GeoTMI:Predicting quantum chemical property with easy-to-obtain geometry via positional denoising
This addresses a practical bottleneck for chemists and materials scientists by enabling accurate property predictions without costly geometry calculations, though it is an incremental improvement over existing 3D GNN methods.
The paper tackles the problem of predicting quantum chemical properties using graph neural networks that require high-level quantum mechanical geometries, which are infeasible for real-world applications, by proposing GeoTMI, a training framework that uses denoising on easy-to-obtain corrupted geometries, resulting in consistent accuracy improvements across multiple tasks.
As quantum chemical properties have a dependence on their geometries, graph neural networks (GNNs) using 3D geometric information have achieved high prediction accuracy in many tasks. However, they often require 3D geometries obtained from high-level quantum mechanical calculations, which are practically infeasible, limiting their applicability to real-world problems. To tackle this, we propose a new training framework, GeoTMI, that employs denoising process to predict properties accurately using easy-to-obtain geometries (corrupted versions of correct geometries, such as those obtained from low-level calculations). Our starting point was the idea that the correct geometry is the best description of the target property. Hence, to incorporate information of the correct, GeoTMI aims to maximize mutual information between three variables: the correct and the corrupted geometries and the property. GeoTMI also explicitly updates the corrupted input to approach the correct geometry as it passes through the GNN layers, contributing to more effective denoising. We investigated the performance of the proposed method using 3D GNNs for three prediction tasks: molecular properties, a chemical reaction property, and relaxed energy in a heterogeneous catalytic system. Our results showed consistent improvements in accuracy across various tasks, demonstrating the effectiveness and robustness of GeoTMI.