def _run_wham_coords(inputs, background_bams, coords, final_file):
"""Run WHAM on a specific set of chromosome, start, end coordinates.
"""
base, ext = os.path.splitext(final_file)
out_file = "%s-%s%s" % (base, region.to_safestr(coords), ext)
if not utils.file_exists(out_file):
with file_transaction(inputs[0], out_file) as tx_out_file:
cores = dd.get_cores(inputs[0])
ref_file = dd.get_ref_file(inputs[0])
all_bams = ",".join([x["align_bam"] for x in inputs] + background_bams)
coord_str = bamprep.region_to_gatk(coords)
opts = "-k -m 30"
cmd = ("WHAM-GRAPHENING {opts} -x {cores} -a {ref_file} -f {all_bams} -r {coord_str} "
"> {tx_out_file}")
do.run(cmd.format(**locals()), "Run WHAM: %s" % region.to_safestr(coords))
return [[coords, out_file]]
def _run_wham_coords(inputs, background_bams, coords, final_file):
"""Run WHAM on a specific set of chromosome, start, end coordinates.
"""
base, ext = utils.splitext_plus(final_file)
raw_file = "%s-%s.vcf" % (base, region.to_safestr(coords))
all_bams = ",".join([x["align_bam"] for x in inputs] + background_bams)
if not utils.file_exists(raw_file):
with file_transaction(inputs[0], raw_file) as tx_raw_file:
cores = dd.get_cores(inputs[0])
ref_file = dd.get_ref_file(inputs[0])
coord_str = bamprep.region_to_gatk(coords)
opts = "-k -m 30"
cmd = ("WHAM-GRAPHENING {opts} -x {cores} -a {ref_file} -f {all_bams} -r {coord_str} "
"> {tx_raw_file}")
do.run(cmd.format(**locals()), "Run WHAM: %s" % region.to_safestr(coords))
merge_vcf = _run_wham_merge(raw_file, inputs[0])
gt_vcf = _run_wham_genotype(merge_vcf, all_bams, coords, inputs[0])
prep_vcf = vcfutils.sort_by_ref(gt_vcf, inputs[0])
return [[coords, prep_vcf]]
def _run_wham_coords(inputs, background_bams, coords, final_file):
"""Run WHAM on a specific set of chromosome, start, end coordinates.
"""
base, ext = os.path.splitext(final_file)
out_file = "%s-%s%s" % (base, region.to_safestr(coords), ext)
if not utils.file_exists(out_file):
with file_transaction(inputs[0], out_file) as tx_out_file:
cores = dd.get_cores(inputs[0])
ref_file = dd.get_ref_file(inputs[0])
all_bams = ",".join([x["align_bam"]
for x in inputs] + background_bams)
coord_str = bamprep.region_to_gatk(coords)
opts = "-k -m 30"
cmd = (
"WHAM-GRAPHENING {opts} -x {cores} -a {ref_file} -f {all_bams} -r {coord_str} "
"> {tx_out_file}")
do.run(cmd.format(**locals()),
"Run WHAM: %s" % region.to_safestr(coords))
return [[coords, out_file]]
def _run_wham_coords(inputs, background_bams, coords, final_file):
"""Run WHAM on a specific set of chromosome, start, end coordinates.
"""
base, ext = utils.splitext_plus(final_file)
raw_file = "%s-%s.vcf" % (base, region.to_safestr(coords))
all_bams = ",".join([x["align_bam"] for x in inputs] + background_bams)
if not utils.file_exists(raw_file):
with file_transaction(inputs[0], raw_file) as tx_raw_file:
cores = dd.get_cores(inputs[0])
ref_file = dd.get_ref_file(inputs[0])
coord_str = bamprep.region_to_gatk(coords)
opts = "-k -m 30"
cmd = ("WHAM-GRAPHENING {opts} -x {cores} -a {ref_file} -f {all_bams} -r {coord_str} "
"> {tx_raw_file}")
do.run(cmd.format(**locals()), "Run WHAM: %s" % region.to_safestr(coords))
merge_vcf = _run_wham_merge(raw_file, inputs[0])
gt_vcf = _run_wham_genotype(merge_vcf, all_bams, coords, inputs[0])
prep_vcf = vcfutils.sort_by_ref(gt_vcf, inputs[0])
return [[coords, prep_vcf]]
def _run_wham_genotype(in_file, all_bams, coords, data):
"""Run genotyping on a prepped, merged VCF file.
