Latent Molecular Optimization for Targeted Therapeutic Design

TL;DR

This paper presents a novel method for designing targeted therapeutics by encoding protein sites and chemicals into a continuous space, enabling gradient-based optimization to generate molecules with high binding affinity and favorable pharmacological properties.

Contribution

The authors introduce a graph convolutional network-based site signature and a continuous latent space for molecules, allowing efficient multi-objective optimization in drug discovery.

Findings

Successfully optimized molecules with high predicted binding affinity

Generated molecules validated by docking simulations

Demonstrated efficient gradient-based optimization in molecular design

Abstract

We devise an approach for targeted molecular design, a problem of interest in computational drug discovery: given a target protein site, we wish to generate a chemical with both high binding affinity to the target and satisfactory pharmacological properties. This problem is made difficult by the enormity and discreteness of the space of potential therapeutics, as well as the graph-structured nature of biomolecular surface sites. Using a dataset of protein-ligand complexes, we surmount these issues by extracting a signature of the target site with a graph convolutional network and by encoding the discrete chemical into a continuous latent vector space. The latter embedding permits gradient-based optimization in molecular space, which we perform using learned differentiable models of binding affinity and other pharmacological properties. We show that our approach is able to efficiently optimize these multiple objectives and discover new molecules with potentially useful binding properties, validated via docking methods.