def split_somatic(items):
"""Split somatic batches, adding a germline target.
Enables separate germline calling of samples using shared alignments.
"""
somatic_groups, somatic, non_somatic = vcfutils.somatic_batches(items)
# extract germline samples to run from normals in tumor/normal pairs
germline_added = set([])
germline = []
for somatic_group in somatic_groups:
paired = vcfutils.get_paired(somatic_group)
if paired and paired.normal_data:
cur = utils.deepish_copy(paired.normal_data)
vc = dd.get_variantcaller(cur)
if isinstance(vc, dict) and "germline" in vc:
cur["description"] = "%s-germline" % cur["description"]
if cur["description"] not in germline_added:
germline_added.add(cur["description"])
cur["rgnames"]["sample"] = cur["description"]
del cur["metadata"]["batch"]
cur["metadata"]["phenotype"] = "germline"
cur = remove_align_qc_tools(cur)
cur["config"]["algorithm"]["variantcaller"] = vc[
"germline"]
germline.append(cur)
# Fix variantcalling specification for only somatic targets
somatic_out = []
for data in somatic:
vc = dd.get_variantcaller(data)
if isinstance(vc, dict) and "somatic" in vc:
data["config"]["algorithm"]["variantcaller"] = vc["somatic"]
somatic_out.append(data)
return non_somatic + somatic_out + germline
def split_somatic(items):
"""Split somatic batches, adding a germline target.
Enables separate germline calling of samples using shared alignments.
"""
items = [_clean_flat_variantcaller(x) for x in items]
somatic_groups, somatic, non_somatic = vcfutils.somatic_batches(items)
# extract germline samples to run from normals in tumor/normal pairs
germline_added = set([])
germline = []
for somatic_group in somatic_groups:
paired = vcfutils.get_paired(somatic_group)
if paired and paired.normal_data:
cur = utils.deepish_copy(paired.normal_data)
vc = dd.get_variantcaller(cur)
if isinstance(vc, dict) and "germline" in vc:
if cur["description"] not in germline_added:
germline_added.add(cur["description"])
cur["rgnames"]["sample"] = cur["description"]
cur["metadata"]["batch"] = "%s-germline" % cur["description"]
cur["metadata"]["phenotype"] = "germline"
cur = remove_align_qc_tools(cur)
cur["config"]["algorithm"]["variantcaller"] = vc["germline"]
germline.append(cur)
# Fix variantcalling specification for only somatic targets
somatic_out = []
for data in somatic:
vc = dd.get_variantcaller(data)
if isinstance(vc, dict) and "somatic" in vc:
data["config"]["algorithm"]["variantcaller"] = vc["somatic"]
somatic_out.append(data)
return non_somatic + somatic_out + germline