Abstract:Molecular property prediction is a crucial foundation for drug discovery. In recent years, pre-trained deep learning models have been widely applied to this task. Some approaches that incorporate prior biological domain knowledge into the pre-training framework have achieved impressive results. However, these methods heavily rely on biochemical experts, and retrieving and summarizing vast amounts of domain knowledge literature is both time-consuming and expensive. Large Language Models (LLMs) have demonstrated remarkable performance in understanding and efficiently providing general knowledge. Nevertheless, they occasionally exhibit hallucinations and lack precision in generating domain-specific knowledge. Conversely, Domain-specific Small Models (DSMs) possess rich domain knowledge and can accurately calculate molecular domain-related metrics. However, due to their limited model size and singular functionality, they lack the breadth of knowledge necessary for comprehensive representation learning. To leverage the advantages of both approaches in molecular property prediction, we propose a novel Molecular Graph representation learning framework that integrates Large language models and Domain-specific small models (MolGraph-LarDo). Technically, we design a two-stage prompt strategy where DSMs are introduced to calibrate the knowledge provided by LLMs, enhancing the accuracy of domain-specific information and thus enabling LLMs to generate more precise textual descriptions for molecular samples. Subsequently, we employ a multi-modal alignment method to coordinate various modalities, including molecular graphs and their corresponding descriptive texts, to guide the pre-training of molecular representations. Extensive experiments demonstrate the effectiveness of the proposed method.

Deep Learning for Multi-Scale Molecular Modeling

Deep Learning for Variational Multi-Scale Molecular Modeling

Reinforcement Learning for Multi-Scale Molecular Modeling

Deep learning for variational multiscale molecular modeling

Simulate Time-integrated Coarse-grained Molecular Dynamics with Multi-Scale Graph Networks

A Perspective on Deep Learning for Molecular Modeling and Simulations

Advances in Deep Learning-Based 3D Molecular Generative Models

Deep Learning for Complex Chemical Systems

Machine Learning of coarse-grained Molecular Dynamics Force Fields

Machine Learning in Molecular Simulations of Biomolecules

Understanding the Limitations of Deep Models for Molecular Property Prediction: Insights and Solutions.

Learning Neural Generative Dynamics for Molecular Conformation Generation

Molecular geometric deep learning

Generative Modeling of Molecular Dynamics Trajectories

Simulate Time-integrated Coarse-grained Molecular Dynamics with Geometric Machine Learning

The confluence of machine learning and multiscale simulations

Machine learning for molecular simulation

Probabilistic Generative Deep Learning for Molecular Design

Molecular Graph Representation Learning Integrating Large Language Models with Domain-specific Small Models

Scalable Geometric Deep Learning on Molecular Graphs

Deep learning algorithms applied to computational chemistry