def insert_sv_variants(adapter: MongoAdapter, case_obj: Case) -> None:
"""Build sv_variant documents and insert them into database on the fly, one at a time"""
for variant in VCF(case_obj.vcf_sv_path, threads=settings.cyvcf_threads):
variant_id = get_variant_id(variant=variant)
ref = variant.REF
alt = variant.ALT[0]
coordinates = get_coords(variant)
chrom = coordinates["chrom"]
pos = coordinates["pos"]
variant_obj = Variant(
variant_id=variant_id,
chrom=chrom,
pos=pos,
end=coordinates["end"],
ref=ref,
alt=alt,
end_chrom=coordinates["end_chrom"],
sv_type=coordinates["sv_type"],
sv_len=coordinates["sv_length"],
case_id=case_obj.case_id,
homozygote=0,
hemizygote=0,
is_sv=True,
id_column=variant.ID,
)
adapter.add_structural_variant(variant=variant_obj, max_window=settings.load_sv_window)
def insert_snv_variants(adapter: MongoAdapter, case_obj: Case) -> None:
"""Build variant documents and bulk insert them into database"""
variants = []
for variant in VCF(case_obj.vcf_path, threads=settings.cyvcf_threads):
variant_id = get_variant_id(variant=variant)
ref = variant.REF
alt = variant.ALT[0]
coordinates = get_coords(variant)
chrom = coordinates["chrom"]
pos = coordinates["pos"]
found_homozygote = 0
found_hemizygote = 0
for ind_obj in case_obj.individuals:
ind_pos = ind_obj["ind_index"]
if int(variant.gt_quals[ind_pos]) < settings.load_gq_threshold:
continue
genotype = GENOTYPE_MAP[variant.gt_types[ind_pos]]
if genotype not in ["het", "hom_alt"]:
continue
if genotype == "hom_alt":
found_homozygote = 1
if (
chrom in ["X", "Y"]
and ind_obj["sex"] == 1
and not check_par(chrom, pos, genome_build=settings.genome_build)
):
found_hemizygote = 1
variant_obj = Variant(
variant_id=variant_id,
chrom=chrom,
pos=pos,
end=coordinates["end"],
ref=ref,
alt=alt,
end_chrom=coordinates["end_chrom"],
sv_type=coordinates["sv_type"],
sv_len=coordinates["sv_length"],
case_id=case_obj.case_id,
homozygote=found_homozygote,
hemizygote=found_hemizygote,
is_sv=False,
id_column=variant.ID,
)
variants.append(variant_obj)
adapter.add_variants(variants=variants)
def annotate_svs(adapter, vcf_obj):
"""Annotate all SV variants in a VCF
Args:
adapter(loqusdb.plugin.adapter)
vcf_obj(cyvcf2.VCF)
Yields:
variant(cyvcf2.Variant)
"""
for nr_variants, variant in enumerate(vcf_obj, 1):
variant_info = get_coords(variant)
match = adapter.get_structural_variant(variant_info)
if match:
annotate_variant(variant, match)
yield variant
def test_get_coords_for_BND(bnd_variant):
coords = get_coords(bnd_variant)
assert coords["pos"] == coords["end"]
assert coords["sv_length"] == float("inf")
assert coords["sv_type"] == "BND"
def test_get_coords_for_BND(bnd_variant):
coords = get_coords(bnd_variant)
assert coords['pos'] == coords['end']
assert coords['sv_length'] == float('inf')
assert coords['sv_type'] == 'BND'