def convertTccFileToGff(tccFileName): tccList = compare.tccFileToList(tccFileName, 0) gffList = convertTccListToBed(tccList) gffFileName = tccFileName + '.gff' gffFile = open(gffFileName, 'w') for line in gffList: gffFile.write(line) gffFile.close()
def convertTccFileToBed(tccFileName): tccList = compare.tccFileToList(tccFileName, 0) bedList = convertTccListToBed(tccList) bedFileName = tccFileName + '.bed' bedFile = open(bedFileName, 'w') for line in bedList: bedFile.write(line) bedFile.close()
def convertTccFileToBed(tccFileName):
tccList = compare.tccFileToList(tccFileName, 0)
bedList = convertTccListToBed(tccList)
bedFileName = tccFileName + '.bed'
bedFile = open(bedFileName, 'w')
for line in bedList:
bedFile.write(line)
bedFile.close()
def convertTccFileToGff(tccFileName):
tccList = compare.tccFileToList(tccFileName, 0)
gffList = convertTccListToBed(tccList)
gffFileName = tccFileName + '.gff'
gffFile = open(gffFileName, 'w')
for line in gffList:
gffFile.write(line)
gffFile.close()
def filterOut(cName = None): #Init conf = cgConfig.getConfig(cName) predictionList = compare.tccFileToList(conf.conf['resultsRaw'], 1) #predictionList = compare.tccFileToList(conf.conf['resultsRaw'], 0) overlapped = compare.filterOutTccs(predictionList, conf.conf['knownDirectory'], True) #True gives me the filtered out ones instead of the list without filtered out matureOverlaps = open(conf.conf['matureOverlaps'], 'w') for tcc in overlapped: matureOverlaps.write(tcc + '\n')
def filterOut(cName=None):
# Init
conf = cgConfig.getConfig(cName)
predictionList = compare.tccFileToList(conf.conf["resultsRaw"], 1)
# predictionList = compare.tccFileToList(conf.conf['resultsRaw'], 0)
overlapped = compare.filterOutTccs(
predictionList, conf.conf["knownDirectory"], True
) # True gives me the filtered out ones instead of the list without filtered out
matureOverlaps = open(conf.conf["matureOverlaps"], "w")
for tcc in overlapped:
matureOverlaps.write(tcc + "\n")
def splitExonsIntrons(cName = None):
mConf = c.cgConfig('Main.conf')
conf = c.getConfig(cName)
#init
organism = conf.conf['organism']
minOverlap = 50
cHairs = getHairpins.getHairpins() #CID: HAIRPIN
exonList = compare.tccFileToList('%sExons.tcc' % organism, 0)
hairpins = []
for CID in cHairs:
hairpins.append(cHairs[CID])
print 'checking overlaps'
#check which hairpins overlap exons and by how much
exonOverlapped = compare.compareTwoTcc(hairpins, exonList, 1, amount = True)
print ' ', len(exonOverlapped)
print 'removing partial introns'
#remove the ones that didn't overlap more than X:
remList = []
for tcc, oAmount in exonOverlapped:
if oAmount < minOverlap:
remList.append([tcc, oAmount])
for item in remList:
exonOverlapped.remove(item)
print ' ', len(exonOverlapped), 'out of', len(cHairs.keys())
#get CIDs of exons
exonCIDs = []
for tcc, oAmount in exonOverlapped:
for CID in cHairs:
if cHairs[CID] == tcc:
exonCIDs.append(str(CID))
#Open sorted predictions and write lines with CIDs to respective files
predFile = open(conf.conf['resultsSorted'], 'r')
exonFile = open(conf.conf['resultsSorted'] + '.exons', 'w')
intronFile = open(conf.conf['resultsSorted'] + '.introns', 'w')
for line in predFile:
if line.split('\t')[7] in exonCIDs:
exonFile.write(line)
else:
intronFile.write(line)
predFile.close()
exonFile.close()
intronFile.close()
