def test_compound_multiple_2(self):
"""Reference case for the following ones"""
genotypes = array(
[[2, 1, 2, 2, 1, 1], [1, 2, 2, 2, 1, 1], [1, 2, 2, 2, 1, 1]],
dtype=DTYPE)
variants = VariantsCollection([
Variant(variant_id=1, start=1, gene_symbol='B'), # mother contrib
Variant(variant_id=2, start=2,
gene_symbol='B'), # father contrib 1, not transmitted
Variant(variant_id=3, start=3,
gene_symbol='B'), # father contrib 2, transmitted
])
var = GenotypesFilterCompoundHeterozygous(self.ss).apply(
variants, genotypes).variants
self.assertEqual(len(var), 3)
def test_compound_healty_homozygous(self):
"""Unaffected homozygous: exclude (because it will affect both alleles)"""
genotypes = array(
[[2, 1, 2, 2, 1, 1], [1, 2, 2, 2, 1, 4], [1, 2, 2, 2, 1, 1]],
dtype=DTYPE)
variants = VariantsCollection([
Variant(variant_id=1, start=1, gene_symbol='B'), # mother
Variant(variant_id=2, start=2,
gene_symbol='B'), # father 1 + unaffected hom
Variant(variant_id=3, start=3, gene_symbol='B'), # father 2
])
var = GenotypesFilterCompoundHeterozygous(self.ss).apply(
variants, genotypes).variants
self.assertEqual(
len(var), 2) # would be 3 but we exclude 1 variant -> 1 pair left
def test_active(self):
"""Should return what is not ref/ref (1) in at least one of the selected samples."""
genotypes = array(
[[2, 1, 2, 1, 1, 1], [1, 2, 2, 2, 1, 1], [1, 2, 2, 2, 1, 1],
[1, 2, 2, 2, 1, 1], [1, 1, 2, 2, 1, 1]],
dtype=DTYPE)
variants = VariantsCollection(
[Variant(variant_id=x + 1) for x in range(len(genotypes))],
db='test')
# the db parameter should be ignored
# pk needs to start at 0 to emulate the array indexing
groups = {"affected": ["Mother"]}
ss = SamplesSelection(self.F.ss.samples, groups)
var = GenotypesFilterActive(ss).apply(variants, genotypes).variants
self.assertEqual(len(var), 1)
self.assertTrue(any([genotypes[v.pk][i] > 1 for i in [0]]
for v in var))
groups = {"affected": ["Mother", "Father"]}
ss = SamplesSelection(self.F.ss.samples, groups)
var = GenotypesFilterActive(ss).apply(variants, genotypes).variants
self.assertEqual(len(var), 4)
self.assertTrue(
any([genotypes[v.pk][i] > 1 for i in [0, 1]] for v in var))
groups = {"affected": ["Dasha", "Lena"], "not_affected": ["Lesha"]}
ss = SamplesSelection(self.F.ss.samples, groups)
var = GenotypesFilterActive(ss).apply(variants, genotypes).variants
self.assertEqual(len(var), 4)
self.assertTrue(
any([genotypes[v.pk][i] > 1 for i in [3, 4, 5]] for v in var))
def test_recessive(self):
"""All affected must be homozygous (4), the parents must be hererozygous,
and all other not affected must be heterozygous or non-carriers."""
genotypes = array(
[
[2, 1, 4, 2, 1, 1], # X - Both parents must have it
[2, 2, 4, 2, 1, 1], # X - Affected but not homozygous
[2, 2, 4, 4, 1, 1], # V - OK
[2, 2, 4, 4, 4, 1], # X - Not affected homozygous
[2, 2, 4, 4, 2, 1]
], # V - Not affected carrier
dtype=DTYPE)
variants = VariantsCollection(
[Variant(variant_id=x + 1) for x in range(len(genotypes))],
db='test')
groups = {
"affected": ["Sasha", "Dasha"],
"not_affected": ["Mother", "Father", "Lena", "Lesha"]
}
ss = SamplesSelection(self.F.samples, groups)
var = GenotypesFilterRecessive(ss).apply(variants, genotypes).variants
self.assertEqual(len(var), 2)
for v in var:
gts = genotypes[v.pk - 1]
self.assertEqual(gts[0], 2)
self.assertEqual(gts[1], 2)
self.assertEqual(gts[2], 4)
self.assertEqual(gts[3], 4)
self.assertLessEqual(gts[4], 2)
self.assertLessEqual(gts[5], 2)
def test_compound_affected_nocompound(self):
"""Affected without the compound: exclude (the disease is not due to a compound).
