Uncertainty quantification with graph neural networks for efficient molecular design
Lung-Yi Chen, Yi‐Pei Li
Abstract
Optimizing molecular design across expansive chemical spaces presents unique challenges, especially in maintaining predictive accuracy under domain shifts. This study integrates uncertainty quantification (UQ), directed message passing neural networks (D-MPNNs), and genetic algorithms (GAs) to address these challenges. We systematically evaluate whether UQ-enhanced D-MPNNs can effectively optimize broad, open-ended chemical spaces and identify the most effective implementation strategies. Using benchmarks from the Tartarus and GuacaMol platforms, our results show that UQ integration via probabilistic improvement optimization (PIO) enhances optimization success in most cases, supporting more reliable exploration of chemically diverse regions. In multi-objective tasks, PIO proves especially advantageous, balancing competing objectives and outperforming uncertainty-agnostic approaches. This work provides practical guidelines for integrating UQ in computational-aided molecular design (CAMD). Optimizing molecular design across chemical spaces is challenging. Here, authors integrate uncertainty quantification with graph neural networks and genetic algorithms, demonstrating that probabilistic improvement optimization enhances success rates in molecular discovery.