def test_partition_into_regions__num_parts(shared_datadir, vcf_file, is_path):
vcf_path = path_for_test(shared_datadir, vcf_file, is_path)
regions = partition_into_regions(vcf_path, num_parts=4)
assert regions is not None
part_variant_counts = [
count_variants(vcf_path, region) for region in regions
]
total_variants = count_variants(vcf_path)
assert sum(part_variant_counts) == total_variants
def test_partition_into_regions__num_parts_large(shared_datadir, is_path):
vcf_path = path_for_test(shared_datadir, "CEUTrio.20.21.gatk3.4.g.vcf.bgz",
is_path)
regions = partition_into_regions(vcf_path, num_parts=100)
assert regions is not None
assert len(regions) == 18
part_variant_counts = [
count_variants(vcf_path, region) for region in regions
]
total_variants = count_variants(vcf_path)
assert sum(part_variant_counts) == total_variants
def test_record_counts_csi(shared_datadir, vcf_file, is_path):
# Check record counts in csi with actual count of VCF
vcf_path = path_for_test(shared_datadir, vcf_file, is_path)
csi_path = get_csi_path(vcf_path)
assert csi_path is not None
csi = read_csi(csi_path)
for i, contig in enumerate(VCF(vcf_path).seqnames):
assert csi.record_counts[i] == count_variants(vcf_path, contig)
def test_record_counts_tbi(shared_datadir, vcf_file, is_path):
# Check record counts in tabix with actual count of VCF
vcf_path = path_for_test(shared_datadir, vcf_file, is_path)
tabix_path = get_tabix_path(vcf_path)
assert tabix_path is not None
tabix = read_tabix(tabix_path)
for i, contig in enumerate(tabix.sequence_names):
assert tabix.record_counts[i] == count_variants(vcf_path, contig)