Diffusion-based generative AI for exploring transition states from 2D molecular graphs

Abstract

The exploration of transition state (TS) geometries is crucial for elucidating chemical reaction mechanisms and modeling their kinetics. Recently, machine learning (ML) models have shown remarkable performance for prediction of TS geometries. However, they require 3D conformations of reactants and products often with their appropriate orientations as input, which demands substantial efforts and computational cost. Here, we propose a generative approach based on the stochastic diffusion method, namely TSDiff, for prediction of TS geometries just from 2D molecular graphs. TSDiff outperforms the existing ML models with 3D geometries in terms of both accuracy and efficiency. Moreover, it enables to sample various TS conformations, because it learns the distribution of TS geometries for diverse reactions in training. Thus, TSDiff finds more favorable reaction pathways with lower barrier heights than those in the reference database. These results demonstrate that TSDiff shows promising potential for an efficient and reliable TS exploration.

3DReact: Geometric Deep Learning for Chemical Reactions

“…Another category of reaction fingerprints arises from discretization of physically inspired functions − ,, constructed using a cheap estimate of the transition state (TS) structure or rather the structures of the reaction components ,, The SLATM d representation , in particular has been shown to yield accurate predictions of reaction barriers, particularly for data sets , relying on subtle changes in the geometry of reactants and/or products. End-to-end models based on three-dimensional structures of reactants and products have also recently emerged. ,, In a different vein, several works − aim to directly predict the TS structure, which together with the reactant structure gives the reaction barrier. These approaches lie outside the scope of the property prediction focus here.…”