def processSample(sampleName, sampleNameDict):
'''pipeline for sample '''
# process communication
mng = Manager()
libraryBamFileList = mng.list()
# This libraryBamFileList is a list contains seval bams from one sample.
# init mult-processer
record = []
for sampleID in sampleNameDict:
Processer = Process(name=sampleID,
target=processSampleFromLibarary,
args=(
sampleID,
sampleNameDict[sampleID],
libraryBamFileList,
))
Processer.start()
record.append(Processer)
# wait for processer
for proc in record:
proc.join()
###################### 2.1 post mapping ####################
bamFileDir = os.path.dirname(libraryBamFileList[0])
finalBamFilePath = os.path.join(bamFileDir, "%s_properly.bam" % sampleName)
mergedBamFilePath = os.path.join(bamFileDir, "%s_merged.bam" % sampleName)
# merge the bam from each line to one final bam
command = NGSTools.picard_merge(libraryBamFileList, mergedBamFilePath, cfg)
NGSTools.writeCommands(command,
bamFileDir + '/picard_mergebam_' + sampleName +
'.sh',
run=_run)
###################### 3. filter bam ######################
#command = 'samtools view -Sb -h -f 2 -q 10 %s > %s ' % (mergedBamFilePath, finalBamFilePath)
command = 'samtools view -Sb -h -q 10 %s > %s ' % (mergedBamFilePath,
finalBamFilePath)
NGSTools.writeCommands(command,
bamFileDir + '/filterBam_' + sampleName + '.sh',
run=_run)
###################### 4. romove duplicates ##################
if RMDUP:
command = NGSTools.picard_rmdup(finalBamFilePath, True, cfg)
NGSTools.writeCommands(command,
bamFileDir + '/picard_rmdup_' + sampleName +
'.sh',
run=_run)
finalBamFilePath = re.sub(r'.bam$', '.rmdup.bam', finalBamFilePath)
def processSample(sampleName, sampleNameDict):
'''pipeline for sample '''
# process communication
mng = Manager()
libraryBamFileList = mng.list()
# This libraryBamFileList is a list contains seval bams from one sample.
# init mult-processer
record = []
for sampleID in sampleNameDict:
Processer = Process(name = sampleID, target = processSampleFromLibarary, args = (sampleID, sampleNameDict[sampleID], libraryBamFileList, ))
Processer.start()
record.append(Processer)
# wait for processer
for proc in record:
proc.join()
###################### 2.1 post mapping ####################
bamFileDir = os.path.dirname(libraryBamFileList[0])
finalBamFilePath = os.path.join(bamFileDir, "%s_properly.bam" % sampleName)
mergedBamFilePath = os.path.join(bamFileDir, "%s_merged.bam" % sampleName)
# merge the bam from each lane to one final bam
command = NGSTools.picard_merge(libraryBamFileList, mergedBamFilePath, cfg)
NGSTools.writeCommands(command, bamFileDir+'/picard_mergebam_'+sampleName+'.sh', run=_run)
###################### 2.2. filter bam ######################
#command = 'samtools view -Sb -h -f 2 -q 10 %s > %s ' % (mergedBamFilePath, finalBamFilePath)
#command = 'samtools view -Sb -h -q 10 %s > %s ' % (mergedBamFilePath, finalBamFilePath)
#NGSTools.writeCommands(command, bamFileDir+'/filterBam_'+sampleName+'.sh', run=_run)
###################### 3. romove duplicates ##################
if RMDUP:
command = NGSTools.picard_rmdup(mergedBamFilePath, True, cfg)
NGSTools.writeCommands(command, bamFileDir+'/picard_rmdup_'+sampleName+'.sh', run=_run)
finalBamFilePath = re.sub(r'.bam$', '.rmdup.bam', finalBamFilePath)
####################### 4. call SNP ######################
if MPILEUP:
SNP_out = os.path.join(args.outDir, "SNP", sampleName)
NGSTools._mkdir(SNP_out)
command, rawVcf = NGSTools.bcftools_call(finalBamFilePath, cfg, outdir=SNP_out, sampleName=sampleName)
NGSTools.writeCommands(command, SNP_out+'/bcftools_call_'+sampleName+'.sh', run=_run)
outVcf = rawVcf.replace("vcf$", "flt.vcf")
command = NGSTools.bcftools_filter(rawVcf, outVcf, cfg)
NGSTools.writeCommands(command, SNP_out+'/bcftools_filter_'+sampleName+'.sh', run=_run)
def processSample(sampleName, sampleNameDict):
'''pipeline for sample '''
# process communication
mng = Manager()
libraryBamFileList = mng.list()
