ChemHGNN: A Hierarchical Hypergraph Neural Network for Reaction Virtual Screening and Discovery

TL;DR

ChemHGNN introduces a hierarchical hypergraph neural network that effectively models multi-reactant reactions, significantly improving reaction virtual screening and discovery by capturing high-order relationships and addressing key challenges.

Contribution

This work presents ChemHGNN, a novel hypergraph neural network framework that naturally models multi-reactant reactions and incorporates hierarchical features, outperforming traditional GNNs in reaction prediction tasks.

Findings

ChemHGNN outperforms GNN and HGNN baselines on the USPTO dataset.

The model maintains interpretability and chemical plausibility.

ChemHGNN is effective in large-scale reaction screening.

Abstract

Reaction virtual screening and discovery are fundamental challenges in chemistry and materials science, where traditional graph neural networks (GNNs) struggle to model multi-reactant interactions. In this work, we propose ChemHGNN, a hypergraph neural network (HGNN) framework that effectively captures high-order relationships in reaction networks. Unlike GNNs, which require constructing complete graphs for multi-reactant reactions, ChemHGNN naturally models multi-reactant reactions through hyperedges, enabling more expressive reaction representations. To address key challenges, such as combinatorial explosion, model collapse, and chemically invalid negative samples, we introduce a reaction center-aware negative sampling strategy (RCNS) and a hierarchical embedding approach combining molecule, reaction and hypergraph level features. Experiments on the USPTO dataset demonstrate that ChemHGNN significantly outperforms HGNN and GNN baselines, particularly in large-scale settings, while maintaining interpretability and chemical plausibility. Our work establishes HGNNs as a superior alternative to GNNs for reaction virtual screening and discovery, offering a chemically informed framework for accelerating reaction discovery.