################################################################################
## Perform RNASeQC analysis
################################################################################
args['tophatBam'] = os.path.join(args['tophatSampleDir'], 'accepted_hits.bam')
args['mdupBam'] = os.path.join(args['tophatSampleDir'],
args['name'] + '_tophat_mdup.bam')
args['mdupLog'] = os.path.join(args['tophatSampleDir'],
args['name'] + '_mdup.log')
args['seqcLog'] = os.path.join(args['tophatSampleDir'],
args['name'] + '_rnaseqc.log')
# Create command to mark duplicates
markDupCommand = picard.markDuplicates(inBam=args['tophatBam'],
outBam=args['mdupBam'],
logFile=args['mdupLog'],
removeDuplicates=False,
picardPath=paths['picard'],
javaPath=paths['java'],
delete=True)
# Create command to peform RNASeqC
seqcCommand = bamQC.RNASeqC(inBam=args['mdupBam'],
fasta=args['<bwt2genindex>'] + '.fa',
gtf=args['<gtf>'],
rRNA=args['<rrna>'],
outDir=args['tophatSampleDir'],
outPrefix=args['name'],
seqcPath=paths['rnaseqc'],
javaPath=paths['java'],
singleEnd=args['--singleend'])
# Combine mark, index and RNASeqC commands
seqcComboCommand = '%s && %s' % (markDupCommand, seqcCommand)
pmDict[('samtools', 'modules')],
memory=7,
stdout=alignLog,
stderr=alignLog)
# Dedup individual files and store outputs
if args['--dedupindv']:
# Create file names
dedupLog = os.path.join(args['<logfolder>'],
'rg{}.dedup.log'.format(readgroup))
dedupBam = args['<outbam>'][:-4] + '.rg{}.dedup.bam'.format(readgroup)
# Create command for marking duplicates
dedupCommand = picard.markDuplicates(inBam=alignBam,
outBam=dedupBam,
logFile=dedupLog + '1',
picardPath=pmDict[('picard',
'path')],
javaPath='java',
removeDuplicates=False,
delete=True,
memory=20)
# Add command to dictionary
dedupJobID = jobObject.add(command=dedupCommand,
processors=4,
modules=pmDict[('picard', 'modules')],
memory=6,
stdout=dedupLog + '2',
stderr=dedupLog + '2',
depend=[alignJobID])
# Store output BAM files and job IDs
bamList.append(dedupBam)
jobList.append(dedupJobID)
################################################################################
## Perform RNASeQC analysis
################################################################################
args['tophatBam'] = os.path.join(args['tophatSampleDir'], 'accepted_hits.bam')
args['mdupBam'] = os.path.join(args['tophatSampleDir'],
args['name'] + '_tophat_mdup.bam')
args['mdupLog'] = os.path.join(args['tophatSampleDir'],
args['name'] + '_mdup.log')
args['seqcLog'] = os.path.join(args['tophatSampleDir'],
args['name'] + '_rnaseqc.log')
# Create command to mark duplicates
markDupCommand = picard.markDuplicates(
inBam = args['tophatBam'],
outBam = args['mdupBam'],
logFile = args['mdupLog'],
removeDuplicates = False,
picardPath = paths['picard'],
javaPath = paths['java'],
delete = True
)
# Create command to peform RNASeqC
seqcCommand = bamQC.RNASeqC(
inBam = args['mdupBam'],
fasta = args['<bwt2genindex>'] + '.fa',
gtf = args['<gtf>'],
rRNA = args['<rrna>'],
outDir = args['tophatSampleDir'],
outPrefix = args['name'],
seqcPath = paths['rnaseqc'],
javaPath = paths['java'],
singleEnd = args['--singleend']
###############################################################################
## Mark duplicates in Tophat output
###############################################################################
args['tophatBam'] = os.path.join(args['tophatSampleDir'], 'accepted_hits.bam')
args['mdupBam'] = os.path.join(args['tophatSampleDir'],
args['name'] + '_tophat2_mdup.bam')
args['mdupLog1'] = os.path.join(args['tophatSampleDir'],
args['name'] + '_mdup.log1')
args['mdupLog2'] = os.path.join(args['tophatSampleDir'],
args['name'] + '_mdup.log2')
# Create command to mark duplicates
markDupCommand = picard.markDuplicates(
