{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,28]],"date-time":"2026-07-28T19:33:27Z","timestamp":1785267207412,"version":"3.55.0"},"reference-count":33,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,5,21]],"date-time":"2026-05-21T00:00:00Z","timestamp":1779321600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100017510","name":"UK Dementia Research Institute","doi-asserted-by":"publisher","award":["MR\/T04327X\/1"],"award-info":[{"award-number":["MR\/T04327X\/1"]}],"id":[{"id":"10.13039\/501100017510","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Digit. Health"],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>The Human Phenotype Ontology (HPO) provides a unified framework cataloguing over 17,500 phenotypic abnormalities across more than 8,600 rare diseases, defining hierarchical relationships between them. For example, classifying missing arms and missing legs as both abnormalities of the limb. This structure enables phenome-wide analyses, including the prioritisation of phenotypes as candidates for gene therapy. However, the HPO currently lacks sufficient metadata describing the clinical severity of these phenotypes. Manual expert curation at this scale would be prohibitively labour-intensive, creating a need for automated approaches to systematically annotate phenotypic severity.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>GPT-4, a large language model (LLM) developed by OpenAI, was employed to annotate the severity of all phenotypic abnormalities catalogued in the HPO. Severity was operationalised using nine clinical characteristics: congenital onset, reduced fertility, sensory impairments, impaired mobility, immunodeficiency, physical malformations, cancer, intellectual disability, and death. Each characteristic was further qualified by frequency of occurrence across four levels: never, rarely, often, and always. To assess annotation quality, GPT-4's outputs were benchmarked against ground-truth labels embedded within the HPO itself. For instance, phenotypes residing in the \u201cCancer\u201d HPO branch were expected to be annotated as cancer-causing. A novel severity scoring system was then developed that integrates both the nature of each clinical characteristic and its frequency of occurrence.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Benchmarking demonstrated strong performance across all clinical characteristics, with true positive recall rates ranging from 89% to 100% (mean = 97%). This indicates that GPT-4 can replicate expert-level curation with high fidelity. The resulting severity scoring system produced quantitative severity metrics for phenotypic abnormalities across the HPO, incorporating both the type and frequency of associated clinical characteristics.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>These findings demonstrate that LLMs can automate the large-scale curation of clinical metadata with a high degree of accuracy, substantially reducing the burden of manual expert annotation. The severity metrics generated here provide a foundation for systematically ranking human phenotypes by their impact on health and quality of life, enabling more principled prioritisation of targets for therapeutic intervention, particularly in the context of rare diseases where evidence is sparse and resources for curation are limited. Future work may extend this framework to incorporate additional clinical dimensions or validate annotations against independent clinical datasets.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fdgth.2026.1794934","type":"journal-article","created":{"date-parts":[[2026,5,21]],"date-time":"2026-05-21T05:45:58Z","timestamp":1779342358000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Using GPT-4 to annotate the severity of all phenotypic abnormalities within the human phenotype ontology"],"prefix":"10.3389","volume":"8","author":[{"given":"Kitty B.","family":"Murphy","sequence":"first","affiliation":[{"name":"Department of Brain Sciences, Imperial College London","place":["London, United Kingdom"]},{"name":"UK Dementia Research Institute, Imperial College London","place":["London, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Brian M.","family":"Schilder","sequence":"additional","affiliation":[{"name":"Department of Brain Sciences, Imperial College London","place":["London, United Kingdom"]},{"name":"UK Dementia Research Institute, Imperial College London","place":["London, United Kingdom"]},{"name":"Icahn School of Medicine at Mount Sinai","place":["New York, NY, United