commit(id_oRNA)
def makeGeneWig(tranFN, wigDir, chrom, strand):
coord_id = {}
f = open(tranFN, 'r')
for line in f:
ls = line.strip().split('\t')
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
if tChrom != chrom or tStrand != strand:
continue
gID = ls[10]
#gID = gID.replace(" ", "_")
tStart, tEnd = int(ls[3]), int(ls[4]) - 1
for i in xrange(tStart, tEnd + 1):
coord_id[i] = coord_id.get(i, '') + '%s$' % gID #$ is used because of spaces
#unique, string
for i, ids in coord_id.iteritems():
coord_id[i] = ','.join([x for x in set(ids.strip().split('$')) if x != ''])
#write wig to file
writeWigDictToWig(coord_id, chrom, strand, 'hg19', 'ALL', wigDir, 'None')
def makeTranscriptWig(tranFN, wigDir, chrom, strand, species='hg19'):
coord_id = {}
f = open(tranFN, 'r')
for line in f:
ls = line.strip().split('\t')
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
if tChrom != chrom or tStrand != strand:
continue
tID = ls[0]
tStart, tEnd = int(ls[3]), int(ls[4]) - 1
for i in xrange(tStart, tEnd + 1):
coord_id[i] = coord_id.get(i, '') + '%s ' % tID
#unique, string
for i, ids in coord_id.iteritems():
coord_id[i] = ','.join([x for x in set(ids.strip().split(' '))])
#write wig to file
writeWigDictToWig(coord_id, chrom, strand, species, 'transcript', wigDir,
'None')
def updateTccAndSNR(mFN, alignmentFN):
mNX = cgNexusFlat.Nexus(mFN, cgOriginRNAFlat.OriginRNA)
mNX.load(['tcc', 'snrSS'])
f = open(alignmentFN, 'r')
i = 0
for line in f:
ls = line.strip().split('\t')
chrom = ls[2]
strand = bioLibCG.switchStrandFormat(ls[1])
start = int(ls[3]) + 1 # 1BASE conversion
end = start + len(ls[4])
newTcc = bioLibCG.makeTcc(chrom, strand, start, end)
newSNR = 10.0
mNX.tcc[i] = newTcc
mNX.snrSS[i] = newSNR
i += 1
mNX.save()
def checkMessy(tranFN):
p = bioLibCG.cgPrint()
f = open(tranFN, 'r')
a = 0
b = 0
c = 0
d = 0
e = 0
for line in f:
ls = line.strip().split('\t')
chrom, strand = ls[1], bioLibCG.switchStrandFormat(ls[2])
tStart, tEnd = int(ls[3]), int(ls[4]) - 1
cStart, cEnd = int(ls[5]), int(ls[6]) - 1
exonStarts = [int(x) for x in ls[8][:-1].split(',')]
exonEnds = [int(x) - 1 for x in ls[9][:-1].split(',')]
exonPairs = zip(exonStarts, exonEnds)
codingStatus = '_coding' in ls[15]
tID = ls[0]
#debug
p.show = False
intronPairs = []
i = 0
for pair in exonPairs:
if i == 0:
i += 1
continue
iStart = exonPairs[i -1][1] + 1
iEnd = exonPairs[i][0] - 1
intronPairs.append((iStart, iEnd))
i += 1
#p.tell(tStart, tEnd, cStart, cEnd, exonPairs, intronPairs)
#take care of messy UTRs and assign utr ranges
#5UTR
if strand == '1':
if cStart == tStart or cStart == tEnd + 1:
p.tell('5 is none')
b += 1
if codingStatus:
d += 1
range5 = ()
else:
range5 = (tStart, cStart - 1)
else:
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
p.tell('5 is none')
b += 1
if codingStatus:
d += 1
range5 = ()
else:
range5 = (cEnd + 1, tEnd)
#3UTR
if strand == '1':
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
p.tell('3 is none')
c += 1
if codingStatus:
e += 1
range3 = ()
else:
range3 = (cEnd + 1, tEnd)
else:
if cStart == tStart or cStart == tEnd + 1:
p.tell('3 is none')
c += 1
if codingStatus:
e += 1
range3 = ()
else:
range3 = (tStart, cStart - 1)
a += 1
print a, b, c, d, e
def createCollapsedGeneSets(tranFN, outFN, acceptableTypes = 'EXON', onlyCoding = True):
'''get areas occupied by all transcripts in a gene'''
acceptableTypes = acceptableTypes.strip().split(',')
geneName_intervalSet = {}
geneName_info = {}
f = open(tranFN, 'r')
for line in f:
ls = line.strip().split('\t')
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
#if tChrom != chrom or tStrand != strand:
#continue
tStart, tEnd = int(ls[3]), int(ls[4]) - 1
cStart, cEnd = int(ls[5]), int(ls[6]) - 1
