{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T14:07:57Z","timestamp":1784815677490,"version":"3.55.0"},"reference-count":25,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T00:00:00Z","timestamp":1784764800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T00:00:00Z","timestamp":1784764800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100001659","name":"Deutsche Forschungsgemeinschaft","doi-asserted-by":"crossref","award":["390874280"],"award-info":[{"award-number":["390874280"]}],"id":[{"id":"10.13039\/501100001659","id-type":"DOI","asserted-by":"crossref"}]},{"name":"German Center for Infection Research (DZIF) Translational Infrastructure Bioresources, Biodata, and Digital Health","award":["TI 12.002"],"award-info":[{"award-number":["TI 12.002"]}]},{"name":"Helmholtz-Zentrum f\u00fcr Infektionsforschung GmbH (HZI)"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Genome Biol"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Large language models (LLMs) are increasingly used to extract knowledge from text, yet their coverage and reliability in biology remain unclear. Microbial phenotypes are especially important to assess, as comprehensive data remain sparse except for well-studied organisms and they underpin our understanding of microbial characteristics, functional roles, and applications.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Here, we systematically assessed the biological knowledge encoded in publicly available LLMs for structured phenotype assignment of microbial species. We evaluated the performance of up to 57 LLMs across different experiments, including state-of-the-art models such as Claude Sonnet 4 and the GPT-5 family of models. Across phenotypes, LLMs reached accurate assignments for many species, but performance varied widely by model and trait, and no single model dominated. Model self-reported confidence is informative, with higher confidence aligning with higher accuracy, and can be used to prioritize phenotype assignment, effectively distinguishing between high- and low-confidence inferences.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Overall, our study outlines the utility and limitations of text-based LLMs for phenotype characterization in microbiology.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1186\/s13059-026-04207-7","type":"journal-article","created":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T13:09:02Z","timestamp":1784812142000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Comparative assessment of large language models for microbial phenotype assignment"],"prefix":"10.1186","volume":"27","author":[{"given":"Philipp C.","family":"M\u00fcnch","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nasim","family":"Safaei","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ren\u00e9","family":"Mreches","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Martin","family":"Binder","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yichen","family":"Han","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gary","family":"Robertson","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Eric A.","family":"Franzosa","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Curtis","family":"Huttenhower","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alice Carolyn","family":"McHardy","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,7,23]]},"reference":[{"issue":"D1","key":"4207_CR1","doi-asserted-by":"publisher","first-page":"D631","DOI":"10.1093\/nar\/gky879","volume":"47","author":"LC Reimer","year":"2019","unstructured":"Reimer LC, Vetcininova A, Carbasse JS, S\u00f6hngen C, Gleim D, Ebeling C, et al. BacDive in 2019: bacterial phenotypic data for high-throughput biodiversity analysis. Nucleic Acids Res. 2019;47(D1):D631\u20136.","journal-title":"Nucleic Acids Res"},{"issue":"D1","key":"4207_CR2","doi-asserted-by":"publisher","first-page":"D801","DOI":"10.1093\/nar\/gkab902","volume":"50","author":"JP Meier-Kolthoff","year":"2022","unstructured":"Meier-Kolthoff JP, Carbasse JS, Peinado-Olarte RL, G\u00f6ker M. TYGS and LPSN: a database tandem for fast and reliable genome-based classification and nomenclature of prokaryotes. Nucleic Acids Res. 2022;50(D1):D801\u20137.","journal-title":"Nucleic Acids Res"},{"key":"4207_CR3","volume-title":"Bergey\u2019s manual of systematics of Archaea and Bacteria","author":"WB Whitman","year":"2015","unstructured":"Whitman WB. Bergey\u2019s manual of systematics of Archaea and Bacteria. Wiley & Sons, Limited, John; 2015."