def test_heterozygosity_observed(self):
# diploid
g = GenotypeArray(
[[[0, 0], [0, 0]], [[1, 1], [1, 1]], [[1, 1], [2, 2]],
[[0, 0], [0, 1]], [[0, 0], [0, 2]], [[1, 1], [1, 2]],
[[0, 1], [0, 1]], [[0, 1], [1, 2]], [[0, 0], [-1, -1]],
[[0, 1], [-1, -1]], [[-1, -1], [-1, -1]]],
dtype='i1')
expect = [0, 0, 0, .5, .5, .5, 1, 1, 0, 1, -1]
actual = allel.heterozygosity_observed(g, fill=-1)
aeq(expect, actual)
# polyploid
g = GenotypeArray(
[[[0, 0, 0], [0, 0, 0]], [[1, 1, 1], [1, 1, 1]],
[[1, 1, 1], [2, 2, 2]], [[0, 0, 0], [0, 0, 1]],
[[0, 0, 0], [0, 0, 2]], [[1, 1, 1], [0, 1, 2]],
[[0, 0, 1], [0, 1, 1]], [[0, 1, 1], [0, 1, 2]],
[[0, 0, 0], [-1, -1, -1]], [[0, 0, 1], [-1, -1, -1]],
[[-1, -1, -1], [-1, -1, -1]]],
dtype='i1')
expect = [0, 0, 0, .5, .5, .5, 1, 1, 0, 1, -1]
actual = allel.heterozygosity_observed(g, fill=-1)
aeq(expect, actual)
def test_slice_types(self):
g = GenotypeArray(diploid_genotype_data, dtype='i1')
# row slice
s = g[1:]
assert isinstance(s, GenotypeArray)
# col slice
s = g[:, 1:]
assert isinstance(s, GenotypeArray)
# row index
s = g[0]
assert isinstance(s, GenotypeVector)
assert not isinstance(s, GenotypeArray)
# col index
s = g[:, 0]
assert isinstance(s, GenotypeVector)
assert not isinstance(s, GenotypeArray)
# ploidy index
s = g[:, :, 0]
assert isinstance(s, np.ndarray)
assert not isinstance(s, GenotypeArray)
# item
s = g[0, 0, 0]
assert isinstance(s, np.int8)
assert not isinstance(s, GenotypeArray)
def test_pairwise_distance_multidim(self):
g = GenotypeArray(
[[[0, 0], [0, 0]], [[1, 1], [1, 1]], [[1, 1], [2, 2]],
[[0, 0], [0, 1]], [[0, 0], [0, 2]], [[1, 1], [1, 2]],
[[0, 1], [0, 1]], [[0, 1], [1, 2]], [[0, 0], [-1, -1]],
[[0, 1], [-1, -1]], [[-1, -1], [-1, -1]]],
dtype='i1')
gac = g.to_allele_counts()
def metric(ac1, ac2):
mpd = allel.mean_pairwise_difference_between(ac1, ac2, fill=0)
return mpd.sum()
expect = [
allel.mean_pairwise_difference_between(gac[:, 0],
gac[:, 1],
fill=0).sum()
]
actual = allel.pairwise_distance(gac, metric)
aeq(expect, actual)
def test_haploidify_samples(self):
# diploid
g = GenotypeArray([[[0, 1], [2, 3]],
[[4, 5], [6, 7]],
[[8, 9], [10, 11]]], dtype='i1')
h = g.haploidify_samples()
assert 2 == h.ndim
assert 3 == h.n_variants
assert 2 == h.n_haplotypes
assert np.int8 == h.dtype
for i in range(g.n_variants):
for j in range(g.n_samples):
self.assertIn(h[i, j], set(g[i, j]))
# triploid
g = GenotypeArray([[[0, 1, 2], [3, 4, 5]],
[[6, 7, 8], [9, 10, 11]],
[[12, 13, 14], [15, 16, 17]]], dtype='i1')
h = g.haploidify_samples()
assert 2 == h.ndim
assert 3 == h.n_variants
assert 2 == h.n_haplotypes
assert np.int8 == h.dtype
for i in range(g.n_variants):
for j in range(g.n_samples):
self.assertIn(h[i, j], set(g[i, j]))
def test_heterozygosity_expected(self):
def refimpl(f, ploidy, fill=0):
"""Limited reference implementation for testing purposes."""
