def merge_bam_files(bam_files, work_dir, config):
"""Merge multiple BAM files from a sample into a single BAM for processing.
"""
out_file = os.path.join(work_dir, os.path.basename(bam_files[0]))
if not os.path.exists(out_file):
picard = PicardRunner(config["program"]["picard"])
with utils.curdir_tmpdir() as tmp_dir:
opts = [("OUTPUT", out_file),
("SORT_ORDER", "coordinate"),
("TMP_DIR", tmp_dir)]
for b in bam_files:
opts.append(("INPUT", b))
picard.run("MergeSamFiles", opts)
return out_file
def do_work(ref, cur_fastq, config, config_file):
picard = PicardRunner(config["program"]["picard"])
align_dir = config["align_dir"]
if not os.path.exists(align_dir):
os.makedirs(align_dir)
fastq_one, fastq_two = _get_align_files(cur_fastq, config)
base = "%s-%s" % (cur_fastq["name"], os.path.basename(ref))
align_file = bowtie_to_sam(fastq_one, fastq_two, ref, base,
align_dir, config)
bam_file = sam_to_bam(align_file, fastq_one, fastq_two, ref,
config_file)
picard.run_fn("picard_index", bam_file)
bed_file = bam_to_bed(bam_file)
if fastq_two is not None:
inferred_bed = generate_inferred_coverage(bam_file)
generate_bigwig(inferred_bed, ref, picard)
else:
generate_bigwig(bed_file, ref, picard)
return cur_fastq["name"], bed_file