美国国立卫生研究院陆致用研究小组的研究开发出了GeneAgent：使用域数据库进行基因集分析的自验证语言代理。该研究于2025年7月28日发表于国际一流学术期刊《自然—方法学》杂志上。

在这里，小组介绍了GeneAgent，一个基于llm的人工智能代理，用于基因集分析，通过自主与生物数据库交互来验证其自己的输出来减少幻觉。对从不同地点收集的1106个基因集的评估表明，GeneAgent始终比GPT-4更准确。该课题组研究人员进一步将GeneAgent应用于来自小鼠B2905黑色素瘤细胞系的七个新基因集。专家评审证实，GeneAgent比GPT-4提供了更相关和全面的功能描述，为基因功能提供了有价值的见解，加快了知识发现。

据了解，基因集分析旨在确定具有共享功能的基因群的生物学机制。大型语言模型（llm）最近在为输入基因集生成功能描述方面表现出了希望，但可能会产生事实错误的陈述，在llm中通常被称为幻觉。

附：英文原文

Title: GeneAgent: self-verification language agent for gene-set analysis using domain databases

Author: Wang, Zhizheng, Jin, Qiao, Wei, Chih-Hsuan, Tian, Shubo, Lai, Po-Ting, Zhu, Qingqing, Day, Chi-Ping, Ross, Christina, Leaman, Robert, Lu, Zhiyong

Issue&Volume: 2025-07-28

Abstract: Gene-set analysis seeks to identify the biological mechanisms underlying groups of genes with shared functions. Large language models (LLMs) have recently shown promise in generating functional descriptions for input gene sets but may produce factually incorrect statements, commonly referred to as hallucinations in LLMs. Here we present GeneAgent, an LLM-based AI agent for gene-set analysis that reduces hallucinations by autonomously interacting with biological databases to verify its own output. Evaluation of 1,106 gene sets collected from different sources demonstrates that GeneAgent is consistently more accurate than GPT-4 by a significant margin. We further applied GeneAgent to seven novel gene sets derived from mouse B2905 melanoma cell lines. Expert review confirmed that GeneAgent produces more relevant and comprehensive functional descriptions than GPT-4, providing valuable insights into gene functions and expediting knowledge discovery.

DOI: 10.1038/s41592-025-02748-6

Source: https://www.nature.com/articles/s41592-025-02748-6