bulk_fraggle_similarity

skfp.distances.bulk_fraggle_similarity(X: list[Mol], Y: list[Mol] | None = None, tversky_threshold: float = 0.8) → ndarray

Bulk Fraggle similarity.

Computes the pairwise Fraggle similarity between lists of molecules. If a single list is passed, similarities are computed between its molecules. For two lists, similarities are between their respective molecules, with i-th row and j-th column in output corresponding to i-th molecule from first list and j-th molecule from second list.

See also fraggle_similarity().

Parameters:

• X (ndarray) – First list of molecules, of length m.

• Y (ndarray, default=None) – First list of molecules, of length n. If not passed, similarities are computed between molecules from X.

• tversky_threshold (float, default=0.8) – Required minimal Tversky similarity between a fragment and the reference molecule.

Returns:

similarities – Array with pairwise Fraggle similarity values. Shape is \(m \times n\) if two lists of molecules are passed, or \(m \times m\) otherwise.

Return type:

ndarray

See also

fraggle_similarity()

Fraggle similarity function for two molecules.

Examples

>>> from skfp.distances import bulk_fraggle_similarity
>>> from rdkit.Chem import MolFromSmiles
>>> mols = [MolFromSmiles("COc1ccccc1"), MolFromSmiles("CN1C=NC2=C1C(=O)N(C(=O)N2C)C")]
>>> sim = bulk_fraggle_similarity(mols)
>>> sim
array([[1.        , 0.23275862],
       [0.23275862, 1.        ]])