Multimodal Molecular Pretraining via Modality Blending

TL;DR

MoleBLEND introduces a novel atomic-relation level self-supervised learning approach that effectively aligns and integrates 2D and 3D molecular structures, enhancing molecular understanding for drug discovery.

Contribution

It proposes a fine-grained, relation-based modality blending method that unifies multiple learning objectives into a cohesive framework for molecular representation.

Findings

Achieves state-of-the-art results on molecular benchmarks.

Effectively aligns 2D and 3D structures at atomic level.

Provides theoretical insights via mutual-information maximization.

Abstract

Self-supervised learning has recently gained growing interest in molecular modeling for scientific tasks such as AI-assisted drug discovery. Current studies consider leveraging both 2D and 3D molecular structures for representation learning. However, relying on straightforward alignment strategies that treat each modality separately, these methods fail to exploit the intrinsic correlation between 2D and 3D representations that reflect the underlying structural characteristics of molecules, and only perform coarse-grained molecule-level alignment. To derive fine-grained alignment and promote structural molecule understanding, we introduce an atomic-relation level "blend-then-predict" self-supervised learning approach, MoleBLEND, which first blends atom relations represented by different modalities into one unified relation matrix for joint encoding, then recovers modality-specific information for 2D and 3D structures individually. By treating atom relationships as anchors, MoleBLEND organically aligns and integrates visually dissimilar 2D and 3D modalities of the same molecule at fine-grained atomic level, painting a more comprehensive depiction of each molecule. Extensive experiments show that MoleBLEND achieves state-of-the-art performance across major 2D/3D molecular benchmarks. We further provide theoretical insights from the perspective of mutual-information maximization, demonstrating that our method unifies contrastive, generative (cross-modality prediction) and mask-then-predict (single-modality prediction) objectives into one single cohesive framework.