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 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 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 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()
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 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 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 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 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')