},{"issue":"21","key":"4207_CR4","first-page":"10074","volume":"44","author":"M Brbi\u0107","year":"2016","unstructured":"Brbi\u0107 M, Pi\u0161korec M, Vidulin V, Kri\u0161ko A, \u0160muc T, Supek F. The landscape of microbial phenotypic traits and associated genes. Nucleic Acids Res. 2016;44(21):10074\u201390.","journal-title":"Nucleic Acids Res"},{"issue":"1","key":"4207_CR5","doi-asserted-by":"publisher","first-page":"160018","DOI":"10.1038\/sdata.2016.18","volume":"3","author":"MD Wilkinson","year":"2016","unstructured":"Wilkinson MD, Dumontier M, Aalbersberg IJJ, Appleton G, Axton M, Baak A, et al. The FAIR guiding principles for scientific data management and stewardship. Sci Data. 2016;3(1):160018.","journal-title":"Sci Data"},{"key":"4207_CR6","doi-asserted-by":"publisher","first-page":"3579","DOI":"10.63317\/3uyduafd26my","volume":"1","author":"C Buck","year":"2014","unstructured":"Buck C, Heafield K, Ooyen BV. N-gram counts and language models from the common crawl. LREC. 2014;1:3579\u201384.","journal-title":"LREC"},{"key":"4207_CR7","unstructured":"Brown TB, Mann B, Ryder N, Subbiah M, Kaplan J, Dhariwal P, et al. Language Models are Few-Shot Learners . arXiv [cs.CL]. 2020. Available from: http:\/\/arxiv.org\/abs\/2005.14165."},{"issue":"D1","key":"4207_CR8","doi-asserted-by":"publisher","first-page":"D748","DOI":"10.1093\/nar\/gkae959","volume":"53","author":"I Schober","year":"2025","unstructured":"Schober I, Koblitz J, Sard\u00e0 Carbasse J, Ebeling C, Schmidt ML, Podstawka A, et al. BacDive in 2025: the core database for prokaryotic strain data. Nucleic Acids Res. 2025;53(D1):D748\u201356.","journal-title":"Nucleic Acids Res"},{"issue":"6","key":"4207_CR9","doi-asserted-by":"publisher","first-page":"1825","DOI":"10.1099\/ijsem.0.002786","volume":"68","author":"AC Parte","year":"2018","unstructured":"Parte AC. LPSN - list of prokaryotic names with standing in nomenclature (bacterio.net), 20 years on. Int J Syst Evol Microbiol. 2018;68(6):1825\u20139.","journal-title":"Int J Syst Evol Microbiol"},{"issue":"4","key":"4207_CR10","doi-asserted-by":"publisher","first-page":"366","DOI":"10.1093\/bfgp\/elt008","volume":"12","author":"BE Dutilh","year":"2013","unstructured":"Dutilh BE, Backus L, Edwards RA, Wels M, Bayjanov JR, van Hijum SAFT. Explaining microbial phenotypes on a genomic scale: GWAS for microbes. Brief Funct Genomics. 2013;12(4):366\u201380.","journal-title":"Brief Funct Genomics"},{"key":"4207_CR11","unstructured":"Banerjee S, Agarwal A, Singla S. LLMs will always hallucinate, and we need to live with this. 2024. Available from: http:\/\/arxiv.org\/abs\/2409.05746. Cited 11 Sep 2024."},{"issue":"3","key":"4207_CR12","doi-asserted-by":"publisher","first-page":"570","DOI":"10.1099\/ijsem.0.001656","volume":"67","author":"SY Lin","year":"2017","unstructured":"Lin SY, Hameed A, Liu YC, Hsu YH, Hsieh YT, Lai WA, et al. Chryseobacterium endophyticum sp. nov., isolated from a maize leaf. Int J Syst Evol Microbiol. 2017;67(3):570\u20135.","journal-title":"Int J Syst Evol Microbiol"},{"issue":"Pt 8","key":"4207_CR13","doi-asserted-by":"publisher","first-page":"2975","DOI":"10.1099\/ijs.0.045427-0","volume":"63","author":"NL Nguyen","year":"2013","unstructured":"Nguyen NL, Kim YJ, Hoang VA, Yang DC. Chryseobacterium ginsengisoli sp. nov., isolated from the rhizosphere of ginseng and emended description of Chryseobacterium gleum. Int J Syst Evol Microbiol. 2013;63(Pt 8):2975\u201380.","journal-title":"Int J Syst Evol Microbiol"},{"issue":"1","key":"4207_CR14","doi-asserted-by":"publisher","first-page":"31","DOI":"10.51585\/gjvr.2023.1.0050","volume":"3","author":"K Holzer","year":"2023","unstructured":"Holzer K, Hoelzle LE, Wareth G. Genetic comparison of Brucella spp. and Ochrobactrum spp. erroneously included into the genus Brucella confirms separate genera. German J Veterinary Res. 2023;3(1):31\u20137.","journal-title":"German J Veterinary Res"},{"issue":"3","key":"4207_CR15","doi-asserted-by":"publisher","DOI":"10.1128\/jcm.01624-24","volume":"63","author":"G Perkins","year":"2025","unstructured":"Perkins G, Anderson NW, Spies NC. Retrieval-augmented generation salvages poor performance from large language models in answering microbiology-specific multiple-choice questions. J Clin Microbiol. 