PROJECT SUMMARY/ABSTRACT Studying genetic contributions to dementia risk from both the population-specific and trans-ancestral perspectives is vital to disease prediction, characterization, and the development of treatment interventions. Late-onset Alzheimer’s disease (LOAD) is the most common form of dementia, costs the United States an estimated $305 billion in associated costs, and affects approximately 6 million individuals nationally, and more than 24 million people globally. Apolipoprotein E (ApoE) is the strongest known genetic risk factor for LOAD. Specifically, ApoE-associated LOAD risk is driven by a risk variant of ApoE (termed ε4 or ApoE4) while ApoE- associated LOAD protection is thought to be driven by a protective variant of the gene (termed ε2 or ApoE2). The allele frequency of these alleles varies across populations, partially explaining higher LOAD prevalence in populations of African descent who carry the highest ApoE4 allele frequencies. In contrast to these findings, evidence indicates ApoE4-associated LOAD risk is significantly lower in individuals of African descent compared to other groups. This effect modification is partially driven by population-specific local ancestry that modifies the effect size of ApoE4. This observation, however, does not fully explain the reduction in effect size in African Americans. I hypothesize that genetic variants (local and global) modify ApoE4 and ApoE2 effect sizes. This proposal describes three aims to address the investigation into genetic contributions to ApoE- associated LOAD risk in multiple, diverse populations. Aim 1 will document the relationship between LOAD risk and the ApoE4 and ApoE2 alleles in the largest and most diverse dataset to be studied to date by estimating effect sizes in each of several diverse populations. Aim 2 will identify SNPs that associate with extreme ApoE4 and ApoE2 effect sizes and test whether these SNPs define specific physiological pathway association. Finally, aim 3 will evaluate the relative roles of race/ethnicity and genetic variation as modifiers of effect sizes within and between populations. As an MD/PhD candidate, I am passionate about this project as it will lead to new insights that can be used to understand, identify, and predict LOAD risk in underserved populations in the United States, especially those that experience increased LOAD burdens. The combined educational support from the Population and Quantitative Health Sciences Department and clinical training from the Medical Scientist Training Program at Case Western Reserve University will prepare me in tackling the genetic and healthcare avenues this work aims to inform. With support from these sources, and committee members with expertise in population genetics, statistical analysis, and bioinformatics, my training will provide the skills necessary to become a physician scientist versed in rigorous statistical approaches to understanding contributions to complex diseases.