def test_simulation_shuffling_perturbation(file_regression, tmpdir):
"""Test simulation with clustering hierarchy"""
func_name = sys._getframe().f_code.co_name
output_dir = "%s/output_dir_%s" % (tmpdir, func_name)
pvalues_filename = "%s/%s" % (output_dir, constants.PVALUES_FILENAME)
cmd = ("python -m mihifepe.simulation -seed 3 -num_instances 100 -num_features 10 -fraction_relevant_features 0.5"
" -contiguous_node_names -hierarchy_type random -perturbation shuffling -num_shuffling_trials 10 -output_dir %s" % output_dir)
pass_args = cmd.split()[2:]
with patch.object(sys, 'argv', pass_args):
simulation.main()
with open(pvalues_filename, "r") as pvalues_file:
pvalues = sorted(pvalues_file.readlines())
file_regression.check("\n".join(pvalues), extension="_pvalues.csv")
fdr_filename = "%s/%s/%s.csv" % (output_dir, constants.HIERARCHICAL_FDR_DIR, constants.HIERARCHICAL_FDR_OUTPUTS)
with open(fdr_filename, "r") as fdr_file:
fdr = sorted(fdr_file.readlines())
file_regression.check("\n".join(fdr), extension="_fdr.json")
def test_simulation_shuffling_interactions(file_regression, tmpdir):
"""Test simulation with interactions and shuffling perturbations"""
func_name = sys._getframe().f_code.co_name
output_dir = "%s/output_dir_%s" % (tmpdir, func_name)
pvalues_filename = "%s/%s" % (output_dir, constants.INTERACTIONS_PVALUES_FILENAME)
cmd = ("python -m mihifepe.simulation -seed 5 -num_instances 1000 -num_features 10 -fraction_relevant_features 0.5"
" -analyze_interactions -hierarchy_type random -perturbation shuffling -num_shuffling_trials 200"
" -noise_multiplier 0.0 -noise_type additive_gaussian -num_interactions 3 -output_dir %s" % output_dir)
pass_args = cmd.split()[2:]
with patch.object(sys, 'argv', pass_args):
simulation.main()
with open(pvalues_filename, "r") as pvalues_file:
pvalues = sorted(pvalues_file.readlines())
file_regression.check("\n".join(pvalues), extension="_pvalues.csv")
fdr_filename = "%s/%s/%s.csv" % (output_dir, constants.INTERACTIONS_FDR_DIR, constants.HIERARCHICAL_FDR_OUTPUTS)
with open(fdr_filename, "r") as fdr_file:
fdr = sorted(fdr_file.readlines())
file_regression.check("\n".join(fdr), extension="_fdr.json")