exonStarts = [int(x) for x in ls[8][:-1].split(',')]
exonEnds = [int(x) - 1 for x in ls[9][:-1].split(',')]
exonPairs = zip(exonStarts, exonEnds)
codingStatus = '_coding' in ls[13]
geneName = ls[10]
tID = ls[0]
#calulate intron pairs
intronPairs = []
i = 0
for pair in exonPairs:
if i == 0:
i += 1
continue
iStart = exonPairs[i -1][1] + 1
iEnd = exonPairs[i][0] - 1
intronPairs.append((iStart, iEnd))
i += 1
#take care of messy UTRs and assign utr ranges
#5UTR
if tStrand == '1':
if cStart == tStart or cStart == tEnd + 1:
range5 = ()
else:
range5 = (tStart, cStart - 1)
else:
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
range5 = ()
else:
range5 = (cEnd + 1, tEnd)
#3UTR
if tStrand == '1':
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
range3 = ()
else:
range3 = (cEnd + 1, tEnd)
else:
if cStart == tStart or cStart == tEnd + 1:
range3 = ()
else:
range3 = (tStart, cStart - 1)
utr5 = compareData.subtractTwoRanges([range5], intronPairs)
utr3 = compareData.subtractTwoRanges([range3], intronPairs)
exonPairs = compareData.subtractTwoRanges(exonPairs, [range5])
exonPairs = compareData.subtractTwoRanges(exonPairs, [range3])
geneName_info.setdefault(geneName, set()).add((tChrom, tStrand))
pairs__type = [ (exonPairs, 'EXON'), (intronPairs, 'INTRON'), (utr5, '5UTR'), (utr3, '3UTR') ]
for pairs, type in pairs__type:
for pair in pairs:
if type in acceptableTypes:
if onlyCoding and not codingStatus: continue
#create geneset/info if does not exist
if geneName not in geneName_intervalSet:
geneName_intervalSet[geneName] = IntervalSet()
geneName_intervalSet[geneName].add(Interval(pair[0], pair[1] + 1))
for geneName, info in geneName_info.iteritems():
if len(info) > 1:
if geneName in geneName_intervalSet:
del geneName_intervalSet[geneName] # if it spans different chromosomes/strands...
print geneName, info, 'FAILED'
fOut = open(outFN, 'w')
for geneName, iSet in geneName_intervalSet.iteritems():
gStarts = []
gEnds = []
for interv in iSet:
gStarts.append(interv.lower_bound)
gEnds.append(interv.upper_bound)
chrom, strand = geneName_info[geneName].pop()
outString = [geneName, chrom, strand, ','.join([str(x) for x in gStarts]), ','.join([str(x) for x in gEnds])]
fOut.write('\t'.join([str(x) for x in outString]) + '\n')
fOut.close()
def getSplicingUnitOccupancy(tranFN, wigDir1, wigDir2, chrom, strand, maxCut):
'''get the number of spots in each data set, and the number that overlap'''
'''wigDir2 has to be hela cuz strand flip'''
maxCut = int(maxCut)
oppStrand = bioLibCG.switchStrand(strand)
coord_value1 = cgWig.loadSingleWig(wigDir1, chrom, strand, 'ALL')
coord_value2 = cgWig.loadSingleWig(wigDir2, chrom, oppStrand, 'ALL')
# 0, 0, 0 = num1, num2, numOverlap
covered = set()
cutoff_overlap = dict( (i, [0, 0, 0]) for i in range(maxCut))
f = open(tranFN, 'r')
for line in f:
ls = line.strip().split('\t')
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
if tChrom != chrom or tStrand != strand:
continue
tStart, tEnd = int(ls[3]), int(ls[4]) - 1
cStart, cEnd = int(ls[5]), int(ls[6]) - 1
exonStarts = [int(x) for x in ls[8][:-1].split(',')]
exonEnds = [int(x) - 1 for x in ls[9][:-1].split(',')]
exonPairs = zip(exonStarts, exonEnds)
codingStatus = '_coding' in ls[13]
tID = ls[0]
#calulate intron pairs
intronPairs = []
i = 0
for pair in exonPairs:
if i == 0:
i += 1
continue
iStart = exonPairs[i -1][1] + 1
iEnd = exonPairs[i][0] - 1
intronPairs.append((iStart, iEnd))
i += 1
#take care of messy UTRs and assign utr ranges
#5UTR
if strand == '1':
if cStart == tStart or cStart == tEnd + 1:
range5 = ()
else:
range5 = (tStart, cStart - 1)
else:
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
range5 = ()
else:
range5 = (cEnd + 1, tEnd)
