Computer-Aided Multi-Objective Optimization in Small Molecule Discovery
arXiv:2210.07209 · doi:10.1016/j.patter.2023.100678
Abstract
Molecular discovery is a multi-objective optimization problem that requires identifying a molecule or set of molecules that balance multiple, often competing, properties. Multi-objective molecular design is commonly addressed by combining properties of interest into a single objective function using scalarization, which imposes assumptions about relative importance and uncovers little about the trade-offs between objectives. In contrast to scalarization, Pareto optimization does not require knowledge of relative importance and reveals the trade-offs between objectives. However, it introduces additional considerations in algorithm design. In this review, we describe pool-based and de novo generative approaches to multi-objective molecular discovery with a focus on Pareto optimization algorithms. We show how pool-based molecular discovery is a relatively direct extension of multi-objective Bayesian optimization and how the plethora of different generative models extend from single-objective to multi-objective optimization in similar ways using non-dominated sorting in the reward function (reinforcement learning) or to select molecules for retraining (distribution learning) or propagation (genetic algorithms). Finally, we discuss some remaining challenges and opportunities in the field, emphasizing the opportunity to adopt Bayesian optimization techniques into multi-objective de novo design.
References in corpus (19)
- Automatic chemical design using a data-driven continuous representation of molecules
- Deep Reinforcement Learning for De-Novo Drug Design
- Less is more: sampling chemical space with active learning
- Optimization of Molecules via Deep Reinforcement Learning
- Resolving transition metal chemical space: feature selection for machine learning and structure-property relationships
- BatchBALD: Efficient and Diverse Batch Acquisition for Deep Bayesian Active Learning
- Conditional molecular design with deep generative models
- Keeping it Simple: Language Models can learn Complex Molecular Distributions
- High-Dimensional Materials and Process Optimization using Data-driven Experimental Design with Well-Calibrated Uncertainty Estimates
- Deep Learning and Knowledge-Based Methods for Computer Aided Molecular Design -- Toward a Unified Approach: State-of-the-Art and Future Directions
- Optimizing Molecules using Efficient Queries from Property Evaluations
- Molecular Design in Synthetically Accessible Chemical Space via Deep Reinforcement Learning
- Sample Efficiency Matters: A Benchmark for Practical Molecular Optimization
- New Paradigms for Exploiting Parallel Experiments in Bayesian Optimization
- A Pareto Algorithm for Efficient De Novo Design of Multi-Functional Molecule
- FastFlows: Flow-Based Models for Molecular Graph Generation
- Retrieval-based Controllable Molecule Generation
- De novo design of protein target specific scaffold-based Inhibitors via Reinforcement Learning
- HelixMO: Sample-Efficient Molecular Optimization in Scene-Sensitive Latent Space