Atom-Motif Contrastive Transformer for Molecular Property Prediction

TL;DR

The paper introduces a novel Transformer model that captures both atom-level and motif-level interactions in molecules, improving molecular property prediction by leveraging contrastive learning and property-aware attention mechanisms.

Contribution

It proposes the Atom-Motif Contrastive Transformer (AMCT), integrating motif interactions and contrastive learning for enhanced molecular property prediction.

Findings

Outperforms state-of-the-art methods on seven benchmark datasets.

Effectively captures motif-level interactions for better property prediction.

Demonstrates the importance of motif-aware representations in molecular modeling.

Abstract

Recently, Graph Transformer (GT) models have been widely used in the task of Molecular Property Prediction (MPP) due to their high reliability in characterizing the latent relationship among graph nodes (i.e., the atoms in a molecule). However, most existing GT-based methods usually explore the basic interactions between pairwise atoms, and thus they fail to consider the important interactions among critical motifs (e.g., functional groups consisted of several atoms) of molecules. As motifs in a molecule are significant patterns that are of great importance for determining molecular properties (e.g., toxicity and solubility), overlooking motif interactions inevitably hinders the effectiveness of MPP. To address this issue, we propose a novel Atom-Motif Contrastive Transformer (AMCT), which not only explores the atom-level interactions but also considers the motif-level interactions. Since the representations of atoms and motifs for a given molecule are actually two different views of the same instance, they are naturally aligned to generate the self-supervisory signals for model training. Meanwhile, the same motif can exist in different molecules, and hence we also employ the contrastive loss to maximize the representation agreement of identical motifs across different molecules. Finally, in order to clearly identify the motifs that are critical in deciding the properties of each molecule, we further construct a property-aware attention mechanism into our learning framework. Our proposed AMCT is extensively evaluated on seven popular benchmark datasets, and both quantitative and qualitative results firmly demonstrate its effectiveness when compared with the state-of-the-art methods.