"""
out_file = "%s-wgts%s" % utils.splitext_plus(in_file)
if not utils.file_exists(out_file):
with file_transaction(data, out_file) as tx_out_file:
cores = dd.get_cores(data)
ref_file = dd.get_ref_file(data)
coord_str = bamprep.region_to_gatk(coords)
cmd = ("WHAM-GRAPHENING -b {in_file} -x {cores} -a {ref_file} -f {all_bams} -r {coord_str} "
"> {tx_out_file}")
do.run(cmd.format(**locals()), "Genotype WHAM: %s" % region.to_safestr(coords))
return out_file
def _run_wham_genotype(in_file, all_bams, coords, data):
"""Run genotyping on a prepped, merged VCF file.
"""
out_file = "%s-wgts%s" % utils.splitext_plus(in_file)
if not utils.file_exists(out_file):
with file_transaction(data, out_file) as tx_out_file:
cores = dd.get_cores(data)
ref_file = dd.get_ref_file(data)
coord_str = bamprep.region_to_gatk(coords)
cmd = ("WHAM-GRAPHENING -b {in_file} -x {cores} -a {ref_file} -f {all_bams} -r {coord_str} "
"> {tx_out_file}")
do.run(cmd.format(**locals()), "Genotype WHAM: %s" % region.to_safestr(coords))
return out_file
def _do_work(data):
if "region" in data:
name = data["group"][0] if "group" in data else data["description"]
out_dir = os.path.join(data["dirs"]["work"], dir_ext_fn(data))
out_file = os.path.join(out_dir, "%s%s" % (name, ext))
out_parts = []
out_region_dir = os.path.join(out_dir, data["region"][0])
out_region_file = os.path.join(out_region_dir,
"%s-%s%s" % (name, region.to_safestr(data["region"]), ext))
out_parts = [(data["region"], out_region_file)]
return out_file, out_parts
else:
return None, []
def _do_work(data):
if "region" in data and not data["region"][0] in ["nochrom", "noanalysis"]:
name = data["group"][0] if "group" in data else data["description"]
out_dir = os.path.join(data["dirs"]["work"], dir_ext_fn(data))
out_file = os.path.join(out_dir, "%s%s" % (name, ext))
out_parts = []
if not utils.file_exists(out_file) or utils.get_in(data, ("metadata", "phenotype")) in batch_drivers:
out_region_dir = os.path.join(out_dir, data["region"][0])
out_region_file = os.path.join(out_region_dir,
"%s-%s%s" % (name, region.to_safestr(data["region"]), ext))
out_parts = [(data["region"], out_region_file)]
return out_file, out_parts
else:
return None, []
def _do_work(data):
if "region" in data:
name = data["group"][0] if "group" in data else data["description"]
out_dir = os.path.join(data["dirs"]["work"], dir_ext_fn(data))
out_file = os.path.join(out_dir, "%s%s" % (name, ext))
assert isinstance(data["region"], (list, tuple))
out_parts = []
for r, work_bams in sorted(_assign_bams_to_regions(data), key=_sort_by_size, reverse=True):
out_region_dir = os.path.join(out_dir, r[0])
out_region_file = os.path.join(out_region_dir,
"%s-%s%s" % (name, region.to_safestr(r), ext))
out_parts.append((r, work_bams, out_region_file))
return out_file, out_parts
else:
return None, []
def _do_work(data):
if "region" in data:
name = data["group"][0] if "group" in data else data["description"]
out_dir = os.path.join(data["dirs"]["work"], dir_ext_fn(data))
out_file = os.path.join(out_dir, "%s%s" % (name, ext))
assert isinstance(data["region"], (list, tuple))
out_parts = []
for r, work_bams in sorted(_assign_bams_to_regions(data), key=_sort_by_size, reverse=True):
out_region_dir = os.path.join(out_dir, r[0])
out_region_file = os.path.join(out_region_dir,
"%s-%s%s" % (name, region.to_safestr(r), ext))
out_parts.append((r, work_bams, out_region_file))
return out_file, out_parts
else:
return None, []
def _split_by_callable_region(data):
"""Split by callable or variant regions.
