def test_view(self):
# data has wrong dtype
data = 'foo bar'
with self.assertRaises(TypeError):
np.array(data).view(HaplotypeArray)
# data has wrong dtype
data = [4., 5., 3.7]
with self.assertRaises(TypeError):
np.array(data).view(HaplotypeArray)
# data has wrong dimensions
data = [1, 2, 3]
with self.assertRaises(TypeError):
np.array(data).view(HaplotypeArray)
# data has wrong dimensions
data = diploid_genotype_data # use GenotypeArray instead
with self.assertRaises(TypeError):
np.array(data).view(HaplotypeArray)
# haploid data
h = np.array(haplotype_data).view(HaplotypeArray)
aeq(haplotype_data, h)
eq(np.int, h.dtype)
eq(2, h.ndim)
eq(4, h.n_variants)
eq(3, h.n_haplotypes)
def test_constructor(self):
# missing data arg
with assert_raises(TypeError):
# noinspection PyArgumentList
AlleleCountsDaskArray.from_array()
# data has wrong dtype
data = 'foo bar'
with assert_raises(ValueError):
AlleleCountsDaskArray.from_array(data)
# data has wrong dtype
data = np.array([4., 5., 3.7])
with assert_raises(ValueError):
AlleleCountsDaskArray.from_array(data)
# data has wrong dimensions
data = np.array([1, 2, 3])
with assert_raises(ValueError):
AlleleCountsDaskArray.from_array(data)
# data has wrong dimensions
data = np.array([[[1, 2], [3, 4]]])
with assert_raises(ValueError):
AlleleCountsDaskArray.from_array(data)
# valid data (typed)
hd = self.setup_instance(np.array(allele_counts_data, dtype='u2'))
aeq(allele_counts_data, hd)
eq(np.uint16, hd.dtype)
def test_to_hdf5_group(self):
# setup HDF5 file
node_path = 'test'
tf = tempfile.NamedTemporaryFile(delete=False)
file_path = tf.name
tf.close()
a = np.rec.array(variant_table_data, dtype=variant_table_dtype)
# reorder columns because will come back out in sorted order
a = a[sorted(a.dtype.names)]
vt = self.setup_instance(a)
# write using file path and node path
vt.to_hdf5_group(file_path, node_path)
with h5py.File(file_path, mode='r') as h5f:
h5g = h5f[node_path]
eq(sorted(a.dtype.names), sorted(h5g.keys()))
for n in a.dtype.names:
aeq(a[n], h5g[n][:])
# write using group and node path
with h5py.File(file_path, mode='w') as h5f:
vt.to_hdf5_group(h5f, node_path)
with h5py.File(file_path, mode='r') as h5f:
h5g = h5f[node_path]
eq(sorted(a.dtype.names), sorted(h5g.keys()))
for n in a.dtype.names:
aeq(a[n], h5g[n][:])
def test_from_hdf5_group(self):
# setup HDF5 file
node_path = 'test'
tf = tempfile.NamedTemporaryFile(delete=False)
file_path = tf.name
tf.close()
a = np.rec.array(variant_table_data, dtype=variant_table_dtype)
# reorder columns because will come back out in sorted order
a = a[sorted(a.dtype.names)]
with h5py.File(file_path, mode='w') as h5f:
h5g = h5f.create_group(node_path)
for n in a.dtype.names:
h5g.create_dataset(n,
data=a[n],
chunks=True,
compression='gzip')
# file and node path
vt = self._class.from_hdf5_group(file_path, node_path)
self.assertIsInstance(vt, self._class)
aeq(a, vt[:])
# dataset
with h5py.File(file_path, mode='r') as h5f:
h5g = h5f[node_path]
vt = self._class.from_hdf5_group(h5g)
self.assertIsInstance(vt, self._class)
aeq(a, vt[:])
def test_mean_pairwise_diversity(self):
# start with simplest case, two haplotypes, one pairwise comparison
h = HaplotypeArray([[0, 0],
[1, 1],
[0, 1],
[1, 2],
[0, -1],
[-1, -1]])
ac = h.count_alleles()
expect = [0, 0, 1, 1, -1, -1]
actual = allel.mean_pairwise_difference(ac, fill=-1)
aeq(expect, actual)
# four haplotypes, 6 pairwise comparison
h = HaplotypeArray([[0, 0, 0, 0],
[0, 0, 0, 1],
[0, 0, 1, 1],
[0, 1, 1, 1],
[1, 1, 1, 1],
[0, 0, 1, 2],
[0, 1, 1, 2],
[0, 1, -1, -1],
[-1, -1, -1, -1]])
ac = h.count_alleles()
expect = [0, 3/6, 4/6, 3/6, 0, 5/6, 5/6, 1, -1]
actual = allel.mean_pairwise_difference(ac, fill=-1)
assert_array_almost_equal(expect, actual)
def test_to_hdf5(self):
# setup HDF5 file
tf = tempfile.NamedTemporaryFile(delete=False)
file_path = tf.name
tf.close()
# setup genotype array
node_path = 'test'
g = GenotypeCArray(diploid_genotype_data, dtype='i1')
