Predicting the binding of small molecules to proteins through invariant representation of the molecular structure

TL;DR

This paper introduces an invariant molecular representation and a machine learning approach to predict ligand-protein binding, demonstrating improved generalization over existing methods.

Contribution

The authors develop a novel invariant molecular representation and apply a non-deep learning classifier for binding prediction, enhancing generalization capabilities.

Findings

Invariant molecular representations improve prediction accuracy.

Non-deep classifiers outperform existing methods.

Better generalization to unseen molecule-pocket pairs.

Abstract

We present a computational scheme for predicting the ligands that bind to a pocket of known structure. It is based on the generation of a general abstract representation of the molecules, which is invariant to rotations, translations and permutations of atoms, and has some degree of isometry with the space of conformations. We use these representations to train a non-deep machine learning algorithm to classify the binding between pockets and molecule pairs, and show that this approach has a better generalization capability than existing methods.