def test_trios_multi_multi3_full(variants_vcf):
fvars = variants_vcf("backends/trios_multi")
vs = list(
fvars.query_variants(
regions=[Region("1", 11505, 11505)],
return_reference=True,
return_unknown=True,
)
)
assert len(vs) == 1
for v in vs:
print(mat2str(v.best_state))
assert v.best_state.shape == (4, 3)
fvars = variants_vcf("backends/trios_multi")
vs = list(
fvars.query_variants(
regions=[Region("1", 11506, 11506)],
return_reference=True,
return_unknown=True,
)
)
assert len(vs) == 1
for v in vs:
print(mat2str(v.best_state))
assert v.best_state.shape == (4, 3)
def test_multi_alt_allele_effects(variants_vcf, region, worst_effect, count):
fvars = variants_vcf("backends/effects_trio_multi")
vs = list(fvars.query_variants(regions=[region], effects=["missense"]))
for v in vs:
print("------------------")
print(mat2str(v.best_state))
print(mat2str(v.gt))
assert len(v.effects) == 2
assert v.effects[0].worst == worst_effect[0]
assert v.effects[1].worst == worst_effect[1]
assert len(vs) == count
def test_inheritance_multi_full(variants_vcf, region, count, inheritance):
fvars = variants_vcf("backends/inheritance_multi")
vs = list(
fvars.query_variants(
regions=[region], return_reference=False, return_unknown=False
)
)
assert len(vs) == count
for v in vs:
for a in v.alt_alleles:
print(a, mat2str(a.gt), a.members_ids, a.inheritance_in_members)
assert len(a.inheritance_in_members) == 7
# assert Inheritance.from_name(inheritance) in v.inheritance_in_members
assert len(mat2str(v.best_state)) == 15
def test_inheritance_quad_full(variants_impl, variants, count, inheritance):
fvars = variants_impl(variants)("backends/inheritance_quad")
vs = list(
fvars.query_variants(inheritance=inheritance, return_reference=False))
assert len(vs) == count
for v in vs:
assert len(mat2str(v.best_state)) == 9
def test_df_query_multiallelic3_families(variants_impl, variants,
fixture_name):
dfvars = variants_impl(variants)(fixture_name)
assert dfvars is not None
regions = [Region("1", 11606, 11606)]
family_ids = ["f1"]
vs = dfvars.query_variants(
regions=regions,
family_ids=family_ids,
return_reference=True,
return_unknown=True,
)
vs = list(vs)
assert len(vs) == 1
v = vs[0]
print(v, mat2str(v.best_state))
fa1 = v.alt_alleles[0]
fa2 = v.alt_alleles[1]
assert len(v.alt_alleles) == 2
assert "mom1" in fa1.variant_in_members
assert "dad1" in fa2.variant_in_members
assert "ch1" not in fa1.variant_in_members
assert "ch1" not in fa2.variant_in_members
def test_family_variant_X_gt2best_st(fvX1, fvX2, gt, bs1, gm1, bs2, gm2,
genome_2013):
v = fvX1(gt, None)
best_state = VariantsGenotypesLoader._calc_best_state(v, genome_2013)
print(v)
print(mat2str(best_state))
assert mat2str(best_state) == bs1
genetic_model = VariantsGenotypesLoader._calc_genetic_model(v, genome_2013)
assert genetic_model == gm1
v = fvX2(gt, None)
best_state = VariantsGenotypesLoader._calc_best_state(v, genome_2013)
print(v)
print(mat2str(best_state))
assert mat2str(best_state) == bs2
genetic_model = VariantsGenotypesLoader._calc_genetic_model(v, genome_2013)
assert genetic_model == gm2
def test_variant_in_members(variants_impl, variants, region, count, members):
fvars = variants_impl(variants)("backends/unknown_trio")
vs = list(fvars.query_variants(regions=[region]))
assert len(vs) == count
for v in vs:
print(v, mat2str(v.best_state))
for aa in v.alt_alleles:
print(aa, aa.variant_in_members)
assert list(aa.variant_in_members) == members
def test_reference_multiallelic_variant(variants_impl, variants, fixture_name):
fvars = variants_impl(variants)(fixture_name)
assert fvars is not None
vs = fvars.query_variants(return_reference=True, return_unknown=True)
vs = list(vs)
print(vs)
assert len(vs) == 2
