Advancements in Artificial Intelligence for Variant Pathogenicity Prediction in Genetic Disorders

The human genome harbors numerous subtle sequence variations, termed variants, which influence protein synthesis within cellular processes. Among these, only a minority exert deleterious effects on protein function, thereby contributing to pathological conditions. Identifying such pathogenic variants amidst the abundance of benign alterations represents a longstanding challenge in genomic medicine.

Prior methodologies, including genome-wide association studies and various computational algorithms, have sought to address this issue. A recent innovation, the population-based Evolutionary Variant Effect (popEVE) model developed by researchers at Harvard Medical School and collaborators, advances this domain by assigning a quantitative pathogenicity score to each genomic variant. This approach situates variants along a continuous gradient of potential disease risk.

As detailed in a study published on November 24 in Nature Genetics, popEVE demonstrates proficiency in distinguishing benign from pathogenic variants, as well as delineating those associated with pediatric lethality from those manifesting in adult-onset disorders. Furthermore, the model has facilitated the discovery of over 100 previously unrecognized variants implicated in unresolved cases of rare genetic pathologies.

Current evaluations are underway to assess popEVE’s utility in clinical environments, with the aim of expediting precise diagnoses for monogenic rare diseases. Beyond diagnostics, this framework holds promise for elucidating novel therapeutic targets in genetically mediated disorders.

Authorship contributions include Mafalda Dias and Jonathan Frazer as co-senior authors, with additional collaborators Courtney A. Shearer, Aaron W. Kollasch, Aviv D. Spinner, Thomas Hopf, Lood van Niekerk, and Dinko Franceschi. Support for this research was derived from a Chan Zuckerberg Initiative Award (Neurodegeneration Challenge Network, CZI2018-191853), a National Institutes of Health Transformational Research Award (TR01CA260415), a National Science Foundation Graduate Research Fellowship, the Spanish Ministry of Science and Innovation (PID2022-140793NA-I00; CEX2020-001049-S; MCIN/AEI/10.13039/501100011033, MCIN/AEI/10.13039/501100011033/FEDER, UE), and the Generalitat de Catalunya (Government of Catalonia) through the CERCA program.