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_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
"""
from bcbio.qc import fastqc, gemini, kraken, qsignature, qualimap, samtools, picard, srna, umi
tools = {
"fastqc": fastqc.run,
"small-rna": srna.run,
"samtools": samtools.run,
"qualimap": qualimap.run,
"qualimap_rnaseq": qualimap.run_rnaseq,
"gemini": gemini.run,
"qsignature": qsignature.run,
"coverage": _run_coverage_qc,
"variants": _run_variants_qc,
"kraken": kraken.run,
"picard": picard.run,
"umi": umi.run
}
qc_dir = utils.safe_makedir(
os.path.join(data["dirs"]["work"], "qc", data["description"]))
metrics = {}
qc_out = {}
for program_name in tz.get_in(["config", "algorithm", "qc"], data):
qc_fn = tools[program_name]
cur_qc_dir = os.path.join(qc_dir, program_name)
out = qc_fn(bam_file, data, cur_qc_dir)
qc_files = None
if out and isinstance(out, dict):
metrics.update(out)
elif out and isinstance(out, basestring) and os.path.exists(out):
qc_files = {"base": out, "secondary": []}
if not qc_files:
qc_files = _organize_qc_files(program_name, cur_qc_dir)
if qc_files:
qc_out[program_name] = qc_files
bam.remove("%s-downsample%s" % os.path.splitext(bam_file))
metrics["Name"] = dd.get_sample_name(data)
metrics["Quality format"] = dd.get_quality_format(data).lower()
return {"qc": qc_out, "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
"""
from bcbio.qc import fastqc, gemini, kraken, qsignature, qualimap, samtools, picard, srna
tools = {"fastqc": fastqc.run,
"small-rna": srna.run,
"samtools": samtools.run,
"qualimap": qualimap.run,
"qualimap_rnaseq": qualimap.run_rnaseq,
"gemini": gemini.run,
"qsignature": qsignature.run,
"coverage": _run_coverage_qc,
"variants": _run_variants_qc,
"kraken": kraken.run,
"picard": picard.run}
qc_dir = utils.safe_makedir(os.path.join(data["dirs"]["work"], "qc", data["description"]))
metrics = {}
qc_out = {}
for program_name in tz.get_in(["config", "algorithm", "qc"], data):
qc_fn = tools[program_name]
cur_qc_dir = os.path.join(qc_dir, program_name)
out = qc_fn(bam_file, data, cur_qc_dir)
qc_files = None
if out and isinstance(out, dict):
metrics.update(out)
elif out and isinstance(out, basestring) and os.path.exists(out):
qc_files = {"base": out, "secondary": []}
if not qc_files:
qc_files = _organize_qc_files(program_name, cur_qc_dir)
if qc_files:
qc_out[program_name] = qc_files
bam.remove("%s-downsample%s" % os.path.splitext(bam_file))
metrics["Name"] = dd.get_sample_name(data)
metrics["Quality format"] = dd.get_quality_format(data).lower()
return {"qc": qc_out, "metrics": metrics}