2025;63(3):e01624-24.","journal-title":"J Clin Microbiol"},{"key":"4207_CR16","unstructured":"Yoo H, Rosen GL. Predicting microbial ontology and pathogen risk from environmental metadata with large language models . arXiv [cs.CL]. 2025. Available from: http:\/\/arxiv.org\/abs\/2507.21980."},{"key":"4207_CR17","doi-asserted-by":"publisher","DOI":"10.1093\/gigascience\/giaf070","volume":"14","author":"S Ikeda","year":"2025","unstructured":"Ikeda S, Zou Z, Bono H, Moriya Y, Kawashima S, Katayama T, et al. Extraction of biological terms using large language models enhances the usability of metadata in the BioSample database. Gigascience. 2025;14:giaf070.","journal-title":"Gigascience"},{"key":"4207_CR18","unstructured":"Uryu S. Evaluating large Language Models for IUCN Red List species information . arXiv [cs.CL]. 2025. Available from: http:\/\/arxiv.org\/abs\/2510.02830."},{"key":"4207_CR19","unstructured":"Carlini N, Ippolito D, Jagielski M, Lee K, Tramer F, Zhang C. Quantifying memorization across neural language models . arXiv [cs.LG]. 2022. Available from: http:\/\/arxiv.org\/abs\/2202.07646. Cited 21 Apr 2026."},{"key":"4207_CR20","unstructured":"Feldman V. Does learning require memorization? A short tale about a long tail . arXiv [cs.LG]. 2019. Available from: http:\/\/arxiv.org\/abs\/1906.05271."},{"issue":"5","key":"4207_CR21","doi-asserted-by":"publisher","first-page":"1008","DOI":"10.1093\/bib\/bbx035","volume":"19","author":"GV Gkoutos","year":"2018","unstructured":"Gkoutos GV, Schofield PN, Hoehndorf R. The anatomy of phenotype ontologies: principles, properties and applications. Brief Bioinform. 2018;19(5):1008\u201321.","journal-title":"Brief Bioinform"},{"issue":"1","key":"4207_CR22","doi-asserted-by":"publisher","DOI":"10.1186\/2041-1480-4-43","volume":"4","author":"PL Buttigieg","year":"2013","unstructured":"Buttigieg PL, Morrison N, Smith B, Mungall CJ, Lewis SE, ENVO Consortium. The environment ontology: contextualising biological and biomedical entities. J Biomed Semantics. 2013;4(1):43.","journal-title":"J Biomed Semantics"},{"key":"4207_CR23","doi-asserted-by":"crossref","unstructured":"M\u00fcnch P, Safaei N, Mreches R, Binder M, Han Y, Robertson G, et al. LLM-BioEval: A Python tool that harnesses the power of large language models (LLMs) for microbial phenotype prediction . Github; 2026. Available from: https:\/\/github.com\/GenomeNet\/LLM-BioEval. Cited 18 Jun 2026.","DOI":"10.1101\/2025.11.24.690272"},{"key":"4207_CR24","unstructured":"M\u00fcnch P, Safaei N, Mreches R, Binder M, Han Y, Robertson G, et al. GenomeNet\/LLM-BioEval: v2.0.0. Zenodo; 2026. Available from: https:\/\/zenodo.org\/doi\/10.5281\/zenodo.13839818. Cited 18 Jun 2026."},{"key":"4207_CR25","doi-asserted-by":"publisher","unstructured":"Gamboa S, Eckenrode K, Ye J, Wokaty J, Mirzayi C, Nachod B, et al. Bugphyzz: A harmonized data resource and software for enrichment analysis of microbial physiologies. Zenodo; 2024. Available from: https:\/\/doi.org\/10.5281\/zenodo.12574596. Cited 18 Jun 2026.","DOI":"10.5281\/zenodo.12574596"}],"container-title":["Genome Biology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s13059-026-04207-7","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13059-026-04207-7.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13059-026-04207-7.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T13:28:36Z","timestamp":1784813316000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1186\/s13059-026-04207-7"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,23]]},"references-count":25,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["4207"],"URL":"https:\/\/doi.org\/10.1186\/s13059-026-04207-7","relation":{},"ISSN":["1474-760X"],"issn-type":[{"value":"1474-760X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,23]]},"assertion":[{"value":"12 January 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"13 July 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"23 July 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"Not applicable. This study did not involve human participants, human data, or animals.","order":1,"name":"Ethics","label":"Ethics approval and consent to participate","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable. This manuscript does not contain any individual person\u2019s data in any form.","order":2,"name":"Ethics","label":"Consent for publication","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests.","order":3,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Declarations"}}],"article-number":"235"}}