def splitExonsIntrons(cName=None):
mConf = c.cgConfig('Main.conf')
conf = c.getConfig(cName)
#init
organism = conf.conf['organism']
minOverlap = 50
cHairs = getHairpins.getHairpins() #CID: HAIRPIN
exonList = compare.tccFileToList('%sExons.tcc' % organism, 0)
hairpins = []
for CID in cHairs:
hairpins.append(cHairs[CID])
print 'checking overlaps'
#check which hairpins overlap exons and by how much
exonOverlapped = compare.compareTwoTcc(hairpins, exonList, 1, amount=True)
print ' ', len(exonOverlapped)
print 'removing partial introns'
#remove the ones that didn't overlap more than X:
remList = []
for tcc, oAmount in exonOverlapped:
if oAmount < minOverlap:
remList.append([tcc, oAmount])
for item in remList:
exonOverlapped.remove(item)
print ' ', len(exonOverlapped), 'out of', len(cHairs.keys())
#get CIDs of exons
exonCIDs = []
for tcc, oAmount in exonOverlapped:
for CID in cHairs:
if cHairs[CID] == tcc:
exonCIDs.append(str(CID))
#Open sorted predictions and write lines with CIDs to respective files
predFile = open(conf.conf['resultsSorted'], 'r')
exonFile = open(conf.conf['resultsSorted'] + '.exons', 'w')
intronFile = open(conf.conf['resultsSorted'] + '.introns', 'w')
for line in predFile:
if line.split('\t')[7] in exonCIDs:
exonFile.write(line)
else:
intronFile.write(line)
predFile.close()
exonFile.close()
intronFile.close()
import cgPeaks
import compareData as compare
import math
import bioLibCG as cg
knowns = compare.tccFileToList('mouseKnownMirs.tcc', 0)
eLevels = []
for known in knowns:
chrom, strand, start, end = cg.tccSplit(known, True) #text...
if strand == '1':
strand = '-1'
else:
strand = '1'
oppTcc = cg.makeTcc(chrom, strand, start, end)
knownStretch = cgPeaks.stretch(known)
knownStretch.createPeaks(1, 20)
kPos = knownStretch.getHighestPeak()
if kPos: eLevels.append(knownStretch.profile[kPos])
oppStretch = cgPeaks.stretch(oppTcc)
oppStretch.createPeaks(1, 20)
oPos = oppStretch.getHighestPeak()
if oPos and kPos:
#determine if they are close enough to be considered mirrored...
if math.fabs(int(kPos) - int(oPos)) < 12:
print known, oPos, kPos, oppStretch.profile[
oPos], knownStretch.profile[kPos]
import profileTargets
import compareData as compare
tccList = compare.tccFileToList('ago.200.tcc', 0)
profileTargets.profileTargets(tccList, 'agoProfile.conf', dir = 'ago', min = 30)
#profileTargets.profileTargetsHistoAS(tccList, 'agoProfile.conf', name = 'agoNEG')
import compareData as compare
import bioLibCG as cg
exonList = compare.tccFileToList('allExons.tcc', 0)
print cg.getTccListTotalLength(exonList)
nonOverlap = compare.collapseOverlaps(exonList)
print cg.getTccListTotalLength(nonOverlap)
o = open('mouseExons.tcc', 'w')
for tcc in nonOverlap:
o.write(tcc + '\n')
o.close()
def intronNoisy(cName = None):
mConf = c.cgConfig('Main.conf')
conf = c.getConfig(cName)
#init
cHairs = getHairpins.getHairpins(conf.conf['resultsIntrons']) #CID: HAIRPIN
organism = conf.conf['organism']
exonList = compare.tccFileToList('%sExons.tcc' % organism, 0)
slide = 1000
#make prediction overlap hitmap
predMap = {}
predList = []
for CID in cHairs:
hPin = cHairs[CID]
predList.append(hPin)
#collapse Overlaps
print ' collapsing predictions'
predList = compare.collapseOverlaps(predList)
print ' collapsing exons'
exonList = compare.collapseOverlaps(exonList)
#collect levels for each hairpin region
