WHIMFingerprint

class skfp.fingerprints.WHIMFingerprint(clip_val: float = 2147483647, sparse: bool = False, n_jobs: int | None = None, batch_size: int | None = None, verbose: int | dict = 0)

WHIM (Weighted Holistic Invariant Molecular descriptors) fingerprint.

The implementation uses RDKit. This is a descriptor-based fingerprint, where bits measure rotation-invariant 3D information regarding size, shape, symmetry and atom distributions.

Features are based on the principal component analysis (PCA) on the centered cartesian coordinates of a molecule by using a weighted covariance matrix. There are two groups of features, each one measuring size, shape, symmetry and density of atoms: - 11 directional, using scores of individual principal axes - 7 global, aggregating information about the whole molecule

Additionally, all directional features and 5 of the global ones are computed using unweighted distances matrix, as well as 6 weighted variants, using atomic features: atomic mass, van der Waals volume, electronegativity, polarizability, ion polarity, and IState [1] [2]. They are relative to the carbon, e.g. molecular weight is: MW(atom_type) / MW(carbon).

This gives 114 features: 11 * 7 weighted directional + 5 * 7 weighted global + 2 unweighted global. See [3] [4] [5] [6] for details.

Typical correct values should be small, but can result in NaN or infinity for some molecules. Value clipping with clip_val parameter, feature selection, and/or imputation should be used.

Parameters:

• clip_val (float or None, default=2147483647) – Value to clip results at, both positive and negative ones.The default value is the maximal value of 32-bit integer, but should often be set lower, depending on the application. None means that no clipping is applied.

• sparse (bool, default=False) – Whether to return dense NumPy array, or sparse SciPy CSR array.

• n_jobs (int, default=None) – The number of jobs to run in parallel. transform() is parallelized over the input molecules. None means 1 unless in a joblib.parallel_backend context. -1 means using all processors. See scikit-learn documentation on n_jobs for more details.

• batch_size (int, default=None) – Number of inputs processed in each batch. None divides input data into equal-sized parts, as many as n_jobs.

• verbose (int or dict, default=0) – Controls the verbosity when computing fingerprints. If a dictionary is passed, it is treated as kwargs for tqdm(), and can be used to control the progress bar.

n_features_out

Number of output features, size of fingerprints.

Type:

int = 114

requires_conformers

Value is always True, as this fingerprint is 3D based. It always requires molecules with conformers as inputs, with conf_id integer property set.

Type:

bool = True

References

[1]

RDKit IState discussion rdkit/rdkit#6122

[2]

RDKit IState implementation rdkit/rdkit

[3]

Rajarshi Guha “Weighted Holistic Invariant Molecular Descriptors”

[4]

Roberto Todeschini and Viviana Consonni “Molecular Descriptors for Chemoinformatics”

[5]

Guillaume Godin “3D descriptors in RDKit” UGM 2017

[6]

Roberto Todeschini and Paola Gramatica “The WHIM Theory: New 3D Molecular Descriptors for QSAR in Environmental Modelling” SAR and QSAR in Environmental Research, 7(1-4), 89-115

Examples

>>> from skfp.fingerprints import WHIMFingerprint
>>> from skfp.preprocessing import MolFromSmilesTransformer, ConformerGenerator
>>> smiles = ["O", "CC", "[C-]#N", "CC=O"]
>>> fp = WHIMFingerprint()
>>> fp
WHIMFingerprint()

>>> mol_from_smiles = MolFromSmilesTransformer()
>>> mols = mol_from_smiles.transform(smiles)
>>> conf_gen = ConformerGenerator()
>>> mols = conf_gen.transform(mols)
>>> fp.transform(mols)  
array([[0.44 , 0.067, 0.   , ..., 0.514, 0.537, 0.537],
       [1.17 , 0.395, 0.393, ..., 2.266, 3.38 , 2.542],
       [0.329, 0.   , 0.   , ..., 0.329, 0.329, 0.329],
       [1.196, 0.507, 0.242, ..., 2.183, 3.285, 3.129]])

Methods

fit(X[, y])	Unused, kept for scikit-learn compatibility.
fit_transform(X[, y])	The same as .transform() method, kept for scikit-learn compatibility.
get_feature_names_out([input_features])	Get fingerprint output feature names.
get_metadata_routing()	Get metadata routing of this object.
get_params([deep])	Get parameters for this estimator.
set_output(*[, transform])	Set output container.
set_params(**params)	Set the parameters of this estimator.
set_transform_request(*[, copy])	Configure whether metadata should be requested to be passed to the transform method.
transform(X[, copy])	Compute WHIM fingerprints.

fit(X: Sequence[str | Mol], y: Any | None = None, **fit_params)

Unused, kept for scikit-learn compatibility.

Parameters:

• X (any) – Unused, kept for scikit-learn compatibility.

• y (any) – Unused, kept for scikit-learn compatibility.

• **fit_params (dict) – Unused, kept for scikit-learn compatibility.

Return type:

self

fit_transform(X: Sequence[str | Mol], y: Any | None = None, **fit_params)

The same as .transform() method, kept for scikit-learn compatibility.

Parameters:

• X (any) – See .transform() method.

• y (any) – See .transform() method.

• **fit_params (dict) – Unused, kept for scikit-learn compatibility.

Returns:

X_new – See .transform() method.

Return type:

any

get_feature_names_out(input_features=None) → ndarray

Get fingerprint output feature names. They correspond to various descriptors derived from weighted covariance matrix. See references given in main body for explanations.

Parameters:

input_features (array-like of str or None, default=None) – Unused, kept for scikit-learn compatibility.

Returns:

feature_names_out – WHIM feature names.

Return type:

ndarray of str objects

get_metadata_routing()

Get metadata routing of this object.

Please check User Guide on how the routing mechanism works.

Returns:

routing – A MetadataRequest encapsulating routing information.

Return type:

MetadataRequest

get_params(deep=True)

Get parameters for this estimator.

Parameters:

deep (bool, default=True) – If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns:

params – Parameter names mapped to their values.

Return type:

dict

set_output(*, transform=None)

Set output container.

See Introducing the set_output API for an example on how to use the API.

Parameters:

transform ({"default", "pandas", "polars"}, default=None) –

Configure output of transform and fit_transform.

• ”default”: Default output format of a transformer

• ”pandas”: DataFrame output

• ”polars”: Polars output

• None: Transform configuration is unchanged

Added in version 1.4: “polars” option was added.

Returns:

self – Estimator instance.

Return type:

estimator instance

set_params(**params)

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters:

**params (dict) – Estimator parameters.

Returns:

self – Estimator instance.

Return type:

estimator instance

set_transform_request(*, copy: bool | None | str = '$UNCHANGED$') → WHIMFingerprint

Configure whether metadata should be requested to be passed to the transform method.

Note that this method is only relevant when this estimator is used as a sub-estimator within a meta-estimator and metadata routing is enabled with enable_metadata_routing=True (see sklearn.set_config()). Please check the User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to transform if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to transform.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Parameters:

copy (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for copy parameter in transform.

Returns:

self – The updated object.

Return type:

object

transform(X: Sequence[str | Mol], copy: bool = False) → ndarray | csr_array

Compute WHIM fingerprints.

Parameters:

• X ({sequence of str or Mol}) – Sequence containing RDKit Mol objects, with conformers generated and conf_id integer property set.

• copy (bool, default=False) – Whether to copy input data.

Returns:

X – Transformed data.

Return type:

{ndarray, sparse matrix} of shape (n_samples, 114)