{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T18:21:07Z","timestamp":1778523667920,"version":"3.51.4"},"reference-count":32,"publisher":"Oxford University Press (OUP)","issue":"2","license":[{"start":{"date-parts":[[2025,4,7]],"date-time":"2025-04-07T00:00:00Z","timestamp":1743984000000},"content-version":"vor","delay-in-days":37,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,3,4]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Understanding causality in medical research is essential for developing effective interventions and diagnostic tools. Mendelian Randomization (MR) is a pivotal method for inferring causality through genetic data. However, MR analysis often requires pre-identification of exposure-outcome pairs from clinical experience or literature, which can be challenging to obtain. This poses difficulties for clinicians investigating causal factors of specific diseases. To address this, we introduce MRAgent, an innovative automated agent leveraging Large Language Models (LLMs) to enhance causal knowledge discovery in disease research. MRAgent autonomously scans scientific literature, discovers potential exposure-outcome pairs, and performs MR causal inference using extensive Genome-Wide Association Study data. We conducted both automated and human evaluations to compare different LLMs in operating MRAgent and provided a proof-of-concept case to demonstrate the complete workflow. MRAgent\u2019s capability to conduct large-scale causal analyses represents a significant advancement, equipping researchers and clinicians with a robust tool for exploring and validating causal relationships in complex diseases. Our code is public at https:\/\/github.com\/xuwei1997\/MRAgent.<\/jats:p>","DOI":"10.1093\/bib\/bbaf140","type":"journal-article","created":{"date-parts":[[2025,4,8]],"date-time":"2025-04-08T02:57:49Z","timestamp":1744081069000},"source":"Crossref","is-referenced-by-count":15,"title":["MRAgent: an LLM-based automated agent for causal knowledge discovery in disease via Mendelian randomization"],"prefix":"10.1093","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3708-6816","authenticated-orcid":false,"given":"Wei","family":"Xu","sequence":"first","affiliation":[{"name":"Centre for Artificial Intelligence Driven Drug Discovery, Macao Polytechnic University , R. de Lu\u00eds Gonzaga Gomes, Macao SAR 999078 ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Gang","family":"Luo","sequence":"additional","affiliation":[{"name":"Centre for Artificial Intelligence Driven Drug Discovery, Macao Polytechnic University , R. de Lu\u00eds Gonzaga Gomes, Macao SAR 999078 ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Weiyu","family":"Meng","sequence":"additional","affiliation":[{"name":"Centre for Artificial Intelligence Driven Drug Discovery, Macao Polytechnic University , R. de Lu\u00eds Gonzaga Gomes, Macao SAR 999078 ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiaobing","family":"Zhai","sequence":"additional","affiliation":[{"name":"Centre for Artificial Intelligence Driven Drug Discovery, Macao Polytechnic University , R. de Lu\u00eds Gonzaga Gomes, Macao SAR 999078 ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Keli","family":"Zheng","sequence":"additional","affiliation":[{"name":"The State Key Laboratory of Computer Science , Institute of Software Chinese Academy of Sciences, No. 4, South 4th Street, Zhongguancun, Haidian District, Beijing 100190,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ji","family":"Wu","sequence":"additional","affiliation":[{"name":"School of Computer Science , Zhuhai College of Science and Technology, No. 8, Anji East Road, Sanzao Town, Jinwan District, Zhuhai 519041,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yanrong","family":"Li","sequence":"additional","affiliation":[{"name":"Centre for Artificial Intelligence Driven Drug Discovery, Macao Polytechnic University , R. de Lu\u00eds Gonzaga Gomes, Macao SAR 999078 ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Abao","family":"Xing","sequence":"additional","affiliation":[{"name":"Centre for Artificial Intelligence Driven Drug Discovery, Macao Polytechnic University , R. de Lu\u00eds Gonzaga Gomes, Macao SAR 999078 ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Junrong","family":"Li","sequence":"additional","affiliation":[{"name":"Centre for Artificial Intelligence Driven Drug Discovery, Macao Polytechnic University , R. de Lu\u00eds Gonzaga Gomes, Macao SAR 999078 ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhifan","family":"Li","sequence":"additional","affiliation":[{"name":"Centre for Artificial Intelligence Driven Drug Discovery, Macao Polytechnic University , R. de Lu\u00eds Gonzaga Gomes, Macao SAR 999078 ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ke","family":"Zheng","sequence":"additional","affiliation":[{"name":"Centre for Artificial Intelligence Driven Drug Discovery, Macao Polytechnic University , R. de Lu\u00eds Gonzaga Gomes, Macao SAR 999078 ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kefeng","family":"Li","sequence":"additional","affiliation":[{"name":"Centre for Artificial Intelligence Driven Drug Discovery, Macao Polytechnic University , R. de Lu\u00eds Gonzaga Gomes, Macao SAR 999078 ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2025,4,7]]},"reference":[{"key":"2025040721502466500_ref1","doi-asserted-by":"publisher","first-page":"157","DOI":"10.1080\/02698590701498084","article-title":"Interpreting causality in the health sciences","volume":"21","author":"Russo","year":"2007","journal-title":"International Studies in the Philosophy of Science"},{"key":"2025040721502466500_ref2","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1093\/ije\/dyg070","article-title":"\u2018Mendelian randomization\u2019: Can genetic epidemiology contribute to understanding environmental determinants of disease? $\\ast $","volume":"32","author":"Smith","year":"2003","journal-title":"Int J Epidemiol"},{"key":"2025040721502466500_ref3","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s43586-021-00092-5","article-title":"Mendelian randomization","volume":"2","author":"Eleanor Sanderson","year":"2022","journal-title":"Nature Reviews Methods Primers"},{"key":"2025040721502466500_ref4","doi-asserted-by":"publisher","first-page":"e1007081","DOI":"10.1371\/journal.pgen.1007081","article-title":"Orienting the causal relationship between imprecisely measured traits using GWAS summary data","volume":"13","author":"Hemani","year":"2017","journal-title":"PLoS Genet"},{"key":"2025040721502466500_ref5","doi-asserted-by":"publisher","DOI":"10.1101\/2020.08.10.244293","article-title":"The MRC IEU OpenGWAS data infrastructure","volume-title":"bioRxiv","author":"Elsworth"},{"key":"2025040721502466500_ref6","volume-title":"GPT-4 Technical Report","author":"OpenAI"},{"key":"2025040721502466500_ref7","volume-title":"A Survey of Large Language Models. arXiv","author":"Zhao"},{"key":"2025040721502466500_ref8","doi-asserted-by":"crossref","volume-title":"The Rise and Potential of Large Language Model Based Agents: A Survey","author":"Xi","DOI":"10.1007\/s11432-024-4222-0"},{"key":"2025040721502466500_ref9","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/JBHI.2024.3417280","article-title":"Dietary assessment with multimodal ChatGPT: A systematic analysis","volume":"28","author":"Lo","year":"2024","journal-title":"IEEE J Biomed Health Inform"},{"key":"2025040721502466500_ref10","volume-title":"ProAgent: From Robotic Process Automation to Agentic Process Automation","author":"Ye"},{"key":"2025040721502466500_ref11","volume-title":"AutoGen: Enabling Next-Gen LLM Applications Via Multi-Agent Conversation","author":"Wu","year":"2023"},{"key":"2025040721502466500_ref12","volume-title":"Agent Hospital: A Simulacrum of Hospital with Evolvable Medical Agents","author":"Li","year":"2024"},{"key":"2025040721502466500_ref13","first-page":"15523","article-title":"EconAgent: Large language model-empowered agents for simulating macroeconomic activities","volume-title":"Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics","author":"Li","year":"2024"},{"key":"2025040721502466500_ref14"},{"key":"2025040721502466500_ref15","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.emnlp-main.552","article-title":"SimCSE: Simple Contrastive Learning of Sentence Embeddings","volume-title":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (pp. 6894\u2013910)","author":"Gao"},{"key":"2025040721502466500_ref16","first-page":"33","article-title":"Further normalization