if bam.startswith(prefix):
sampleDict[individual][tissue].append(bam)
# Check each idividual has a normal and every tissue has a BAM
for individual in sampleDict:
# Check for normal
if not 'NORM' in sampleDict[individual]:
raise IOError('no normal for %s' % (individual))
# check for individual BAM files
for tissue in sampleDict[individual]:
if len(sampleDict[individual][tissue]) < 2:
raise IOError('no bam found for %s %s' % (individual, tissue))
if len(sampleDict[individual][tissue]) > 2:
raise IOError('multiple bams found for %s %s' %
(individual, tissue))
# Loop through individuals data
slurmJobs = slurm.submitJobs()
for individual, indvData in sampleDict.items():
# Create lists to store job IDs
pileupFileList = []
pileupJobList = []
varscanJobList = []
# Create log directory
logDir = os.path.join(args['<outdir>'], individual + '.log')
if not os.path.isdir(logDir):
os.mkdir(logDir)
# Create and store pileup commands
for tissue, tissueData in indvData.items():
# Extract sample name and BAM file
name, bam = tissueData
# Create and store file names
pileup = os.path.join(args['<outdir>'], name + '.mpileup')
read1List.extend(read1)
read1List = [os.path.abspath(x) for x in read1List]
else:
for prefix in args['<prefix>']:
read1, read2 = fastqFind.findFastq(prefix=prefix,
dirList=args['<indir>'],
pair=True)
read1List.extend(read1)
read2List.extend(read2)
read1List = [os.path.abspath(x) for x in read1List]
read2List = [os.path.abspath(x) for x in read2List]
# Raise Error if no Fastq files identified
if len(read1List) == 0:
raise IOError('Failed to find FASTQ files')
# Create object to store jobs and create log folder
jobObject = slurm.submitJobs()
bamList = []
jobList = []
args['<logfolder>'] = args['<outbam>'][:-4]
os.mkdir(args['<logfolder>'])
# Perform alignment for each pair of FASTQ files
for readgroup, (read1, read2) in enumerate(
itertools.izip_longest(read1List, read2List)):
# Modify and format read group and library id
readgroup = format(readgroup + 1, '03d')
# Create file names for alignment
alignLog = os.path.join(args['<logfolder>'],
'rg{}.align.log'.format(readgroup))
alignBam = args['<outbam>'][:-4] + '.rg{}.sort.bam'.format(readgroup)
# Generate command for alignment
alignCommand = fastqAlign.bwaMemAlign(
# Generate and store standard output directories
args['fastqDir'] = os.path.join(args['<outdir>'], 'trimFastq')
args['fastqSampleDir'] = os.path.join(args['fastqDir'], args['name'])
args['rsemTranDir'] = os.path.join(args['<outdir>'], 'rsemAlign')
args['rsemTranSampleDir'] = os.path.join(args['rsemTranDir'], args['name'])
args['tophatDir'] = os.path.join(args['<outdir>'], 'tophat2Align')
args['tophatSampleDir'] = os.path.join(args['tophatDir'], args['name'])
dirList = [args['fastqDir'], args['fastqSampleDir'], args['rsemTranDir'],
args['rsemTranSampleDir'], args['tophatDir'], args['tophatSampleDir']]
# Create output directories
for directory in dirList:
if not os.path.exists(directory):
os.mkdir(directory)
# Extract path and module data and create object to submit commands
pmDict = slurm.parsePathModule(args['<paths>'])
slurmJobs = slurm.submitJobs()
###############################################################################
## Find and process fastq files
###############################################################################
# Extract fastq files and check
if args['--singleend']:
args['read1'] = fastqFind.findFastq(
prefix = args['prefix'],
dirList = args['<indir>'].split(','),
pair = False
)
else:
args['read1'], args['read2'] = fastqFind.findFastq(
prefix = args['prefix'],
dirList = args['<indir>'].split(','),