def byGene(geneSpanFN, wigDir1, wigDir2, chrom, strand, outFN, simulation = False):
'''hela must be 2nd wigDir2 cuz strand flip'''
strand = str(strand) #undo autocast
print 'loading wigs'
oppStrand = bioLibCG.switchStrand(strand)
coord_value1 = cgWig.loadSingleWig(wigDir1, chrom, strand, 'ALL')
coord_value2 = cgWig.loadSingleWig(wigDir2, chrom, oppStrand, 'ALL')
print 'calculating bin values'
f = open(geneSpanFN, 'r')
fOut = open(outFN, 'w')
for line in f:
ls = line.strip().split('\t')
sChrom, sStrand = ls[1], ls[2]
if sChrom != chrom or sStrand != strand:
continue
geneName = ls[0]
geneStarts = [int(x) for x in ls[3].split(',')]
geneEnds = [int(x) for x in ls[4].split(',')]
spanPairs = zip(geneStarts, geneEnds)
frameLength = 10
skipAmount = 2
theSpan = fullSpanFromPairs(spanPairs)
spanLength = len(theSpan)
binAvgs1 = []
binAvgs2 = []
for theBinAvg, theCoord_Val in [(binAvgs1, coord_value1), (binAvgs2, coord_value2)]:
#mix up bins if simulation
if simulation:
newSpan = mixSpanByBin(theSpan, frameLength)
else:
newSpan = theSpan
i = 0
while (i+frameLength) < (spanLength+1):
binNums = newSpan[i:(i + frameLength)]
theBinAvg.append(binAvg(theCoord_Val, binNums))
i = i + skipAmount
#get rid of all 0,0 pairs for correlation
editPairs = zip(binAvgs1, binAvgs2)
newPairs = [pair for pair in editPairs if not (pair[0] == 0 and pair[1] == 0)]
newX = [pair[0] for pair in newPairs]
newY = [pair[1] for pair in newPairs]
dataLoad = sum(binAvgs1) + sum(binAvgs2)
dataLoad = float(dataLoad)/2
pcc = pStats.pearsonr(binAvgs1, binAvgs2)
scc, pVal = pStats.spearmanr(binAvgs1, binAvgs2)
outString = [geneName, pcc[0], ','.join([str(x) for x in binAvgs1]), ','.join([str(x) for x in binAvgs2]), '%s:%s:%s' % (sChrom, sStrand, theSpan[0]), dataLoad, scc]
fOut.write('\t'.join([str(x) for x in outString]) + '\n')
fOut.close()
f.close()
def getPlotData(aSites, wigDir, outFN):
'''get box plot data from sites in degradome'''
#load and init
spreadRange = range(-200, 201) #200 +/- ... might want to check distance each AAUAAA is from each other
relCoord_degVals = dict( (i, []) for i in spreadRange )
for chrom in bioLibCG.humanChromosomes:
for strand in ('1', '-1'):
print chrom, strand
coord_value = cgWig.loadSingleWig(wigDir, chrom, strand, 'ALL')
f = open(aSites, 'r')
for line in f:
ls = line.strip().split('\t')
ichrom, istrand, start, end = bioLibCG.tccSplit(ls[0])
if ichrom != chrom or istrand != strand: continue
for i in spreadRange:
degVal = coord_value.get(end + i, 0)
relCoord_degVals[i].append(degVal)
f.close()
#output box data
#each row is a histogram of spread position (e.g., first row is -200)
f = open(outFN, 'w')
outLines = []
for i in spreadRange:
l = [str(x) for x in relCoord_degVals[i]]
outLines.append('\t'.join(l) + '\n')
f.writelines(outLines)
f.close()
def test(tcc, wigDir):
chrom, strand, start, end = bioLibCG.tccSplit(tcc)
print chrom, strand
coord_eLevel = cgWig.loadSingleWig(wigDir, chrom, strand, 'ALL')
sKeys = sorted(coord_eLevel.keys())
for i in range(start, end + 1):
print i, coord_eLevel.get(i, 0)
def updateRepeatStatus(fN, fF, wigDir, chrom, strand):
#load oRNAs
NX = Nexus(fN, fF)
NX.load(['repeat', 'tcc'])
#load wig file for chrom, strand
coord_value = cgWig.loadSingleWig(wigDir, chrom, strand, 'REPEAT')
while NX.nextID():
oChrom, oStrand, start, end = bioLibCG.tccSplit(NX.tcc)
if oChrom != chrom or oStrand != strand: continue
NX.repeat = False
for i in range(start, end + 1):
if i in coord_value:
NX.repeat = True
break
NX.save()
def updateRepeatStatus(oFN, wigDir, chrom, strand, rn=None, tn=None):
#load oRNAs
oNX = cgNexusFlat.Nexus(oFN, degPeak.degPeak)
oNX.load(['repeatStatus', 'tcc'])