for v in vs:
print(mat2str(v.best_state))
def test_trios_multi_single_allele1(variants_vcf):
fvars = variants_vcf("backends/trios_multi")
vs = list(
fvars.query_variants(
regions=[Region("1", 11500, 11500)],
return_reference=True,
return_unknown=True,
))
assert len(vs) == 1
for v in vs:
assert v.best_state.shape == (3, 3)
assert len(mat2str(v.best_state)) == 11
def test_inheritance_trio_full(variants_vcf, region, count, inheritance):
fvars = variants_vcf("backends/inheritance_trio")
vs = list(
fvars.query_variants(
inheritance=inheritance, regions=[region], return_reference=True
)
)
for v in vs:
# assert Inheritance.from_name(inheritance) in v.inheritance_in_members
assert len(mat2str(v.best_state)) == 7
for a in v.alleles:
print(">>>", a, a.inheritance_in_members)
assert len(a.inheritance_in_members) == 3
assert len(vs) == count
def test_variant_frequency_single(variants_impl, variants, region, count,
freq0, freq1):
fvars = variants_impl(variants)("backends/trios2")
vs = list(
fvars.query_variants(regions=[region],
return_reference=False,
return_unknown=False))
assert len(vs) == count
for v in vs:
print(v, mat2str(v.best_state))
print(v.frequencies)
assert freq0 == pytest.approx(v.frequencies[0], 1e-2)
if len(v.frequencies) == 2:
assert freq1 == pytest.approx(v.frequencies[1], 1e-2)
def test_f1_check_all_variants_effects(variants_vcf):
vvars = variants_vcf("backends/f1_test")
assert vvars is not None
vs = vvars.query_variants(return_reference=True, return_unknown=True)
vs = list(vs)
for v in vs:
print(v, v.effects, mat2str(v.best_state))
summary_alleles = v.summary_alleles
sa1 = summary_alleles[1]
sa2 = summary_alleles[2]
assert sa1.effect.worst == "synonymous"
assert sa2.effect.worst == "missense"
assert len(vs) == 7
def test_query_strange_multiallelic_families(variants_impl, variants,
fixture_name):
fvars = variants_impl(variants)(fixture_name)
assert fvars is not None
vs = fvars.query_variants()
vs = list(vs)
assert len(vs) == 2
for v in vs:
# v = vs[0]
print(v, mat2str(v.best_state))
print(v.alt_alleles)
fa1 = v.alt_alleles[0]
fa2 = v.alt_alleles[1]
assert len(v.alt_alleles) == 2
print(fa1.variant_in_members)
print(fa2.variant_in_members)
def test_reference_variant_single_allele(variants_impl, variants,
fixture_name):
dfvars = variants_impl(variants)(fixture_name)
assert dfvars is not None
vs = dfvars.query_variants(family_ids=["f1"],
return_reference=True,
return_unknown=True)
vs = list(vs)
assert len(vs) == 1
v = vs[0]
print(v)
print("best_st:", mat2str(v.best_state))
print("freq: ", v.frequencies)
print("effects:", v.effects)
print("alleles:", v.alleles)
# print("summary:", v.summary_variant)
assert len(v.frequencies) == 1
assert v.frequencies[0] == pytest.approx(87.5)
assert not v.effects
def test_best2gt(bs, gt):
res = best2gt(bs)
print(mat2str(res))
assert np.all(res == gt)
def test_family_variant_unknown_simple_best_st(fv1, gt, bs):
v = fv1(gt, calculate_simple_best_state(gt, 3))
print(v)
print(mat2str(v.best_state))
assert mat2str(v.best_state) == bs
def test_omission(variants_impl, variants):
fvars = variants_impl(variants)("backends/inheritance_omission")
for fv in fvars.query_variants(inheritance="any(mendelian,denovo)"):
print(fv, fv.matched_alleles)
for fa in fv.alleles:
print(fa, mat2str(fa.best_state), fa.inheritance_in_members)
def test_family_variant_unknown_full_best_st(fv1, gt, bs, genome_2013):
v = fv1(gt, None)
best_state = VariantsGenotypesLoader._calc_best_state(v, genome_2013)
print(v)
print(mat2str(best_state))
assert mat2str(best_state) == bs
def test_dae_transmitted_loader_simple(dae_transmitted):
for fv in dae_transmitted.family_variants_iterator():
print(fv, mat2str(fv.best_state), mat2str(fv.gt))
