Python Pipeline.varsPipeline方法代码示例

# 需要导入模块: import Pipeline [as 别名]
# 或者: from Pipeline import varsPipeline [as 别名]
def runCharacterize(cwd, rabin, refcmap, contigdir, contigbase, runaligns, xmappath, optargs, nthreads):
    '''Load Pipeline files from first arg; configure CharacterizeModule; run alignments if runaligns;
    report on those alignments or the xmap provided as xmappath.
    '''

    printargs = True

    if not os.path.isfile(os.path.join(cwd,"utilities.py")):
        print "utilities.py missing in dir", cwd, "check -p argument, or run this script in Pipeline dir"
        sys.exit(1)
    import utilities as util

    if not util.checkFile(os.path.join(cwd,"Pipeline.py")):
        print "Pipeline.py missing in dir", cwd, "check -p argument, or run this script in Pipeline dir"
        sys.exit(1)
    import Pipeline

    if not util.checkFile(os.path.join(cwd,"CharacterizeModule.py")):
        print "CharacterizeModule.py missing in dir", cwd, "check -p argument, or run this script in Pipeline dir"
        sys.exit(1)
    import CharacterizeModule as cm

    #if not util.checkFile(os.path.join(cwd,"MapClassesRev.py")):
    #    print "MapClassesRev.py missing in dir", cwd, "check -p argument, or run this script in Pipeline dir"
    #    sys.exit(1)
    #import MapClassesRev

    #use Pipeline objects

    varsP = Pipeline.varsPipeline()

    varsP.optArgumentsFileIn   = optargs
    varsP.RefAlignerBin        = rabin
    varsP.latestMergedCmap     = os.path.join(contigdir, contigbase+".cmap") #file suffix required to be .cmap
    varsP.contigFolder         = contigdir
    varsP.nThreads             = nthreads #necessary otherwise job won't start
    varsP.ref                  = refcmap
    varsP.stdoutlog            = True #enable -stdout -stderr args to RefAligner
    varsP.curCharacterizeCmaps = [varsP.latestMergedCmap]

    if runaligns :
        varsP.contigAlignTarget = contigdir+"/alignref" #this is output dir
        varsP.runSV = False
        varsP.groupContigs = False
        varsP.stageComplete = contigbase
        varsP.outputContigFolder = contigdir
        varsP.memoryLogpath  = os.path.join(contigdir, "memory_log.txt")
        varsP.pipeReportFile = os.path.join(contigdir, "pipeReport.txt")
        varsP.parseArguments() #parses optArgumentsFile
        if printargs :
            print "\nRunning Characterization with arguments:\n" + " ".join(varsP.argsListed('characterizeDefault')) + '\n'
        if hasattr(util, "InitStatus") : #if old version, skip
            util.InitStatus(os.path.join(contigdir, "status.xml")) #needed otherwise call to status_log fails
        charmod = cm.Characterize(varsP) #create Characterize object from CharacterizeModule -- this also calls generateJobList
        xmappath = charmod.xmapTarget #set in Characterize.generateJobList
        charmod.runJobs()
    else :
        #varsP.contigAlignTarget = contigdir #this is dir in which _q and _r cmaps must be located -- contigdir is from cmap; this should be from xmap
        varsP.contigAlignTarget = os.path.split(xmappath)[0]
        print "Loading alignments from\n" + xmappath + "\n"

    #no longer using this in Pipeline
    #print MapClassesRev.TopLevelCharacterization(varsP, [os.path.join(varsP.contigAlignTarget, contigbase)])

    print cm.characterizeContigs(varsP, xmappath) 

# 需要导入模块: import Pipeline [as 别名]
# 或者: from Pipeline import varsPipeline [as 别名]
Usage: python pipelineCL.py -h
"""

    
#this is moved here from utilities/Pipeline.py because those files are used outside of the pipeline
@atexit.register
def on_exit():
	try:
            util.LogStatus("pipeline", "exit", "%d" % os.getpid())
	except:
            pass


if __name__ == "__main__":
    import Pipeline
    varsP = Pipeline.varsPipeline()
    varsP.prerunChecks()
    
    print('  Prerun Tests:\n\t%d ERRORS\n\t%d WARNINGS\n' % (varsP.error, varsP.warning))
    if varsP.error or varsP.warning:
        #print(varsP.message)
	varsP.printMessage()
    if varsP.error:
        print('  EXITING: See errors') 
        sys.exit()
    
    dnpipeline = Pipeline.DNPipeline()
    dnpipeline.run(varsP)
    

