Exploring the strengths and limitations of AI-driven variant prioritization versus manual curation in inborn errors of immunity
Our results show that AI-driven tools are powerful for detecting clearly pathogenic variants and can markedly accelerate the diagnostic process. However, their strong reliance on curated databases, limited incorporation of phenotypic data, and challenges in handling VUS may reduce their effectiveness. Enhancing phenotype integration, expanding annotations (including non-coding regions), and incorporating up-to-date literature could improve their performance. Ultimately, AI tools should complement expert curation, with future models evolving toward integrative approaches that better capture the complexity of human disorders.
Authors
- Elena Pérez-Estevez
- Laith Ibrahim Moushib
- Janire Perurena-Prieto (ORCID: https://orcid.org/0000-0002-3048-1058)
- Andrea Martín-Nalda (ORCID: https://orcid.org/0000-0002-3590-0186)
- Blanca Urban
- Roger Colobran
- Pere Soler-Palacín
- Aina Aguiló-Cucurull
- Romina Dieli-Crimi
- Nerea Moreno-Ruiz
- Xavier Solanich
- Laura Batlle-Masó
Institutions
- Hebron University (PS)
- Universitat Autònoma de Barcelona (ES)
- Mustansiriyah University (IQ)
- Bellvitge University Hospital (ES)
- Vall d'Hebron Institut de Recerca (ES)
- Vall d'Hebron Hospital Universitari (ES)
- Institut de Biologia Evolutiva (ES)
- Hospital de Cruces (ES)
- Jeffrey Modell Foundation (AU)
- Universitat de Barcelona (ES)
Publication Details
- Journal
- Frontiers in Genetics
- Published
- 2026-03-03
- DOI
- https://doi.org/10.3389/fgene.2026.1713299
- Primary Topic
- Immunodeficiency and Autoimmune Disorders
- Type
- article
- Field-Weighted Citation Impact
- 0.00