# write using file path and node path
g.to_hdf5(file_path, node_path)
# test outcome
with h5py.File(file_path, mode='r') as h5f:
h5d = h5f[node_path]
aeq(g[:], h5d[:])
# write using group
with h5py.File(file_path, mode='w') as h5f:
g.to_hdf5(h5f, node_path)
# test outcome
with h5py.File(file_path, mode='r') as h5f:
h5d = h5f[node_path]
aeq(g[:], h5d[:])
def test_constructor(self):
# missing data arg
with self.assertRaises(TypeError):
# noinspection PyArgumentList
HaplotypeArray()
# data has wrong dtype
data = 'foo bar'
with self.assertRaises(TypeError):
HaplotypeArray(data)
# data has wrong dtype
data = [4., 5., 3.7]
with self.assertRaises(TypeError):
HaplotypeArray(data)
# data has wrong dimensions
data = [1, 2, 3]
with self.assertRaises(TypeError):
HaplotypeArray(data)
# data has wrong dimensions
data = diploid_genotype_data # use GenotypeArray instead
with self.assertRaises(TypeError):
HaplotypeArray(data)
# haploid data (typed)
h = HaplotypeArray(haplotype_data, dtype='i1')
aeq(haplotype_data, h)
eq(np.int8, h.dtype)
def test_constructor(self):
# missing data arg
with assert_raises(TypeError):
# noinspection PyArgumentList
AlleleCountsChunkedArray()
# data has wrong dtype
data = 'foo bar'
with assert_raises(TypeError):
AlleleCountsChunkedArray(data)
# data has wrong dtype
data = np.array([4., 5., 3.7])
with assert_raises(TypeError):
AlleleCountsChunkedArray(data)
# data has wrong dimensions
data = np.array([1, 2, 3])
with assert_raises(TypeError):
AlleleCountsChunkedArray(data)
# data has wrong dimensions
data = np.array([[[1, 2], [3, 4]]])
with assert_raises(TypeError):
AlleleCountsChunkedArray(data)
# typed data (typed)
ac = AlleleCountsChunkedArray(np.array(allele_counts_data, dtype='u1'))
aeq(allele_counts_data, ac)
eq(np.uint8, ac.dtype)
def test_constructor(self):
# missing data arg
with assert_raises(TypeError):
# noinspection PyArgumentList
AlleleCountsDaskArray.from_array()
# data has wrong dtype
data = 'foo bar'
with assert_raises(ValueError):
AlleleCountsDaskArray.from_array(data)
# data has wrong dtype
data = np.array([4., 5., 3.7])
with assert_raises(ValueError):
AlleleCountsDaskArray.from_array(data)
# data has wrong dimensions
data = np.array([1, 2, 3])
with assert_raises(ValueError):
AlleleCountsDaskArray.from_array(data)
# data has wrong dimensions
data = np.array([[[1, 2], [3, 4]]])
with assert_raises(ValueError):
AlleleCountsDaskArray.from_array(data)
# valid data (typed)
hd = self.setup_instance(np.array(allele_counts_data, dtype='u2'))
aeq(allele_counts_data, hd)
eq(np.uint16, hd.dtype)
def test_from_hdf5_condition(self):
# setup HDF5 file
node_path = 'test'
tf = tempfile.NamedTemporaryFile(delete=False)
file_path = tf.name
tf.close()
with h5py.File(file_path, mode='w') as h5f:
h5f.create_dataset(node_path,
data=diploid_genotype_data,
chunks=(2, 3, 2))
# selection
condition = [False, True, False, True, False]
# file and node path
g = GenotypeCArray.from_hdf5(file_path, node_path, condition=condition)
expect = GenotypeArray(diploid_genotype_data).compress(condition,
axis=0)
aeq(expect, g)
# dataset
with h5py.File(file_path, mode='r') as h5f:
dataset = h5f[node_path]
g = GenotypeCArray.from_hdf5(dataset, condition=condition)
aeq(expect, g)
def test_constructor(self):
# missing data arg
with self.assertRaises(TypeError):
# noinspection PyArgumentList
AlleleCountsArray()
# data has wrong dtype
data = 'foo bar'
with self.assertRaises(TypeError):
AlleleCountsArray(data)
# data has wrong dtype
data = [4., 5., 3.7]
with self.assertRaises(TypeError):
AlleleCountsArray(data)
# data has wrong dimensions
data = [1, 2, 3]
with self.assertRaises(TypeError):
AlleleCountsArray(data)
# data has wrong dimensions
data = diploid_genotype_data
with self.assertRaises(TypeError):
AlleleCountsArray(data)
# valid data (typed)
ac = AlleleCountsArray(allele_counts_data, dtype='u1')
aeq(allele_counts_data, ac)
eq(np.uint8, ac.dtype)
def test_to_hdf5(self):
# setup HDF5 file
tf = tempfile.NamedTemporaryFile(delete=False)
