Reaction Prediction via Interaction Modeling of Symmetric Difference Shingle Sets

Runhan Shi, Letian Chen, Gufeng Yu, Yang Yang

yield prediction molecular representation learning chemical reaction prediction

⋅ NeurIPS ⋅ Project Page ⋅Poster ⋅OpenReview

Abstract

Chemical reaction prediction remains a fundamental challenge in organic chemistry, where existing machine learning models face two critical limitations: sensitivity to input permutations (molecule/atom orderings) and inadequate modeling of substructural interactions governing reactivity. These shortcomings lead to inconsistent predictions and poor generalization to real-world scenarios.

To address these challenges, we propose ReaDISH, a novel reaction prediction model that learns permutation-invariant representations while incorporating interaction-aware features. It introduces two innovations:

symmetric difference shingle encoding, which computes molecular shingle differences to capture reaction-specific structural changes while eliminating order sensitivity; and
geometry-structure interaction attention, a mechanism that models intra- and inter-molecular interactions at the shingle level.

Extensive experiments demonstrate that ReaDISH improves reaction prediction performance across diverse benchmarks. It shows enhanced robustness with an average improvement of 8.76% on R

^{2}

under permutation perturbations.