Bidirectional Generation of Structure and Properties Through a Single Molecular Foundation Model

TL;DR

This paper introduces a novel multimodal molecular pre-trained model that aligns molecular structure and biochemical properties in a shared space, enabling versatile chemical task solutions with a single model.

Contribution

It presents the first multimodal pre-training approach for molecules that jointly models structure and properties, improving performance on diverse chemical tasks.

Findings

Effective in conditional molecule generation

Accurate in property prediction tasks

Versatile across multiple chemical applications

Abstract

The recent success of large foundation models in artificial intelligence has prompted the emergence of chemical pre-trained models. Despite the growing interest in large molecular pre-trained models that provide informative representations for downstream tasks, attempts for multimodal pre-training approaches on the molecule domain were limited. To address this, we present a novel multimodal molecular pre-trained model that incorporates the modalities of structure and biochemical properties, drawing inspiration from recent advances in multimodal learning techniques. Our proposed model pipeline of data handling and training objectives aligns the structure/property features in a common embedding space, which enables the model to regard bidirectional information between the molecules' structure and properties. These contributions emerge synergistic knowledge, allowing us to tackle both multimodal and unimodal downstream tasks through a single model. Through extensive experiments, we demonstrate that our model shows remarkable capabilities in solving various meaningful chemical challenges, including conditional molecule generation, property prediction, molecule classification, and reaction prediction.