(!? When one affected has it but not the other, the disease cannot be due to this compound)."""
genotypes = array(
[[2, 1, 2, 2, 1, 1], [1, 2, 2, 2, 1, 1], [1, 2, 2, 1, 1, 1]],
dtype=DTYPE)
variants = VariantsCollection([
Variant(variant_id=1, start=1,
gene_symbol='B'), # } } father + aff1 + aff2
Variant(variant_id=2, start=2,
gene_symbol='B'), # } } mother + (aff1) + aff2
Variant(variant_id=3, start=3,
gene_symbol='B'), # } mother + aff1
])
var = GenotypesFilterCompoundHeterozygous(self.ss).apply(
variants, genotypes).variants
self.assertEqual(
len(var), 2) # would be 3, but we exclude 1 variant -> 1 pair left
def test_compound_unaffected_compound(self):
"""Unaffected compound: exclude (the disease is not due to a compound)"""
genotypes = array([[2, 1, 2, 2, 1, 1], [2, 1, 2, 2, 2, 1],
[1, 2, 2, 2, 1, 1], [1, 2, 2, 2, 2, 1]],
dtype=DTYPE)
variants = VariantsCollection([
Variant(variant_id=1, start=0, gene_symbol='B'), # mother 1
Variant(variant_id=2, start=1,
gene_symbol='B'), # mother 2 + unaffected 1
Variant(variant_id=3, start=2, gene_symbol='B'), # father 1
Variant(variant_id=4, start=3,
gene_symbol='B'), # father 2 + unaffected 2
])
var = GenotypesFilterCompoundHeterozygous(self.ss).apply(
variants, genotypes).variants
self.assertEqual(
len(var),
2) # would be 4, but here we exclude an entire pair (2 variants):
def setUp(self):
self.F = Family()
groups = {
"affected": ["Sasha", "Dasha"],
"not_affected": ["Mother", "Father", "Lesha", "Lena"]
}
self.ss = SamplesSelection(self.F.samples, groups)
# Most standard case:
self.genotypes = array(
[
[2, 1, 2, 2, 1, 1], # mother and child carry var 1
[1, 2, 2, 2, 1, 1]
], # father and child carry var 2
dtype=DTYPE)
self.variants = VariantsCollection([
Variant(variant_id=1, start=1, gene_symbol='B'), # mother contrib
Variant(variant_id=2, start=2, gene_symbol='B'), # father contrib
])
def test_compound_same_parent(self):
"""One of the affected gets a pair from the same parent: exclude"""
genotypes = array([[2, 1, 2, 1, 1, 1], [1, 2, 2, 2, 1, 1],
[1, 2, 2, 2, 1, 1], [1, 2, 2, 2, 1, 1]],
dtype=DTYPE)
variants = VariantsCollection([
Variant(variant_id=1, start=1,
gene_symbol='B'), # mother contrib } }
Variant(
variant_id=2, start=2,
gene_symbol='B'), # father contrib 1 } father + affected
Variant(
variant_id=3, start=3,
gene_symbol='B'), # father contrib 2 } father + affected
Variant(variant_id=4, start=4,
gene_symbol='B'), # father contrib 3
])
var = GenotypesFilterCompoundHeterozygous(self.ss).apply(
variants, genotypes).variants
self.assertEqual(len(var), 0)
def test_compound_multiple_1(self):
"""There can be multiple combinations"""
genotypes = array(
[[2, 1, 2, 2, 1, 1], [2, 1, 2, 2, 1, 1], [1, 2, 2, 2, 1, 1],
[1, 2, 2, 2, 1, 1], [1, 2, 2, 2, 1, 1]],
dtype=DTYPE)
variants = VariantsCollection([
Variant(variant_id=1, start=1,
gene_symbol='B'), # mother contrib 1
Variant(variant_id=2, start=1,
gene_symbol='B'), # mother contrib 2
Variant(variant_id=3, start=2,
gene_symbol='B'), # father contrib 1
Variant(variant_id=4, start=3,