States"]},{"name":"Cold Spring Harbor Laboratory","place":["Cold Spring Harbor, NY, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nathan G.","family":"Skene","sequence":"additional","affiliation":[{"name":"Department of Brain Sciences, Imperial College London","place":["London, United Kingdom"]},{"name":"UK Dementia Research Institute, Imperial College London","place":["London, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,5,21]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"D1207","DOI":"10.1093\/nar\/gkaa1043","article-title":"The human phenotype ontology in 2021","volume":"49","author":"K\u00f6hler","year":"2021","journal-title":"Nucleic Acids Res"},{"key":"B2","doi-asserted-by":"publisher","first-page":"D1333","DOI":"10.1093\/nar\/gkad1005","article-title":"The human phenotype ontology in 2024: phenotypes around the world","volume":"52","author":"Gargano","year":"2024","journal-title":"Nucleic Acids Res"},{"key":"B3","doi-asserted-by":"publisher","first-page":"1134","DOI":"10.1038\/s41591-024-02855-5","article-title":"Adapted large language models can outperform medical experts in clinical text summarization","volume":"30","author":"Van Veen","year":"2024","journal-title":"Nat Med"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2403.18421","article-title":"BioMedLM: a 2.7B parameter language model trained on biomedical text","author":"Bolton","year":"2024","journal-title":"arXiv"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2305.17100","article-title":"BiomedGPT: a unified and generalist biomedical generative Pre-trained transformer for vision, language, and multimodal tasks","author":"Zhang","year":"2023","journal-title":"arXiv"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2402.10373","article-title":"Biomistral: a collection of open-source pretrained large language models for medical domains","author":"Labrak","year":"2024","journal-title":"arXiv"},{"key":"B7","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3458754","article-title":"Domain-Specific language model pretraining for biomedical natural language processing","volume":"3","author":"Gu","year":"2021","journal-title":"ACM Trans Comput Healthcare"},{"key":"B8","doi-asserted-by":"publisher","first-page":"172","DOI":"10.1038\/s41586-023-06291-2","article-title":"\u2018Large language models encode clinical knowledge","volume":"620","author":"Singhal","year":"2023","journal-title":"Nature"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2305.09617","article-title":"Towards expert-level medical question answering with large language models","author":"Singhal","year":"","journal-title":"arXiv"},{"key":"B10","first-page":"4700","article-title":"Biomegatron: larger biomedical domain language model","author":"Shin","year":"2020"},{"key":"B11","doi-asserted-by":"publisher","first-page":"1645","DOI":"10.1007\/s10439-023-03221-1","article-title":"Exploring the potential of GPT-4 in biomedical engineering: the dawn of a new era","volume":"51","author":"Cheng","year":"2023","journal-title":"Ann Biomed Eng"},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.1101\/2024.01.31.578275","article-title":"Phenomics assistant: an interface for LLM-based biomedical knowledge graph exploration","author":"O\u2019Neil","year":"2024","journal-title":"bioRxiv"},{"key":"B13","doi-asserted-by":"publisher","first-page":"bbac409","DOI":"10.1093\/bib\/bbac409","article-title":"BioGPT: generative Pre-trained transformer for biomedical text generation and mining","volume":"23","author":"Luo","year":"2022","journal-title":"Brief Bioinformatics"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2312.00164","article-title":"Towards accurate differential diagnosis with large language models","author":"McDuff","year":"2023","journal-title":"arXiv"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2303.08774","article-title":"GPT-4 technical report","author":"OpenAI","year":"2024","journal-title":"arXiv"},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2311.16452","article-title":"Can generalist foundation models outcompete special-purpose tuning? Case study in medicine","author":"Nori","year":"2023","journal-title":"arXiv"},{"key":"B17","doi-asserted-by":"publisher","first-page":"e114391","DOI":"10.1371\/journal.pone.0114391","article-title":"Systematic classification of disease severity for evaluation of expanded carrier screening panels","volume":"9","author":"Lazarin","year":"2014","journal-title":"PLoS One"},{"key":"B18","first-page":"1","article-title":"Syndromes associated with