AtomPairFingerprint#

class skfp.fingerprints.AtomPairFingerprint(fp_size: int = 2048, min_distance: int = 1, max_distance: int = 30, use_pharmacophoric_invariants: bool = False, include_chirality: bool = False, use_3D: bool = False, count_simulation: bool = True, scale_by_hac: bool | int = False, count: bool = False, sparse: bool = False, n_jobs: int | None = None, batch_size: int | None = None, verbose: int = 0)#

Atom Pair fingerprint.

The implementation uses RDKit. This is a hashed fingerprint [1], where fragments are computed based on pairs of atoms and distance between them.

Concretely, the hashed fragment is a triplet: (atom type 1, atom type 2, distance)

Atom type takes into consideration:

  • atomic number

  • number of pi electrons

  • degree (number of bonds)

Distance is normally the topological distance, i.e. length of the shortest path in the molecular graph (number of bonds between atoms). Only pairs with distance between min_distance and max_distance (both inclusive) are used.

If use_3D is True, then the Euclidean distance between atoms in a conformation is used. Note that this uses conf_id property of input molecules, and requires them to have this property set.

Values of count version are sensitive to the molecule size, since the number of shortest paths scales with square of heavy atom count (HAC). This can be offset by setting scale_by_hac to True (divide counts by HAC), or integer value greater than 1, which divides by HAC to the given power. Setting scale_by_hac=2 makes valeus independent of molecule size.

Parameters:

  • fp_size (int, default=2048) – Size of output vectors, i.e. number of bits for each fingerprint. Must be positive.

  • min_distance (int, default = 1) – Minimal distance between atoms. Must be positive and less or equal to max_distance.

  • max_distance (int, default = 30) – Maximal distance between atoms. Must be positive and greater or equal to min_distance.

  • use_pharmacophoric_invariants (bool, default=False) – Whether to use pharmacophoric invariants (atom types) instead of default ones. They are the same as in the FCFP fingerprint: Donor, Acceptor, Aromatic, Halogen, Basic, Acidic.

  • include_chirality (bool, default=False) – Whether to include chirality information when computing atom types.

  • use_3D (bool, default=False) – Whether to use 3D Euclidean distance matrix. If False, uses topological distances on molecular graph.

  • count_simulation (bool, default=True) – Whether to use count simulation for approximating feature counts [3].

  • count (bool, default=False) – Whether to return binary (bit) features, or their counts.

  • scale_by_hac (bool or int, default=False) – Whether to scale count fingerprint by the heavy atom count (HAC) to obtain a proportionality to molecule size [2]. If integer value is given, scaling uses given power of HAC, e.g. scale_by_hac=2 divides counts by squared HAC. Using squared HAC results in values range [0, 1].

  • 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, default=0) – Controls the verbosity when computing fingerprints.

n_features_out#

Number of output features, size of fingerprints. Equal to fp_size.

Type:

int

requires_conformers#

Whether the fingerprint is 3D-based and requires molecules with conformers as inputs, with conf_id integer property set. This depends on the use_3D parameter, and has the same value.

Type:

bool

See also

TopologicalTorsionFingerprint

Related fingerprint, but uses 4-atom paths.

SECFPFingerprint

Related fingerprint, which additionally uses circular substructures around each atom like ECFP fingerprint.

References

Examples

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

>>> fp.transform(smiles)
array([[0, 0, 0, ..., 0, 0, 0],
       [0, 0, 0, ..., 0, 0, 0],
       [0, 0, 0, ..., 0, 0, 0],
       [0, 0, 0, ..., 0, 0, 0]], dtype=uint8)

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 output feature names for transformation.

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])

Request metadata passed to the transform method.

transform(X[, copy])

Compute Atom Pair 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)#

Get output feature names for transformation.

The feature names out will prefixed by the lowercased class name. For example, if the transformer outputs 3 features, then the feature names out are: [“class_name0”, “class_name1”, “class_name2”].

Parameters:

input_features (array-like of str or None, default=None) – Only used to validate feature names with the names seen in fit.

Returns:

feature_names_out – Transformed 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 = '$UNCHANGED$') → AtomPairFingerprint#

Request metadata passed to the transform method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see 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.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

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 Atom Pair fingerprints.

Parameters:

  • X ({sequence, array-like} of shape (n_samples,)) – Sequence containing SMILES strings or RDKit Mol objects. If use_3D is True, only Mol objects with computed conformations and with conf_id property are allowed.

  • copy (bool, default=False) – Copy the input X or not.

Returns:

X – Array with fingerprints.

Return type:

{ndarray, sparse matrix} of shape (n_samples, self.fp_size)