def test_seeding_is_consistent(self):
hl.set_global_seed(0)
a = [Env.next_seed() for _ in range(10)]
hl.set_global_seed(0)
b = [Env.next_seed() for _ in range(10)]
self.assertEqual(len(set(a)), 10)
self.assertEqual(a, b)
def test_seeding_is_consistent(self):
hl.set_global_seed(0)
a = [Env.next_seed() for _ in range(10)]
hl.set_global_seed(0)
b = [Env.next_seed() for _ in range(10)]
self.assertEqual(len(set(a)), 10)
self.assertEqual(a, b)
def test_king_homo_estimator():
hl.set_global_seed(1)
mt = hl.balding_nichols_model(2, 5, 5)
mt = mt.select_entries(genotype_score=hl.float(mt.GT.n_alt_alleles()))
da = hl.experimental.dnd.array(mt, 'genotype_score', block_size=3)
def sqr(x):
return x * x
score_difference = da.T.inner_product(
da, lambda l, r: sqr(l - r), lambda l, r: l + r, hl.float(0),
hl.agg.sum).checkpoint(new_temp_file())
assert np.array_equal(
score_difference.collect(),
np.array([[0., 6., 4., 2., 4.], [6., 0., 6., 4., 6.],
[4., 6., 0., 6., 0.], [2., 4., 6., 0., 6.],
[4., 6., 0., 6., 0.]]))
import hail as hl hl.set_global_seed(0) mt = hl.balding_nichols_model(n_populations=3, n_variants=(1 << 10), n_samples=4) mt = mt.key_cols_by(s='s' + hl.str(mt.sample_idx)) mt = mt.annotate_entries(GT=hl.or_missing(hl.rand_bool(0.99), mt.GT)) hl.export_plink(mt, 'balding-nichols-1024-variants-4-samples-3-populations', fam_id='f' + mt.s)
h2 = 0.6
pi = 0.01
K = 0.05
n_pops = 3
fst = [0.1] * n_pops
n_vars = int(100e3)
n_cas_list = [5000] * 3
n_list = [2 * x for x in n_cas_list]
n_sim = int(320e3) # over simulate to be able to have ascertainment
else:
raise ValueError(f'sim_name="{sim_name}" does not match any models')
hl.set_global_seed(seed)
gt_sim_suffix = f'bn.npops_{n_pops}.nvars_{n_vars}.nsim_{n_sim}' if sim_name[:
3] == 'bn_' else '' # suffix for genotype simulation (empty string if using ukb data)
param_suffix = f'{gt_sim_suffix}.h2_{h2}.pi_{pi}.K_{K}.seed_{seed}'
betas_path = f'{smiles_wd}/betas.{param_suffix}.tsv.gz'
phens_path = f'{smiles_wd}/phens.{param_suffix}.tsv.gz'
if sim_name[:3] == 'bn_':
mt = hl.balding_nichols_model(n_populations=n_pops,
n_samples=n_sim,
n_variants=n_vars,
fst=fst)
mt = mt.filter_rows(
(hl.abs(hl.agg.mean(mt.GT.n_alt_alleles()) / 2 - 0.5) <
