Scalar Coupling Constant Prediction Using Graph Embedding Local Attention Encoder

TL;DR

This paper introduces a graph embedding local self-attention encoder (GELAE) for efficient and accurate scalar coupling constant prediction, leveraging invariant structure representations and a novel attention mechanism.

Contribution

The paper proposes a new GELAE model with invariant structure representation and a local self-attention module for improved SCC prediction accuracy.

Findings

MAE reduced from 0.1603 Hz to 0.1067 Hz with the new attention module

Invariant structure representations improve prediction accuracy

Classification loss function yields MAE close to quantum chemistry standards

Abstract

Scalar coupling constant (SCC) plays a key role in the analysis of three-dimensional structure of organic matter, however, the traditional SCC prediction using quantum mechanical calculations is very time-consuming. To calculate SCC efficiently and accurately, we proposed a graph embedding local self-attention encoder (GELAE) model, in which, a novel invariant structure representation of the coupling system in terms of bond length, bond angle and dihedral angle was presented firstly, and then a local self-attention module embedded with the adjacent matrix of a graph was designed to extract effectively the features of coupling systems, finally, with a modified classification loss function, the SCC was predicted. To validate the superiority of the proposed method, we conducted a series of comparison experiments using different structure representations, different attention modules, and different losses. The experimental results demonstrate that, compared to the traditional chemical bond structure representations, the rotation and translation invariant structure representations proposed in this work can improve the SCC prediction accuracy; with the graph embedded local self-attention, the mean absolute error (MAE) of the prediction model in the validation set decreases from 0.1603 Hz to 0.1067 Hz; using the classification based loss function instead of the scaled regression loss, the MAE of the predicted SCC can be decreased to 0.0963 HZ, which is close to the quantum chemistry standard on CHAMPS dataset.