def _remove_variant_set(self, variant_set_name):
vs = VariantSet.objects.get(name=variant_set_name,
reference_set=self.reference_set)
for call_set in VariantCallSet.objects(variant_sets=vs):
call_set.variant_sets.remove(vs)
call_set.save()
# Remove calls from callsets that only have this variantset
if len(call_set.variant_sets) < 2:
VariantCall.objects(call_set=call_set).delete()
call_set.delete()
# Remove variants that are ONLY from this variant set
Variant.objects(variant_sets=vs, variant_sets__size=2).delete()
VariantSetMetadata.objects(variant_set=vs).delete()
vs.delete()
def test_add_add_variants_and_calls(self):
vcf = VCF(f="tests/vcf_tests/test.vcf",
reference_set_id=self.reference_set.id,
method="CORTEX")
vcf.add_to_database()
assert VariantCall.objects().count() == 21
assert Variant.objects().count() == 21
def test_create_SNP_het_call(self):
c1 = VariantCall.create(variant=self.variant_snp,
call_set=self.call_set,
genotype=[0, 1],
genotype_likelihoods=[0.1, 0.9, 0.12])
assert c1.call_set_name == "C00123"
assert c1.genotype == [0, 1]
self.variant_snp.save()
c1.save()
assert c1 in self.variant_snp.calls
c2 = VariantCall.create(variant=self.variant_snp,
call_set=self.call_set,
genotype="1/1",
genotype_likelihoods=[0.01, 0.1, 0.9])
assert c2.call_set_name == "C00123"
assert c2.genotype == [1, 1]
c2.save()
assert c2 in self.variant_snp.calls
def test_create_SNP_het_call(self):
c1 = VariantCall.create(variant=self.variant_snp,
call_set=self.call_set,
genotype=[0, 1],
genotype_likelihoods=[0.1, 0.9, 0.12])
assert c1.call_set_name == "C00123"
assert c1.genotype == [0, 1]
self.variant_snp.save()
c1.save()
assert c1 in self.variant_snp.calls
c2 = VariantCall.create(variant=self.variant_snp,
call_set=self.call_set,
genotype="1/1",
genotype_likelihoods=[0.01, 0.1, 0.9])
assert c2.call_set_name == "C00123"
assert c2.genotype == [1, 1]
c2.save()
assert c2 in self.variant_snp.calls
def _create_call_sets(self):
try:
self.call_set = VariantCallSet.create_and_save(
name=self.call_set_name,
variant_sets=[self.global_variant_set],
sample_id=self.sample,
info={"variant_caller": self.method})
except NotUniqueError:
if self.force:
# logger.info("There is already a call set for sample %s with method %s but add has been called with --force so this callset is being removed from the DB" %
# (self.sample, self.method))
cs = VariantCallSet.objects.get(
name="_".join([self.sample, self.method]))
VariantCall.objects(call_set=cs).delete()
cs.delete()
self._create_call_sets()
else:
raise ValueError(
"There is already a call set for sample %s with method %s "
% (self.sample, self.method))
def test_create_complex_call(self):
c1 = VariantCall.create(
variant=self.variant_snp_mult_alts,
call_set=self.call_set,
genotype="2/1",
genotype_likelihoods=[0.01, 0.1, 0.9, 0.1, 0.2, 0.6])
self.variant_snp.save()
self.variant_snp_mult_alts.save()
c1.save()
assert c1.call_set_name == "C00123"
assert c1.genotype == [2, 1]
assert c1 not in self.variant_snp.calls
assert c1 in self.variant_snp_mult_alts.calls
def _create_variants_and_calls(self):
for record in self.vcf_reader:
if not record.FILTER and self._is_record_valid(record):
v = self._get_or_create_variant(record)
for call in record.samples:
genotype_likelihoods = self._get_genotype_likelihoods(call)
c = VariantCall.create(
variant=v,
call_set=self.call_sets[call.sample],
genotype=call['GT'],
genotype_likelihoods=genotype_likelihoods)
self.calls.append(c)
VariantCall.objects.insert(self.calls)
def _create_calls(self):
calls = []
for var_hash, call in self.data["variant_calls"].items():
if sum(call["genotype"]) > 0:
var_hash = var_hash[:64]
v = Variant.objects.get(var_hash=var_hash)
c = VariantCall.create(
variant=v,
call_set=self.call_set,
genotype=call["genotype"],
genotype_likelihoods=call["genotype_likelihoods"],
info=call["info"])
calls.append(c)
VariantCall.objects.insert(calls)
def test_add_second_vcf_variant_set(self):
# This VCF only has one Variant which is not in the first VCF
vcf = VCF(f="tests/vcf_tests/test2.vcf",
reference_set_id=self.reference_set.id,
method="CORTEX")
vcf.add_to_database()
assert VariantSet.objects().count() == 3
assert VariantCallSet.objects().count() == 2
assert VariantCall.objects().count() == 42
assert Variant.objects().count() == 22
assert len(Variant.objects()[0].variant_sets) == 3
assert len(
Variant.objects.get(
names="UNION_BC_k31_var_147").variant_sets) == 3
def test_add_second_vcf_variant_set(self):
# This VCF only has one Variant which is not in the first VCF
vcf = VCF(f="tests/vcf_tests/test3.vcf",
reference_set_id=self.reference_set.id,
method="CORTEX")
vcf.add_to_database()
assert VariantSet.objects().count() == 2
assert VariantCallSet.objects().count() == 1
assert VariantCall.objects().count() == 106
assert Variant.objects().count() == 106
assert Variant.snps().count() == 89
assert Variant.indels().count() == 17
assert Variant.insertions().count() == 8
assert Variant.deletions().count() == 8
assert Variant.ph_snps.count() == 1
def test_create_complex_call(self):
c1 = VariantCall.create(
variant=self.variant_snp_mult_alts,
call_set=self.call_set,
genotype="2/1",
genotype_likelihoods=[
0.01,
0.1,
0.9,
0.1,
0.2,
0.6])
self.variant_snp.save()
self.variant_snp_mult_alts.save()
c1.save()
assert c1.call_set_name == "C00123"
assert c1.genotype == [2, 1]
assert c1 not in self.variant_snp.calls
assert c1 in self.variant_snp_mult_alts.calls