MolBind: Multimodal Alignment of Language, Molecules, and Proteins

TL;DR

MolBind is a novel multi-modal framework that aligns language, molecules, and proteins in a shared space, improving zero-shot learning in drug discovery by integrating four modalities with a new dataset.

Contribution

The paper introduces MolBind, a multi-modal contrastive learning framework that unifies diverse biological data modalities and provides a high-quality dataset for pre-training.

Findings

Superior zero-shot performance across multiple tasks

Effective semantic alignment of diverse modalities

Introduction of a new multi-modal dataset MolBind-M4

Abstract

Recent advancements in biology and chemistry have leveraged multi-modal learning, integrating molecules and their natural language descriptions to enhance drug discovery. However, current pre-training frameworks are limited to two modalities, and designing a unified network to process different modalities (e.g., natural language, 2D molecular graphs, 3D molecular conformations, and 3D proteins) remains challenging due to inherent gaps among them. In this work, we propose MolBind, a framework that trains encoders for multiple modalities through contrastive learning, mapping all modalities to a shared feature space for multi-modal semantic alignment. To facilitate effective pre-training of MolBind on multiple modalities, we also build and collect a high-quality dataset with four modalities, MolBind-M4, including graph-language, conformation-language, graph-conformation, and conformation-protein paired data. MolBind shows superior zero-shot learning performance across a wide range of tasks, demonstrating its strong capability of capturing the underlying semantics of multiple modalities.