gene_symbol='B'), # father contrib 2
Variant(variant_id=5, start=4,
gene_symbol='B'), # father contrib 3
])
var = GenotypesFilterCompoundHeterozygous(self.ss).apply(
variants, genotypes).variants
self.assertEqual(len(var), 5) # 6 pairs but 5 variants
def test_compound_same_gene(self):
"""Compounds be on the same gene"""
genotypes = array(
[[2, 1, 2, 2, 1, 1], [1, 2, 2, 2, 1, 1], [1, 2, 2, 2, 1, 1]],
dtype=DTYPE)
variants = VariantsCollection([
Variant(variant_id=1, start=1, gene_symbol='B'), # mother contrib
Variant(variant_id=2, start=2, gene_symbol='B'), # father contrib
Variant(variant_id=3, start=2,
gene_symbol='C'), # X - not the same transcript
])
var = GenotypesFilterCompoundHeterozygous(self.ss).apply(
variants, genotypes).variants
self.assertEqual(
len(var), 2) # not 1 because they are reported on 2 separate lines
for v in var:
gts = genotypes[v.pk]
self.assertTrue((gts[0] == 1 and gts[1] == 2)
or (gts[0] == 2 and gts[1] == 1)) # parents
self.assertGreaterEqual(gts[2], 2)
self.assertGreaterEqual(gts[3], 2)
self.assertLessEqual(gts[4], 2)
self.assertLessEqual(gts[5], 1)
def test_not_on_chrom_X(self):
"""Only variants on chrom X must pass"""
genotypes = array([[2, 2, 2, 4, 1, 1]] * 10, dtype=DTYPE)
groups = {
"affected": ["Father", "Sasha", "Dasha"],
"not_affected": ["Mother", "Lena", "Lesha"]
}
ss = SamplesSelection(self.F.samples, groups)
variants = VariantsCollection([
Variant(variant_id=x + 1, chrom=('chrX' if x > 5 else 'chrU'))
for x in range(len(genotypes))
],
db='test')
var = GenotypesFilterXLinked(ss).apply(variants, genotypes).variants
self.assertEqual(len(var), 4)
variants = VariantsCollection([
Variant(variant_id=x + 1, chrom=('chrX' if x % 2 else 'chrU'))
for x in range(len(genotypes))
],
db='test')
var = GenotypesFilterXLinked(ss).apply(variants, genotypes).variants
self.assertEqual(len(var), 5)
def setUp(self):
self.F = Family()
self.genotypes = array(
[
[2, 1, 2, 2, 1, 1], # X - Mother carrier
[1, 2, 2, 2, 1, 1], # X - Father carrier
[1, 1, 2, 2, 1, 1], # V - OK
[1, 1, 2, 2, 1, 2], # X - Not affected mutant
[1, 1, 2, 1, 1, 1], # X - Affected non carrier
[1, 1, 2, 2, 1, 1]
], # V - Actually the only working situation
dtype=DTYPE)
self.variants = VariantsCollection(
[Variant(variant_id=x + 1) for x in range(len(self.genotypes))],
db='test')
def setUp(self):
self.F = Family()
self.genotypes = array(
[
[2, 1, 2, 2, 1, 1], # X - from unaffected mother
[1, 2, 2, 2, 1, 1], # V - from affected father
[1, 2, 2, 1, 1, 1], # X - affected without the mutation
[1, 2, 2, 2, 2, 1], # X - not affected with the mutation
[1, 1, 2, 1, 1, 1], # X - de novo
[2, 2, 4, 1, 1, 1
], # X - recessive, both parents have it (unaffected)
[1, 2, 2, 2, 1, 1]
], # V - Actually the only working situation
dtype=DTYPE)
self.variants = VariantsCollection(
[Variant(variant_id=x + 1) for x in range(len(self.genotypes))],
db='test')
def test_both_parents_carriers_unaffected(self):
"""Both parents are carriers"""
genotypes = array(
[
[2, 2, 2, 4, 1, 1
], # V - Both parents carriers, affected children have it