inBam = args['tophatBam'],
outBam = args['mdupBam'],
logFile = args['mdupLog1'],
removeDuplicates = False,
picardPath = pmDict[('picard', 'path')],
delete = True,
memory = 10
)
# Submit mark duplicates job ID
markDupJobID = slurmJobs.add(
command = markDupCommand,
stdout = args['mdupLog2'],
stderr = args['mdupLog2'],
depend = [tophat2AlignJobID],
memory = 12,
modules = pmDict[('picard', 'modules')]
)
###############################################################################
modules=pmDict[('bowtie2', 'modules')] +
pmDict[('samtools', 'modules')],
depend=trimJobList,
stderr=bowtie2Log)
# Delete trim files if generated
if args['--trim']:
rmCommand = 'rm {}'.format(' '.join(read1List + read2List))
slurmJobs.add(rmCommand, depend=qcJobList + [bowtie2JobID])
# Mark duplicates in BAM files
dedupLog = args['<logprefix>'] + '.dedup.log'
dedupCommand = picard.markDuplicates(inBam=unmarkedBAM,
outBam=args['<outbam>'],
logFile=dedupLog + '1',
picardPath=pmDict[('picard', 'path')],
javaPath='java',
removeDuplicates=False,
delete=True,
memory=30)
# Add job to queue
dedupJobID = slurmJobs.add(command=dedupCommand,
processors=6,
memory=6,
stdout=dedupLog + '2',
stderr=dedupLog + '2',
modules=pmDict[('picard', 'modules')],
depend=[bowtie2JobID])
#Submit jobs
slurmJobs.submit(verbose=True, check_sub=False)
###############################################################################
## Mark duplicates in Tophat output
###############################################################################
args['tophatBam'] = os.path.join(args['tophatSampleDir'], 'accepted_hits.bam')
args['mdupBam'] = os.path.join(args['tophatSampleDir'],
args['name'] + '_tophat2_mdup.bam')
args['mdupLog1'] = os.path.join(args['tophatSampleDir'],
args['name'] + '_mdup.log1')
args['mdupLog2'] = os.path.join(args['tophatSampleDir'],
args['name'] + '_mdup.log2')
# Create command to mark duplicates
markDupCommand = picard.markDuplicates(
inBam = args['tophatBam'],
outBam = args['mdupBam'],
logFile = args['mdupLog1'],
removeDuplicates = False,
picardPath = pmDict[('picard', 'path')],
delete = True,
memory = 10
)
# Submit mark duplicates job ID
markDupJobID = slurmJobs.add(
command = markDupCommand,
stdout = args['mdupLog2'],
stderr = args['mdupLog2'],
depend = [tophat2AlignJobID],
memory = 12,
modules = pmDict[('picard', 'modules')]
)
###############################################################################
'realignlog' : logPrefix + '_realign.log',
'recallog' : logPrefix + '_recal.log'
}
# Generate command for alignment
alignCommand = fastqAlign.bwaMemAlign(
index = args['<index>'], outFile = outfiles['initialbam'],
read1 = read1[0], read2 = read2[0], bwaPath = paths['bwa'],
threads = args['--threads'], sampleName = args['name'],
libraryID = args['prefix'], readGroup = 1, platform = 'ILLUMINA',
markSecondary = True, check = True, samtoolsPath = paths['samtools'],
memory = 2, nameSort = False
)
# Mark duplicates using picard
dedupCommand = picard.markDuplicates(
inBam = outfiles['initialbam'], outBam = outfiles['dedupbam'],
logFile = outfiles['deduplog1'], picardPath = paths['picard'],
javaPath = paths['java'], removeDuplicates = True, delete = True
)
# Perform local realignment
realignCommand = gatk.gatkRealign(
inBam = outfiles['dedupbam'], outBam = outfiles['realignbam'],
inVcf = args['<indelvcf>'], reference = args['<index>'],
javaPath = paths['java'], gatkPath = paths['gatk'], delete = True,
threads = 4, listFile = outfiles['listfile']
)
recalCommand = gatk.bsqr(
inBam = outfiles['realignbam'], outBam = outfiles['recalbam'],
inVcf = args['<snpvcf>'], reference = args['<index>'],
bsqrTable = outfiles['bsqrfile'], javaPath = paths['java'],
gatkPath = paths['gatk'], delete = True
)