file_path = tf.name
tf.close()
# setup genotype array
node_path = 'test'
g = GenotypeCArray(diploid_genotype_data, dtype='i1')
# write using file path and node path
g.to_hdf5(file_path, node_path)
# test outcome
with h5py.File(file_path, mode='r') as h5f:
h5d = h5f[node_path]
aeq(g[:], h5d[:])
# write using group
with h5py.File(file_path, mode='w') as h5f:
g.to_hdf5(h5f, node_path)
# test outcome
with h5py.File(file_path, mode='r') as h5f:
h5d = h5f[node_path]
aeq(g[:], h5d[:])
def test_constructor(self):
# missing data arg
with assert_raises(TypeError):
# noinspection PyArgumentList
HaplotypeChunkedArray()
# data has wrong dtype
data = 'foo bar'
with assert_raises(TypeError):
HaplotypeChunkedArray(data)
# data has wrong dtype
data = np.array([4., 5., 3.7])
with assert_raises(TypeError):
HaplotypeChunkedArray(data)
# data has wrong dimensions
data = np.array([1, 2, 3])
with assert_raises(TypeError):
HaplotypeChunkedArray(data)
# data has wrong dimensions
data = np.array([[[1, 2], [3, 4]]]) # use GenotypeCArray instead
with assert_raises(TypeError):
HaplotypeChunkedArray(data)
# typed data (typed)
h = HaplotypeChunkedArray(np.array(haplotype_data, dtype='i1'))
aeq(haplotype_data, h)
eq(np.int8, h.dtype)
def test_constructor(self):
# missing data arg
with assert_raises(TypeError):
# noinspection PyArgumentList
HaplotypeDaskArray.from_array()
# data has wrong dtype
data = 'foo bar'
with assert_raises(ValueError):
HaplotypeDaskArray.from_array(data)
# data has wrong dtype
data = np.array([4., 5., 3.7])
with assert_raises(ValueError):
HaplotypeDaskArray.from_array(data)
# data has wrong dimensions
data = np.array([1, 2, 3])
with assert_raises(ValueError):
HaplotypeDaskArray.from_array(data)
# data has wrong dimensions
data = np.array([[[1, 2], [3, 4]]]) # use GenotypeDaskArray instead
with assert_raises(ValueError):
HaplotypeDaskArray.from_array(data)
# valid data (typed)
hd = self.setup_instance(np.array(haplotype_data, dtype='i1'))
aeq(haplotype_data, hd)
eq(np.int8, hd.dtype)
def test_from_hdf5_group(self):
# setup HDF5 file
node_path = 'test'
tf = tempfile.NamedTemporaryFile(delete=False)
file_path = tf.name
tf.close()
a = np.rec.array(variant_table_data, dtype=variant_table_dtype)
# reorder columns because will come back out in sorted order
a = a[sorted(a.dtype.names)]
with h5py.File(file_path, mode='w') as h5f:
h5g = h5f.create_group(node_path)
for n in a.dtype.names:
h5g.create_dataset(n, data=a[n], chunks=True,
compression='gzip')
# file and node path
vt = self._class.from_hdf5_group(file_path, node_path)
self.assertIsInstance(vt, self._class)
aeq(a, vt[:])
# dataset
with h5py.File(file_path, mode='r') as h5f:
h5g = h5f[node_path]
vt = self._class.from_hdf5_group(h5g)
self.assertIsInstance(vt, self._class)
aeq(a, vt[:])
def test_constructor(self):
# missing data arg
with self.assertRaises(TypeError):
# noinspection PyArgumentList
HaplotypeArray()
# data has wrong dtype
data = 'foo bar'
with self.assertRaises(TypeError):
HaplotypeArray(data)
# data has wrong dtype
data = [4., 5., 3.7]
with self.assertRaises(TypeError):
HaplotypeArray(data)
# data has wrong dimensions
data = [1, 2, 3]
with self.assertRaises(TypeError):
HaplotypeArray(data)
# data has wrong dimensions
data = diploid_genotype_data # use GenotypeArray instead
with self.assertRaises(TypeError):
HaplotypeArray(data)
# haploid data (typed)
h = HaplotypeArray(haplotype_data, dtype='i1')
aeq(haplotype_data, h)
eq(np.int8, h.dtype)
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.stats.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.stats.heterozygosity_observed(g, fill=-1)
aeq(expect, actual)
def test_view(self):
# data has wrong dtype
data = 'foo bar'
with self.assertRaises(TypeError):
np.array(data).view(HaplotypeArray)
# data has wrong dtype
data = [4., 5., 3.7]
with self.assertRaises(TypeError):
np.array(data).view(HaplotypeArray)
# data has wrong dimensions
data = [1, 2, 3]