#load wig file for chrom, strand
coord_value = cgWig.loadSingleWig(wigDir, chrom, strand, 'REPEAT')
for oID in oNX.repeatStatus:
oChrom, oStrand, start, end = bioLibCG.tccSplit(oNX.tcc[oID])
if oChrom != chrom or oStrand != strand:
continue
oNX.repeatStatus[oID] = False
for i in range(start, end + 1):
if i in coord_value:
oNX.repeatStatus[oID] = True
break
oNX.save()
def updateRepeatStatus(oFN, wigDir, chrom, strand):
#load oRNAs
oNX = cgNexusFlat.Nexus(oFN, cgDegPeak.Peak)
oNX.load(['repeatStatus', 'tcc'])
#load wig file for chrom, strand
coord_value = cgWig.loadSingleWig(wigDir, chrom, strand, 'REPEAT')
for oID in oNX.repeatStatus:
oChrom, oStrand, start, end = bioLibCG.tccSplit(oNX.tcc[oID])
if oChrom != chrom or oStrand != strand:
continue
oNX.repeatStatus[oID] = False
for i in range(start, end + 1):
if i in coord_value:
oNX.repeatStatus[oID] = True
break
oNX.save()
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 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 plotAllDegOverlap(inFile, chrom, strand, wigDir1, wigDir2, outDir, withIntrons = False, flipStrand = True):
'''hela must be 2nd wigDir2 cuz strand flip'''
oppStrand = strand
if flipStrand:
oppStrand = bioLibCG.switchStrand(strand)
print 'loading Wigs', chrom, strand
coord_value1 = cgWig.loadSingleWig(wigDir1, chrom, strand, 'ALL')
coord_value2 = cgWig.loadSingleWig(wigDir2, chrom, oppStrand, 'ALL')
f = open(inFile, 'r')
for line in f:
gName, dChrom, dStrand, exonStarts, exonEnds = line.strip().split('\t')
if dChrom != chrom or dStrand != strand:
continue
exonStarts = [int(x) for x in exonStarts.split(',')]
exonEnds = [int(x) for x in exonEnds.split(',')]
print 'Plotting', gName
#create the span info for boxplots (JUST EXONS!!!)
exons = zip(exonStarts, exonEnds)
introns = [(x[0] + 1, x[1] - 1) for x in zip(exonEnds[:-1], exonStarts[1:])]
iLengths = [x[0] - x[1] + 1 for x in zip(exonStarts[1:], exonEnds[:-1])]
all = exons[:]
if withIntrons:
all.extend(introns)
all.sort()
tSpan = [('exon', x) if x in exons else ('intron', x) for x in all]
#gather expression data
c_v = {}
c_v2 = {}
for type, (eStart, eEnd) in tSpan:
for i in range(eStart, eEnd + 1):
if i in coord_value1:
c_v[i] = coord_value1[i]
if i in coord_value2:
c_v2[i] = coord_value2[i]
#intron displacement for ONLY EXONS
if not withIntrons:
iCumulativeLengths = [sum(iLengths[:x]) for x in range(1,len(introns) + 1)]
for i, (eStart, eEnd) in enumerate(exons):
if i == 0: continue
dAmount = iCumulativeLengths[i - 1]
for j in range(eStart, eEnd + 1):
if j in c_v:
c_v[j - dAmount] = c_v[j]
del c_v[j]
if j in c_v2:
c_v2[j - dAmount] = c_v2[j]
del c_v2[j]
#get overall max
overMax = max([max(x) for x in [c_v.values(), c_v2.values()]])
a, b = set(c_v.keys()), set(c_v2.keys())
overlap = a.intersection(b)
colors_a = ['r' if x in overlap else 'k' for x in sorted(a)]
colors_b = ['r' if x in overlap else 'k' for x in sorted(b)]
plotGrassTrack(c_v, [9, 15], manualMax = overMax, flip = False, colors = colors_a)
plotGrassTrack(c_v2, [-3, 3], manualMax = overMax,flip = True, colors = colors_b)
xStart = plotGeneTrack(tSpan, 0)
#labels and axes
plt.figtext(.05, .5, gName)
plt.figtext(.05, .62, '0 -')
plt.figtext(.05, .89, '%s -' % overMax)
plt.figtext(.05, 1 - .62, '0 -')
plt.figtext(.05, 1 - .89, '%s -' % overMax)
plt.ylim(-3,15)
frame1 = plt.gca()
frame1.axes.get_yaxis().set_visible(False)
if dStrand == '1':
plt.title('Degradome Comparison (5-->3)')
else:
plt.title('Degradome Comparison (3-->5)')
imgName = outDir + '/' + gName + '.degOverlapPlot.png'
plt.savefig(imgName, bbox_inches='tight', pad_inches=1)
#plt.show()
plt.close('all')
f.close()