Refining Human Phenotype Ontology (HPO) to enable better phenotype-genotype integration in systemic autoimmune...

The Human Phenotype Ontology (HPO) provides a standardised framework for disease-phenotype associations. Given the complexity of systemic autoimmune rheumatic diseases (SARDs) and the absence of prior evaluations, we aimed to assess the completeness of HPO terms capturing SARD-related phenotypic features and SARD-phenotype associations.

Whole-exome sequencing was performed in 111 individuals from the LEAP (Lupus Extended Autoimmune Phenotype) cohort, a cohort of adult and paediatric SARD patients, who had at least one feature suggestive of a monogenic disease. We evaluated the completeness of HPO annotations relative to documented phenotypic data and SARD diagnoses. We also explored the potential added value of these terms descriptively and by using a phenotype-based genomic tool, Emedgene.

At least one missing HPO term occurred in 36.9% of our patients. Five patient features could not be mapped to existing HPO terms: anti-chromatin antibody, autoimmune hepatitis, scleroderma renal crisis, shrinking lung syndrome, and tenosynovitis. The HPO term "scleroderma" was misdefined, and six HPO terms did not include six linked SARDs in the respective disease-association section. In a random sample of 10 SARD patients, we also demonstrated that HPO terms affect the gene variant prioritisation on the Emedgene platform.

This is the first study to examine the HPO term coverage in SARDs and assess the potential impact of missing terms. We have proposed to the HPO hub the inclusion of absent or misdefined terms and the addition of missing SARD-HPO term associations. We anticipate that expanding the HPO will enhance our ability to characterise the genetic basis of SARDs.