#3UTR
if strand == '1':
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
range3 = ()
else:
range3 = (cEnd + 1, tEnd)
else:
if cStart == tStart or cStart == tEnd + 1:
range3 = ()
else:
range3 = (tStart, cStart - 1)
utr5 = compareData.subtractTwoRanges([range5], intronPairs)
utr3 = compareData.subtractTwoRanges([range3], intronPairs)
exonPairs = compareData.subtractTwoRanges(exonPairs, [range5])
exonPairs = compareData.subtractTwoRanges(exonPairs, [range3])
pairs__type = [ (exonPairs, 'C_EXON'), (intronPairs, 'C_INTRON') ]
for pairs, type in pairs__type:
for pair in pairs:
for i in xrange(pair[0], pair[1] + 1):
if codingStatus:
if type == 'C_EXON':
if i in covered: continue #multiple transcripts will have same exons
covered.add(i)
val1 = coord_value1.get(i, 0)
val2 = coord_value2.get(i, 0)
for cut in range(1, maxCut):
#in1 = (val1 >= cut)
#in2 = (val2 >= cut)
in1 = (val1 == cut)
in2 = (val2 == cut)
if in1 and in2:
cutoff_overlap[cut][2] += 1
if in1:
cutoff_overlap[cut][0] += 1
if in2:
cutoff_overlap[cut][1] += 1
elif type == 'C_INTRON':
#intronChr_strand_coord.setdefault(tChrom, {}).setdefault(tStrand, set()).add(i)
pass
for i in range(1, maxCut):
cutoff_overlap[i].extend(['%s:%s' % (chrom, strand), i])
pString = '\t'.join([ str(x) for x in cutoff_overlap[i] ])
print pString
def makeContextWig(tranFN, wigDir, chrom, strand, species = 'hg19'):
p = bioLibCG.cgPrint()
coord_id = {}
f = open(tranFN, 'r')
for line in f:
ls = line.strip().split('\t')
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
if tChrom != chrom or tStrand != strand:
continue
tStart, tEnd = int(ls[3]), int(ls[4]) - 1
cStart, cEnd = int(ls[5]), int(ls[6]) - 1
exonStarts = [int(x) for x in ls[8][:-1].split(',')]
exonEnds = [int(x) - 1 for x in ls[9][:-1].split(',')]
exonPairs = zip(exonStarts, exonEnds)
codingStatus = '_coding' in ls[13]
tID = ls[0]
#debug
p.show = False
intronPairs = []
i = 0
for pair in exonPairs:
if i == 0:
i += 1
continue
iStart = exonPairs[i -1][1] + 1
iEnd = exonPairs[i][0] - 1
intronPairs.append((iStart, iEnd))
i += 1
#p.tell(tStart, tEnd, cStart, cEnd, exonPairs, intronPairs)
#take care of messy UTRs and assign utr ranges
#5UTR
if strand == '1':
if cStart == tStart or cStart == tEnd + 1:
p.tell('5 is none')
range5 = ()
else:
range5 = (tStart, cStart - 1)
else:
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
p.tell('5 is none')
range5 = ()
else:
range5 = (cEnd + 1, tEnd)
#3UTR
if strand == '1':
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
p.tell('3 is none')
range3 = ()
else:
range3 = (cEnd + 1, tEnd)
else:
if cStart == tStart or cStart == tEnd + 1:
p.tell('3 is none')
range3 = ()
else:
range3 = (tStart, cStart - 1)
p.tell('ranges', range5, range3)
p.tell('intronRange', intronPairs)
utr5 = compareData.subtractTwoRanges([range5], intronPairs)
utr3 = compareData.subtractTwoRanges([range3], intronPairs)
p.tell('utr', utr5, utr3)
p.tell('exon before', exonPairs)
exonPairs = compareData.subtractTwoRanges(exonPairs, [range5])
exonPairs = compareData.subtractTwoRanges(exonPairs, [range3])
p.tell('exon after', exonPairs)
debugSpot = 23631989
#5UTR
for pair in utr5:
p.tell('filling utr5', pair[0], pair[1])
for i in xrange(pair[0], pair[1] + 1):
if i == debugSpot: p.tell('*** 5UTR', codingStatus, tID)
if codingStatus:
coord_id[i] = coord_id.get(i, '') + 'C_5UTR '
else:
coord_id[i] = coord_id.get(i, '') + 'NC_5UTR '
#Exons
for pair in exonPairs:
p.tell('filling exons', pair[0], pair[1])
for i in xrange(pair[0], pair[1] + 1):
if i == debugSpot: p.tell('*** exon', codingStatus, tID)
if codingStatus:
coord_id[i] = coord_id.get(i, '') + 'C_EXON '
else:
coord_id[i] = coord_id.get(i, '') + 'NC_EXON '
#Introns
