def main(args):
# debugging info
logfile = args.outDirName + "/" + args.outBaseName + "/logs/%d" % os.getpid() + "." + args.outBaseName + ".mergepairs.log"
logging.basicConfig(format='%(asctime)s %(message)s',filename=logfile,level=logging.DEBUG)
logging.info("\ninDir1=%s\ninDir2=%s\noutBaseName=%s\nconfigFileName=%s"
% (args.inDir1,args.inDir2,args.outBaseName,args.configFileName))
# create output directory
discordant.prepOutDir(args.outBaseName,args.outDirName,args.overwrite)
# make sure input sources exist
peakparser.checkOutDir(args.inDir1,args.outDirName)
peakparser.checkOutDir(args.inDir2,args.outDirName)
# make sure config files exist
configPath1 = args.outDirName + "/" + args.inDir1 + "/" + args.configFileName
configPath2 = args.outDirName + "/" + args.inDir2 + "/" + args.configFileName
discordant.checkfile(configPath1)
discordant.checkfile(configPath2)
# read parameters for both inputs
configDict1 = peakparser.readConfig(configPath1,args.inDir1,args.outDirName)
configDict2 = peakparser.readConfig(configPath2,args.inDir2,args.outDirName)
maxDist = int(configDict1['insertSize']) + 2*int(configDict1['readLength'])
eltLenDict = discordant.makeEltLenDict(configDict1['eltLenFileName'])
# merge readfiles
outReadFileName = args.outDirName + "/" + args.outBaseName + "/" + args.outBaseName + ".readpairs.txt"
readFileName1 = args.outDirName + "/" + args.inDir1 + "/" + args.inDir1 + ".readpairs.txt"
readFileName2 = args.outDirName + "/" + args.inDir2 + "/" + args.inDir2 + ".readpairs.txt"
logging.info("merging readfiles (%s, %s)" % (readFileName1,readFileName2))
# print "merging readfiles (%s, %s)" % (readFileName1,readFileName2)
mergeChrPosFiles(readFileName1,readFileName2,outReadFileName,maxDist,eltLenDict)
# merge bedfiles
outBedFileName = args.outDirName + "/" + args.outBaseName + "/" + args.outBaseName + ".reads.bed"
bedFileName1 = args.outDirName + "/" + args.inDir1 + "/" + args.inDir1 + ".reads.bed"
bedFileName2 = args.outDirName + "/" + args.inDir2 + "/" + args.inDir2 + ".reads.bed"
logging.info("merging bedfiles (%s,%s)" % (bedFileName1,bedFileName2))
# print "merging bedfiles (%s,%s)" % (bedFileName1,bedFileName2)
mergeChrPosFiles(bedFileName1,bedFileName2,outBedFileName,maxDist,eltLenDict)
# write new config file
configPath = args.outDirName + "/" + args.outBaseName + "/" + args.configFileName
configDict = configDict1
configDict['bamFileName1'] = configDict1['bamFileName']
configDict['bamFileName2'] = configDict2['bamFileName']
configDict['merged'] = 'True'
configDict['outBaseName'] = args.outBaseName
configDict['outDirName'] = args.outDirName
configDict['readFileName'] = outReadFileName
del configDict['bamFileName']
f = open(configPath, 'w')
for k,v in configDict.iteritems():
f.write(k + "=" + v + "\n")
f.close()
def main(args):
peakparser.checkOutDir(args.outBaseName,args.outDirName)
configPath = args.outDirName + "/" + args.outBaseName + "/" + args.configFileName
discordant.checkfile(configPath)
configDict = peakparser.readConfig(configPath,args.outBaseName,args.outDirName)
refGenome = configDict['refGenome']
checkAnnotDir(refGenome, args.annotDir)
debugFile = args.outDirName + "/" + args.outBaseName + "/" + args.outBaseName + ".debug.txt"
discordant.checkfile(debugFile)
# can filter on retroelement subfamilies
discordant.checkfile(args.eltFile)
eltDict = {}
f = open(args.eltFile,'r')
for elt in f:
eltDict[elt.strip()]=1
germlineIns = []
cancerIns = []
normalIns = []
otherIns = []
f = open(debugFile, 'r')
for line in f: # loop over lines
if re.search('valid=1', line) and (re.search('olClN=0', line) or args.allowOverlap):
insdata = {}
for c in line.rstrip().rsplit(' '): # loop over columns
(key,value) = c.rsplit('=')
insdata[key] = value
if int(insdata['np']) >= int(args.minPeakSize) and eltDict.has_key(insdata['pME']):
insSum = insertionSummary()
insSum.setPos(insdata['pos'])
insSum.setEltPos(insdata['eltExt'])
insSum.setBestStrand(insdata['eInv'],insdata['eS'])
insSum.setSources(insdata['sources'])
insSum.setClassIndex(insdata['index'])
insSum.eltFam = insdata['pME']
insSum.numreads = insdata['np']
if len(insSum.sources) > 1:
germlineIns.append(insSum)
else:
if re.search('CANCER', insSum.sources[0]):
cancerIns.append(insSum)
elif re.search('NORMAL', insSum.sources[0]):
normalIns.append(insSum)
else:
otherIns.append(insSum)
f.close()
annotatePos(germlineIns,refGenome,args.annotDir,args.outDirName + "/" + args.outBaseName + "/germline.tab.txt",args.printout)
annotatePos(cancerIns,refGenome,args.annotDir,args.outDirName + "/" + args.outBaseName + "/canceronly.tab.txt",args.printout)
annotatePos(normalIns,refGenome,args.annotDir,args.outDirName + "/" + args.outBaseName + "/normalonly.tab.txt",args.printout)