# check allele frequencies sum to 1
af_sum = np.sum(f, axis=1)
# assume three alleles
p = f[:, 0]
q = f[:, 1]
r = f[:, 2]
out = 1 - p**ploidy - q**ploidy - r**ploidy
with ignore_invalid():
out[(af_sum < 1) | np.isnan(af_sum)] = fill
return out
# diploid
g = GenotypeArray([[[0, 0], [0, 0]],
[[1, 1], [1, 1]],
[[1, 1], [2, 2]],
[[0, 0], [0, 1]],
[[0, 0], [0, 2]],
[[1, 1], [1, 2]],
[[0, 1], [0, 1]],
[[0, 1], [1, 2]],
[[0, 0], [-1, -1]],
[[0, 1], [-1, -1]],
[[-1, -1], [-1, -1]]], dtype='i1')
expect1 = [0, 0, 0.5, .375, .375, .375, .5, .625, 0, .5, -1]
af = g.count_alleles().to_frequencies()
expect2 = refimpl(af, ploidy=g.ploidy, fill=-1)
actual = allel.heterozygosity_expected(af, ploidy=g.ploidy, fill=-1)
assert_array_almost_equal(expect1, actual)
assert_array_almost_equal(expect2, actual)
expect3 = [0, 0, 0.5, .375, .375, .375, .5, .625, 0, .5, 0]
actual = allel.heterozygosity_expected(af, ploidy=g.ploidy, fill=0)
assert_array_almost_equal(expect3, actual)
# polyploid
g = GenotypeArray([[[0, 0, 0], [0, 0, 0]],
[[1, 1, 1], [1, 1, 1]],
[[1, 1, 1], [2, 2, 2]],
[[0, 0, 0], [0, 0, 1]],
[[0, 0, 0], [0, 0, 2]],
[[1, 1, 1], [0, 1, 2]],
[[0, 0, 1], [0, 1, 1]],
[[0, 1, 1], [0, 1, 2]],
[[0, 0, 0], [-1, -1, -1]],
[[0, 0, 1], [-1, -1, -1]],
[[-1, -1, -1], [-1, -1, -1]]], dtype='i1')
af = g.count_alleles().to_frequencies()
expect = refimpl(af, ploidy=g.ploidy, fill=-1)
actual = allel.heterozygosity_expected(af, ploidy=g.ploidy, fill=-1)
assert_array_almost_equal(expect, actual)
def build_genotype_array(genotypes, pop_samples, markers):
g = list()
for marker in markers:
mgt = list()
for population, sample_list in pop_samples.items():
for sample in sorted(sample_list):
mgt.append(genotypes[sample][marker])
g.append(mgt)
gt = GenotypeArray(g)
print('GenotypeArray construction complete:',
gt.n_variants,
'markers,',
gt.n_samples,
'samples, and a ploidy of',
gt.ploidy,
file=sys.stderr)
return gt
def test_inbreeding_coefficient(self):
# diploid
g = GenotypeArray(
[[[0, 0], [0, 0]], [[1, 1], [1, 1]], [[1, 1], [2, 2]],
[[0, 0], [0, 1]], [[0, 0], [0, 2]], [[1, 1], [1, 2]],
[[0, 1], [0, 1]], [[0, 1], [1, 2]], [[0, 0], [-1, -1]],
[[0, 1], [-1, -1]], [[-1, -1], [-1, -1]]],
dtype='i1')
# ho = np.array([0, 0, 0, .5, .5, .5, 1, 1, 0, 1, -1])
# he = np.array([0, 0, 0.5, .375, .375, .375, .5, .625, 0, .5, -1])
# expect = 1 - (ho/he)
expect = [
-1, -1, 1 - 0, 1 - (.5 / .375), 1 - (.5 / .375), 1 - (.5 / .375),
1 - (1 / .5), 1 - (1 / .625), -1, 1 - (1 / .5), -1
]
actual = allel.inbreeding_coefficient(g, fill=-1)
assert_array_almost_equal(expect, actual)
def test_constructor(self):
# missing data arg
with pytest.raises(TypeError):
# noinspection PyArgumentList
GenotypeArray()
# data has wrong dtype
data = 'foo bar'
with pytest.raises(TypeError):
GenotypeArray(data)
# data has wrong dtype
data = [4., 5., 3.7]
with pytest.raises(TypeError):
GenotypeArray(data)
# data has wrong dimensions
data = [1, 2, 3]
with pytest.raises(TypeError):
GenotypeArray(data)
# data has wrong dimensions
data = [[1, 2], [3, 4]] # use HaplotypeArray instead
with pytest.raises(TypeError):
GenotypeArray(data)
# diploid data (typed)
g = GenotypeArray(diploid_genotype_data, dtype='i1')
aeq(diploid_genotype_data, g)
assert np.int8 == g.dtype
# polyploid data (typed)
g = GenotypeArray(triploid_genotype_data, dtype='i1')
aeq(triploid_genotype_data, g)
assert np.int8 == g.dtype
def setup_instance(self, data, dtype=None):
return GenotypeArray(data, dtype=dtype)