The Committee congratulates the representatives of the nine phenotype terminologies represented on the International Consortium of Human Phenotype Terminologies (ICHPT; http://www.irdirc.org/ichpt/) on the recent release of a core set of 2,086 terms that represent the major phenotypic abnormalities encountered in persons with rare diseases.
The Committee notes that the Human Variome Project exists to build capacity in the practice of responsible genomics and focusses on increasing both the quality and quantity of genomic knowledge that is collected, curated, interpreted and shared for clinical practice, and that the use of a common language to unambiguously refer to concepts is vital to this effort. Such common languages are already in place for some aspects of the work relevant to the Human Variome Project—the Committee especially notes the HGVS Nomenclature for the description of sequence variants—and now, with the release of this work by the International Consortium of Human Phenotype Terminologies, the beginnings of a similar language are in place for phenotypic features.
The ICHPT core terms provide both terms and definitions for the most common phenotypic features represented in the ICHPT member terminologies as well as cross references to a number of existing terminologies, including the Human Phenotype Ontology, which has become the de-facto standard for representing human phenotypes. This provides researchers that require resources offered by only one of the terminologies that have agreed on the core set with the ability to use their preferred terminology but still be able to integrate their data at a certain level of granularity by subsuming annotations to the level of the common terms.
The Committee believes that widespread adoption of this set of terms will allow easier exchange and integration of data from specialist and generalist databases, sophisticated search algorithms for biomedical literature, and the comprehensive and accurate documentation of phenotypic abnormalities, especially in hospital information systems. We therefore make the following recommendations:
- Phenotypic features should be represented in all genetic variation databases using terms from a terminology that has agreed to and defined cross-references to the ICHPT core set of phenotypic feature terms; and
- Existing phenotype terminologies that have not yet agreed to the ICHPT core set of terms and defined cross-references between their terms and the ICHPT core set should do so at the earliest possible opportunity.