Enhancing Molecular Property Predictions by Learning from Bond Modelling and Interactions

TL;DR

DeMol introduces a dual-graph framework that explicitly models bonds and atoms in molecules, significantly improving the accuracy of molecular property predictions by capturing complex bond phenomena.

Contribution

The paper presents DeMol, a novel dual-graph architecture with bond-centric modeling and multi-scale interaction learning, advancing molecular property prediction accuracy.

Findings

DeMol achieves state-of-the-art results on multiple benchmarks.

Explicit bond modeling improves predictive performance.

The framework effectively captures complex chemical interactions.

Abstract

Molecule representation learning is crucial for understanding and predicting molecular properties. However, conventional atom-centric models, which treat chemical bonds merely as pairwise interactions, often overlook complex bond-level phenomena like resonance and stereoselectivity. This oversight limits their predictive accuracy for nuanced chemical behaviors. To address this limitation, we introduce \textbf{DeMol}, a dual-graph framework whose architecture is motivated by a rigorous information-theoretic analysis demonstrating the information gain from a bond-centric perspective. DeMol explicitly models molecules through parallel atom-centric and bond-centric channels. These are synergistically fused by multi-scale Double-Helix Blocks designed to learn intricate atom-atom, atom-bond, and bond-bond interactions. The framework's geometric consistency is further enhanced by a regularization term based on covalent radii to enforce chemically plausible structures. Comprehensive evaluations on diverse benchmarks, including PCQM4Mv2, OC20 IS2RE, QM9, and MoleculeNet, show that DeMol establishes a new state-of-the-art, outperforming existing methods. These results confirm the superiority of explicitly modelling bond information and interactions, paving the way for more robust and accurate molecular machine learning.