# 需要导入模块: import Pipeline [as 别名]
# 或者: from Pipeline import varsPipeline [as 别名]

#.........这里部分代码省略.........
            sys.exit(1)

    #.err file
    errfile = result.errFile
    if errfile and errbinfile :
        print "Warning: .err and .errbin arguments supplied; ignoring .err file"
        errfile = ""
    elif errfile :
        errfile = os.path.realpath(result.errFile)
        if not util.checkFile(errfile, ".err") :
            print "err file supplied but not found or incorrect suffix:", errfile
            sys.exit(1)

    if errfile and not util.checkFile(os.path.join(cwd,"SampleCharModule.py")):
        print "SampleCharModule.py missing in dir", cwd, "check -p argument, or run this script in Pipeline dir"
        sys.exit(1)
    elif errfile :
        import SampleCharModule as scm

    doref = result.ref

    #DONE checking arguments

    print "Using output dir", outdir
    if runaligns :
        print "Aligning", bnxfile, "\nTo", qrypath, "\n"
    else :
        print "Merging", qrypath, "\n"

    startTime = time.time() #time since Epoch
    memory_log = os.path.join(outdir, "memory_log.txt")
    util.initMemoryLog(memory_log)

    varsP = Pipeline.varsPipeline()
    varsP.RefAlignerBin        = rabin
    varsP.contigFolder         = "" #not used but needs to be an attr
    varsP.outputContigFolder   = "" #not used but needs to be a string attr
    varsP.pipeReportFile = os.path.join(outdir, "alignmol_jobs_log.txt")
    varsP.infoReportFile = os.path.join(outdir, "alignmol_log.txt")
    util.InitStatus( os.path.join(outdir, "status.xml") )

    if runaligns :
        varsP.optArgumentsFileIn   = optargs
        varsP.latestMergedCmap     = qrypath #if !doref, need this one
        varsP.ref                  = qrypath #and if doref, need this one
        varsP.nThreads             = nthreads #necessary otherwise job won't start -- max threads per node
        varsP.maxthreads           = maxthreads #threads per job
        p = os.path.split(qrypath)[1]
        varsP.outputContigPrefix   = p[:p.rfind(".")] #filename prefix
        varsP.stdoutlog    = True #use -stdout -stderr

        varsP.memoryLogpath  = os.path.join(outdir, "memory_log.txt")
        varsP.parseArguments() #parses optArgumentsFile
        varsP.replaceParam("alignmol", "-T", pvalue)
        varsP.checkDependencies()
        varsP.RefAlignerBinOrig = rabin
        varsP.prerunLog() #general information in log -- needed for refaligner_version

        noisep = {}
        if errbinfile :
            noisep = {"readparameters": errbinfile}
            #print "Using noise parameters from "+errbinfile+"\n" #move below
        elif errfile :
            noisep = scm.readNoiseParameters(errfile.replace(".err",""))
            if noisep.has_key('readparameters') : #remove this because it's redundant, and it can cause problems with RefAligner compatibility
                del noisep['readparameters']

# 需要导入模块: import Pipeline [as 别名]
# 或者: from Pipeline import varsPipeline [as 别名]
def runSV(cwd, rabin, refcmap, contigdir, contigbase, runaligns, xmappath, optargs, nthreads, maxthreads, bedfile, errfile, outdir, errbinfile, clustargs, groupsv):
    '''Load Pipeline files from first arg; configure CharacterizeModule; run alignments if runaligns;
    report on those alignments or the xmap provided as xmappath.
    '''

    printargs = True

    if not os.path.isfile(os.path.join(cwd,"utilities.py")):
        print "utilities.py missing in dir", cwd, "check -p argument, or run this script in Pipeline dir"
        sys.exit(1)
    import utilities as util

    if not util.checkFile(os.path.join(cwd,"Pipeline.py")):
        print "Pipeline.py missing in dir", cwd, "check -p argument, or run this script in Pipeline dir"
        sys.exit(1)
    import Pipeline

    if not util.checkFile(os.path.join(cwd,"SVModule.py")):
        print "SVModule.py missing in dir", cwd, "check -p argument, or run this script in Pipeline dir"
        sys.exit(1)
    import SVModule as svm

    if errfile and not util.checkFile(os.path.join(cwd,"SampleCharModule.py")):
        print "SampleCharModule.py missing in dir", cwd, "check -p argument, or run this script in Pipeline dir"
        sys.exit(1)
    elif errfile :
        import SampleCharModule as scm

