Automated Molecular Concept Generation and Labeling with Large Language Models

TL;DR

This paper presents AutoMolCo, a framework that uses large language models to automatically generate and label molecular concepts, enabling more explainable and effective models in molecular science.

Contribution

AutoMolCo introduces an automated, explainable approach to generate molecular concepts using LLMs, surpassing traditional GNNs without manual labeling.

Findings

AutoMolCo outperforms GNNs and LLM in-context learning on benchmarks.

The framework operates without human input, reducing manual effort.

Explainable concepts improve molecular research insights.

Abstract

Artificial intelligence (AI) is transforming scientific research, with explainable AI methods like concept-based models (CMs) showing promise for new discoveries. However, in molecular science, CMs are less common than black-box models like Graph Neural Networks (GNNs), due to their need for predefined concepts and manual labeling. This paper introduces the Automated Molecular Concept (AutoMolCo) framework, which leverages Large Language Models (LLMs) to automatically generate and label predictive molecular concepts. Through iterative concept refinement, AutoMolCo enables simple linear models to outperform GNNs and LLM in-context learning on several benchmarks. The framework operates without human knowledge input, overcoming limitations of existing CMs while maintaining explainability and allowing easy intervention. Experiments on MoleculeNet and High-Throughput Experimentation (HTE) datasets demonstrate that AutoMolCo-induced explainable CMs are beneficial for molecular science research.