cidLevels = {}
for CID in cHairs:
print CID
hPin = cHairs[CID]
chrom = ss(hPin, ':')[0]
strand = ss(hPin, ':')[1]
start = int(ss(hPin, ':')[2])
end = int(ss(hPin, ':')[3])
scanStart = start - slide
scanEnd = end + slide
scanRange = []
scanRange.append('%s:%s:%s:%s' % (chrom, strand, scanStart, start))
scanRange.append('%s:%s:%s:%s' % (chrom, strand, end, scanEnd))
print scanRange
scanRange = compare.subtractTwoTccLists(scanRange, predList)
scanRange = compare.subtractTwoTccLists(scanRange, exonList)
levels = []
print ' Retrieving Expression levels:', cg.getTccListTotalLength(scanRange)
levels = []
hPinLevels = stepVectorScan.scanVectorsHist(scanRange, cName)
for hPin in hPinLevels:
levels.extend(hPinLevels[hPin])
cidLevels[CID] = levels
#output levels to file
#find longest
longest = 0
for CID in cidLevels:
length = len(cidLevels[CID])
if length > longest:
longest = length
sortedKeys = cidLevels.keys()
sortedKeys.sort()
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['intronNoiseData']
outFile = open(outFileN, 'w')
outFile.write('\t'.join(sortedKeys) + '\n')
outFile.writelines(newLines)
outFile.close()
import profileTargets
import compareData as compare
tccList = compare.tccFileToList('ago.200.tcc', 0)
profileTargets.profileTargets(tccList, 'agoProfile.conf', dir='ago', min=30)
#profileTargets.profileTargetsHistoAS(tccList, 'agoProfile.conf', name = 'agoNEG')
#given tcc, return best peak combo
import bioLibCG as cg
import cgConfig as c
import wigValue
import compareData as compare
#init
mConf = c.cgConfig('Main.conf')
conf = c.cgConfig()
pRange = 100
tccList = ['chr3:-1:96042576:96042685', 'chr3:-1:96042576:96042685']
tccList = compare.tccFileToList('mouseKnownMirs.tcc', 0)
timer = cg.cgTimer()
timer.start()
#put peaks in memory
print 'loading peak data'
peakFilesNames = cg.recurseDir(mConf.conf['wigMouse'], end = '.peaks')
peaks = {} # chr:peak:value
for pN in peakFilesNames:
chrom = pN.strip().split('.')[4]
strand = pN.strip().split('.')[2]
#init dictionary
if chrom not in peaks:
peaks[chrom] = {}
if strand not in peaks[chrom]:
import compareData as compare
import bioLibCG as cg
exonList = compare.tccFileToList('allExons.tcc', 0)
print cg.getTccListTotalLength(exonList)
nonOverlap = compare.collapseOverlaps(exonList)
print cg.getTccListTotalLength(nonOverlap)
o = open('mouseExons.tcc', 'w')
for tcc in nonOverlap:
o.write(tcc + '\n')
o.close()
'''For all things testing''' import bioLibCG as cg import compareData as compare fileName = '/u/home8/gxxiao/chrisgre/scripts/FilterKnownMirs/ensemblHumanData/ensemblData .dblColonDash' dcdList = compare.tccFileToList(fileName, 0) tccList = cg.convertDcdToTcc(dcdList) for x in tccList: print x
##Clusters are based off of overlapping neighbors, if you have an overlapping neighbor than you are part of that cluster. import bioLibCG as cg import subprocess import compareData as compare import cgConfig #Start Timer timer = cg.cgTimer() timer.start() #Get list of mature tccs conf = cgConfig.returnConfDict() finalMirFileName = '/u/home8/gxxiao/chrisgre/projects/PipeRuns/LanderHuman/out/LanderHuman-s3k8b17.ALL.FINAL.mirs.tsv' finalMirFileName = conf['resultsRaw'] matureTccs = compare.tccFileToList(finalMirFileName, 1) # list of all mature micro in tcc print 'List getting', timer.split() #make connections dict matureConnections = compare.makeConnectionsDict(matureTccs) print 'Make connections:', timer.split() #Now have to define Clusters... clusters = [] addedList = [] #I don't think python passes by reference? also I think this function is in the middle because it uses a global variable :P def createClusters(item = None, mode = None): if item in addedList:
