cmiles package¶

Subpackages¶

Submodules¶

cmiles.generator module¶

Generate canonical identifiers for chemical databases, specifically quantum chemical data.

cmiles.generator.get_inchi_and_key(molecule)[source]¶

Generate inchi and inchikey. Uses RDKit which uses the inchi API

Parameters:	molecule: rdkit.Chem.Mol If an oechem.OEMol is provided, will convert it to an rdkit.Chem.Mol
Returns:	tuple (inchi, inchi_key)

cmiles.generator.get_iupac(molecule)[source]¶

Generate IUPAC name

Parameters:	molecule : oechem.OEMol
Returns:	str: iupac name

Notes

Will only be generated if has openeye license

cmiles.generator.get_molecule_ids(molecule_input, toolkit='openeye', strict=True, **kwargs)[source]¶

Generate a dictionary of canonical identifiers

The molecule_input can be either a JSON serialised molecule (see QCSchema) or an isomeric SMILES with all steroechemistry defined.

Required fields for the QCSchema molecule:

symbols
geometry
connectivity

Parameters:

molecule_input: dict or str

A JSON serialized QC molecule or an isomeric SMILES

toolkit: str, optional, default ‘openeye’

toolkit to use for canonicalization. Currently supports openeye and rdkit

strict: bool, optional. Default True

If true, will raise an exception if SMILES is missing explicit H.

**permute_xyz: bool, optional, default False

Only use if input molecule is in QCSchema format.

If True, the geometry will be permuted to reflect the canononical atom order in the mapped SMILES. get_molecule_ids will return the permuted QCSchema. cmiles identifiers will be in the identifiers field

If False, the map indices in the mapped SMILES will reflect the order of the atoms in the input QCSchema.

Returns:

dict: If permute_xyz=True, will return permuted qcschema with cmiles identifiers in identifiers field.

cmiles.generator.get_unique_protomer(molecule)[source]¶

Generate unique protomer for all tuatomers and charge states of the moelcule.

Requires openeye license

Parameters:	molecule: oechem.OEMol Will convert rdkit.Chem.Mol to oechem.OEMol if openeye is installed and license is valid
Returns:	str unique protomer

cmiles.generator.standardize_tautomer(iso_can_smi)[source]¶

Standardize tautomer to one universal tautomer.

Parameters:	iso_can_smi: str isomeric SMILES
Returns:	str: standardized tautomer

Notes

Does not standardize for ionization states. In some cases preforms better than oequacpac.OEGetUniqueProtomer. See notebook

cmiles.utils module¶

Utility functions for cmiles generator

cmiles.utils.add_explicit_hydrogen(molecule)[source]¶

Add explicit hydrogen to molecule

Parameters:	molecule: oechem.OEMol or rdkit.Chem.Mol
Returns:	molecule oechem.OEMol or rdkit.Chem.Mol with explict hydrogen

cmiles.utils.get_atom_map(molecule, mapped_smiles, **kwargs)[source]¶

Get mapping of map index -> atom index

Parameters:	molecule: oechem.OEMol or rdkit.Chem.Mol mapped_smiles: str explicit hydrogen mapped SMILES
Returns:	atom_map: dict dictionary mapping {map_index: atom_index}

cmiles.utils.get_charge(molecule)[source]¶

Get charge state of molecule

Parameters:	molecule: oechem.OEMol or rdkit.Chem.Mol
Returns:	int total charge of molecule

cmiles.utils.get_connectivity_table(molecule, atom_map)[source]¶

Generate connectivity table

Parameters:	molecule: oechem.Mol or rdkit.Chem.Mol atom_map: dict `{map_idx : atom_idx}`
Returns:	list: list of lists lists of atoms bonded and the bond order [[map_idx_1, map_idx_2, bond_order] …]

cmiles.utils.has_atom_map(molecule)[source]¶

Check if molecule has atom map indices. Will return True even if only one atom has map index

Parameters:	molecule: oechem.Mol or rdkit.Chem.Mol
Returns:	bool True if has one map index. False if molecule has no map indices

cmiles.utils.has_explicit_hydrogen(molecule)[source]¶

Check if molecule has explicit hydrogen.

Parameters:	molecule: oechem.OEMol or rdkit.Chem.Mol
Returns:	bool True if has all explicit H. False otherwise.

cmiles.utils.has_stereo_defined(molecule)[source]¶

Checks if molecule has all stereo defined.

