def main():
mol_list = parse_sdf_utils.get_sdf_to_mol('/mnt/storage/NIST_zipped/NIST17/replib_mend.sdf')
inchikey_dict = train_test_split_utils.make_inchikey_dict(mol_list)
spectra_for_one_mol = make_spectra_array(inchikey_dict['PDACHFOTOFNHBT-UHFFFAOYSA-N'])
distance_matrix = get_similarities(spectra_for_one_mol)
print('distance for spectra in PDACHFOTOFNHBT-UHFFFAOYSA-N', distance_matrix)
def setUp(self):
test_data_directory = test_utils.test_dir('testdata/')
self.temp_dir = tempfile.mkdtemp(
dir=absltest.get_default_test_tmpdir())
test_sdf_file_large = os.path.join(test_data_directory,
'test_14_mend.sdf')
test_sdf_file_small = os.path.join(test_data_directory,
'test_2_mend.sdf')
max_atoms = ms_constants.MAX_ATOMS
self.mol_list_large = parse_sdf_utils.get_sdf_to_mol(
test_sdf_file_large, max_atoms=max_atoms)
self.mol_list_small = parse_sdf_utils.get_sdf_to_mol(
test_sdf_file_small, max_atoms=max_atoms)
self.inchikey_dict_large = train_test_split_utils.make_inchikey_dict(
self.mol_list_large)
self.inchikey_dict_small = train_test_split_utils.make_inchikey_dict(
self.mol_list_small)
self.inchikey_list_large = self.inchikey_dict_large.keys()
self.inchikey_list_small = self.inchikey_dict_small.keys()
def make_mainlib_replicates_train_test_split(
mainlib_mol_list,
replicates_mol_list,
splitting_type,
mainlib_fractions,
replicates_fractions,
mainlib_maximum_num_molecules_to_use=None,
replicates_maximum_num_molecules_to_use=None,
rseed=42):
"""Makes train/validation/test inchikey lists from two lists of rdkit.Mol.
Args:
mainlib_mol_list : list of molecules from main library
replicates_mol_list : list of molecules from replicates library
splitting_type : type of splitting to use for validation splits.
mainlib_fractions : TrainValTestFractions namedtuple
holding desired fractions for train/val/test split of mainlib
replicates_fractions : TrainValTestFractions namedtuple
holding desired fractions for train/val/test split of replicates.
For the replicates set, the train fraction should be set to 0.
mainlib_maximum_num_molecules_to_use : Largest number of molecules to use
when making datasets from mainlib
replicates_maximum_num_molecules_to_use : Largest number of molecules to use
when making datasets from replicates
rseed : random seed for shuffling
Returns:
main_inchikey_dict : Dict that is keyed by inchikey, containing a list of
rdkit.Mol objects corresponding to that inchikey from the mainlib
replicates_inchikey_dict : Dict that is keyed by inchikey, containing a list
of rdkit.Mol objects corresponding to that inchikey from the replicates
library
main_replicates_split_inchikey_lists_dict : dict with keys :
'mainlib_train', 'mainlib_validation', 'mainlib_test',
'replicates_train', 'replicates_validation', 'replicates_test'
Values are lists of inchikeys corresponding to each dataset.
"""
random.seed(rseed)
main_inchikey_dict = train_test_split_utils.make_inchikey_dict(
mainlib_mol_list)
main_inchikey_list = main_inchikey_dict.keys()
if mainlib_maximum_num_molecules_to_use is not None:
main_inchikey_list = random.sample(
main_inchikey_list, mainlib_maximum_num_molecules_to_use)
replicates_inchikey_dict = train_test_split_utils.make_inchikey_dict(
replicates_mol_list)
replicates_inchikey_list = replicates_inchikey_dict.keys()
if replicates_maximum_num_molecules_to_use is not None:
replicates_inchikey_list = random.sample(
replicates_inchikey_list, replicates_maximum_num_molecules_to_use)
# Make train/val/test splits for main dataset.
main_train_validation_test_inchikeys = (
train_test_split_utils.make_train_val_test_split_inchikey_lists(
main_inchikey_list,
main_inchikey_dict,
mainlib_fractions,
holdout_inchikey_list=replicates_inchikey_list,
splitting_type=splitting_type))
# Make train/val/test splits for replicates dataset.
replicates_validation_test_inchikeys = (
train_test_split_utils.make_train_val_test_split_inchikey_lists(
replicates_inchikey_list,
replicates_inchikey_dict,
replicates_fractions,
splitting_type=splitting_type))
component_inchikey_dict = {
ds_constants.MAINLIB_TRAIN_BASENAME:
main_train_validation_test_inchikeys.train,
ds_constants.MAINLIB_VALIDATION_BASENAME:
main_train_validation_test_inchikeys.validation,
ds_constants.MAINLIB_TEST_BASENAME:
main_train_validation_test_inchikeys.test,
ds_constants.REPLICATES_TRAIN_BASENAME:
replicates_validation_test_inchikeys.train,
ds_constants.REPLICATES_VALIDATION_BASENAME:
replicates_validation_test_inchikeys.validation,
ds_constants.REPLICATES_TEST_BASENAME:
replicates_validation_test_inchikeys.test
}
train_test_split_utils.assert_all_lists_mutally_exclusive(
component_inchikey_dict.values())
# Test that the set of the 5 component inchikey lists is equal to the set of
# inchikeys in the main library.
all_inchikeys_in_components = [
ikey for ikey_list in component_inchikey_dict.values()
for ikey in ikey_list
]
assert set(main_inchikey_list + replicates_inchikey_list) == set(
all_inchikeys_in_components
), ('The inchikeys in the original inchikey dictionary are not all included'
' in the train/val/test component libraries')
return (main_inchikey_dict, replicates_inchikey_dict,
component_inchikey_dict)