with self.assertRaises(TypeError):
np.array(data).view(HaplotypeArray)
# data has wrong dimensions
data = diploid_genotype_data # use GenotypeArray instead
with self.assertRaises(TypeError):
np.array(data).view(HaplotypeArray)
# haploid data
h = np.array(haplotype_data).view(HaplotypeArray)
aeq(haplotype_data, h)
eq(np.int, h.dtype)
eq(2, h.ndim)
eq(4, h.n_variants)
eq(3, h.n_haplotypes)
def test_constructor(self):
# missing data arg
with self.assertRaises(TypeError):
# noinspection PyArgumentList
AlleleCountsArray()
# data has wrong dtype
data = 'foo bar'
with self.assertRaises(TypeError):
AlleleCountsArray(data)
# data has wrong dtype
data = [4., 5., 3.7]
with self.assertRaises(TypeError):
AlleleCountsArray(data)
# data has wrong dimensions
data = [1, 2, 3]
with self.assertRaises(TypeError):
AlleleCountsArray(data)
# data has wrong dimensions
data = diploid_genotype_data
with self.assertRaises(TypeError):
AlleleCountsArray(data)
# valid data (typed)
ac = AlleleCountsArray(allele_counts_data, dtype='u1')
aeq(allele_counts_data, ac)
eq(np.uint8, ac.dtype)
def test_view(self):
# data has wrong dtype
data = 'foo bar'
with self.assertRaises(TypeError):
np.array(data).view(AlleleCountsArray)
# data has wrong dtype
data = [4., 5., 3.7]
with self.assertRaises(TypeError):
np.array(data).view(AlleleCountsArray)
# data has wrong dimensions
data = [1, 2, 3]
with self.assertRaises(TypeError):
np.array(data).view(AlleleCountsArray)
# data has wrong dimensions
data = diploid_genotype_data
with self.assertRaises(TypeError):
np.array(data).view(AlleleCountsArray)
# valid data
ac = np.array(allele_counts_data).view(AlleleCountsArray)
aeq(allele_counts_data, ac)
eq(np.int, ac.dtype)
eq(2, ac.ndim)
eq(6, ac.n_variants)
eq(3, ac.n_alleles)
def test_mean_pairwise_diversity(self):
# start with simplest case, two haplotypes, one pairwise comparison
h = HaplotypeArray([[0, 0],
[1, 1],
[0, 1],
[1, 2],
[0, -1],
[-1, -1]])
ac = h.count_alleles()
expect = [0, 0, 1, 1, -1, -1]
actual = allel.stats.mean_pairwise_difference(ac, fill=-1)
aeq(expect, actual)
# four haplotypes, 6 pairwise comparison
h = HaplotypeArray([[0, 0, 0, 0],
[0, 0, 0, 1],
[0, 0, 1, 1],
[0, 1, 1, 1],
[1, 1, 1, 1],
[0, 0, 1, 2],
[0, 1, 1, 2],
[0, 1, -1, -1],
[-1, -1, -1, -1]])
ac = h.count_alleles()
expect = [0, 3/6, 4/6, 3/6, 0, 5/6, 5/6, 1, -1]
actual = allel.stats.mean_pairwise_difference(ac, fill=-1)
assert_array_close(expect, actual)
def test_constructor(self):
# missing data arg
with self.assertRaises(TypeError):
# noinspection PyArgumentList
UniqueIndex()
# data has wrong dimensions
data = [['A', 'C'], ['B', 'F']]
with self.assertRaises(TypeError):
UniqueIndex(data)
# labels are not unique
data = ['A', 'B', 'D', 'B']
with self.assertRaises(ValueError):
UniqueIndex(data)
# valid data
data = ['A', 'C', 'B', 'F']
lbl = UniqueIndex(data)
aeq(data, lbl)
eq(1, lbl.ndim)
eq(4, len(lbl))
# valid data (typed)
data = np.array(['A', 'C', 'B', 'F'], dtype='S1')
lbl = UniqueIndex(data, dtype='S1')
aeq(data, lbl)
def test_view(self):
# data has wrong dtype
data = 'foo bar'
with self.assertRaises(TypeError):
np.array(data).view(AlleleCountsArray)
# data has wrong dtype
data = [4., 5., 3.7]
with self.assertRaises(TypeError):
np.array(data).view(AlleleCountsArray)
# data has wrong dimensions
data = [1, 2, 3]
with self.assertRaises(TypeError):
np.array(data).view(AlleleCountsArray)
# data has wrong dimensions
data = diploid_genotype_data
with self.assertRaises(TypeError):
np.array(data).view(AlleleCountsArray)
# valid data
ac = np.array(allele_counts_data).view(AlleleCountsArray)
aeq(allele_counts_data, ac)
eq(np.int, ac.dtype)
eq(2, ac.ndim)
eq(5, ac.n_variants)
eq(3, ac.n_alleles)
def test_from_hdf5_condition(self):
# setup HDF5 file
node_path = 'test'
tf = tempfile.NamedTemporaryFile(delete=False)
file_path = tf.name
tf.close()
with h5py.File(file_path, mode='w') as h5f:
h5f.create_dataset(node_path,
data=diploid_genotype_data,
