NAGLRDKitToolkitWrapper

class openff.nagl.toolkits.NAGLRDKitToolkitWrapper[source]

Bases: NAGLToolkitWrapperBase, RDKitToolkitWrapper

Methods

assign_partial_charges

Compute partial charges with RDKit, and assign the new values to the partial_charges attribute.

calculate_circular_fingerprint_similarity

Compute the similarity between two molecules using a fingerprinting method.

get_atoms_are_in_ring_size

Determine whether each atom in a molecule is in a ring of a given size.

get_best_rmsd

Compute the lowest all-atom RMSD between a reference and target conformer, allowing for symmetry-equivalent atoms to be permuted.

get_bonds_are_in_ring_size

Determine whether each bond in a molecule is in a ring of a given size.

get_molecule_hybridizations

Get the hybridization of each atom in a molecule.

mapped_smiles_to_smiles

Convert a mapped SMILES string to a SMILES string.

smiles_to_mapped_smiles

Convert a SMILES string to a mapped SMILES string.

stream_molecules_from_sdf_file

Stream molecules from an SDF file.

stream_molecules_to_file

Stream molecules to an SDF file using a context manager.

to_rdkit

Create an RDKit molecule Requires the RDKit to be installed.

Attributes

name

assign_partial_charges(molecule, partial_charge_method=None, use_conformers=None, strict_n_conformers=False, normalize_partial_charges=True, _cls=None)[source]

Compute partial charges with RDKit, and assign the new values to the partial_charges attribute.

Warning

This API is experimental and subject to change.

Parameters:

  • molecule (openff.toolkit.topology.Molecule) – Molecule for which partial charges are to be computed

  • partial_charge_method (str, optional, default None) –

    The charge model to use. One of [‘mmff94’, ‘gasteiger’]. If None, ‘mmff94’ will be used.

    • ’mmff94’: Applies partial charges using the Merck Molecular Force Field

      (MMFF). This method does not make use of conformers, and hence use_conformers and strict_n_conformers will not impact the partial charges produced.

  • use_conformers (iterable of unit-wrapped numpy arrays, each with) – shape (n_atoms, 3) and dimension of distance. Optional, default = None Coordinates to use for partial charge calculation. If None, an appropriate number of conformers will be generated.

  • strict_n_conformers (bool, default False) – Whether to raise an exception if an invalid number of conformers is provided for the given charge method. If this is False and an invalid number of conformers is found, a warning will be raised.

  • normalize_partial_charges (bool, default True) – Whether to offset partial charges so that they sum to the total formal charge of the molecule. This is used to prevent accumulation of rounding errors when the partial charge generation method has low precision.

  • _cls (class) – Molecule constructor

Raises:

  • ChargeMethodUnavailableError – if the requested charge method can not be handled by this toolkit

  • ChargeCalculationError – if the charge method is supported by this toolkit, but fails

calculate_circular_fingerprint_similarity(molecule: Molecule, reference_molecule: Molecule, radius: int = 3, nbits: int = 2048) → float[source]

Compute the similarity between two molecules using a fingerprinting method. Uses a Morgan fingerprint with RDKit and a Circular fingerprint with OpenEye.

Parameters:

  • molecule (openff.toolkit.topology.Molecule) – The molecule to compute the fingerprint for.

  • reference_molecule (openff.toolkit.topology.Molecule) – The molecule to compute the fingerprint for.

  • radius (int, default 3) – The radius of the fingerprint to use.

  • nbits (int, default 2048) – The length of the fingerprint to use. Not used in RDKit.

Returns:

similarity (float) – The Dice similarity between the two molecules.

get_atoms_are_in_ring_size(molecule: Molecule, ring_size: int) → List[bool][source]

Determine whether each atom in a molecule is in a ring of a given size.

Parameters:
Returns:

in_ring_size (List[bool])

get_best_rmsd(molecule: Molecule, reference_conformer: ndarray | Quantity, target_conformer: ndarray | Quantity) → Quantity[source]

Compute the lowest all-atom RMSD between a reference and target conformer, allowing for symmetry-equivalent atoms to be permuted.

