Multi-objective latent space optimization of generative molecular design models

A N M Nafiz Abeer; Nathan M Urban; M Ryan Weil; Francis J Alexander; Byung-Jun Yoon

doi:10.1016/j.patter.2024.101042

Multi-objective latent space optimization of generative molecular design models

Patterns (N Y). 2024 Aug 12;5(10):101042. doi: 10.1016/j.patter.2024.101042. eCollection 2024 Oct 11.

Authors

A N M Nafiz Abeer¹, Nathan M Urban², M Ryan Weil³, Francis J Alexander⁴, Byung-Jun Yoon^{1

2}

Affiliations

¹ Department of Electrical and Computer Engineering, Texas A&M University, College Station, TX 77843, USA.
² Computational Science Initiative, Brookhaven National Laboratory, Upton, NY 11973, USA.
³ Strategic and Data Science Initiatives, Frederick National Laboratory, Frederick, MD 21702, USA.
⁴ Computing, Environment and Life Sciences, Argonne National Laboratory, Lemont, IL 60439, USA.

Abstract

Molecular design based on generative models, such as variational autoencoders (VAEs), has become increasingly popular in recent years due to its efficiency for exploring high-dimensional molecular space to identify molecules with desired properties. While the efficacy of the initial model strongly depends on the training data, the sampling efficiency of the model for suggesting novel molecules with enhanced properties can be further enhanced via latent space optimization (LSO). In this paper, we propose a multi-objective LSO method that can significantly enhance the performance of generative molecular design (GMD). The proposed method adopts an iterative weighted retraining approach, where the respective weights of the molecules in the training data are determined by their Pareto efficiency. We demonstrate that our multi-objective GMD LSO method can significantly improve the performance of GMD for jointly optimizing multiple molecular properties.

Keywords: GMD; LSO; VAE; drug discovery; generative molecular design; latent space optimization; multi-objective optimization; variational autoencoder.