annotatePos(otherIns,refGenome,args.annotDir,args.outDirName + "/" + args.outBaseName + "/other.tab.txt",args.printout)
def main(args):
checkOutDir(args.outBaseName,args.outDirName)
configPath = args.outDirName + "/" + args.outBaseName + "/" + args.configFileName
discordant.checkfile(configPath)
configDict = readConfig(configPath,args.outBaseName,args.outDirName)
# debugging info
logfile = args.outDirName + "/" + args.outBaseName + "/logs/%d" % os.getpid() + "." + args.outBaseName +".peakparser.log"
logging.basicConfig(format='%(asctime)s %(message)s',filename=logfile,level=logging.DEBUG)
# log parameters
logging.info("\noutBaseName=%s\nbamFileName=%s\ntabixFileName=%s\neltFileName=%s"
% (args.outBaseName,configDict['readFileName'],configDict['tabixFileName'],
configDict['eltFileName']))
discordant.checkfile(configDict['readFileName'])
discordant.checkfile(configDict['tabixFileName'])
discordant.checkfile(configDict['eltFileName'])
tabixFile = pysam.Tabixfile(configDict['tabixFileName'], 'r')
eltDict = makeEltDict(configDict['eltFileName'])
# structure for keeping element classes seperate
eltPeakDict = {}
for (eltName,eltClass) in eltDict.iteritems():
eltPeakDict[eltClass] = PeakBuilder(eltClass)
f = open(configDict['readFileName'])
for line in f:
fields = line.rsplit()
eltClass = ''
try:
eltClass = fields[13].rstrip()
except:
print "bad line: " + line
chrom = fields[0]
readPos = int(fields[1])
readStrand = fields[2]
mateElt = fields[3]
mateChrom = fields[4]
mateStrand = fields[5]
matePos = int(fields[6])
eltStart = int(fields[7])
eltEnd = int(fields[8])
eltStrand = fields[9]
eltFullLen = fields[10]
genomeName = fields[11]
peakIndex = fields[12]
pair = pairinfo.Pair(chrom,readPos,readStrand,mateElt,mateChrom,
mateStrand,matePos,eltStart,eltEnd,eltStrand,
eltFullLen,genomeName,peakIndex,eltClass)
if eltPeakDict[eltClass].lastPeakNum != peakIndex:
eltPeakDict[eltClass].peaks.append(eltPeakDict[eltClass].peak)
eltPeakDict[eltClass].peak = pairinfo.Peak()
eltPeakDict[eltClass].peak.addpair(pair)
eltPeakDict[eltClass].lastPeakNum = peakIndex
else:
eltPeakDict[eltClass].peak.addpair(pair)
logging.info("starting long output...")
longOutput(eltPeakDict,args.outBaseName,args.outDirName,tabixFile,eltDict)
logging.info("starting bed output...")
bedOutput(eltPeakDict,args.outBaseName,args.outDirName,configDict['refGenome'],tabixFile,eltDict)
def checkAnnotDir(refGenome,annotDir):
if not os.path.exists(annotDir + "/" + refGenome):
raise IOError("cannot find genome annotation directory for " + refGenome)
return 0
discordant.checkfile(annotDir + "/" + refGenome + "/names.txt")
def main(args):
# create output directory
discordant.prepOutDir(args.outBaseName,args.outDirName,args.overwrite)
sampleList = []
try:
sampleList = args.sampleList
except AttributeError:
discordant.checkfile(args.sampleListFile)
sampleList = open(args.sampleListFile, 'r')
readFileNames = []
bamFileNames = []
sampleNames = []
insertSizes = []
readLengths = []
eltLenDict = None
lastConfig = None
for sampleLine in sampleList:
if not re.search("^#", sampleLine):
(sampleBam,sampleSubDir,refGenome,groupName) = sampleLine.strip().split()
peakparser.checkOutDir(sampleSubDir,args.outDirName)
configPath = args.outDirName + "/" + sampleSubDir + "/" + args.configFileName
discordant.checkfile(configPath)
configDict = peakparser.readConfig(configPath,sampleSubDir,args.outDirName)
eltLenDict = discordant.makeEltLenDict(configDict['eltLenFileName'])
lastConfig = configDict
insertSizes.append(int(configDict['insertSize']))
readLengths.append(int(configDict['readLength']))
readFileName = args.outDirName + "/" + sampleSubDir + "/" + sampleSubDir + ".readpairs.txt"
readFileNames.append(readFileName)
bamFileNames.append(configDict['bamFileName'])
sampleNames.append(sampleSubDir)
maxDist = max(insertSizes) + 2*max(readLengths)
# merge readfiles
outReadFileName = args.outDirName + "/" + args.outBaseName + "/" + args.outBaseName + ".readpairs.txt"
mergeChrPosFiles(readFileNames,outReadFileName,maxDist,eltLenDict)
# write new config file
configPath = args.outDirName + "/" + args.outBaseName + "/" + args.configFileName
configDict = lastConfig
assert len(bamFileNames) == len(sampleNames)
bfnum = 0
for bamFileName in bamFileNames:
bfvname = "bamFileName" + str(bfnum)
configDict[bfvname] = bamFileName
bfnum += 1
snum = 0
for sampleName in sampleNames:
samname = "sampleName" + str(snum)
configDict[samname] = sampleName
snum += 1
configDict['merged'] = 'True'
configDict['outBaseName'] = args.outBaseName
configDict['outDirName'] = args.outDirName
configDict['readFileName'] = outReadFileName
del configDict['bamFileName']
f = open(configPath, 'w')
for k,v in configDict.iteritems():
f.write(k + "=" + v + "\n")
f.close()
