Computable Definitions of HPO Terms

Unless we instruct the computer otherwise, there is no implication that an HPO term such as Cerebral calcification is somehow related to the human brain. The HPO term can be related to other terms in the ontology by subclass relations, but it is not explicitly related to concepts from anatomy, histology, pathology, biochemistry, and cellular physiology. For this reason, a consortium including Sandra Dölken, Sebastian Köhler, Sebastian Bauer and Peter N Robinson from the Institute of Human Genetics and Medical Genetics of the Charité Berlin, Chris Mungall and Suzi Lewis from the Berkely Bioinformatics Open Source Projects (BBOP) group, Barbara Ruef and Monte Westerfield from the Zebrafish Model Organism Database (ZFIN), and George Ghoutos, Robert Hoehndorf, and Paul Schofield from the University of Cambridge has joined forces to develop computer-readable logical definitions of HPO terms that will allow human phenotypic abnormalities to be related to entities from anatomy, pathology, physiology, biochemistry, and other areas.

PATO

We are creating the definitions using the Phenotypic Quality Ontology PATO. We logically define phenotypes by stating that classes in the HPO are logically equivalent to Entity/Quality descriptions, with each such description consisting of the following elements: Q, the type of quality (characteristic) that the genotype affects; E, the type of entity that bears the quality; E2, an additional optional entity type, for relational qualities; M, a modifier. The basic methodology has been described in the paper Integrating phenotype ontologies across multiple species by Mungall and coathors (Genome Biology 2010, 11:R2).