of the data base relational model","volume":"6","author":"Codd","year":"1972","journal-title":"Data base systems"},{"key":"2025040721502466500_ref17","doi-asserted-by":"publisher","first-page":"1614","DOI":"10.1001\/jama.2021.18236","article-title":"Strengthening the reporting of observational studies in epidemiology using mendelian randomization: The STROBE-MR statement","volume":"326","author":"Skrivankova","year":"2021","journal-title":"JAMA"},{"key":"2025040721502466500_ref18","doi-asserted-by":"publisher","DOI":"10.1186\/1472-6947-12-12","article-title":"Performance evaluation of unified medical language system$\\circledR $\u2018s synonyms expansion to query ref14","volume":"12","author":"Griffon","year":"2012","journal-title":"BMC Med Inform Decis Mak"},{"key":"2025040721502466500_ref19","doi-asserted-by":"crossref","DOI":"10.1101\/173682","volume-title":"Automating Mendelian Randomization through Machine Learning to Construct a Putative Causal Map of the Human Phenome","author":"Hemani","year":"2017"},{"key":"2025040721502466500_ref20","doi-asserted-by":"publisher","first-page":"314","DOI":"10.1002\/gepi.22522","article-title":"Bias correction for inverse variance weighting mendelian randomization","volume":"47","author":"Mounier","year":"2023","journal-title":"Genet Epidemiol"},{"key":"2025040721502466500_ref21","article-title":"ReAct: Synergizing reasoning and acting in language models","volume-title":"The Eleventh International Conference on Learning Representations","author":"Yao","year":"2022"},{"key":"2025040721502466500_ref22","first-page":"1877","article-title":"Language models are few-shot learners","volume-title":"Advances in Neural Information Processing Systems","author":"Brown","year":"2020"},{"key":"2025040721502466500_ref23","article-title":"The Claude 3 model family: Opus, sonnet, haiku","volume":"1","author":"Anthropic","year":"2024","journal-title":"Claude-3 Model Card"},{"key":"2025040721502466500_ref24","volume-title":"Qwen Technical Report","author":"Bai"},{"key":"2025040721502466500_ref25","volume-title":"Mixtral of Experts","author":"Jiang","year":"2024"},{"key":"2025040721502466500_ref26","volume-title":"Meta-Llama\/llama3. Github","author":"Meta Llama"},{"key":"2025040721502466500_ref27","volume-title":"LLaMA: Open and Efficient Foundation Language Models","author":"Touvron","year":"2023"},{"key":"2025040721502466500_ref28","first-page":"39","article-title":"A survey on evaluation of large language models","volume":"15","author":"Yupeng Chang","year":"2024","journal-title":"ACM Trans Intell Syst Technol"},{"key":"2025040721502466500_ref29","doi-asserted-by":"publisher","first-page":"108","DOI":"10.1007\/s00262-024-03689-3","article-title":"A2ar-mediated cxcl5 upregulation on macrophages promotes nsclc progression via netosis","volume":"73","author":"Lei","year":"2024","journal-title":"Cancer Immunol Immunother"},{"key":"2025040721502466500_ref30","volume-title":"Large Language Models Are Zero-Shot Reasoners","author":"Kojima"},{"key":"2025040721502466500_ref31","volume-title":"GPT-4o System Card","author":"OpenAI"},{"key":"2025040721502466500_ref32","doi-asserted-by":"crossref","first-page":"baz068","DOI":"10.1093\/database\/baz068","article-title":"Re-curation and rational enrichment of knowledge graphs in biological expression language","volume":"2019","author":"Hoyt","year":"2019","journal-title":"Database"}],"container-title":["Briefings in Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/26\/2\/bbaf140\/62885664\/bbaf140.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/26\/2\/bbaf140\/62885664\/bbaf140.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,4,8]],"date-time":"2025-04-08T02:59:13Z","timestamp":1744081153000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bib\/article\/doi\/10.1093\/bib\/bbaf140\/8107848"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3]]},"references-count":32,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2025,3,4]]}},"URL":"https:\/\/doi.org\/10.1093\/bib\/bbaf140","relation":{},"ISSN":["1467-5463","1477-4054"],"issn-type":[{"value":"1467-5463","type":"print"},{"value":"1477-4054","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2025,3]]},"published":{"date-parts":[[2025,3]]},"article-number":"bbaf140"}}