def intronNoisy(cName=None):
mConf = c.cgConfig('Main.conf')
conf = c.getConfig(cName)
#init
cHairs = getHairpins.getHairpins(
conf.conf['resultsIntrons']) #CID: HAIRPIN
organism = conf.conf['organism']
exonList = compare.tccFileToList('%sExons.tcc' % organism, 0)
slide = 1000
#make prediction overlap hitmap
predMap = {}
predList = []
for CID in cHairs:
hPin = cHairs[CID]
predList.append(hPin)
#collapse Overlaps
print ' collapsing predictions'
predList = compare.collapseOverlaps(predList)
print ' collapsing exons'
exonList = compare.collapseOverlaps(exonList)
#collect levels for each hairpin region
cidLevels = {}
for CID in cHairs:
print CID
hPin = cHairs[CID]
chrom = ss(hPin, ':')[0]
strand = ss(hPin, ':')[1]
start = int(ss(hPin, ':')[2])
end = int(ss(hPin, ':')[3])
scanStart = start - slide
scanEnd = end + slide
scanRange = []
scanRange.append('%s:%s:%s:%s' % (chrom, strand, scanStart, start))
scanRange.append('%s:%s:%s:%s' % (chrom, strand, end, scanEnd))
print scanRange
scanRange = compare.subtractTwoTccLists(scanRange, predList)
scanRange = compare.subtractTwoTccLists(scanRange, exonList)
levels = []
print ' Retrieving Expression levels:', cg.getTccListTotalLength(
scanRange)
levels = []
hPinLevels = stepVectorScan.scanVectorsHist(scanRange, cName)
for hPin in hPinLevels:
levels.extend(hPinLevels[hPin])
cidLevels[CID] = levels
#output levels to file
#find longest
longest = 0
for CID in cidLevels:
length = len(cidLevels[CID])
if length > longest:
longest = length
sortedKeys = cidLevels.keys()
sortedKeys.sort()
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['intronNoiseData']
outFile = open(outFileN, 'w')
outFile.write('\t'.join(sortedKeys) + '\n')
outFile.writelines(newLines)
outFile.close()
def defineClusters(cName=None):
#Start Timer
timer = cg.cgTimer()
timer.start()
#Get list of mature tccs
conf = cgConfig.getConfig(cName) #passed or default
finalMirFileName = conf.conf['resultsRaw']
matureTccs = compare.tccFileToList(finalMirFileName,
1) # list of all mature micro in tcc
print 'List getting', timer.split()
#make connections dict
matureConnections = compare.makeConnectionsDict(matureTccs)
print 'Make connections:', timer.split()
#Now have to define Clusters...
clusters = []
addedList = []
#I don't think python passes by reference? also I think this function is in the middle because it uses a global variable :P
def createClusters(item=None, mode=None):
if item in addedList:
return 0
elif mode == "top":
clusters.append([item])
addedList.append(
item) ##creates new cluster with the item already stored in it
for connectedItem in matureConnections[item]:
createClusters(connectedItem, "neighbor")
elif mode == "neighbor":
clusters[-1].append(
item) #add this item to the last cluster created
addedList.append(item)
for connectedItem in matureConnections[item]:
createClusters(connectedItem, "neighbor")
for tcc in matureTccs:
createClusters(tcc, "top")
print 'Make Clusters', timer.split()
#Sort Clusters.