chunks=(2, 3, 2))
# selection
condition = [False, True, False, True, False]
# file and node path
g = GenotypeCArray.from_hdf5(file_path, node_path, condition=condition)
expect = GenotypeArray(diploid_genotype_data).compress(condition,
axis=0)
aeq(expect, g)
# dataset
with h5py.File(file_path, mode='r') as h5f:
dataset = h5f[node_path]
g = GenotypeCArray.from_hdf5(dataset, condition=condition)
aeq(expect, g)
def test_constructor(self):
# missing data arg
with pytest.raises(TypeError):
# noinspection PyArgumentList
UniqueIndex()
# data has wrong dimensions
data = [['A', 'C'], ['B', 'F']]
with pytest.raises(TypeError):
UniqueIndex(data)
# labels are not unique
data = ['A', 'B', 'D', 'B']
with pytest.raises(ValueError):
UniqueIndex(data)
# valid data
data = ['A', 'C', 'B', 'F']
lbl = UniqueIndex(data)
aeq(data, lbl)
assert 1 == lbl.ndim
assert 4 == len(lbl)
# valid data (typed)
data = np.array(['A', 'C', 'B', 'F'], dtype='S1')
lbl = UniqueIndex(data, dtype='S1')
aeq(data, lbl)
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_to_hdf5_group(self):
# setup HDF5 file
node_path = 'test'
tf = tempfile.NamedTemporaryFile(delete=False)
file_path = tf.name
tf.close()
a = np.rec.array(variant_table_data, dtype=variant_table_dtype)
# reorder columns because will come back out in sorted order
a = a[sorted(a.dtype.names)]
vt = self.setup_instance(a)
# write using file path and node path
vt.to_hdf5_group(file_path, node_path)
with h5py.File(file_path, mode='r') as h5f:
h5g = h5f[node_path]
eq(sorted(a.dtype.names), sorted(h5g.keys()))
for n in a.dtype.names:
aeq(a[n], h5g[n][:])
# write using group and node path
with h5py.File(file_path, mode='w') as h5f:
vt.to_hdf5_group(h5f, node_path)
with h5py.File(file_path, mode='r') as h5f:
h5g = h5f[node_path]
eq(sorted(a.dtype.names), sorted(h5g.keys()))
for n in a.dtype.names:
aeq(a[n], h5g[n][:])
def test_constructor(self):
# missing data arg
with assert_raises(TypeError):
# noinspection PyArgumentList
HaplotypeDaskArray.from_array()
# data has wrong dtype
data = 'foo bar'
with assert_raises(ValueError):
HaplotypeDaskArray.from_array(data)
# data has wrong dtype
data = np.array([4., 5., 3.7])
with assert_raises(ValueError):
HaplotypeDaskArray.from_array(data)
# data has wrong dimensions
data = np.array([1, 2, 3])
with assert_raises(ValueError):
HaplotypeDaskArray.from_array(data)
# data has wrong dimensions
data = np.array([[[1, 2], [3, 4]]]) # use GenotypeDaskArray instead
with assert_raises(ValueError):
HaplotypeDaskArray.from_array(data)
# valid data (typed)
hd = self.setup_instance(np.array(haplotype_data, dtype='i1'))
aeq(haplotype_data, hd)
eq(np.int8, hd.dtype)
def test_locate_unlinked(self):
gn = [[0, 1, 2], [0, 1, 2]]
expect = [True, False]
actual = allel.locate_unlinked(gn, size=2, step=2, threshold=.5)
aeq(expect, actual)
gn = [[0, 1, 1, 2], [0, 1, 1, 2], [1, 1, 0, 2], [1, 1, 0, 2]]
actual = allel.locate_unlinked(gn, size=2, step=1, threshold=.5)
expect = [True, False, True, False]
aeq(expect, actual)
gn = [[0, 1, 1, 2], [0, 1, 1, 2], [0, 1, 1, 2], [1, 1, 0, 2],
[1, 1, 0, 2]]
actual = allel.locate_unlinked(gn, size=2, step=1, threshold=.5)
expect = [True, False, True, True, False]
aeq(expect, actual)
actual = allel.locate_unlinked(gn, size=3, step=1, threshold=.5)
expect = [True, False, False, True, False]
aeq(expect, actual)
# test with bcolz carray
import bcolz
gnz = bcolz.carray(gn, chunklen=2)
actual = allel.locate_unlinked(gnz,
size=2,
step=1,
threshold=.5,
blen=2)
expect = [True, False, True, True, False]
aeq(expect, actual)
def test_constructor(self):
# missing data arg
with assert_raises(TypeError):
# noinspection PyArgumentList
AlleleCountsChunkedArray()
# data has wrong dtype
data = 'foo bar'
with assert_raises(ValueError):
AlleleCountsChunkedArray(data)
# data has wrong dtype
data = np.array([4., 5., 3.7])
with assert_raises(ValueError):
AlleleCountsChunkedArray(data)
# data has wrong dimensions
data = np.array([1, 2, 3])