[4, 2, 2, 4, 1, 1
], # X - Both parents carriers, mother carrier hom
[2, 2, 2, 4, 2, 1
], # X - Both parents carriers, non-affected son carrier
[
2, 2, 2, 4, 1, 4
], # X - Both parents carriers, non-affected external female carrier hom
[
2, 4, 2, 4, 1, 2
], # X - Both parents carriers, mother is carrier hom (should be affected)
[
4, 2, 2, 4, 1, 2
], # X - Both parents carriers, father is carrier hom (can't be)
[
2, 2, 2, 2, 1, 1
], # X - Both parents carriers, affected daughter is carrier het only
[2, 2, 4, 4, 1, 1]
], # X - Both parents carriers, affected male is carrier hom
dtype=DTYPE)
variants = VariantsCollection([
Variant(variant_id=x + 1, chrom='chrX')
for x in range(len(genotypes))
],
db='test')
groups = {
"affected": ["Father", "Sasha", "Dasha"],
"not_affected": ["Mother", "Lena", "Lesha"]
}
ss = SamplesSelection(self.F.samples, groups)
var = GenotypesFilterXLinked(ss).apply(variants, genotypes).variants
self.assertEqual(len(var), 1)
for v in var:
gts = genotypes[v.pk - 1]
self.assertLessEqual(gts[0], 6) # mother
self.assertEqual(gts[1], 2) # father
self.assertEqual(gts[2], 2) # Sasha
self.assertEqual(gts[3], 4) # Dasha
self.assertEqual(gts[4], 1) # Lesha
self.assertLessEqual(gts[5], 3) # Lena
def test_mother_only_carrier(self):
"""Only the mother carries the variant. Father is unaffected."""
variants = VariantsCollection([Variant(variant_id=1, chrom='chrX')],
db='test')
# X - Mother carrier, father not, affected daughter (cannot be, she needs 2 copies)
groups = {
"affected": ["Dasha"],
"not_affected": ["Mother", "Father", "Sasha", "Lena", "Lesha"]
}
ss = SamplesSelection(self.F.samples, groups)
genotypes = array([[2, 1, 1, 4, 1, 1]], dtype=DTYPE)
var = GenotypesFilterXLinked(ss).apply(variants, genotypes).variants
self.assertEqual(len(var), 0)
# V - Mother carrier, father not, affected son carrier het
groups = {
"affected": ["Sasha"],
"not_affected": ["Mother", "Father", "Dasha", "Lena", "Lesha"]
}
ss = SamplesSelection(self.F.samples, groups)
genotypes = array([[2, 1, 2, 1, 1, 1]], dtype=DTYPE)
var = GenotypesFilterXLinked(ss).apply(variants, genotypes).variants
self.assertEqual(len(var), 1)
def test_father_only_carrier(self):
"""Only the father carries the variant - so he is affected."""
variants = VariantsCollection([Variant(variant_id=1, chrom='chrX')],
db='test')
# X - Father carrier, mother not, daughter affected
groups = {
"affected": ["Father", "Dasha"],
"not_affected": ["Mother", "Sasha", "Lena", "Lesha"]
}
ss = SamplesSelection(self.F.samples, groups)
genotypes = array([[1, 2, 1, 4, 1, 1]], dtype=DTYPE)
var = GenotypesFilterXLinked(ss).apply(variants, genotypes).variants
self.assertEqual(len(var), 0)
# X - Father carrier, mother not, son affected
groups = {
"affected": ["Father", "Sasha"],
"not_affected": ["Mother", "Dasha", "Lena", "Lesha"]
}
ss = SamplesSelection(self.F.samples, groups)
genotypes = array([[1, 2, 2, 1, 1, 1]], dtype=DTYPE)
var = GenotypesFilterXLinked(ss).apply(variants, genotypes).variants
self.assertEqual(len(var), 0)