EPSCoR Research Fellows: NSF: A Genotype-Forward Framework for Mechanistic Penetrance Estimation in Alzheimer's Disease

This Research Infrastructure Improvement (RII) EPSCoR Research Fellows project provides a fellowship to an Assistant professor and training for a graduate student at the New Mexico Institute of Mining and Technology. This work is conducted in collaboration with Dr. Jason H. Moore at Cedars-Sinai Medical Center. Through the fellowship, the PI will develop new computational methods that predict whether a genetic variant will lead to Alzheimer's disease, even when that variant's effect is currently unknown. Many variants found in patients cannot be interpreted because their effects on health are unknown, leaving families and clinicians without clear answers. The project will combine artificial intelligence with curated biological knowledge to trace the biological routes by which such variants might cause disease, producing estimates even for variants that current methods cannot evaluate. These estimates could inform future variant interpretation for clinicians and genetic counselors, and will strengthen biomedical research capacity in New Mexico. This project will develop the first genotype-forward framework for mechanistic penetrance estimation, the probability that a carrier of a genetic variant will develop disease. Its contribution is to treat penetrance not as a single number inferred from patient outcomes, but as the result of genetic uncertainty propagating through competing biological mechanisms. The framework will begin from a variant's structured biological uncertainty, represented by the principal investigator's disease-specific genomic model, and apply revision-capable reasoning operators over a curated Alzheimer's disease knowledge graph to produce calibrated penetrance estimates validated against longitudinal patient outcomes accessed through the host site. The project will advance the faculty member's independent research program and train a graduate student in AI-driven genomic analysis through host-site visits. It will also release open-source tools and establish a lasting partnership that strengthens AI-driven biomedical research capacity at the home institution. This project is supported by the EPSCoR Research Infrastructure Improvement Program: EPSCoR Research Fellows (ERF). The ERF program supports early- and mid-career investigators in eligible jurisdictions to develop collaborations at the nation’s private, government or academic research institutions. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.