ChemBO: Bayesian Optimization of Small Organic Molecules with Synthesizable Recommendations

TL;DR

ChemBO is a Bayesian optimization framework that efficiently generates synthesizable organic molecules with desired properties, incorporating novel graph-aware kernels for improved molecular similarity assessment.

Contribution

It introduces a sample-efficient Bayesian optimization method with a new optimal-transport based kernel that explicitly considers molecular graph structures.

Findings

ChemBO outperforms existing methods in molecular property optimization

The new kernel improves similarity assessment for molecular graphs

ChemBO produces synthesizable candidate molecules effectively

Abstract

In applications such as molecule design or drug discovery, it is desirable to have an algorithm which recommends new candidate molecules based on the results of past tests. These molecules first need to be synthesized and then tested for objective properties. We describe ChemBO, a Bayesian optimization framework for generating and optimizing organic molecules for desired molecular properties. While most existing data-driven methods for this problem do not account for sample efficiency or fail to enforce realistic constraints on synthesizability, our approach explores the synthesis graph in a sample-efficient way and produces synthesizable candidates. We implement ChemBO as a Gaussian process model and explore existing molecular kernels for it. Moreover, we propose a novel optimal-transport based distance and kernel that accounts for graphical information explicitly. In our experiments, we demonstrate the efficacy of the proposed approach on several molecular optimization problems.