def get_bond_graph_distance_one_hot(
bond: RDKitBond,
graph_dist_matrix: np.ndarray,
allowable_set: List[int] = DEFAULT_GRAPH_DISTANCE_SET,
include_unknown_set: bool = True) -> List[float]:
"""Get an one-hot feature of graph distance.
Parameters
---------
bond: rdkit.Chem.rdchem.Bond
RDKit bond object
graph_dist_matrix: np.ndarray
The return value of `Chem.GetDistanceMatrix(mol)`. The shape is `(num_atoms, num_atoms)`.
allowable_set: List[int]
The graph distance types to consider. The default set is `[1, 2, ..., 7]`.
include_unknown_set: bool, default False
If true, the index of all types not in `allowable_set` is `len(allowable_set)`.
Returns
-------
List[float]
A one-hot vector of the graph distance.
If `include_unknown_set` is False, the length is `len(allowable_set)`.
If `include_unknown_set` is True, the length is `len(allowable_set) + 1`.
"""
graph_dist = graph_dist_matrix[bond.GetBeginAtomIdx(), bond.GetEndAtomIdx()]
return one_hot_encode(graph_dist, allowable_set, include_unknown_set)
def get_bond_is_in_same_ring_one_hot(bond: RDKitBond) -> List[float]:
"""Get an one-hot feature about whether atoms of a bond is in the same ring or not.
Parameters
---------
bond: rdkit.Chem.rdchem.Bond
RDKit bond object
Returns
-------
List[float]
A one-hot vector of whether a bond is in the same ring or not.
"""
return [int(bond.IsInRing())]
def get_bond_is_conjugated_one_hot(bond: RDKitBond) -> List[float]:
"""Get an one-hot feature about whether a bond is conjugated or not.
Parameters
---------
bond: rdkit.Chem.rdchem.Bond
RDKit bond object
Returns
-------
List[float]
A one-hot vector of whether a bond is conjugated or not.
"""
return [int(bond.GetIsConjugated())]
def get_bond_type_one_hot(bond: RDKitBond,
allowable_set: List[str] = DEFAULT_BOND_TYPE_SET,
include_unknown_set: bool = False) -> List[float]:
"""Get an one-hot feature of bond type.
Parameters
---------
bond: rdkit.Chem.rdchem.Bond
RDKit bond object
allowable_set: List[str]
The bond types to consider. The default set is `["SINGLE", "DOUBLE", "TRIPLE", "AROMATIC"]`.
include_unknown_set: bool, default False
If true, the index of all types not in `allowable_set` is `len(allowable_set)`.
Returns
-------
List[float]
A one-hot vector of the bond type.
If `include_unknown_set` is False, the length is `len(allowable_set)`.
If `include_unknown_set` is True, the length is `len(allowable_set) + 1`.
"""
return one_hot_encode(
str(bond.GetBondType()), allowable_set, include_unknown_set)
def get_bond_stereo_one_hot(bond: RDKitBond,
allowable_set: List[str] = DEFAULT_BOND_STEREO_SET,
include_unknown_set: bool = True) -> List[float]:
"""Get an one-hot feature of the stereo configuration of a bond.
Parameters
---------
bond: rdkit.Chem.rdchem.Bond
RDKit bond object
allowable_set: List[str]
The stereo configuration types to consider.
The default set is `["STEREONONE", "STEREOANY", "STEREOZ", "STEREOE"]`.
include_unknown_set: bool, default True
If true, the index of all types not in `allowable_set` is `len(allowable_set)`.
Returns
-------
List[float]
A one-hot vector of the stereo configuration of a bond.
If `include_unknown_set` is False, the length is `len(allowable_set)`.
If `include_unknown_set` is True, the length is `len(allowable_set) + 1`.
"""
return one_hot_encode(
str(bond.GetStereo()), allowable_set, include_unknown_set)