for pair in intronPairs:
p.tell('filling introns', pair[0], pair[1])
for i in xrange(pair[0], pair[1] + 1):
if i == debugSpot: p.tell('*** INTRON', codingStatus, tID)
if codingStatus:
coord_id[i] = coord_id.get(i, '') + 'C_INTRON '
else:
coord_id[i] = coord_id.get(i, '') + 'NC_INTRON '
#3UTR
for pair in utr3:
p.tell('filling utr3', pair[0], pair[1])
for i in xrange(pair[0], pair[1] + 1):
if i == debugSpot: p.tell(' *** 3UTR', codingStatus, tID)
if codingStatus:
coord_id[i] = coord_id.get(i, '') + 'C_3UTR '
else:
coord_id[i] = coord_id.get(i, '') + 'NC_3UTR '
p.show = False
#uniqify, stringify
for i, ids in coord_id.iteritems():
coord_id[i] = ','.join([x for x in set(ids.strip().split(' '))])
#p.tell('finalInfo', utr5, exonPairs, utr3)
#write wig to file
writeWigDictToWig(coord_id, chrom, strand, species, 'context', wigDir, 'INTER')
def getSplicingUnitLengths(tranFN, wigDir, chrom, strand):
exonChr_strand_coord = {}
intronChr_strand_coord = {}
f = open(tranFN, 'r')
for line in f:
ls = line.strip().split('\t')
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
#if tChrom != chrom or tStrand != strand:
#continue
tStart, tEnd = int(ls[3]), int(ls[4]) - 1
cStart, cEnd = int(ls[5]), int(ls[6]) - 1
exonStarts = [int(x) for x in ls[8][:-1].split(',')]
exonEnds = [int(x) - 1 for x in ls[9][:-1].split(',')]
exonPairs = zip(exonStarts, exonEnds)
codingStatus = '_coding' in ls[13]
tID = ls[0]
#calulate intron pairs
intronPairs = []
i = 0
for pair in exonPairs:
if i == 0:
i += 1
continue
iStart = exonPairs[i - 1][1] + 1
iEnd = exonPairs[i][0] - 1
intronPairs.append((iStart, iEnd))
i += 1
#take care of messy UTRs and assign utr ranges
#5UTR
if strand == '1':
if cStart == tStart or cStart == tEnd + 1:
range5 = ()
else:
range5 = (tStart, cStart - 1)
else:
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
range5 = ()
else:
range5 = (cEnd + 1, tEnd)
#3UTR
if strand == '1':
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
range3 = ()
else:
range3 = (cEnd + 1, tEnd)
else:
if cStart == tStart or cStart == tEnd + 1:
range3 = ()
else:
range3 = (tStart, cStart - 1)
utr5 = compareData.subtractTwoRanges([range5], intronPairs)
utr3 = compareData.subtractTwoRanges([range3], intronPairs)
exonPairs = compareData.subtractTwoRanges(exonPairs, [range5])
exonPairs = compareData.subtractTwoRanges(exonPairs, [range3])
pairs__type = [(exonPairs, 'C_EXON'), (intronPairs, 'C_INTRON')]
for pairs, type in pairs__type:
for pair in pairs:
for i in xrange(pair[0], pair[1] + 1):
if codingStatus:
if type == 'C_EXON':
exonChr_strand_coord.setdefault(
tChrom, {}).setdefault(tStrand, set()).add(i)
elif type == 'C_INTRON':
#intronChr_strand_coord.setdefault(tChrom, {}).setdefault(tStrand, set()).add(i)
pass
iLength = 0
for chrom in intronChr_strand_coord:
for strand in intronChr_strand_coord[chrom]:
iLength += len(intronChr_strand_coord[chrom][strand])
eLength = 0
for chrom in exonChr_strand_coord:
for strand in exonChr_strand_coord[chrom]:
eLength += len(exonChr_strand_coord[chrom][strand])
print 'total Exon Length (all exons overlapped)', eLength
print 'total intron Length (all introns overlapped)', iLength
def getSplicingUnitOccupancy(tranFN, wigDir1, wigDir2, chrom, strand, maxCut):
"""get the number of spots in each data set, and the number that overlap"""
"""wigDir2 has to be hela cuz strand flip"""
maxCut = int(maxCut)
oppStrand = bioLibCG.switchStrand(strand)
coord_value1 = cgWig.loadSingleWig(wigDir1, chrom, strand, "ALL")
coord_value2 = cgWig.loadSingleWig(wigDir2, chrom, oppStrand, "ALL")
# 0, 0, 0 = num1, num2, numOverlap
covered = set()
cutoff_overlap = dict((i, [0, 0, 0]) for i in range(maxCut))
f = open(tranFN, "r")
for line in f:
