A 3D-Shape Similarity-based Contrastive Approach to Molecular Representation Learning

TL;DR

This paper introduces MolCLaSS, a contrastive learning method for graph neural networks that captures 3D molecular shape information indirectly through similarity matching, improving molecular property prediction.

Contribution

It presents a novel contrastive learning framework that encodes 3D shape similarity without directly modeling 3D poses, enhancing molecular representations.

Findings

MolCLaSS effectively captures 3D shape features.

It improves scaffold hopping capabilities.

The method outperforms traditional 2D-based models.

Abstract

Molecular shape and geometry dictate key biophysical recognition processes, yet many graph neural networks disregard 3D information for molecular property prediction. Here, we propose a new contrastive-learning procedure for graph neural networks, Molecular Contrastive Learning from Shape Similarity (MolCLaSS), that implicitly learns a three-dimensional representation. Rather than directly encoding or targeting three-dimensional poses, MolCLaSS matches a similarity objective based on Gaussian overlays to learn a meaningful representation of molecular shape. We demonstrate how this framework naturally captures key aspects of three-dimensionality that two-dimensional representations cannot and provides an inductive framework for scaffold hopping.