def _run_qc_tools(bam_file, data):
"""Run a set of third party quality control tools, returning QC directory and metrics.
"""
metrics = {}
to_run = [("fastqc", _run_fastqc)]
if data["analysis"].lower().startswith("rna-seq"):
to_run.append(("rnaseqc", bcbio.rnaseq.qc.sample_summary))
# to_run.append(("coverage", _run_gene_coverage))
to_run.append(("complexity", _run_complexity))
elif data["analysis"].lower().startswith("chip-seq"):
to_run.append(["bamtools", _run_bamtools_stats])
else:
to_run += [("bamtools", _run_bamtools_stats), ("gemini", _run_gemini_stats)]
qc_dir = utils.safe_makedir(os.path.join(data["dirs"]["work"], "qc", data["description"]))
metrics = {}
for program_name, qc_fn in to_run:
cur_qc_dir = os.path.join(qc_dir, program_name)
cur_metrics = qc_fn(bam_file, data, cur_qc_dir)
metrics.update(cur_metrics)
ratio = bam.get_aligned_reads(bam_file,data)
if ratio < 0.60 and data['config']["algorithm"].get("kraken", False) and data["analysis"].lower() == "rna-seq":
cur_metrics =_run_kraken(data, ratio)
metrics.update(cur_metrics)
metrics["Name"] = data["name"][-1]
metrics["Quality format"] = utils.get_in(data,
("config", "algorithm",
"quality_format"),
"standard").lower()
return {"qc": qc_dir, "metrics": metrics}
def _run_qc_tools(bam_file, data):
"""Run a set of third party quality control tools, returning QC directory and metrics.
:param bam_file: alignments in bam format
:param data: dict with all configuration information
:returns: dict with output of different tools
"""
metrics = {}
to_run = []
if "fastqc" not in tz.get_in(("config", "algorithm", "tools_off"), data,
[]):
to_run.append(("fastqc", _run_fastqc))
if data["analysis"].lower().startswith("rna-seq"):
# to_run.append(("rnaseqc", bcbio.rnaseq.qc.sample_summary))
# to_run.append(("coverage", _run_gene_coverage))
# to_run.append(("complexity", _run_complexity))
to_run.append(("qualimap", _rnaseq_qualimap))
elif data["analysis"].lower().startswith("chip-seq"):
to_run.append(["bamtools", _run_bamtools_stats])
elif not data["analysis"].lower().startswith("smallrna-seq"):
to_run += [("bamtools", _run_bamtools_stats),
("gemini", _run_gemini_stats)]
if data["analysis"].lower().startswith(("standard", "variant2")):
to_run.append(["qsignature", _run_qsignature_generator])
if "qualimap" in tz.get_in(("config", "algorithm", "tools_on"), data,
[]):
to_run.append(("qualimap", _run_qualimap))
qc_dir = utils.safe_makedir(
os.path.join(data["dirs"]["work"], "qc", data["description"]))
metrics = {}
for program_name, qc_fn in to_run:
cur_qc_dir = os.path.join(qc_dir, program_name)
cur_metrics = qc_fn(bam_file, data, cur_qc_dir)
metrics.update(cur_metrics)
# if (ratio < 0.60 and data['config']["algorithm"].get("kraken", None) and
# (data["analysis"].lower().startswith("rna-seq") or
# data["analysis"].lower().startswith("standard"))):
if data['config']["algorithm"].get("kraken", None):
ratio = bam.get_aligned_reads(bam_file, data)
cur_metrics = _run_kraken(data, ratio)
metrics.update(cur_metrics)
bam.remove("%s-downsample%s" % os.path.splitext(bam_file))
metrics["Name"] = data["name"][-1]
metrics["Quality format"] = utils.get_in(
data, ("config", "algorithm", "quality_format"), "standard").lower()
return {"qc": qc_dir, "metrics": metrics}
def _run_qc_tools(bam_file, data):
"""Run a set of third party quality control tools, returning QC directory and metrics.
:param bam_file: alignments in bam format
:param data: dict with all configuration information
:returns: dict with output of different tools
"""
metrics = {}
to_run = []
if "fastqc" not in tz.get_in(("config", "algorithm", "tools_off"), data, []):
to_run.append(("fastqc", _run_fastqc))
if data["analysis"].lower().startswith("rna-seq"):
# to_run.append(("rnaseqc", bcbio.rnaseq.qc.sample_summary))
# to_run.append(("coverage", _run_gene_coverage))
# to_run.append(("complexity", _run_complexity))
to_run.append(("qualimap", _rnaseq_qualimap))
elif data["analysis"].lower().startswith("chip-seq"):
to_run.append(["bamtools", _run_bamtools_stats])
elif not data["analysis"].lower().startswith("smallrna-seq"):
to_run += [("bamtools", _run_bamtools_stats), ("gemini", _run_gemini_stats)]
if data["analysis"].lower().startswith(("standard", "variant2")):
to_run.append(["qsignature", _run_qsignature_generator])
if "qualimap" in tz.get_in(("config", "algorithm", "tools_on"), data, []):
to_run.append(("qualimap", _run_qualimap))
qc_dir = utils.safe_makedir(os.path.join(data["dirs"]["work"], "qc", data["description"]))
metrics = {}
for program_name, qc_fn in to_run:
cur_qc_dir = os.path.join(qc_dir, program_name)
cur_metrics = qc_fn(bam_file, data, cur_qc_dir)
metrics.update(cur_metrics)
# if (ratio < 0.60 and data['config']["algorithm"].get("kraken", None) and
# (data["analysis"].lower().startswith("rna-seq") or
# data["analysis"].lower().startswith("standard"))):
if data['config']["algorithm"].get("kraken", None):
ratio = bam.get_aligned_reads(bam_file, data)
cur_metrics = _run_kraken(data, ratio)
metrics.update(cur_metrics)
bam.remove("%s-downsample%s" % os.path.splitext(bam_file))
metrics["Name"] = data["name"][-1]
metrics["Quality format"] = utils.get_in(data,
("config", "algorithm",
"quality_format"),
"standard").lower()
return {"qc": qc_dir, "metrics": metrics}
def run(bam_file, data, out_dir):
"""Run kraken, generating report in specified directory and parsing metrics.
