def fastq_convertion_and_pairs_capture(self):
"""
Convert cram2.0 file to fastq file with samtools1.1 and picard
"""
jobs = []
cff_file = self.args.cff.name
cram_file = self.args.cff.name
#out_dir = os.path.join("fusion_pairs_capture", "cram_fastq")
for readset in self.readsets:
out_dir = os.path.join("fusion_pairs_capture", "captured_bam", readset.sample.name)
if not readset.fastq1:
if readset.cram:
# convert cram to bam then to fastq, fastq and bam are saved on localhd
out_bam = os.path.join("$TMPDIR", os.path.basename(readset.cram)+".bam")
fastq1 = out_bam + ".1.fastq"
fastq2 = out_bam + ".2.fastq"
cram2bam_job = samtools_1_1.view(readset.cram, out_bam, "-b")
bam2fastq_job = picard.sam_to_fastq(out_bam, fastq1, fastq2)
# bwa aln fastqs to capture reference
out_bam = os.path.join(out_dir, "captured.bam")
ref = os.path.join("fusion_pairs_capture", "fusion_refs", os.path.basename(cff_file)+".fa")
capture_job = bwa_fusion_reads_capture.bwa_fusion_reads_capture(fastq1, fastq2, ref, out_bam, read_group=None, ini_section='bwa_fusion_reads_capture')
job = concat_jobs([
Job(command="mkdir -p " + out_dir),
cram2bam_job,
bam2fastq_job,
capture_job
], name="convert_cram_to_fastq")
jobs.append(job)
else:
raise Exception("Error: CRAM file not available for readset \"" + readset.name + "\"!")
else:
fastq1 = readset.fastq1
fastq2 = readset.fastq2
# bwa aln fastqs to capture reference
out_bam = os.path.join(out_dir, "captured.bam")
ref = os.path.join("fusion_pairs_capture", "fusion_refs", os.path.basename(cff_file)+".genes.fa")
capture_job = bwa_fusion_reads_capture.bwa_fusion_reads_capture(fastq1, fastq2, ref, out_bam, read_group=None, ini_section='bwa_fusion_reads_capture')
job = concat_jobs([
Job(command="mkdir -p " + out_dir),
capture_job
], name="bwa_fusion_reads_capture")
jobs.append(job)
return jobs
def picard_sam_to_fastq(self):
"""
Convert SAM/BAM files from the input readset file into FASTQ format
if FASTQ files are not already specified in the readset file. Do nothing otherwise.
rerwritten from common.Illumina.picard_sam_to_fastq, make directory for this step under result folder in case the orginal bam file directory is not writtable
"""
jobs = []
for readset in self.readsets:
# If readset FASTQ files are available, skip this step
if not readset.fastq1:
if readset.cram:
# convert cram to bam then to fastq. fastq and bam are saved on localhd
out_bam = os.path.join("$TMPDIR", os.path.basename(readset.cram)+".bam")
cram2bam_job = samtools_1_1.view(readset.cram, out_bam)
if readset.run_type == "PAIRED_END":
out_dir = os.path.join("fusions", "picard_sam_to_fastq", readset.sample.name)
fastq1 = os.path.join(out_dir, os.path.basename(re.sub("\.bam$", ".pair1.fastq.gz", out_bam)))
fastq2 = os.path.join(out_dir, os.path.basename(re.sub("\.bam$", ".pair2.fastq.gz", out_bam)))
else:
raise Exception("Error: run type \"" + readset.run_type +
"\" is invalid for readset \"" + readset.name + "\" (should be PAIRED_END or SINGLE_END)!")
picard_job = picard.sam_to_fastq(out_bam, fastq1, fastq2)
job = concat_jobs([
Job(command="mkdir -p " + out_dir),
cram2bam_job,
picard_job
], name= "picard_sam_to_fastq." + readset.name)
jobs.append(job)
elif readset.bam:
if readset.run_type == "PAIRED_END":
out_dir = os.path.join("fusions", "picard_sam_to_fastq", readset.sample.name)
fastq1 = os.path.join(out_dir, os.path.basename(re.sub("\.bam$", ".pair1.fastq.gz", readset.bam)))
fastq2 = os.path.join(out_dir, os.path.basename(re.sub("\.bam$", ".pair2.fastq.gz", readset.bam)))
else:
raise Exception("Error: run type \"" + readset.run_type +
"\" is invalid for readset \"" + readset.name + "\" (should be PAIRED_END or SINGLE_END)!")
picard_job = picard.sam_to_fastq(readset.bam, fastq1, fastq2)
job = concat_jobs([
Job(command="mkdir -p " + out_dir),
picard_job
], name= "picard_sam_to_fastq." + readset.name)
jobs.append(job)
else:
raise Exception("Error: BAM file not available for readset \"" + readset.name + "\"!")
return jobs