ChemBO: Bayesian Optimization of Small Organic Molecules with Synthesizable Recommendations

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.

References in corpus (11)

• Practical Bayesian Optimization of Machine Learning Algorithms
• A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning
• Molecular Graph Convolutions: Moving Beyond Fingerprints
• Objective-Reinforced Generative Adversarial Networks (ORGAN) for Sequence Generation Models
• Predictive Entropy Search for Efficient Global Optimization of Black-box Functions
• Conditional molecular design with deep generative models
• Predicting Organic Reaction Outcomes with Weisfeiler-Lehman Network
• Learning Multimodal Graph-to-Graph Translation for Molecular Optimization
• Contextual Decision Processes with Low Bellman Rank are PAC-Learnable
• Constrained Bayesian Optimization for Automatic Chemical Design
• Chemi-net: a graph convolutional network for accurate drug property prediction

Cited by in corpus (15)

• On-the-fly Closed-loop Autonomous Materials Discovery via Bayesian Active Learning
• Optimizing Molecules using Efficient Queries from Property Evaluations
• On-the-fly Autonomous Control of Neutron Diffraction via Physics-Informed Bayesian Active Learning
• Learning To Navigate The Synthetically Accessible Chemical Space Using Reinforcement Learning
• Constrained Bayesian Optimization for Automatic Chemical Design
• Sample-Efficient Optimization in the Latent Space of Deep Generative Models via Weighted Retraining
• Bridging the Gap between Chemical Reaction Pretraining and Conditional Molecule Generation with a Unified Model
• Gaussian Process Molecule Property Prediction with FlowMO
• Differentiable Scaffolding Tree for Molecular Optimization
• Amortized Tree Generation for Bottom-up Synthesis Planning and Synthesizable Molecular Design
• Achieving Robustness to Aleatoric Uncertainty with Heteroscedastic Bayesian Optimisation
• Continuous surrogate-based optimization algorithms are well-suited for expensive discrete problems
• Maximum Reward Formulation In Reinforcement Learning
• Discovering Diverse Athletic Jumping Strategies
• Mixed Variable Bayesian Optimization with Frequency Modulated Kernels