Mechanistic Reaction Prediction via Discrete Flow Matching on Graph-Structured Electron Occupation
cs.AI
Submitted: 2026-08-27
Updated: 2026-08-28
License: http://creativecommons.org/licenses/by/4.0/
The gist: Chemical reactions are fundamentally transformations in electron space, yet most machine learning approaches model them either through de novo generation of product molecules or through heuristic
Terminology
Abstract
Chemical reactions are fundamentally transformations in electron space, yet most machine learning approaches model them either through de novo generation of product molecules or through heuristic graph edits that operate directly on molecular topology. We introduce MAELLE (MechAnistic Edit fLow-matching on eLectron rEarrangements), which instead models reactions as discrete flow matching over electron occupation vectors. Concretely, we formulate the reactant-to-product mapping as a Continuous-time Markov Chain (CTMC) over the graph-structured integer-valued electron occupation space defined on all bonding, non-bonding, and hydrogen sites. To construct the interpolants between the reactants and products, we generalize the discrete flow matching mixture path to an edit-based formulation, where the electron moves are interpolated using Optimal Transport, yielding a mechanism-like set of moves without elementary step annotations. MAELLE achieves competitive performance on the USPTO-480K benchmark compared with leading reaction prediction models. Beyond in-distribution learning, we evaluate robustness across two out-of-distribution settings - structural complexity and reaction type - and find that MAELLE maintains strong performance where existing methods degrade. Finally, because the learned flow operates over the full electron redistribution, MAELLE naturally recovers mechanistic trajectories that align with known chemistry and can predict side products of a reaction.
Sources
- TempRe: Template generation for single and direct multi-step retrosynthesis
- Holistic chemical evaluation reveals pitfalls in reaction prediction models
- A Generative Model For Electron Paths
- DiGress: Discrete Denoising diffusion for graph generation
- DeFoG: Discrete Flow Matching for Graph Generation
- Generative Flows on Discrete State-Spaces: Enabling Multimodal Flows with Applications to Protein Co-Design
- Edit Flows: Flow Matching with Edit Operations
- Predicting Chemical Reaction Outcomes Based on Electron Movements Using Machine Learning
- Generator Matching: Generative modeling with arbitrary Markov processes
- Chemical reasoning in LLMs unlocks strategy-aware synthesis planning and reaction mechanism elucidation
- Can LLMs Generate Diverse Molecules? Towards Alignment with Structural Diversity
- A Hitchhiker's Guide to Geometric GNNs for 3D Atomic Systems
- Graph Attention Networks
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
- VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection