def align(fastq_file, pair_file, ref_file, names, align_dir, data):
config = data["config"]
out_prefix = path.join(align_dir, names["lane"])
out_file = out_prefix + "Aligned.out.sam"
if file_exists(out_file):
return out_file
star_path = config_utils.get_program("STAR", config)
fastq = " ".join([fastq_file, pair_file]) if pair_file else fastq_file
num_cores = config["algorithm"].get("num_cores", 1)
safe_makedir(align_dir)
cmd = ("{star_path} --genomeDir {ref_file} --readFilesIn {fastq} "
"--runThreadN {num_cores} --outFileNamePrefix {out_prefix} "
"--outReadsUnmapped Fastx --outFilterMultimapNmax 10 "
"--outSAMunmapped Within")
cmd += _read_group_option(names)
fusion_mode = get_in(data, ("config", "algorithm", "fusion_mode"), False)
if fusion_mode:
cmd += " --chimSegmentMin 15 --chimJunctionOverhangMin 15"
strandedness = get_in(data, ("config", "algorithm", "strandedness"),
"unstranded").lower()
if strandedness == "unstranded":
cmd += " --outSAMstrandField intronMotif"
run_message = "Running STAR aligner on %s and %s." % (pair_file, ref_file)
do.run(cmd.format(**locals()), run_message, None)
picard = broad.runner_from_config(config)
out_file = bam.sam_to_bam(out_file, config)
out_file = _fix_sam_header(out_file, config)
return out_file
def align(fastq_file, pair_file, ref_file, names, align_dir, data):
config = data["config"]
out_prefix = os.path.join(align_dir, names["lane"])
out_file = out_prefix + "Aligned.out.sam"
out_dir = os.path.join(align_dir, "%s_star" % names["lane"])
final_out = os.path.join(out_dir, "{0}.bam".format(names["sample"]))
if file_exists(final_out):
return final_out
star_path = config_utils.get_program("STAR", config)
fastq = " ".join([fastq_file, pair_file]) if pair_file else fastq_file
num_cores = config["algorithm"].get("num_cores", 1)
safe_makedir(align_dir)
cmd = ("{star_path} --genomeDir {ref_file} --readFilesIn {fastq} "
"--runThreadN {num_cores} --outFileNamePrefix {out_prefix} "
"--outReadsUnmapped Fastx --outFilterMultimapNmax 10 "
"--outSAMunmapped Within")
cmd += _read_group_option(names)
fusion_mode = get_in(data, ("config", "algorithm", "fusion_mode"), False)
if fusion_mode:
cmd += " --chimSegmentMin 15 --chimJunctionOverhangMin 15"
strandedness = get_in(data, ("config", "algorithm", "strandedness"),
"unstranded").lower()
if strandedness == "unstranded":
cmd += " --outSAMstrandField intronMotif"
run_message = "Running STAR aligner on %s and %s." % (pair_file, ref_file)
do.run(cmd.format(**locals()), run_message, None)
out_file = bam.sam_to_bam(out_file, config)
out_file = _fix_sam_header(out_file, config)
if not file_exists(final_out):
symlink_plus(out_file, final_out)
return final_out
def align(fastq_file, pair_file, ref_file, names, align_dir, data):
config = data["config"]
out_prefix = os.path.join(align_dir, names["lane"])
out_file = out_prefix + "Aligned.out.sam"
out_dir = os.path.join(align_dir, "%s_star" % names["lane"])
final_out = os.path.join(out_dir, "{0}.bam".format(names["sample"]))
if file_exists(final_out):
return final_out
star_path = config_utils.get_program("STAR", config)
fastq = " ".join([fastq_file, pair_file]) if pair_file else fastq_file
num_cores = config["algorithm"].get("num_cores", 1)
safe_makedir(align_dir)
cmd = ("{star_path} --genomeDir {ref_file} --readFilesIn {fastq} "
"--runThreadN {num_cores} --outFileNamePrefix {out_prefix} "
"--outReadsUnmapped Fastx --outFilterMultimapNmax 10 "
"--outSAMunmapped Within --outSAMattributes %s" %
" ".join(ALIGN_TAGS))
cmd += _read_group_option(names)
fusion_mode = get_in(data, ("config", "algorithm", "fusion_mode"), False)
if fusion_mode:
cmd += " --chimSegmentMin 15 --chimJunctionOverhangMin 15"
strandedness = get_in(data, ("config", "algorithm", "strandedness"),
"unstranded").lower()
if strandedness == "unstranded":
cmd += " --outSAMstrandField intronMotif"
run_message = "Running STAR aligner on %s and %s." % (pair_file, ref_file)
do.run(cmd.format(**locals()), run_message, None)
out_file = bam.sam_to_bam(out_file, config)
if not file_exists(final_out):
symlink_plus(out_file, final_out)
return final_out
def _align_from_fastq(fastq1, fastq2, aligner, align_ref, sam_ref, names, align_dir, data):
"""Align from fastq inputs, producing sorted BAM output.