with assert_raises(ValueError):
AlleleCountsChunkedArray(data)
# data has wrong dimensions
data = np.array([[[1, 2], [3, 4]]])
with assert_raises(ValueError):
AlleleCountsChunkedArray(data)
# typed data (typed)
ac = AlleleCountsChunkedArray(np.array(allele_counts_data, dtype='u1'))
aeq(allele_counts_data, ac)
eq(np.uint8, ac.dtype)
def test_constructor(self):
# missing data arg
with assert_raises(TypeError):
# noinspection PyArgumentList
HaplotypeChunkedArray()
# data has wrong dtype
data = 'foo bar'
with assert_raises(ValueError):
HaplotypeChunkedArray(data)
# data has wrong dtype
data = np.array([4., 5., 3.7])
with assert_raises(ValueError):
HaplotypeChunkedArray(data)
# data has wrong dimensions
data = np.array([1, 2, 3])
with assert_raises(ValueError):
HaplotypeChunkedArray(data)
# data has wrong dimensions
data = np.array([[[1, 2], [3, 4]]]) # use GenotypeCArray instead
with assert_raises(ValueError):
HaplotypeChunkedArray(data)
# typed data (typed)
h = HaplotypeChunkedArray(np.array(haplotype_data, dtype='i1'))
aeq(haplotype_data, h)
eq(np.int8, h.dtype)
def test_view(self):
a = np.rec.array(variant_table_data,
dtype=variant_table_dtype)
vt = a.view(VariantTable)
aeq(a, vt)
eq(1, vt.ndim)
eq(5, vt.n_variants)
eq(variant_table_names, vt.names)
def test_view(self):
a = np.rec.array(feature_table_data,
dtype=feature_table_dtype)
ft = a.view(FeatureTable)
aeq(a, ft)
eq(1, ft.ndim)
eq(6, ft.n_features)
eq(feature_table_names, ft.names)
def test_take(self):
g = np.array(diploid_genotype_data)
gd = self.setup_instance(g)
# take variants not in original order
indices = [2, 0]
expect = g.take(indices, axis=0)
actual = gd.take(indices, axis=0)
aeq(expect, actual)
def test_view(self):
a = np.rec.array(variant_table_data,
dtype=variant_table_dtype)
vt = a.view(VariantTable)
aeq(a, vt)
eq(1, vt.ndim)
eq(5, vt.n_variants)
eq(variant_table_names, vt.names)
def test_view(self):
a = np.rec.array(feature_table_data,
dtype=feature_table_dtype)
ft = a.view(FeatureTable)
aeq(a, ft)
eq(1, ft.ndim)
eq(6, ft.n_features)
eq(feature_table_names, ft.names)
def test_take(self):
g = np.array(diploid_genotype_data)
gd = self.setup_instance(g)
# take variants not in original order
indices = [2, 0]
expect = g.take(indices, axis=0)
actual = gd.take(indices, axis=0)
aeq(expect, actual)
def test_sfs_scaled(self):
dac = [0, 1, 2, 1]
expect = [0, 2, 2]
actual = allel.sfs_scaled(dac)
aeq(expect, actual)
for dtype in 'u2', 'i2', 'u8', 'i8':
daca = np.asarray(dac, dtype=dtype)
actual = allel.sfs_scaled(daca)
aeq(expect, actual)
def test_eval_vm(self):
a = np.rec.array(variant_table_data, dtype=variant_table_dtype)
vt = self.setup_instance(a)
expr = '(DP > 30) & (QD < 4)'
r = vt.eval(expr, vm='numexpr')
aeq([False, False, True, False, True], r)
r = vt.eval(expr, vm='python')
aeq([False, False, True, False, True], r)
def test_sfs_folded(self):
ac = [[0, 3], [1, 2], [2, 1]]
expect = [1, 2]
actual = allel.sfs_folded(ac)
aeq(expect, actual)
for dtype in 'u2', 'i2', 'u8', 'i8':
aca = np.asarray(ac, dtype=dtype)
actual = allel.sfs_folded(aca)
aeq(expect, actual)
def test_take(self):
a = np.rec.array(variant_table_data, dtype=variant_table_dtype)
vt = VariantTable(a)
# take variants not in original order
indices = [2, 0]
t = vt.take(indices)
eq(2, t.n_variants)
expect = a.take(indices)
aeq(expect, t)
def test_eval_vm(self):
a = np.rec.array(variant_table_data, dtype=variant_table_dtype)
vt = self.setup_instance(a)
expr = '(DP > 30) & (QD < 4)'
r = vt.eval(expr, vm='numexpr')
aeq([False, False, True, False, True], r)
r = vt.eval(expr, vm='python')
aeq([False, False, True, False, True], r)
def test_pdist(self):
from allel.stats.distance import pdist
h = HaplotypeArray([[0, 0, 0, 0], [0, 0, 0, 1], [0, 0, 1, 1],
[0, 1, 1, 1], [1, 1, 1, 1], [0, 0, 1, 2],
[0, 1, 1, 2], [0, 1, -1, -1], [-1, -1, -1, -1]])
import scipy.spatial
d1 = scipy.spatial.distance.pdist(h.T, 'hamming')
d2 = pdist(h, 'hamming')
aeq(d1, d2)
def test_roh_mhmm_100pct(self):