ls = line.strip().split("\t")
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
if tChrom != chrom or tStrand != strand:
continue
tStart, tEnd = int(ls[3]), int(ls[4]) - 1
cStart, cEnd = int(ls[5]), int(ls[6]) - 1
exonStarts = [int(x) for x in ls[8][:-1].split(",")]
exonEnds = [int(x) - 1 for x in ls[9][:-1].split(",")]
exonPairs = zip(exonStarts, exonEnds)
codingStatus = "_coding" in ls[13]
tID = ls[0]
# calulate intron pairs
intronPairs = []
i = 0
for pair in exonPairs:
if i == 0:
i += 1
continue
iStart = exonPairs[i - 1][1] + 1
iEnd = exonPairs[i][0] - 1
intronPairs.append((iStart, iEnd))
i += 1
# take care of messy UTRs and assign utr ranges
# 5UTR
if strand == "1":
if cStart == tStart or cStart == tEnd + 1:
range5 = ()
else:
range5 = (tStart, cStart - 1)
else:
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
range5 = ()
else:
range5 = (cEnd + 1, tEnd)
# 3UTR
if strand == "1":
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
range3 = ()
else:
range3 = (cEnd + 1, tEnd)
else:
if cStart == tStart or cStart == tEnd + 1:
range3 = ()
else:
range3 = (tStart, cStart - 1)
utr5 = compareData.subtractTwoRanges([range5], intronPairs)
utr3 = compareData.subtractTwoRanges([range3], intronPairs)
exonPairs = compareData.subtractTwoRanges(exonPairs, [range5])
exonPairs = compareData.subtractTwoRanges(exonPairs, [range3])
pairs__type = [(exonPairs, "C_EXON"), (intronPairs, "C_INTRON")]
for pairs, type in pairs__type:
for pair in pairs:
for i in xrange(pair[0], pair[1] + 1):
if codingStatus:
if type == "C_EXON":
if i in covered:
continue # multiple transcripts will have same exons
covered.add(i)
val1 = coord_value1.get(i, 0)
val2 = coord_value2.get(i, 0)
for cut in range(1, maxCut):
# in1 = (val1 >= cut)
# in2 = (val2 >= cut)
in1 = val1 == cut
in2 = val2 == cut
if in1 and in2:
cutoff_overlap[cut][2] += 1
if in1:
cutoff_overlap[cut][0] += 1
if in2:
cutoff_overlap[cut][1] += 1
elif type == "C_INTRON":
# intronChr_strand_coord.setdefault(tChrom, {}).setdefault(tStrand, set()).add(i)
pass
for i in range(1, maxCut):
cutoff_overlap[i].extend(["%s:%s" % (chrom, strand), i])
pString = "\t".join([str(x) for x in cutoff_overlap[i]])
print pString
def makeContextWig(tranFN, wigDir, chrom, strand, species='hg19'):
p = bioLibCG.cgPrint()
coord_id = {}
f = open(tranFN, 'r')
for line in f:
ls = line.strip().split('\t')
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
if tChrom != chrom or tStrand != strand:
continue
tStart, tEnd = int(ls[3]), int(ls[4]) - 1
cStart, cEnd = int(ls[5]), int(ls[6]) - 1
exonStarts = [int(x) for x in ls[8][:-1].split(',')]
exonEnds = [int(x) - 1 for x in ls[9][:-1].split(',')]
exonPairs = zip(exonStarts, exonEnds)
codingStatus = '_coding' in ls[13]
tID = ls[0]
#debug
p.show = False
intronPairs = []
i = 0
for pair in exonPairs:
if i == 0:
i += 1
continue
iStart = exonPairs[i - 1][1] + 1
iEnd = exonPairs[i][0] - 1
intronPairs.append((iStart, iEnd))
i += 1
#p.tell(tStart, tEnd, cStart, cEnd, exonPairs, intronPairs)
#take care of messy UTRs and assign utr ranges
#5UTR
if strand == '1':
if cStart == tStart or cStart == tEnd + 1:
p.tell('5 is none')
range5 = ()
else:
range5 = (tStart, cStart - 1)
else:
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
p.tell('5 is none')
range5 = ()
else:
range5 = (cEnd + 1, tEnd)
#3UTR
if strand == '1':
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
p.tell('3 is none')
range3 = ()
else:
range3 = (cEnd + 1, tEnd)
else:
if cStart == tStart or cStart == tEnd + 1:
p.tell('3 is none')
range3 = ()
else:
range3 = (tStart, cStart - 1)
p.tell('ranges', range5, range3)
p.tell('intronRange', intronPairs)
utr5 = compareData.subtractTwoRanges([range5], intronPairs)
