A 3D-Shape Similarity-based Contrastive Approach to Molecular Representation Learning
This addresses the limitation of graph neural networks in ignoring 3D information for molecular property prediction, offering a domain-specific advancement in computational chemistry.
The authors tackled the problem of molecular property prediction by proposing MolCLaSS, a contrastive-learning method that uses 3D shape similarity to learn molecular representations, achieving improved performance in scaffold hopping tasks.
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.