Electron flow matching for generative reaction mechanism prediction obeying conservation laws

2025-02-18

AI Summary

This work addresses the prevalent neglect of mass conservation in data-driven reaction prediction, which leads to physically inconsistent predictions and hallucinated mechanisms. We propose the first generative reaction mechanism prediction framework that strictly enforces mass conservation. Methodologically, we model chemical reactions as electron redistribution processes and develop a constrained generative model grounded in flow matching and graph neural networks, explicitly encoding atomic conservation and electron flow directionality. Key contributions include: (1) the first enforcement of exact mass conservation in generative reaction prediction; (2) support for mechanism inference on unseen substrate scaffolds, cross-reaction-class generalization, and estimation of thermodynamic and kinetic feasibility; and (3) rapid adaptation to novel reaction types with minimal fine-tuning—yielding substantial improvements in mechanism recovery and product prediction accuracy on unseen classes, while ensuring interpretability and chemical plausibility.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Generation
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Central to our understanding of chemical reactivity is the principle of mass conservation, which is fundamental for ensuring physical consistency, balancing equations, and guiding reaction design. However, data-driven computational models for tasks such as reaction product prediction rarely abide by this most basic constraint. In this work, we recast the problem of reaction prediction as a problem of electron redistribution using the modern deep generative framework of flow matching. Our model, FlowER, overcomes limitations inherent in previous approaches by enforcing exact mass conservation, thereby resolving hallucinatory failure modes, recovering mechanistic reaction sequences for unseen substrate scaffolds, and generalizing effectively to out-of-domain reaction classes with extremely data-efficient fine-tuning. FlowER additionally enables estimation of thermodynamic or kinetic feasibility and manifests a degree of chemical intuition in reaction prediction tasks. This inherently interpretable framework represents a significant step in bridging the gap between predictive accuracy and mechanistic understanding in data-driven reaction outcome prediction.

Problem

• Predicting chemical reactions accurately
• Ensuring mass conservation in models
• Generalizing to unseen reaction classes

Innovation

• Electron redistribution via flow matching
• Enforces exact mass conservation
• Enables thermodynamic feasibility estimation