utr3 = compareData.subtractTwoRanges([range3], intronPairs)
p.tell('utr', utr5, utr3)
p.tell('exon before', exonPairs)
exonPairs = compareData.subtractTwoRanges(exonPairs, [range5])
exonPairs = compareData.subtractTwoRanges(exonPairs, [range3])
p.tell('exon after', exonPairs)
debugSpot = 23631989
#5UTR
for pair in utr5:
p.tell('filling utr5', pair[0], pair[1])
for i in xrange(pair[0], pair[1] + 1):
if i == debugSpot: p.tell('*** 5UTR', codingStatus, tID)
if codingStatus:
coord_id[i] = coord_id.get(i, '') + 'C_5UTR '
else:
coord_id[i] = coord_id.get(i, '') + 'NC_5UTR '
#Exons
for pair in exonPairs:
p.tell('filling exons', pair[0], pair[1])
for i in xrange(pair[0], pair[1] + 1):
if i == debugSpot: p.tell('*** exon', codingStatus, tID)
if codingStatus:
coord_id[i] = coord_id.get(i, '') + 'C_EXON '
else:
coord_id[i] = coord_id.get(i, '') + 'NC_EXON '
#Introns
for pair in intronPairs:
p.tell('filling introns', pair[0], pair[1])
for i in xrange(pair[0], pair[1] + 1):
if i == debugSpot: p.tell('*** INTRON', codingStatus, tID)
if codingStatus:
coord_id[i] = coord_id.get(i, '') + 'C_INTRON '
else:
coord_id[i] = coord_id.get(i, '') + 'NC_INTRON '
#3UTR
for pair in utr3:
p.tell('filling utr3', pair[0], pair[1])
for i in xrange(pair[0], pair[1] + 1):
if i == debugSpot: p.tell(' *** 3UTR', codingStatus, tID)
if codingStatus:
coord_id[i] = coord_id.get(i, '') + 'C_3UTR '
else:
coord_id[i] = coord_id.get(i, '') + 'NC_3UTR '
p.show = False
#uniqify, stringify
for i, ids in coord_id.iteritems():
coord_id[i] = ','.join([x for x in set(ids.strip().split(' '))])
#p.tell('finalInfo', utr5, exonPairs, utr3)
#write wig to file
writeWigDictToWig(coord_id, chrom, strand, species, 'context', wigDir,
'INTER')
def makeContextDB(tranFN, chrom, strand, outFN):
f = open(tranFN, 'r')
fOut = open(outFN, 'w')
id = 0
for line in f:
ls = line.strip().split('\t')
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
if tChrom != chrom or tStrand != strand:
continue
tStart, tEnd = int(ls[3]), int(ls[4]) - 1
cStart, cEnd = int(ls[5]), int(ls[6]) - 1
exonStarts = [int(x) for x in ls[8][:-1].split(',')]
exonEnds = [int(x) - 1 for x in ls[9][:-1].split(',')]
exonPairs = zip(exonStarts, exonEnds)
codingStatus = '_coding' in ls[13]
tID = ls[0]
intronPairs = []
i = 0
for pair in exonPairs:
if i == 0:
i += 1
continue
iStart = exonPairs[i - 1][1] + 1
iEnd = exonPairs[i][0] - 1
intronPairs.append((iStart, iEnd))
i += 1
#p.tell(tStart, tEnd, cStart, cEnd, exonPairs, intronPairs)
#take care of messy UTRs and assign utr ranges
#5UTR
if strand == '1':
if cStart == tStart or cStart == tEnd + 1:
range5 = ()
else:
range5 = (tStart, cStart - 1)
else:
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
range5 = ()
else:
range5 = (cEnd + 1, tEnd)
#3UTR
if strand == '1':
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
range3 = ()
else:
range3 = (cEnd + 1, tEnd)
else:
if cStart == tStart or cStart == tEnd + 1:
range3 = ()
else:
range3 = (tStart, cStart - 1)
utr5 = compareData.subtractTwoRanges([range5], intronPairs)
utr3 = compareData.subtractTwoRanges([range3], intronPairs)
exonPairs = compareData.subtractTwoRanges(exonPairs, [range5])
exonPairs = compareData.subtractTwoRanges(exonPairs, [range3])
#5UTR
for pair in utr5:
myTcc = bioLibCG.makeTcc(chrom, strand, pair[0] + 1, pair[1] + 1)
if codingStatus:
pString = [str(id), 'C_5UTR', myTcc]
pString = '\t'.join(pString) + '\n'
fOut.write(pString)
else:
pString = [str(id), 'NC_5UTR', myTcc]
pString = '\t'.join(pString) + '\n'
fOut.write(pString)
id += 1
#Exons
for pair in exonPairs:
myTcc = bioLibCG.makeTcc(chrom, strand, pair[0] + 1, pair[1] + 1)
if codingStatus:
pString = [str(id), 'C_EXON', myTcc]