sortedClusters = []
for cluster in clusters:
sortedClusters.append(cg.sortTccList(cluster))
print 'Sort Clusters:', timer.split()
#Output sorted cluster file
clusterFileName = conf.conf['sortedClusters']
clusterFile = open(clusterFileName, 'w')
for cluster in sortedClusters:
for hit in cluster:
clusterFile.write('%s,' % hit)
clusterFile.write('\n')
clusterFile.close()
'''
#re-create sortedClusters list:
clusterFileName = 'sortedClusters.data'
clusterFile = open(clusterFileName, 'r')
sortedClusters = []
for line in clusterFile:
sortedClusters.append([])
line = line.strip()[0:-1] #take off last comma ;P
for hit in (line.strip().split(',')):
sortedClusters[-1].append(hit)
'''
print 'Store intermediate data:', timer.split()
#output hitsAround file
outputFile = open(conf.conf['hitsPerFrame'], 'w')
frameLength = 200
frameShift = 1
for cluster in sortedClusters:
#grab first and last coordinate from cluster, for each cluster deduce how many theoretical microRNAs were in hitScope
clusterChrom = cluster[0].split(":")[0]
clusterStrand = cluster[0].split(":")[1]
firstCoord = int(cluster[0].split(":")[2])
#print cluster[-1]
lastCoord = int(cluster[-1].split(":")[3])
startCoord = firstCoord
while startCoord < lastCoord:
#count how many hits there are in this range
rangeStart = startCoord - (frameLength / 2)
rangeEnd = startCoord + (frameLength / 2)
rangeTcc = '%s:%s:%s:%s' % (clusterChrom, clusterStrand,
rangeStart, rangeEnd)
overlappedList = compare.compareTwoTcc([rangeTcc], cluster, 2)
hitCount = len(overlappedList)
#output
outputFile.write('%s\t%s\n' % (rangeTcc, hitCount))
startCoord = startCoord + frameShift #check overlap with range
outputFile.close()
print 'Output Hits per Frame:', timer.split()
print 'Overall Time:', timer.report()
def defineClusters(cName = None):
#Start Timer
timer = cg.cgTimer()
timer.start()
#Get list of mature tccs
conf = cgConfig.getConfig(cName) #passed or default
finalMirFileName = conf.conf['resultsRaw']
matureTccs = compare.tccFileToList(finalMirFileName, 1) # list of all mature micro in tcc
print 'List getting', timer.split()
#make connections dict
matureConnections = compare.makeConnectionsDict(matureTccs)
print 'Make connections:', timer.split()
#Now have to define Clusters...
clusters = []
addedList = []
#I don't think python passes by reference? also I think this function is in the middle because it uses a global variable :P
def createClusters(item = None, mode = None):
if item in addedList:
return 0
elif mode == "top":
clusters.append([item])
addedList.append(item) ##creates new cluster with the item already stored in it
for connectedItem in matureConnections[item]:
createClusters(connectedItem, "neighbor")
elif mode == "neighbor":
clusters[-1].append(item) #add this item to the last cluster created
addedList.append(item)
for connectedItem in matureConnections[item]:
createClusters(connectedItem, "neighbor")
for tcc in matureTccs:
createClusters(tcc, "top")
print 'Make Clusters', timer.split()
#Sort Clusters.