"""
config = data["config"]
align_fn = TOOLS[aligner].align_fn
out = align_fn(fastq1, fastq2, align_ref, names, align_dir, data)
# handle align functions that update the main data dictionary in place
if isinstance(out, dict):
assert "work_bam" in out
return out
# handle output of raw SAM files that need to be converted to BAM
else:
work_bam = bam.sam_to_bam(out, config)
data["work_bam"] = bam.sort(work_bam, config)
return data
def _align_from_fastq(fastq1, fastq2, aligner, align_ref, sam_ref, names,
align_dir, data):
"""Align from fastq inputs, producing sorted BAM output.
"""
config = data["config"]
align_fn = TOOLS[aligner].align_fn
out = align_fn(fastq1, fastq2, align_ref, names, align_dir, data)
# handle align functions that update the main data dictionary in place
if isinstance(out, dict):
assert "work_bam" in out
return out
# handle output of raw SAM files that need to be converted to BAM
else:
work_bam = bam.sam_to_bam(out, config)
data["work_bam"] = bam.sort(work_bam, config)
return data
def merge_unmapped(mapped_sam, unmapped_bam, config):
merged_bam = os.path.join(os.path.dirname(mapped_sam), "merged.bam")
bam_file = bam.sam_to_bam(mapped_sam, config)
if not file_exists(merged_bam):
merged_bam = bam.merge([bam_file, unmapped_bam], merged_bam, config)
return merged_bam
def merge_unmapped(mapped_sam, unmapped_bam, config):
merged_bam = os.path.join(os.path.dirname(mapped_sam), "merged.bam")
bam_file = bam.sam_to_bam(mapped_sam, config)
if not file_exists(merged_bam):
merged_bam = bam.merge([bam_file, unmapped_bam], merged_bam, config)
return merged_bam
cmd += f"--fastqfile2 {fq2} --pairedend TRUE"
check_call(cmd, shell=True)
if __name__ == "__main__":
parser = ArgumentParser()
parser.add_argument("bcbio_config", help="final bcbio configuration file.")
parser.add_argument("bowtie_index", help="location of bowtie2 index")
parser.add_argument("annotation", help="repeat annotation")
parser.add_argument("rep2_setup", help="repenrich2 setup directory")
parser.add_argument("--threads", default=1, help="Number of threads to use.")
parser.add_argument("--outdir", default="align", help="output directory")
parser.add_argument("--rep2_path", default="RepEnrich2", help="path to RepEnrich2 code")
args = parser.parse_args()
filedict = get_files(args.bcbio_config)
safe_makedir(args.outdir)
config = config["algorithm"] = {"num_cores": args.threads}
repenrich = "RepEnrich2/RepEnrich.py"
annotation = "metadata/hg38_repeatmasker_clean.txt"
repenrich_setup = "metadata/RepEnrich2_setup_hg38"
for samplename, files in filedict.items():
out_dir = os.path.join(args.outdir, "repenrich", samplename)
logging.info(f"Aligning {samplename} to {args.bowtie_index}.")
out_sam = run_bowtie2(files, samplename, args.bowtie_index, args.threads, args.outdir)
logging.info(f"Converting {out_sam} to BAM format.")
out_bam = sam_to_bam(out_sam, config)
logging.info(f"Subsetting {out_bam} into unique and multimapped reads.")
subset_files = subset_reads(out_bam, samplename, args.rep2_path)
logging.info(f"Running RepEnrich2 on {samplename}.")
run_repenrich2(subset_files, samplename, args.annotation, args.rep2_path, args.rep2_setup, args.threads)
logging.info(f"Finished {samplename}.")