pString = '\t'.join(pString) + '\n'
fOut.write(pString)
else:
pString = [str(id), 'NC_EXON', myTcc]
pString = '\t'.join(pString) + '\n'
fOut.write(pString)
id += 1
#Introns
for pair in intronPairs:
myTcc = bioLibCG.makeTcc(chrom, strand, pair[0] + 1, pair[1] + 1)
if codingStatus:
pString = [str(id), 'C_INTRON', myTcc]
pString = '\t'.join(pString) + '\n'
fOut.write(pString)
else:
pString = [str(id), 'NC_INTRON', myTcc]
pString = '\t'.join(pString) + '\n'
fOut.write(pString)
id += 1
#3UTR
for pair in utr3:
myTcc = bioLibCG.makeTcc(chrom, strand, pair[0] + 1, pair[1] + 1)
if codingStatus:
pString = [str(id), 'C_3UTR', myTcc]
pString = '\t'.join(pString) + '\n'
fOut.write(pString)
else:
pString = [str(id), 'NC_3UTR', myTcc]
pString = '\t'.join(pString) + '\n'
fOut.write(pString)
id += 1
f.close()
fOut.close()
def getSplicingUnitLengths(tranFN, wigDir, chrom, strand):
exonChr_strand_coord = {}
intronChr_strand_coord = {}
f = open(tranFN, 'r')
for line in f:
ls = line.strip().split('\t')
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
#if tChrom != chrom or tStrand != strand:
#continue
tStart, tEnd = int(ls[3]), int(ls[4]) - 1
cStart, cEnd = int(ls[5]), int(ls[6]) - 1
exonStarts = [int(x) for x in ls[8][:-1].split(',')]
exonEnds = [int(x) - 1 for x in ls[9][:-1].split(',')]
exonPairs = zip(exonStarts, exonEnds)
codingStatus = '_coding' in ls[13]
tID = ls[0]
#calulate intron pairs
intronPairs = []
i = 0
for pair in exonPairs:
if i == 0:
i += 1
continue
iStart = exonPairs[i -1][1] + 1
iEnd = exonPairs[i][0] - 1
intronPairs.append((iStart, iEnd))
i += 1
#take care of messy UTRs and assign utr ranges
#5UTR
if strand == '1':
if cStart == tStart or cStart == tEnd + 1:
range5 = ()
else:
range5 = (tStart, cStart - 1)
else:
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
range5 = ()
else:
range5 = (cEnd + 1, tEnd)
#3UTR
if strand == '1':
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
range3 = ()
else:
range3 = (cEnd + 1, tEnd)
else:
if cStart == tStart or cStart == tEnd + 1:
range3 = ()
else:
range3 = (tStart, cStart - 1)
utr5 = compareData.subtractTwoRanges([range5], intronPairs)
utr3 = compareData.subtractTwoRanges([range3], intronPairs)
exonPairs = compareData.subtractTwoRanges(exonPairs, [range5])
exonPairs = compareData.subtractTwoRanges(exonPairs, [range3])
pairs__type = [ (exonPairs, 'C_EXON'), (intronPairs, 'C_INTRON') ]
for pairs, type in pairs__type:
for pair in pairs:
for i in xrange(pair[0], pair[1] + 1):
if codingStatus:
if type == 'C_EXON':
exonChr_strand_coord.setdefault(tChrom, {}).setdefault(tStrand, set()).add(i)
elif type == 'C_INTRON':
#intronChr_strand_coord.setdefault(tChrom, {}).setdefault(tStrand, set()).add(i)
pass
iLength = 0
for chrom in intronChr_strand_coord:
for strand in intronChr_strand_coord[chrom]:
iLength += len(intronChr_strand_coord[chrom][strand])
eLength = 0
for chrom in exonChr_strand_coord:
for strand in exonChr_strand_coord[chrom]:
eLength += len(exonChr_strand_coord[chrom][strand])
print 'total Exon Length (all exons overlapped)', eLength
print 'total intron Length (all introns overlapped)', iLength
def checkMessy(tranFN):
p = bioLibCG.cgPrint()
f = open(tranFN, 'r')
a = 0
b = 0
c = 0
d = 0
e = 0
for line in f:
ls = line.strip().split('\t')
chrom, strand = ls[1], bioLibCG.switchStrandFormat(ls[2])
tStart, tEnd = int(ls[3]), int(ls[4]) - 1
cStart, cEnd = int(ls[5]), int(ls[6]) - 1
exonStarts = [int(x) for x in ls[8][:-1].split(',')]
exonEnds = [int(x) - 1 for x in ls[9][:-1].split(',')]
exonPairs = zip(exonStarts, exonEnds)
codingStatus = '_coding' in ls[15]
tID = ls[0]
#debug
p.show = False
intronPairs = []
i = 0
for pair in exonPairs:
if i == 0:
i += 1
continue
iStart = exonPairs[i - 1][1] + 1
