concept · intermediate · partly written
The Human Phenotype Ontology (HPO)
HPO is a vocabulary of abnormal human features, linked to diseases, that lets software compare a patient with disease profiles.
Also called: HPO
What it is
HPO was launched in 2008 with over 8,000 terms and annotated all clinical entries in OMIM (Online Mendelian Inheritance in Man, a catalogue of genetic diseases). A specific term such as "seizure" implies all its ancestors in an is-a hierarchy, and diseases carry curated lists of terms. 12
Reported, not yet replicated
Size and activity more
By a 2022 review it held 199,197 annotations to 7,902 Mendelian diseases. The 2024 update added 2,239 terms since 2020, 49,235 annotations and 10 languages, and its trackers logged 3,715 issues from 177 people since 2020. Total term counts are quoted differently by differently dated sources, so no single current size is given here. 23
Reported, not yet replicated
How widely it is used more
The HPO authors state that almost all clinical genetics tools use it. That is an assertion by the maintainers. The founding paper has 792 citing records in Europe PMC (read 2026-11-18, not audited); citations measure reach, not quality. 34
Reported, not yet replicated
Where it does not match clinical terminology more
In 2016, 30% of 10,454 HPO classes had an equivalent SNOMED CT concept, a further 62% a partial match, and 92% a complete or partial one. 5
Reported, not yet replicated
Annotation noise more
An independent study found that the share of patients with the right disease in the top five was 0.993 with annotations inferred from published clinical studies, 0.901 with electronic annotations and 0.580 with traceable-author statements, and that performance fell significantly beyond 40% noisy annotations. 6
Reported, not yet replicated
Coverage limits more
Annotations are biased toward Mendelian disease, and the prenatal part of the hierarchy is thin, about 0.5% of annotations. 2
Reported, not yet replicated
A consequence of how scores are weighted more
Similarity scores weight a term by its information content, the negative log of how often it or its descendants occur in annotations. So adding many deeply annotated diseases would change the weight of every term they use, and so shift the scores of diseases already ranked. This is our reading of the definition, not a measured result. 7
Our inference
Plain answer
Why it matters
It is the best-documented case of a bottom-up ontology that gained an algorithm, an exchange format and a benchmark, and it shows what rare-disease ranking depends on. 2
Our inference
Still open here
- Is orthology a first-class relation inside HPO itself? Not shown in the sources read.
- How many of the roughly 177 contributors in the 2024 paper are curators? Not found.
Not yet written: Equation, Real numbers.
Linked from
Sources
- Robinson PN, Köhler S, Bauer S, Seelow D, et al.. The Human Phenotype Ontology: a tool for annotating and analyzing human hereditary disease. American journal of human genetics (2008). paper; opened 2026-11-18 all records citing this
- Jacobsen JOB, Kelly C, Cipriani V, Research Consortium GE, et al.. Phenotype-driven approaches to enhance variant prioritization and diagnosis of rare disease. Human mutation (2022). paper; opened 2026-11-18 all records citing this
- Gargano MA, Matentzoglu N, Coleman B, Addo-Lartey EB, et al.. The Human Phenotype Ontology in 2024: phenotypes around the world. Nucleic acids research (2024). paper; opened 2026-11-18 all records citing this
- Europe PMC: citation counts for individual ontology papers, as read on 2026-11-18.. database, not peer reviewed; opened 2026-11-18 all records citing this
- Dhombres F, Bodenreider O. Interoperability between phenotypes in research and healthcare terminologies--Investigating partial mappings between HPO and SNOMED CT. Journal of biomedical semantics (2016). paper; opened 2026-11-18 all records citing this
- Peng J, Li Q, Shang X. Investigations on factors influencing HPO-based semantic similarity calculation. Journal of biomedical semantics (2017). paper; opened 2026-11-18 all records citing this
- Philip Resnik. Using Information Content to Evaluate Semantic Similarity in a Taxonomy. arXiv (1995). preprint, not peer reviewed; opened 2026-11-18 all records citing this
Where it sits on the map
History of this record (5)
This record has been changed 5 times, on the dates below. The note kept against each change is written for the people who maintain the atlas, in their own shorthand, so it is not shown here; it is in the record itself.
- Status
- partly written
- Last updated
- 2026-11-19
- Record
- content/concepts/human-phenotype-ontology.json