MOLUCINATE: A Generative Model for Molecules in 3D Space

TL;DR

MOLUCINATE is a new generative model that creates 3D molecular structures with both topology and atom positions, enabling advanced drug design applications like optimizing molecular properties.

Contribution

It introduces a novel architecture that simultaneously generates molecular topology and 3D atom positions, advancing 3D molecule generation methods.

Findings

Successfully optimized molecules for radius of gyration

Demonstrated potential for binding affinity optimization

Outperformed previous 2D-based models in 3D generation

Abstract

Recent advances in machine learning have enabled generative models for both optimization and de novo generation of drug candidates with desired properties. Previous generative models have focused on producing SMILES strings or 2D molecular graphs, while attempts at producing molecules in 3D have focused on reinforcement learning (RL), distance matrices, and pure atom density grids. Here we present MOLUCINATE (MOLecUlar ConvolutIoNal generATive modEl), a novel architecture that simultaneously generates topological and 3D atom position information. We demonstrate the utility of this method by using it to optimize molecules for desired radius of gyration. In the future, this model can be used for more useful optimization such as binding affinity for a protein target.