Abstracts

Rationale: The human phenotype ontology (HPO) incorporates a phenotype vocabulary and disease-phenotype annotations, thereby facilitating deep-phenotyping of genetic diseases. A numerically coded representation of a disease phenotype permits an algorithmic or computational analysis with genomic data, enabling precision diagnostics and ultimately therapeutics. The phenotypic features in a person with epilepsy are often complex with regards to seizure presentations, which is acknowledged by the most recent revision of the classification seizure types by the International League Against Epilepsy (ILAE). The ILAE seizure classification scheme also emphasizes the importance of elucidating the aetiology of each individual’s epilepsy, with a growing recognition of the diverse range of genetically-mediated epilepsies. We provide updated seizure-related HPO terms including neonatal, childhood and adult seizures, to facilitate a deep phenotypic interpretation of heretofore unexplained genetic epilepsies. Methods: The Epilepsiome project is a Task Force of the Genetics Commission of the International League Against Epilepsy (ILAE) and represent the link to the gene curation efforts within the ClinGen Epilepsy Clinical Domain Working Group (CDWG). Within the efforts to align terminology used in the diagnostic space, the Epilepsiome Project revised HPO terms for epileptic seizures. Seizure HPO terminology created is in line with current ILAE seizure classification with deviation for particular seizure types, such as epileptic spams and reflex epilepsies, where greater precision for genomic association is required or where needed by the constraints of the ontology. The updated classification was built through an online portal (Web Protégé) and consensus was achieved through biweekly conference calls. The updated HPO terms are linked to the existing HPO structure, either directly, or via synonyms for anachronistic terminology. Results: Focal, generalized and neonatal HPO seizure terminologies were constructed according to the most recent ILAE 2017 classification and aligned with the existing HPO structure. This ontology allows capture of clinical information at various levels of detail and aims to preserve both the onset, awareness and motor/non-motor nature of each seizure type, using multiple parentages for the terms. This enables a deep phenotypic description of an observed seizure at different degrees of detail. For focal and generalised seizures, the revised HPO terminology created was in line with the current ILAE classification of epileptic seizures. For generalised seizures we created 13 seizure terms including motor and non-motor onset, for focal seizures 64 terms, and for focal seizures of unknown onset we created 5 terms. We integrated other frequently observed seizure types currently not included in the ILAE 2017, which required a separate branch within the ontology due to biological peculiarity of their age of onset (neonatal seizures, 36 seizure terms), their clinical significance (status epilepticus, 30 seizure terms) or their genetic architecture (febrile seizures 4 seizure terms, febrile seizures plus 12 seizure terms, and reflex seizures 11 seizure terms created). Conclusions: Exponential increases in genomic and functional data in the epilepsies mandate a solid and common basis for interpretation and computation of phenotypic data. Improvements in HPO terms for epileptic seizures will enable a more versatile seizure ontology leading to deep phenotyping of people with epilepsy to improve associations with genomic data in both a research and diagnostic setting. This may facilitate more widespread usage of precision medicine in the treatment of the epilepsies where a genetic diagnosis can be ascertained. Funding: No funding