Parameters:

  • molecule (openff.toolkit.topology.Molecule) – The molecule to compute the RMSD for

  • reference_conformer (np.ndarray or openff.units.unit.Quantity) – The reference conformer to compare to the target conformer. If a numpy array, it is assumed to be in units of angstrom.

  • target_conformer (np.ndarray or openff.units.unit.Quantity) – The target conformer to compare to the reference conformer. If a numpy array, it is assumed to be in units of angstrom.

Returns:

rmsd (unit.Quantity)

Examples

>>> from openff.units import unit
>>> from openff.toolkit.topology import Molecule
>>> from openff.toolkit.utils.toolkits import RDKitToolkitWrapper
>>> toolkit_wrapper = RDKitToolkitWrapper()
>>> molecule = Molecule.from_smiles("CCCCO")
>>> molecule.generate_conformers(n_conformers=2)
>>> rmsd = toolkit_wrapper.get_best_rmsd(molecule, molecule.conformers[0], molecule.conformers[1])
>>> print(f"RMSD in angstrom: {rmsd.m_as(unit.angstrom)}")
get_bonds_are_in_ring_size(molecule: Molecule, ring_size: int) → List[bool][source]

Determine whether each bond in a molecule is in a ring of a given size.

Parameters:
Returns:

in_ring_size (List[bool]) – Bonds are in the same order as the molecule’s bonds attribute.

get_molecule_hybridizations(molecule: Molecule) → List[HybridizationType][source]

Get the hybridization of each atom in a molecule.

Parameters:

molecule (openff.toolkit.topology.Molecule) – The molecule to get the hybridizations of.

Returns:

hybridizations (List[HybridizationType]) – The hybridization of each atom in the molecule.

classmethod mapped_smiles_to_smiles(mapped_smiles: str) → str[source]

Convert a mapped SMILES string to a SMILES string.

Parameters:

mapped_smiles (str) – The mapped SMILES string to convert.

Returns:

smiles (str) – The SMILES string.

name = 'rdkit'
classmethod smiles_to_mapped_smiles(smiles: str) → str[source]

Convert a SMILES string to a mapped SMILES string.

Parameters:

smiles (str) – The SMILES string to convert.

Returns:

mapped_smiles (str) – The mapped SMILES string.

classmethod stream_molecules_from_sdf_file(file: str, as_smiles: bool = False, mapped_smiles: bool = False, explicit_hydrogens: bool = True, **kwargs)[source]

Stream molecules from an SDF file.

Parameters:

  • file (str) – The path to the SDF file to stream molecules from.

  • as_smiles (bool, default False) – If True, return a SMILES string instead of an OpenFF Molecule

  • mapped_smiles (bool, default False) – If True, return a SMILES string with atom indices as atom map numbers.

  • explicit_hydrogens (bool, default True) – If True, keep explicit hydrogens in SMILES outputs.

Returns:

molecules (Generator[openff.toolkit.topology.Molecule or str])

stream_molecules_to_file(file: str)[source]

Stream molecules to an SDF file using a context manager.

Parameters:

file (str) – The path to the SDF file to stream molecules to.

Examples

>>> from openff.toolkit.topology import Molecule
>>> from openff.toolkit.utils.toolkits import RDKitToolkitWrapper
>>> toolkit_wrapper = RDKitToolkitWrapper()
>>> molecule1 = Molecule.from_smiles("CCO")
>>> molecule2 = Molecule.from_smiles("CCC")
>>> with toolkit_wrapper.stream_molecules_to_file("molecules.sdf") as writer:
...     writer(molecule1)
...     writer(molecule2)
to_rdkit(molecule: Molecule)[source]

Create an RDKit molecule Requires the RDKit to be installed.

Warning

This API is experimental and subject to change.

Parameters:

aromaticity_model (str, optional, default "OEAroModel_MDL") – The aromaticity model to use. Only OEAroModel_MDL is supported.

Returns:

rdmol (rkit.RDMol) – An RDKit molecule

Examples

Convert a molecule to RDKit >>> from openff.toolkit import Molecule >>> ethanol = Molecule.from_smiles(‘CCO’) >>> rdmol = ethanol.to_rdkit()