    #use Pipeline objects

    varsP = Pipeline.varsPipeline()

    varsP.optArgumentsFileIn   = optargs
    varsP.RefAlignerBin        = rabin
    varsP.latestMergedCmap     = os.path.join(contigdir, contigbase+".cmap") #file suffix required to be .cmap
    varsP.contigFolder         = os.path.split(contigdir)[0]
    varsP.nThreads             = nthreads #necessary otherwise job won't start -- max threads per node
    varsP.maxthreads           = maxthreads #threads per job
    varsP.ref                  = refcmap
    varsP.stdoutlog            = True #enable -stdout -stderr args to RefAligner
    varsP.curCharacterizeCmaps = [varsP.latestMergedCmap]
    varsP.contigSubDirectories = True #needed for prepareContigIO
    varsP.doAlignMolvRef       = False #do not look for copy number
    varsP.groupSV              = groupsv #mimic Pipeline behavior: group or not 

    if runaligns :
        #varsP.contigAlignTarget = outdir
        varsP.runSV = False
        varsP.groupContigs = False
        varsP.stdoutlog    = True #use -stdout -stderr
        varsP.stageComplete = contigbase
        varsP.outputContigPrefix = getContigPrefix(util, contigdir) #if outdir is not supplied, this is used as dir prefix; also used as file pref for -o arg
        varsP.outputContigFolder = contigdir #cmaps are copied from here

        if not outdir :
            outdir = contigdir+"_sv" #this will be outdir of sv jobs
        if os.path.isdir(outdir) :
            if not util.checkDir(outdir) : #check writeable
                print "\nERROR: Output dir is not writeable:\n", outdir, "\n"                
                sys.exit(1)
            elif outdir == contigdir :
                print "\nERROR: Output dir cannot be same as input dir:\n", outdir, "\n"                
                sys.exit(1)                
            print "\nWARNING: Output dir already exists, results will be overwritten:\n", outdir, "\n"
        elif not util.checkDir(outdir) : #does not exist, make, if False, can't make or not writeable
            print "\nERROR: Output dir cannot be created or is not writeable:\n", outdir, "\n"
            sys.exit(1)

        if clustargs :
            os.putenv('SGE_ROOT', '/var/lib/gridengine') #do I want this???
            varsP.onCluster = True
            varsP.clusterLogDir = os.path.join(outdir, 'ClusterLogs')
            util.checkDir(varsP.clusterLogDir) #make it
            varsP.checkCluster()
            varsP.clusterArgumentsFileIn = clustargs #required for parseArguments
            varsP.parseArguments(readingClusterFile=True)
            if varsP.error :
                print varsP.message
                sys.exit(1)
            varsP.RefAlignerBin += "${BINARY_SUFFIX:=}" #copy from varsPipeline, handled by external script on phi host

        varsP.pipeReportFile = os.path.join(outdir, "sv_jobs_log.txt")
        varsP.infoReportFile = os.path.join(outdir, "sv_log.txt")
        varsP.memoryLogpath  = os.path.join(outdir, "memory_log.txt")
        if bedfile :
            varsP.bedFile = bedfile
        util.InitStatus( os.path.join(outdir, "status.xml") )
        varsP.parseArguments() #parses optArgumentsFile
        varsP.checkDependencies()
        varsP.RefAlignerBinOrig = rabin
        varsP.prerunLog() #general information in log -- needed for refaligner_version
        if printargs :
            print "\nRunning SV detection with arguments ("+os.path.split(optargs)[1]+"):\n" + " ".join(varsP.argsListed('svdetect')) + '\n'

        noisep = {}
        if errbinfile :
            noisep = {"readparameters": errbinfile}
            print "Using noise parameters from "+errbinfile+"\n"
        elif errfile :
            noisep = scm.readNoiseParameters(errfile.replace(".err",""))
#.........这里部分代码省略.........