iEnd = exonPairs[i][0] - 1
intronPairs.append((iStart, iEnd))
i += 1
#p.tell(tStart, tEnd, cStart, cEnd, exonPairs, intronPairs)
#take care of messy UTRs and assign utr ranges
#5UTR
if strand == '1':
if cStart == tStart or cStart == tEnd + 1:
p.tell('5 is none')
b += 1
if codingStatus:
d += 1
range5 = ()
else:
range5 = (tStart, cStart - 1)
else:
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
p.tell('5 is none')
b += 1
if codingStatus:
d += 1
range5 = ()
else:
range5 = (cEnd + 1, tEnd)
#3UTR
if strand == '1':
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
p.tell('3 is none')
c += 1
if codingStatus:
e += 1
range3 = ()
else:
range3 = (cEnd + 1, tEnd)
else:
if cStart == tStart or cStart == tEnd + 1:
p.tell('3 is none')
c += 1
if codingStatus:
e += 1
range3 = ()
else:
range3 = (tStart, cStart - 1)
a += 1
print a, b, c, d, e
def makeContextDB(tranFN, chrom, strand, outFN):
'''outputs cID TYPE TCC, used for making wigs'''
f = open(tranFN, 'r')
fOut = open(outFN, 'w')
id = 0
for line in f:
ls = line.strip().split('\t')
tChrom, tStrand = ls[1], bioLibCG.switchStrandFormat(ls[2])
if tChrom != chrom or tStrand != strand:
continue
tStart, tEnd = int(ls[3]), int(ls[4]) - 1
cStart, cEnd = int(ls[5]), int(ls[6]) - 1
exonStarts = [int(x) for x in ls[8][:-1].split(',')]
exonEnds = [int(x) - 1 for x in ls[9][:-1].split(',')]
exonPairs = zip(exonStarts, exonEnds)
codingStatus = '_coding' in ls[13]
tID = ls[0]
intronPairs = []
i = 0
for pair in exonPairs:
if i == 0:
i += 1
continue
iStart = exonPairs[i -1][1] + 1
iEnd = exonPairs[i][0] - 1
intronPairs.append((iStart, iEnd))
i += 1
#p.tell(tStart, tEnd, cStart, cEnd, exonPairs, intronPairs)
#take care of messy UTRs and assign utr ranges
#5UTR
if strand == '1':
if cStart == tStart or cStart == tEnd + 1:
range5 = ()
else:
range5 = (tStart, cStart - 1)
else:
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
range5 = ()
else:
range5 = (cEnd + 1, tEnd)
#3UTR
if strand == '1':
if cEnd + 1 == tStart or cEnd + 1 == tEnd + 1:
range3 = ()
else:
range3 = (cEnd + 1, tEnd)
else:
if cStart == tStart or cStart == tEnd + 1:
range3 = ()
else:
range3 = (tStart, cStart - 1)
utr5 = compareData.subtractTwoRanges([range5], intronPairs)
utr3 = compareData.subtractTwoRanges([range3], intronPairs)
exonPairs = compareData.subtractTwoRanges(exonPairs, [range5])
exonPairs = compareData.subtractTwoRanges(exonPairs, [range3])
#5UTR
for pair in utr5:
myTcc = bioLibCG.makeTcc(chrom, strand, pair[0] + 1, pair[1] + 1)
if codingStatus:
pString = [str(id), 'C_5UTR', myTcc]
pString = '\t'.join(pString) + '\n'
fOut.write(pString)
else:
pString = [str(id), 'NC_5UTR', myTcc]
pString = '\t'.join(pString) + '\n'
fOut.write(pString)
id += 1
#Exons
for pair in exonPairs:
myTcc = bioLibCG.makeTcc(chrom, strand, pair[0] + 1, pair[1] + 1)
if codingStatus:
pString = [str(id), 'C_EXON', myTcc]
pString = '\t'.join(pString) + '\n'
fOut.write(pString)
else:
pString = [str(id), 'NC_EXON', myTcc]
pString = '\t'.join(pString) + '\n'
fOut.write(pString)
id += 1
#Introns
for pair in intronPairs:
myTcc = bioLibCG.makeTcc(chrom, strand, pair[0] + 1, pair[1] + 1)
if codingStatus:
pString = [str(id), 'C_INTRON', myTcc]
pString = '\t'.join(pString) + '\n'
fOut.write(pString)
else:
pString = [str(id), 'NC_INTRON', myTcc]
pString = '\t'.join(pString) + '\n'
fOut.write(pString)
id += 1
#3UTR
for pair in utr3:
myTcc = bioLibCG.makeTcc(chrom, strand, pair[0] + 1, pair[1] + 1)
if codingStatus:
pString = [str(id), 'C_3UTR', myTcc]
pString = '\t'.join(pString) + '\n'
fOut.write(pString)
else:
pString = [str(id), 'NC_3UTR', myTcc]
pString = '\t'.join(pString) + '\n'
fOut.write(pString)
id += 1
f.close()
fOut.close()
