gt4sd.algorithms.prediction.paccmann.implementation module

Implementation of the zero-shot classifier.

Summary

Classes:

BimodalMCAAffinityPredictor	Bimodal MCA (Multiscale Convolutional Attention) affinity prediction model.
MCAPredictor	Base implementation of an MCAPredictor.

Reference

class MCAPredictor

Bases: object

Base implementation of an MCAPredictor.

predict()

Get prediction.

Return type

Any

Returns

predicted affinity

predict_values()

Get prediction for algorithm sample method.

Return type

Any

Returns

predicted values as list.

__dict__ = mappingproxy({'__module__': 'gt4sd.algorithms.prediction.paccmann.implementation', '__doc__': 'Base implementation of an MCAPredictor.', 'predict': <function MCAPredictor.predict>, 'predict_values': <function MCAPredictor.predict_values>, '__dict__': <attribute '__dict__' of 'MCAPredictor' objects>, '__weakref__': <attribute '__weakref__' of 'MCAPredictor' objects>, '__annotations__': {}})

__doc__ = 'Base implementation of an MCAPredictor.'

__module__ = 'gt4sd.algorithms.prediction.paccmann.implementation'

__weakref__

list of weak references to the object (if defined)

class BimodalMCAAffinityPredictor(resources_path, protein_targets, ligands, confidence, device=None)

Bases: MCAPredictor

Bimodal MCA (Multiscale Convolutional Attention) affinity prediction model.

For details see: https://pubs.acs.org/doi/10.1021/acs.molpharmaceut.9b00520 and https://iopscience.iop.org/article/10.1088/2632-2153/abe808.

__init__(resources_path, protein_targets, ligands, confidence, device=None)

Initialize BimodalMCAAffinityPredictor.

Parameters

• resources_path (str) – path where to load model weights and cofiguration.

• protein_targets (List[str]) – list of protein targets as AA sequences.

• ligands (List[str]) – list of ligands in SMILES format.

• confidence (bool) – whether the confidence for the prediction should be returned.

• device (Union[device, str, None]) – device where the inference is running either as a dedicated class or a string. If not provided is inferred.

predict()

Get predicted affinity.

Return type

Any

Returns

predicted affinity.

predict_values()

Get prediction for algorithm sample method.

Return type

List[float]

Returns

predicted values as list.

__annotations__ = {}

__doc__ = 'Bimodal MCA (Multiscale Convolutional Attention) affinity prediction model.\n\n    For details see: https://pubs.acs.org/doi/10.1021/acs.molpharmaceut.9b00520\n    and https://iopscience.iop.org/article/10.1088/2632-2153/abe808.\n    '

__module__ = 'gt4sd.algorithms.prediction.paccmann.implementation'