# This libraryBamFileList is a list contains seval bams from one sample.
# init mult-processer
record = []
for sampleID in sampleNameDict:
Processer = Process(name=sampleID,
target=processSampleFromLibarary,
args=(
sampleID,
sampleNameDict[sampleID],
libraryBamFileList,
))
Processer.start()
record.append(Processer)
# wait for processer
for proc in record:
proc.join()
###################### 2.1 post mapping ####################
bamFileDir = os.path.dirname(libraryBamFileList[0])
finalBamFilePath = os.path.join(bamFileDir, "%s_final.bam" % sampleName)
# merge the bam from each line to one final bam
command = NGSTools.picard_merge(libraryBamFileList, finalBamFilePath, cfg)
NGSTools.writeCommands(command,
args.outDir + '/mapping/picard_mergebam_' +
sampleName + '.sh',
run=_run)
###################### 3. romove duplicates ##################
if RMDUP:
command = NGSTools.picard_rmdup(finalBamFilePath, True, cfg)
NGSTools.writeCommands(command,
args.outDir + '/mapping/picard_rmdup_' +
sampleName + '.sh',
run=_run)
finalBamFilePath = re.sub(r'.bam$', '.rmdup.bam', finalBamFilePath)
##################### 4. DMR calling ##################
if Methylation_extractor:
NGSTools.methylation_extractor(finalBamFilePath,
bamFileDir + '/' + sampleName, cfg)
def processSample(sampleName, sampleNameDict): '''pipeline for sample ''' # process communication mng = Manager() libraryBamFileList = mng.list() # This libraryBamFileList is a list contains seval bams from one sample. # init mult-processer record = [] for sampleID in sampleNameDict: Processer = Process(name = sampleID, target = processSampleFromLibarary, args = (sampleID, sampleNameDict[sampleID], libraryBamFileList, )) Processer.start() record.append(Processer) # wait for processer for proc in record: proc.join() ###################### 2.1 post mapping #################### bamFileDir = os.path.dirname(libraryBamFileList[0]) finalBamFilePath = os.path.join(bamFileDir, "%s_properly.bam" % sampleName) mergedBamFilePath = os.path.join(bamFileDir, "%s_merged.bam" % sampleName) # merge the bam from each line to one final bam command = NGSTools.picard_merge(libraryBamFileList, mergedBamFilePath, cfg) NGSTools.writeCommands(command, bamFileDir+'/picard_mergebam_'+sampleName+'.sh', run=_run) ###################### 3. filter bam ###################### command = 'samtools view -Sb -h -f 2 -q 10 %s > %s ' % (mergedBamFilePath, finalBamFilePath) NGSTools.writeCommands(command, bamFileDir+'/filterBam_'+sampleName+'.sh', run=_run) ###################### 4. romove duplicates ################## if RMDUP: command = NGSTools.picard_rmdup(finalBamFilePath, True, cfg) NGSTools.writeCommands(command, bamFileDir+'/picard_rmdup_'+sampleName+'.sh', run=_run) finalBamFilePath = re.sub(r'.bam$', '.rmdup.bam', finalBamFilePath)
def processSample(sampleName, sampleNameDict): '''pipeline for sample ''' # process communication mng = Manager() libraryBamFileList = mng.list() # This libraryBamFileList is a list contains seval bams from one sample. # init mult-processer record = [] for sampleID in sampleNameDict: Processer = Process(name = sampleID, target = processSampleFromLibarary, args = (sampleID, sampleNameDict[sampleID], libraryBamFileList, )) Processer.start() record.append(Processer) # wait for processer for proc in record: proc.join() ###################### 2.1 post mapping #################### bamFileDir = os.path.dirname(libraryBamFileList[0]) finalBamFilePath = os.path.join(bamFileDir, "%s_final.bam" % sampleName) # merge the bam from each line to one final bam command = NGSTools.picard_merge(libraryBamFileList, finalBamFilePath, cfg) NGSTools.writeCommands(command, args.outDir+'/mapping/picard_mergebam_'+sampleName+'.sh', run=_run) ###################### 3. romove duplicates ################## if RMDUP: command = NGSTools.picard_rmdup(finalBamFilePath, True, cfg) NGSTools.writeCommands(command, args.outDir+'/mapping/picard_rmdup_'+sampleName+'.sh', run=_run) finalBamFilePath = re.sub(r'.bam$', '.rmdup.bam', finalBamFilePath) ##################### 4. DMR calling ################## if Methylation_extractor: NGSTools.methylation_extractor(finalBamFilePath, bamFileDir+'/'+sampleName, cfg)
