Molecular Hypergraph Grammar with its Application to Molecular Optimization

TL;DR

This paper introduces MHG-VAE, a novel variational autoencoder that uses molecular hypergraph grammar to ensure 100% valid molecule generation, simplifying training and improving molecular optimization.

Contribution

It develops a molecular hypergraph grammar-guided VAE that guarantees validity with a single model, streamlining molecular generation and optimization processes.

Findings

Achieves 100% validity in molecule generation.

Simplifies training by using a single VAE model.

Provides an algorithm to construct hypergraph grammar from molecules.

Abstract

Molecular optimization aims to discover novel molecules with desirable properties. Two fundamental challenges are: (i) it is not trivial to generate valid molecules in a controllable way due to hard chemical constraints such as the valency conditions, and (ii) it is often costly to evaluate a property of a novel molecule, and therefore, the number of property evaluations is limited. These challenges are to some extent alleviated by a combination of a variational autoencoder (VAE) and Bayesian optimization (BO). VAE converts a molecule into/from its latent continuous vector, and BO optimizes a latent continuous vector (and its corresponding molecule) within a limited number of property evaluations. While the most recent work, for the first time, achieved 100% validity, its architecture is rather complex due to auxiliary neural networks other than VAE, making it difficult to train. This paper presents a molecular hypergraph grammar variational autoencoder (MHG-VAE), which uses a single VAE to achieve 100% validity. Our idea is to develop a graph grammar encoding the hard chemical constraints, called molecular hypergraph grammar (MHG), which guides VAE to always generate valid molecules. We also present an algorithm to construct MHG from a set of molecules.