# values correspond to start/stop/length/is_marginal
roh_expected = np.array([[1, 100, 100, True]], dtype=object)
fraction_expected = 1.0
gv = np.zeros((4, 2), dtype=np.int16)
pos = [1, 10, 50, 100]
roh, fraction = allel.roh_mhmm(gv, pos, contig_size=100)
aeq(roh.values, roh_expected)
assert fraction == fraction_expected
def test_take(self):
g = self.setup_instance(diploid_genotype_data)
# take variants not in original order
indices = [2, 0]
t = g.take(indices, axis=0)
eq(2, t.n_variants)
eq(g.n_samples, t.n_samples)
eq(g.ploidy, t.ploidy)
expect = np.array(diploid_genotype_data).take(indices, axis=0)
aeq(expect, t)
def test_mask_inaccessible(self):
np.random.seed(2837)
for n_vars in [5, 50, 500]:
pos = np.arange(1, n_vars + 1)
ac = np.random.randint(1, 40, n_vars * 2).reshape((n_vars, 2))
mask = np.random.randint(2, size=n_vars).astype(bool)
mpos, mac = mask_inaccessible(mask, pos, ac)
aeq(mac, ac[mask])
aeq(mpos, pos[mask])
def test_take(self):
a = np.rec.array(variant_table_data,
dtype=variant_table_dtype)
vt = VariantTable(a)
# take variants not in original order
indices = [2, 0]
t = vt.take(indices)
eq(2, t.n_variants)
expect = a.take(indices)
aeq(expect, t)
def test_take(self):
g = self.setup_instance(diploid_genotype_data)
# take variants not in original order
indices = [2, 0]
t = g.take(indices, axis=0)
eq(2, t.n_variants)
eq(g.n_samples, t.n_samples)
eq(g.ploidy, t.ploidy)
expect = np.array(diploid_genotype_data).take(indices, axis=0)
aeq(expect, t)
def test_moving_statistic(self):
f = allel.moving_statistic
values = [2, 5, 8, 16]
expect = [7, 24]
actual = f(values, statistic=np.sum, size=2)
aeq(expect, actual)
values = [2, 5, 8, 16]
expect = [7, 13, 24]
actual = f(values, statistic=np.sum, size=2, step=1)
aeq(expect, actual)
def test_moving_statistic(self):
f = allel.stats.moving_statistic
values = [2, 5, 8, 16]
expect = [7, 24]
actual = f(values, statistic=np.sum, size=2)
aeq(expect, actual)
values = [2, 5, 8, 16]
expect = [7, 13, 24]
actual = f(values, statistic=np.sum, size=2, step=1)
aeq(expect, actual)
def test_count_alleles_subpops(self):
data = chunked.storage_registry['default'].array(diploid_genotype_data, chunklen=2)
g = GenotypeChunkedArray(data)
subpops = {'foo': [0, 2], 'bar': [1]}
ac_subpops = g.count_alleles_subpops(subpops)
for p in subpops.keys():
ac = g.take(subpops[p], axis=1).count_alleles()
aeq(ac, ac_subpops[p])
loc = np.array([True, False, True, False, True])
t = ac_subpops.compress(loc)
eq(3, len(t))
def test_mean_pairwise_divergence(self):
# simplest case, two haplotypes in each population
h = HaplotypeArray([[0, 0, 0, 0], [0, 0, 0, 1], [0, 0, 1, 1],
[0, 1, 1, 1], [1, 1, 1, 1], [0, 0, 1, 2],
[0, 1, 1, 2], [0, 1, -1, -1], [-1, -1, -1, -1]])
h1 = h.take([0, 1], axis=1)
h2 = h.take([2, 3], axis=1)
ac1 = h1.count_alleles()
ac2 = h2.count_alleles()
expect = [0 / 4, 2 / 4, 4 / 4, 2 / 4, 0 / 4, 4 / 4, 3 / 4, -1, -1]
actual = allel.mean_pairwise_difference_between(ac1, ac2, fill=-1)
aeq(expect, actual)
def test_to_n_ref_array_like(self):
# see also https://github.com/cggh/scikit-allel/issues/66
gn = self.setup_instance(diploid_genotype_data).to_n_ref(fill=-1)
t = gn > 0
eq(4, np.count_nonzero(t))
expect = np.array([[1, 1, 0],
[1, 0, 0],
[1, 0, 0],
[0, 0, 0],
[0, 0, 0]], dtype='b1')
aeq(expect, t)
# numpy reductions trigger the issue
expect = np.array([2, 1, 1, 0, 0])
actual = np.sum(t, axis=1)
aeq(expect, actual)
expect = np.array([0, 0, 0, 0, 0])
actual = np.min(t, axis=1)
aeq(expect, actual)
expect = np.array([1, 1, 1, 0, 0])
actual = np.max(t, axis=1)
aeq(expect, actual)
def test_pdist(self):
h = HaplotypeArray([[0, 0, 0, 0],
[0, 0, 0, 1],
[0, 0, 1, 1],
[0, 1, 1, 1],
[1, 1, 1, 1],
[0, 0, 1, 2],
[0, 1, 1, 2],
[0, 1, -1, -1],
[-1, -1, -1, -1]])
import scipy.spatial
d1 = scipy.spatial.distance.pdist(h.T, 'hamming')
import allel.stats.distance
d2 = allel.stats.distance.pdist(h, 'hamming')
aeq(d1, d2)
def test_constructor(self):
