def eg_extract_one_bond_break(molecule_file, reaction_file="reactions.pkl"):
molecules = pickle_load(molecule_file)
print("number of moles:", len(molecules))
extractor = ReactionExtractorFromMolSet(molecules)
extractor.extract_one_bond_break(find_one=False)
extractor.to_file(reaction_file)
def write_dataset_edge_label():
filename = "~/Applications/db_access/mol_builder/molecules_n200.pkl"
mols = pickle_load(filename)
struct_file = "~/Applications/db_access/mol_builder/struct_mols_bond_annotation.sdf"
label_file = "~/Applications/db_access/mol_builder/label_mols_bond_annotation.yaml"
feature_file = (
"~/Applications/db_access/mol_builder/feature_mols_bond_annotation.yaml"
)
write_edge_label_based_on_bond(mols, struct_file, label_file, feature_file)
def eg_buckets(molecule_file):
molecules = pickle_load(molecule_file)
print("number of moles:", len(molecules))
extractor = ReactionExtractorFromMolSet(molecules)
buckets = extractor.bucket_molecules(
keys=["formula", "charge", "spin_multiplicity"])
pprint(buckets)
buckets = extractor.bucket_molecules(keys=["formula"])
pprint(buckets)
def get_single_atom_energy():
filename = "~/Applications/db_access/mol_builder/molecules_unfiltered.pkl"
# filename = "~/Applications/db_access/mol_builder/molecules.pkl"
# filename = "~/Applications/db_access/mol_builder/molecules_n200.pkl"
mols = pickle_load(filename)
formula = ["H1", "Li1", "C1", "O1", "F1", "P1"]
print("# formula free energy charge")
for m in mols:
if m.formula in formula:
print(m.formula, m.free_energy, m.charge)
def number_of_bonds():
filename = "~/Applications/db_access/mol_builder/molecules.pkl"
mols = pickle_load(filename)
nbonds = []
for m in mols:
nbonds.append(len(m.bonds))
mean = np.mean(nbonds)
median = np.median(nbonds)
print("### number of bonds mean:", mean)
print("### number of bonds median:", median)
def write_dataset():
# filename = "~/Applications/db_access/mol_builder/molecules.pkl"
filename = "~/Applications/db_access/mol_builder/molecules_n200.pkl"
mols = pickle_load(filename)
# mols = mols[len(mols) * 739 // 2048 : len(mols) * 740 // 2048]
# #######################
# # filter charge 0 mols
# #######################
# new_mols = []
# for m in mols:
# if m.charge == 1:
# new_mols.append(m)
# mols = new_mols
struct_file = "~/Applications/db_access/mol_builder/struct_mols_n200.sdf"
label_file = "~/Applications/db_access/mol_builder/label_mols_n200.csv"
feature_file = "~/Applications/db_access/mol_builder/feature_mols_n200.yaml"
write_sdf_csv_dataset(mols, struct_file, label_file, feature_file)
def from_file(cls, filename):
molecules = pickle_load(to_path(filename))
logger.info(f"{len(molecules)} molecules loaded from file: {filename}")
return cls(molecules)