Non-equilibrium molecular geometries in graph neural networks

TL;DR

This paper explores the use of non-DFT 3D geometries in graph neural network models for chemical property prediction, aiming to reduce computational costs while maintaining accuracy.

Contribution

It introduces a data augmentation method to enhance model performance when using less-accurate, forcefield-derived geometries.

Findings

Non-DFT geometries can be effectively used in training and testing.

Data augmentation improves prediction accuracy with forcefield geometries.

The approach reduces reliance on computationally expensive DFT calculations.

Abstract

Graph neural networks have become a powerful framework for learning complex structure-property relationships and fast screening of chemical compounds. Recently proposed methods have demonstrated that using 3D geometry information of the molecule along with the bonding structure can lead to more accurate prediction on a wide range of properties. A common practice is to use 3D geometries computed through density functional theory (DFT) for both training and testing of models. However, the computational time needed for DFT calculations can be prohibitively large. Moreover, many of the properties that we aim to predict can often be obtained with little or no overhead on top of the DFT calculations used to produce the 3D geometry information, voiding the need for a predictive model. To be practically useful for high-throughput chemical screening and drug discovery, it is desirable to work with 3D geometries obtained using less-accurate but much more efficient non-DFT methods. In this work we investigate the impact of using non-DFT conformations in the training and the testing of existing models and propose a data augmentation method for improving the prediction accuracy of classical forcefield-derived geometries.