# GATK HC (in RNA-seq mode) call variations
outdir = os.path.join(args.outDir, 'variation')
NGSTools._mkdir(outdir)
for condition in set(finalBam.values()):
bams = []
for bam in finalBam.keys():
if finalBam[bam] == condition:
bams.append(bam)
out = os.path.join(outdir, condition)
NGSTools._mkdir(out)
mergedBam = '%s/%s.bam' % (out, condition)
script = NGSTools.picard_merge(bams, mergedBam, cfg)
NGSTools.writeCommands(script,
'%s/picard_merge_%s.sh' % (out, condition),
False)
script = NGSTools.GATK_HC(mergedBam, cfg, out, condition)
NGSTools.writeCommands(script, '%s/gatk_HC_%s.sh' % (out, condition),
False)
rawVcf = '%s/%s.raw.snps.indels.vcf' % (out, condition)
fltVcf = '%s/%s.flt.snps.indels.vcf' % (out, condition)
script = NGSTools.GATK_filter(mergedBam, rawVcf, fltVcf, cfg)
NGSTools.writeCommands(script,
# GATK HC (in RNA-seq mode) call variations outdir = os.path.join(args.outDir, 'variation') NGSTools._mkdir(outdir) for condition in set(finalBam.values()): bams = [] for bam in finalBam.keys(): if finalBam[bam] == condition: bams.append(bam) out = os.path.join(outdir, condition) NGSTools._mkdir(out) mergedBam = '%s/%s.bam' % (out, condition) script = NGSTools.picard_merge(bams, mergedBam, cfg) NGSTools.writeCommands(script, '%s/picard_merge_%s.sh' % (out, condition), False) script = NGSTools.GATK_HC(mergedBam, cfg, out, condition) NGSTools.writeCommands(script, '%s/gatk_HC_%s.sh' % (out, condition), False) rawVcf = '%s/%s.raw.snps.indels.vcf' % (out, condition) fltVcf = '%s/%s.flt.snps.indels.vcf' % (out, condition) script = NGSTools.GATK_filter(mergedBam, rawVcf, fltVcf, cfg) NGSTools.writeCommands(script, '%s/gatk_filter_%s.sh' % (out, condition), False)
def processSample(sampleName, sampleNameDict):
'''pipeline for sample '''
# process communication
mng = Manager()
libraryBamFileList = mng.list()
# This libraryBamFileList is a list contains seval bams from one sample.
# init mult-processer
record = []
for sampleID in sampleNameDict:
Processer = Process(name=sampleID,
target=processSampleFromLibarary,
args=(
sampleID,
sampleNameDict[sampleID],
libraryBamFileList,
))
Processer.start()
record.append(Processer)
# wait for processer
for proc in record:
proc.join()
###################### 2.1 post mapping ####################
bamFileDir = os.path.dirname(libraryBamFileList[0])
mergedBamFilePath = os.path.join(bamFileDir, "%s_merged.bam" % sampleName)
finalBamFilePath = mergedBamFilePath
# merge the bam from each lane to one final bam
command = NGSTools.picard_merge(libraryBamFileList, mergedBamFilePath, cfg)
NGSTools.writeCommands(command,
bamFileDir + '/picard_mergebam_' + sampleName +
'.sh',
run=_run)
###################### 2.2. filter bam ######################
#command = 'samtools view -Sb -h -f 2 -q 10 %s > %s ' % (mergedBamFilePath, finalBamFilePath)
#command = 'samtools view -Sb -h -q 10 %s > %s ' % (mergedBamFilePath, finalBamFilePath)
#NGSTools.writeCommands(command, bamFileDir+'/filterBam_'+sampleName+'.sh', run=_run)
###################### 3. romove duplicates ##################
if RMDUP:
command = NGSTools.picard_rmdup(mergedBamFilePath, True, cfg)
NGSTools.writeCommands(command,
bamFileDir + '/picard_rmdup_' + sampleName +
'.sh',
run=_run)
finalBamFilePath = re.sub(r'.bam$', '.rmdup.bam', finalBamFilePath)
####################### 4. call SNP ######################
if MPILEUP:
SNP_out = os.path.join(args.outDir, "SNP", sampleName)
NGSTools._mkdir(SNP_out)
command, rawVcf = NGSTools.bcftools_call(finalBamFilePath,
cfg,
outdir=SNP_out,
sampleName=sampleName)
NGSTools.writeCommands(command,
SNP_out + '/bcftools_call_' + sampleName +
'.sh',
run=_run)
outVcf = rawVcf.replace("vcf$", "flt.vcf")
command = NGSTools.bcftools_filter(rawVcf, outVcf, cfg)
NGSTools.writeCommands(command,
SNP_out + '/bcftools_filter_' + sampleName +
'.sh',
run=_run)