# missing data arg
with self.assertRaises(TypeError):
# noinspection PyArgumentList
SortedIndex()
# data has wrong dtype
data = 'foo bar'
with self.assertRaises(TypeError):
SortedIndex(data)
# data has wrong dimensions
data = [[1, 2], [3, 4]]
with self.assertRaises(TypeError):
SortedIndex(data)
# values are not sorted
data = [2, 1, 3, 5]
with self.assertRaises(ValueError):
SortedIndex(data)
# values are not sorted
data = [4., 5., 3.7]
with self.assertRaises(ValueError):
SortedIndex(data)
# valid data (unique)
data = [1, 4, 5, 7, 12]
idx = SortedIndex(data)
aeq(data, idx)
eq(np.int, idx.dtype)
eq(1, idx.ndim)
eq(5, len(idx))
assert idx.is_unique
# valid data (non-unique)
data = [1, 4, 5, 5, 7, 12]
idx = SortedIndex(data)
aeq(data, idx)
eq(np.int, idx.dtype)
eq(1, idx.ndim)
eq(6, len(idx))
assert not idx.is_unique
# valid data (typed)
data = [1, 4, 5, 5, 7, 12]
idx = SortedIndex(data, dtype='u4')
aeq(data, idx)
eq(np.uint32, idx.dtype)
# valid data (non-numeric)
data = ['1', '12', '4', '5', '5', '7']
idx = SortedIndex(data)
aeq(data, idx)
def test_constructor(self):
# missing data arg
with self.assertRaises(TypeError):
# noinspection PyArgumentList
FeatureChunkedTable()
# recarray
ra = np.rec.array(feature_table_data, dtype=feature_table_dtype)
ft = FeatureChunkedTable(ra)
eq(6, len(ft))
aeq(ra, ft)
# dict
d = {n: ra[n] for n in feature_table_names}
ft = FeatureChunkedTable(d, names=feature_table_names)
eq(6, len(ft))
aeq(ra, ft)
def test_constructor(self):
# missing data arg
with self.assertRaises(TypeError):
# noinspection PyArgumentList
VariantChunkedTable()
# recarray
ra = np.rec.array(variant_table_data, dtype=variant_table_dtype)
vt = VariantChunkedTable(ra)
eq(5, len(vt))
aeq(ra, vt)
# dict
d = {n: ra[n] for n in variant_table_names}
vt = VariantChunkedTable(d, names=variant_table_names)
eq(5, len(vt))
aeq(ra, vt)
def test_from_hdf5(self):
# setup HDF5 file
node_path = 'test'
tf = tempfile.NamedTemporaryFile(delete=False)
file_path = tf.name
tf.close()
with h5py.File(file_path, mode='w') as h5f:
h5f.create_dataset(node_path,
data=haplotype_data,
chunks=(2, 3))
# file and node path
h = HaplotypeCArray.from_hdf5(file_path, node_path)
aeq(haplotype_data, h)
# dataset
with h5py.File(file_path, mode='r') as h5f:
dataset = h5f[node_path]
h = HaplotypeCArray.from_hdf5(dataset)
aeq(haplotype_data, h)
def test_mean_pairwise_divergence(self):
# simplest case, two haplotypes in each population
h = HaplotypeArray([[0, 0, 0, 0],
[0, 0, 0, 1],
[0, 0, 1, 1],
[0, 1, 1, 1],
[1, 1, 1, 1],
[0, 0, 1, 2],
[0, 1, 1, 2],
[0, 1, -1, -1],
[-1, -1, -1, -1]])
h1 = h.take([0, 1], axis=1)
h2 = h.take([2, 3], axis=1)
ac1 = h1.count_alleles()
ac2 = h2.count_alleles()
expect = [0/4, 2/4, 4/4, 2/4, 0/4, 4/4, 3/4, -1, -1]
actual = allel.stats.mean_pairwise_difference_between(ac1, ac2,
fill=-1)
aeq(expect, actual)
def test_view(self):
# data has wrong dtype
data = 'foo bar'
with self.assertRaises(TypeError):
np.asarray(data).view(SortedIndex)
# data has wrong dimensions
data = [[1, 2], [3, 4]]
with self.assertRaises(TypeError):
np.asarray(data).view(SortedIndex)
# values are not sorted
data = [2, 1, 3, 5]
with self.assertRaises(ValueError):
np.asarray(data).view(SortedIndex)
# values are not sorted
data = [4., 5., 3.7]
with self.assertRaises(ValueError):
np.asarray(data).view(SortedIndex)
# valid data (unique)
data = [1, 4, 5, 7, 12]
idx = np.asarray(data).view(SortedIndex)
aeq(data, idx)
eq(np.int, idx.dtype)
eq(1, idx.ndim)
eq(5, len(idx))
assert idx.is_unique
# valid data (non-unique)
data = [1, 4, 5, 5, 7, 12]
idx = np.asarray(data).view(SortedIndex)
aeq(data, idx)
eq(np.int, idx.dtype)
eq(1, idx.ndim)
eq(6, len(idx))
assert not idx.is_unique
# valid data (typed)
data = np.array([1, 4, 5, 5, 7, 12], dtype='u4')
idx = np.asarray(data).view(SortedIndex)
aeq(data, idx)
eq(np.uint32, idx.dtype)
# valid data (non-numeric)
data = ['1', '12', '4', '5', '5', '7']
idx = np.asarray(data).view(SortedIndex)
aeq(data, idx)