syndactyly","volume-title":"The Pediatric Upper Extremity","author":"Garagnani","year":"2013"},{"key":"B19","doi-asserted-by":"publisher","first-page":"D938","DOI":"10.1093\/nar\/gkad1082","article-title":"The monarch initiative in 2024: an analytic platform integrating phenotypes, genes and diseases across species","volume":"52","author":"Putman","year":"2024","journal-title":"Nucleic Acids Res"},{"key":"B20","doi-asserted-by":"publisher","first-page":"D1302","DOI":"10.1093\/nar\/gkaa1027","article-title":"Open targets platform: supporting systematic drug\u2013target identification and prioritisation","volume":"49","author":"Ochoa","year":"2021","journal-title":"Nucleic Acids Res"},{"key":"B21","doi-asserted-by":"publisher","first-page":"D712","DOI":"10.1093\/nar\/gkw1128","article-title":"The monarch initiative: an integrative data and analytic platform connecting phenotypes to genotypes across species","volume":"45","author":"Mungall","year":"2017","journal-title":"Nucleic Acids Res"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2312.10904","article-title":"Dynamic retrieval augmented generation of ontologies using artificial intelligence (DRAGON-AI)","author":"Toro","year":"2023","journal-title":"arXiv"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2308.06374","article-title":"Large language models and knowledge graphs: opportunities and challenges","author":"Pan","year":"2023","journal-title":"arXiv"},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2304.02711","article-title":"Structured prompt interrogation and recursive extraction of semantics (SPIRES): a method for populating knowledge bases using zero-shot learning","author":"Caufield","year":"2023","journal-title":"arXiv"},{"key":"B25","doi-asserted-by":"publisher","first-page":"125","DOI":"10.1002\/ajmg.a.20437","article-title":"Molar tooth sign of the midbrain\u2013hindbrain junction: occurrence in multiple distinct syndromes","author":"Gleeson","year":"2004","journal-title":"Am J Med Genet Part A"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2311.03839","article-title":"Aspects of human memory and large language models","author":"Janik","year":"2024","journal-title":"arXiv"},{"key":"B27","doi-asserted-by":"publisher","first-page":"11","DOI":"10.1108\/eb026526","article-title":"A statistical interpretation of term specificity and its application in retrieval","volume":"28","author":"Jones","year":"1972","journal-title":"J Doc"},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1301.3781","article-title":"Efficient estimation of word representations in vector space","author":"Mikolov","year":"2013","journal-title":"arXiv"},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2012.01805","article-title":"Interpretability and explainability: a machine learning zoo mini-tour","author":"Marcinkevi\u010ds","year":"2023","journal-title":"arXiv"},{"key":"B30","doi-asserted-by":"publisher","first-page":"12","DOI":"10.1016\/j.tjog.2012.01.004","article-title":"Prenatal diagnosis and genetic analysis of fetal akinesia deformation sequence and multiple pterygium syndrome associated with neuromuscular junction disorders: a review","volume":"51","author":"Chen","year":"2012","journal-title":"Taiwan J Obstet Gynecol"},{"key":"B31","doi-asserted-by":"publisher","DOI":"10.1101\/2023.02.13.23285820","article-title":"Cell type-specific contextualisation of the human phenome: towards the systematic treatment of all rare diseases","author":"Schilder","year":"2025","journal-title":"medRxiv"},{"key":"B32","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1038\/s41572-018-0008-5","article-title":"Endometriosis","volume":"4","author":"Zondervan","year":"2018","journal-title":"Nat Rev Dis Primers"},{"key":"B33","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2403.18802","article-title":"Long-form factuality in large language models","author":"Wei","year":"2024","journal-title":"arXiv"}],"container-title":["Frontiers in Digital Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1794934\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,21]],"date-time":"2026-05-21T05:46:04Z","timestamp":1779342364000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1794934\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,21]]},"references-count":33,"alternative-id":["10.3389\/fdgth.2026.1794934"],"URL":"https:\/\/doi.org\/10.3389\/fdgth.2026.1794934","relation":{},"ISSN":["2673-253X"],"issn-type":[{"value":"2673-253X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,21]]},"article-number":"1794934"}}