We expect joint calling to be deep in numbers of samples per region, so prefer
splitting aggressively by regions.
"""
batch = tz.get_in(("metadata", "batch"), data)
name = batch if batch else tz.get_in(("rgnames", "sample"), data)
out_dir = utils.safe_makedir(
os.path.join(data["dirs"]["work"], "joint", name))
parts = []
for feat in _get_callable_regions(data):
region_dir = utils.safe_makedir(os.path.join(out_dir, feat[0]))
region_outfile = os.path.join(
region_dir, "%s-%s.vcf.gz" % (batch, region.to_safestr(feat)))
parts.append(
(feat, data["work_bams"], data["vrn_files"], region_outfile))
out_file = os.path.join(out_dir, "%s-joint.vcf.gz" % name)
return out_file, parts
def _split_by_callable_region(data):
"""Split by callable or variant regions.
We expect joint calling to be deep in numbers of samples per region, so prefer
splitting aggressively by regions.
"""
batch = tz.get_in(("metadata", "batch"), data)
name = batch if batch else tz.get_in(("rgnames", "sample"), data)
out_dir = utils.safe_makedir(os.path.join(data["dirs"]["work"], "joint", name))
utils.safe_makedir(os.path.join(out_dir, "inprep"))
parts = []
for feat in _get_callable_regions(data):
region_dir = utils.safe_makedir(os.path.join(out_dir, feat[0]))
region_prep_dir = os.path.join(region_dir, "inprep")
if not os.path.exists(region_prep_dir):
os.symlink(os.path.join(os.pardir, "inprep"), region_prep_dir)
region_outfile = os.path.join(region_dir, "%s-%s.vcf.gz" % (batch, region.to_safestr(feat)))
parts.append((feat, data["work_bams"], data["vrn_files"], region_outfile))
out_file = os.path.join(out_dir, "%s-joint.vcf.gz" % name)
return out_file, parts
def _do_work(data):
if "region" in data:
name = data["group"][0] if "group" in data else data["description"]
out_dir = os.path.join(data["dirs"]["work"], dir_ext_fn(data))
out_file = os.path.join(out_dir, "%s%s" % (name, ext))
assert isinstance(data["region"], (list, tuple))
out_parts = []
for i, r in enumerate(data["region"]):
out_region_dir = os.path.join(out_dir, r[0])
out_region_file = os.path.join(out_region_dir, "%s-%s%s" % (name, region.to_safestr(r), ext))
work_bams = []
for xs in data["region_bams"]:
if len(xs) == 1:
work_bams.append(xs[0])
else:
work_bams.append(xs[i])
for work_bam in work_bams:
assert os.path.exists(work_bam), work_bam
out_parts.append((r, work_bams, out_region_file))
return out_file, out_parts
else:
return None, []
def _do_work(data):
if "region" in data:
name = data["group"][0] if "group" in data else data["description"]
out_dir = os.path.join(data["dirs"]["work"], dir_ext_fn(data))
out_file = os.path.join(out_dir, "%s%s" % (name, ext))
assert isinstance(data["region"], (list, tuple))
out_parts = []
for i, r in enumerate(data["region"]):
out_region_dir = os.path.join(out_dir, r[0])
out_region_file = os.path.join(out_region_dir,
"%s-%s%s" % (name, region.to_safestr(r), ext))
work_bams = []
for xs in data["region_bams"]:
if len(xs) == 1:
work_bams.append(xs[0])
else:
work_bams.append(xs[i])
for work_bam in work_bams:
assert os.path.exists(work_bam), work_bam
out_parts.append((r, work_bams, out_region_file))
return out_file, out_parts
else:
return None, []