sortedClusters = []
for cluster in clusters:
sortedClusters.append(cg.sortTccList(cluster))
print 'Sort Clusters:', timer.split()
#Output sorted cluster file
clusterFileName = conf.conf['sortedClusters']
clusterFile = open(clusterFileName, 'w')
for cluster in sortedClusters:
for hit in cluster:
clusterFile.write('%s,' % hit)
clusterFile.write('\n')
clusterFile.close()
'''
#re-create sortedClusters list:
clusterFileName = 'sortedClusters.data'
clusterFile = open(clusterFileName, 'r')
sortedClusters = []
for line in clusterFile:
sortedClusters.append([])
line = line.strip()[0:-1] #take off last comma ;P
for hit in (line.strip().split(',')):
sortedClusters[-1].append(hit)
'''
print 'Store intermediate data:', timer.split()
#output hitsAround file
outputFile = open(conf.conf['hitsPerFrame'], 'w')
frameLength = 200
frameShift = 1
for cluster in sortedClusters:
#grab first and last coordinate from cluster, for each cluster deduce how many theoretical microRNAs were in hitScope
clusterChrom = cluster[0].split(":")[0]
clusterStrand = cluster[0].split(":")[1]
firstCoord = int(cluster[0].split(":")[2])
#print cluster[-1]
lastCoord = int(cluster[-1].split(":")[3])
startCoord = firstCoord
while startCoord < lastCoord:
#count how many hits there are in this range
rangeStart = startCoord - (frameLength/2)
rangeEnd = startCoord + (frameLength/2)
rangeTcc = '%s:%s:%s:%s' % (clusterChrom, clusterStrand, rangeStart, rangeEnd)
overlappedList = compare.compareTwoTcc([rangeTcc], cluster, 2)
hitCount = len(overlappedList)
#output
outputFile.write('%s\t%s\n' % (rangeTcc, hitCount))
startCoord = startCoord + frameShift #check overlap with range
outputFile.close()
print 'Output Hits per Frame:', timer.split()
print 'Overall Time:', timer.report()
import compareData as compare
tccList = compare.tccFileToList('snos.tcc', 0)
collapsed = compare.collapseOverlaps(tccList)
for tcc in collapsed:
print tcc
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 cgPeaks
import compareData as compare
import math
import bioLibCG as cg
knowns = compare.tccFileToList("mouseKnownMirs.tcc", 0)
eLevels = []
for known in knowns:
chrom, strand, start, end = cg.tccSplit(known, True) # text...
if strand == "1":
strand = "-1"
else:
strand = "1"
oppTcc = cg.makeTcc(chrom, strand, start, end)
knownStretch = cgPeaks.stretch(known)
knownStretch.createPeaks(1, 20)
kPos = knownStretch.getHighestPeak()
if kPos:
eLevels.append(knownStretch.profile[kPos])
oppStretch = cgPeaks.stretch(oppTcc)
oppStretch.createPeaks(1, 20)
oPos = oppStretch.getHighestPeak()
if oPos and kPos:
# determine if they are close enough to be considered mirrored...
if math.fabs(int(kPos) - int(oPos)) < 12:
print known, oPos, kPos, oppStretch.profile[oPos], knownStretch.profile[kPos]
import compareData as compare
tccList = compare.tccFileToList('snos.tcc', 0)
collapsed = compare.collapseOverlaps(tccList)
for tcc in collapsed:
print tcc
#get results that are only noncoding
import bioLibCG as cg
import compareData as compare
predName = '/home/chrisgre/projects/NoncodingMouse/results/NCmouse-s3k8b17.bothNCandC.results'
keepList = compare.tccFileToList('keepNoncoding.tcc', 0)
predList = compare.tccFileToList(predName, 1)
keepers = compare.compareTwoTcc(predList, keepList, 1)
print len(keepers)
#now go back through pred file and create a new file with only lines that have noncoding in them
predFile = open(predName, 'r')
outFile = open('NCmouse.noncoding.results', 'w')
predLines = predFile.readlines()
predFile.close()
newLines = {}
for keeper in keepers:
for line in predLines:
if keeper in line:
newLines[line] = 1
for line in newLines:
outFile.write(line)
'''For all things testing'''
import bioLibCG as cg
import compareData as compare
fileName = '/u/home8/gxxiao/chrisgre/scripts/FilterKnownMirs/ensemblHumanData/ensemblData .dblColonDash'
dcdList = compare.tccFileToList(fileName, 0)
tccList = cg.convertDcdToTcc(dcdList)
for x in tccList:
print x
