Enhancing molecular property prediction via transformer with dual graph representation

Accurate prediction of molecular properties is central to advancing chemistry, materials science, and drug discovery. Attention-based transformers have recently emerged as an effective infrastructure for learning on molecular graphs, yet their full potential depends on representations that capture the rich topology and structure of molecules. To this end, we propose the dual graph transformer (DGT), a self-attention architecture that jointly models atom and bond graphs to achieve comprehensive molecular representation. DGT facilitates the fusion of atom and bond features, graph topology and structure, and three-dimensional information within the self-attention module to learn an effective molecular representation. We benchmark DGT across a range of datasets for molecular property prediction, showing that it considerably outperforms the current state of the art. DGT demonstrates performance contributions from its dual graph representation, relative positional and structural encodings, and spatial information incorporation, while also offering interpretability at the molecular structural level. We envision DGT advancing molecular machine learning by improving both the prediction accuracy and interpretability of molecular properties.