Using only first paired reads.
"""
# logger.info("Number of aligned reads < than 0.60 in %s: %s" % (dd.get_sample_name(data), ratio))
logger.info("Running kraken to determine contaminant: %s" %
dd.get_sample_name(data))
ratio = bam.get_aligned_reads(bam_file, data)
out = out_stats = None
db = data['config']["algorithm"]["kraken"]
kraken_cmd = config_utils.get_program("kraken", data["config"])
if db == "minikraken":
db = os.path.join(install._get_data_dir(), "genomes", "kraken",
"minikraken")
if not os.path.exists(db):
logger.info("kraken: no database found %s, skipping" % db)
return {"kraken_report": "null"}
if not os.path.exists(os.path.join(out_dir, "kraken_out")):
work_dir = os.path.dirname(out_dir)
utils.safe_makedir(work_dir)
num_cores = data["config"]["algorithm"].get("num_cores", 1)
fn_file = data["files_orig"][0] if dd.get_save_diskspace(
data) else data["files"][0]
if fn_file.endswith("bam"):
logger.info("kraken: need fasta files as input")
return {"kraken_report": "null"}
with tx_tmpdir(data) as tx_tmp_dir:
with utils.chdir(tx_tmp_dir):
out = os.path.join(tx_tmp_dir, "kraken_out")
out_stats = os.path.join(tx_tmp_dir, "kraken_stats")
cat = "zcat" if fn_file.endswith(".gz") else "cat"
cl = ("{cat} {fn_file} | {kraken_cmd} --db {db} --quick "
"--preload --min-hits 2 "
"--threads {num_cores} "
"--out {out} --fastq-input /dev/stdin 2> {out_stats}"
).format(**locals())
do.run(cl, "kraken: %s" % dd.get_sample_name(data))
if os.path.exists(out_dir):
shutil.rmtree(out_dir)
shutil.move(tx_tmp_dir, out_dir)
metrics = _parse_kraken_output(out_dir, db, data)
return metrics
def run(bam_file, data, out_dir):
"""Run kraken, generating report in specified directory and parsing metrics.
Using only first paired reads.
"""
# logger.info("Number of aligned reads < than 0.60 in %s: %s" % (dd.get_sample_name(data), ratio))
logger.info("Running kraken to determine contaminant: %s" % dd.get_sample_name(data))
ratio = bam.get_aligned_reads(bam_file, data)
out = out_stats = None
db = data['config']["algorithm"]["kraken"]
kraken_cmd = config_utils.get_program("kraken", data["config"])
if db == "minikraken":
db = os.path.join(install._get_data_dir(), "genomes", "kraken", "minikraken")
if not os.path.exists(db):
logger.info("kraken: no database found %s, skipping" % db)
return {"kraken_report": "null"}
if not os.path.exists(os.path.join(out_dir, "kraken_out")):
work_dir = os.path.dirname(out_dir)
utils.safe_makedir(work_dir)
num_cores = data["config"]["algorithm"].get("num_cores", 1)
fn_file = data["files_orig"][0] if dd.get_save_diskspace(data) else data["files"][0]
if fn_file.endswith("bam"):
logger.info("kraken: need fastq files as input")
return {"kraken_report": "null"}
with tx_tmpdir(data) as tx_tmp_dir:
with utils.chdir(tx_tmp_dir):
out = os.path.join(tx_tmp_dir, "kraken_out")
out_stats = os.path.join(tx_tmp_dir, "kraken_stats")
cat = "zcat" if fn_file.endswith(".gz") else "cat"
cl = ("{cat} {fn_file} | {kraken_cmd} --db {db} --quick "
"--preload --min-hits 2 "
"--threads {num_cores} "
"--out {out} --fastq-input /dev/stdin 2> {out_stats}").format(**locals())
do.run(cl, "kraken: %s" % dd.get_sample_name(data))
if os.path.exists(out_dir):
shutil.rmtree(out_dir)
shutil.move(tx_tmp_dir, out_dir)
metrics = _parse_kraken_output(out_dir, db, data)
return metrics