Parameters:	molecule: oechem.OEMol or rdkit.Chem.Mol
Returns:	bool True if all stereo defined, False otherwise

Notes

This does not check if all chirality or bond stereo are consistent. The best way to check is to try to generate a 3D conformer. If stereo information is inconsistent, this will fail.

cmiles.utils.invert_atom_map(atom_map)[source]¶

Invert atom map {map_idx:atom_idx} –> {atom_idx:map_idx}

Parameters:	atom_map: dict {map_idx:atom_idx}
Returns:	dict {atom_idx:map_idx}

cmiles.utils.is_map_canonical(molecule)[source]¶

Check if map indices on molecule is in caononical order

Parameters:	molecule: oechem.Mol or rdkit.Chem.Mol
Returns:	bool

cmiles.utils.is_missing_atom_map(molecule)[source]¶

Check if any atom in molecule is missing atom map index

Parameters:	molecule: oechem.Mol or rdkit.Chem.Mol
Returns:	bool True if even if only one atom map is missing. False if all atoms have atom maps.

cmiles.utils.load_molecule(inp_molecule, toolkit='openeye', **kwargs)[source]¶

Load molecule.

Input is restrictive. Allowed inputs are:

Isomeric SMILES
JSON serialized molecule

Parameters:	inp_molecule: str or dict isomeric SMILES or QCSChema toolkit: str, optional, default openeye. cheminformatics toolkit to use
Returns:	molecule: oechem.OEMOl or rdkit.Chem.Mol

cmiles.utils.mol_from_json(inp_molecule, toolkit='openeye', **kwargs)[source]¶

Load a molecule from QCSchema

see QCSchema

Required fields for the QCSchema molecule:

symbols
geometry
connectivity

Parameters:

inp_molecule: dict: QCSchema molecule with symbols, geometry and connectivity
toolkit: str, optional. Default openeye: cheminformatics toolkit to use. Currently supports openeye and rdkit
**permute_xyz: bool, optional, default False: If False, will add flag to molecule such that the mapped SMILES retains the order of serialized geometry. If True, mapped SMILES will be in canonical order and serialized geometry will have to be reordered.

Returns:

molecule: oechem.OEMol or rdkit.Chem.Mol

cmiles.utils.mol_to_hill_molecular_formula(molecule)[source]¶

Generate Hill sorted empirical formula.

Hill sorted first lists C and H and then all other symbols in alphabetical order

Parameters:	molecule: `oechem.OEMol` or `rdkit.Chem.Mol`
Returns:	str hill sorted empirical formula

cmiles.utils.mol_to_map_ordered_qcschema(molecule, molecule_ids, multiplicity=1, **kwargs)[source]¶

Genereate JSON serialize following QCSchema specs

Geometry, symbols and connectivity table ordered according to map indices in mapped SMILES

Parameters:	molecule: oechem.OEMol or rdkit.Chem.Mol molecuel must have a conformer. molecule_ids: dict cmiles generated molecular ids. multiplicity: int, optional, defualt 1 multiplicity of molecule
Returns:	dict JSON serialized molecule following QCSchema specs

cmiles.utils.mol_to_smiles(molecule, **kwargs)[source]¶

Generate canonical smiles from molecule

Parameters:	molecule: oechem.OEMol or rdkit.Chem.Mol isomeric: bool, optional, default True If False, SMILES will not include stereo information explicit_hydrogen: bool, optional, default True If True, SMILES will have explicit hydrogen. **mapped: bool, optional, default True If True, SMILES will have map indices Example: O=O will be `[O:1]=[O:2]`
Returns:	str SMILES

cmiles.utils.permute_qcschema(json_mol, molecule_ids, **kwargs)[source]¶

permute geometry and symbols to correspond to map indices on mapped SMILES

Parameters:	json_mol: dict JSON serialized molecule. Required fields: symbols, geometry, connectivity and multiplicity molecule_ids: dict cmiles generated molecular ids
Returns:	dict JSON serialized molecule. symbols, geometry, and connectivity ordered according to map indices on mapped SMILES. Also includes identifiers field with cmiles generated identifiers.

cmiles.utils.remove_atom_map(molecule, keep_map_data=True)[source]¶

Remove atom map from molecule

Parameters:	molecule oechem.OEMol or rdkit.Chem.Mol keep_map_data: bool, optional, default True If True, will save map indices in atom data

cmiles.utils.restore_atom_map(molecule)[source]¶

Restore atom map from atom data in place

Parameters:	molecule oechem.OEMol or rdkit.Chem.Mol

Module contents¶

Generate canonical identifiers for chemical databases, specifically quantum chemical data.