def __init__(self, **keywords): """ 2011-7-11 """ CompareAlleleFrequencyOfTwoPopulationFromOneVCFFolder.__init__(self, **keywords) """
def addAllJobs(self, workflow=None, inputVCFData=None, chr2IntervalDataLs=None, \
GenomeAnalysisTKJar=None, samtools=None, \
CreateSequenceDictionaryJava=None, CreateSequenceDictionaryJar=None, \
BuildBamIndexFilesJava=None, BuildBamIndexJar=None,\
mv=None, \
refFastaFList=None, \
needFastaIndexJob=False, needFastaDictJob=False, \
data_dir=None, no_of_gatk_threads = 1, \
intervalSize=3000, intervalOverlapSize=0, \
outputDirPrefix="", transferOutput=True, job_max_memory=2000, **keywords):
"""
2012.10.15
architect of the whole map-reduce framework
call the parent's addAllJobs in a loop
"""
samplingReturnDataLs = []
for i in xrange(self.noOfSamplings):
oneSamplingReturnData = CompareAlleleFrequencyOfTwoPopulationFromOneVCFFolder.addAllJobs(self, \
workflow=workflow, inputVCFData=inputVCFData, \
chr2IntervalDataLs=chr2IntervalDataLs, samtools=samtools, \
GenomeAnalysisTKJar=GenomeAnalysisTKJar, \
CreateSequenceDictionaryJava=CreateSequenceDictionaryJava, CreateSequenceDictionaryJar=CreateSequenceDictionaryJar, \
BuildBamIndexFilesJava=BuildBamIndexFilesJava, BuildBamIndexJar=BuildBamIndexJar,\
mv=mv, \
refFastaFList=refFastaFList,\
needFastaIndexJob=needFastaIndexJob, needFastaDictJob=needFastaDictJob, \
data_dir=data_dir, no_of_gatk_threads = 1, \
intervalSize=intervalSize, intervalOverlapSize=intervalOverlapSize, \
outputDirPrefix='%s_%s_'%(outputDirPrefix, i), transferOutput=transferOutput, job_max_memory=job_max_memory,\
**keywords)
samplingReturnDataLs.append(oneSamplingReturnData)
topOutputDir = "%sFinalReduce"%(outputDirPrefix)
topOutputDirJob = yh_pegasus.addMkDirJob(workflow, mkdir=workflow.mkdirWrap, outputDir=topOutputDir)
#a ReduceMatrixByAverageColumnsWithSameKey job
outputFile = File(os.path.join(topOutputDir, 'medianAlleleSharingStatAcrossAllSampling.tsv'))
medianReduceJob = self.addStatMergeJob(workflow, statMergeProgram=workflow.ReduceMatrixByAverageColumnsWithSameKey, \
outputF=outputFile, extraArguments='--keyColumnLs 0 -v 1-8', parentJobLs=[topOutputDirJob], \
extraDependentInputLs=None, transferOutput=True)
#a MergeSameHeaderTablesIntoOne job
outputFile = File(os.path.join(topOutputDir, 'alleleSharingStatAcrossAllSampling.tsv'))
mergeJob = self.addStatMergeJob(workflow, statMergeProgram=workflow.MergeSameHeaderTablesIntoOne, \
outputF=outputFile, extraArguments=None, parentJobLs=[topOutputDirJob], \
extraDependentInputLs=None, transferOutput=True)
for oneSamplingReturnData in samplingReturnDataLs:
self.addInputToStatMergeJob(workflow=workflow, statMergeJob=medianReduceJob, parentJobLs=[oneSamplingReturnData.estimateOutlierJob])
self.addInputToStatMergeJob(workflow=workflow, statMergeJob=mergeJob, parentJobLs=[oneSamplingReturnData.estimateOutlierJob])
outputFile = File( os.path.join(topOutputDirJob.output, 'outlierFraction_Hist.png'))
#no spaces or parenthesis or any other shell-vulnerable letters in the x or y axis labels (whichColumnPlotLabel, xColumnPlotLabel)
self.addDrawHistogramJob(workflow=workflow, executable=workflow.DrawHistogram, inputFileList=[mergeJob.output], \
outputFile=outputFile, \
whichColumn=None, whichColumnHeader="outlierFraction", whichColumnPlotLabel="outlierFraction", \
logY=False, positiveLog=True, logCount=False, valueForNonPositiveYValue=-1,\
minNoOfTotal=5,\
figureDPI=100, samplingRate=1,\
parentJobLs=[topOutputDirJob, mergeJob], \
extraDependentInputLs=None, \
extraArguments=None, transferOutput=True, job_max_memory=2000)
outputFile = File( os.path.join(topOutputDirJob.output, 'AFS_cor_Hist.png'))
#no spaces or parenthesis or any other shell-vulnerable letters in the x or y axis labels (whichColumnPlotLabel, xColumnPlotLabel)
self.addDrawHistogramJob(workflow=workflow, executable=workflow.DrawHistogram, inputFileList=[mergeJob.output], \
outputFile=outputFile, \
whichColumn=None, whichColumnHeader="corr", whichColumnPlotLabel="AFSCorrelation", \
logY=False, positiveLog=True, logCount=False, valueForNonPositiveYValue=-1,\
minNoOfTotal=5,\
figureDPI=100, samplingRate=1,\
parentJobLs=[topOutputDirJob, mergeJob], \
extraDependentInputLs=None, \
extraArguments=None, transferOutput=True, job_max_memory=2000)
sys.stderr.write("%s jobs.\n"%(self.no_of_jobs))
