Genomic, Epigenomic, and Transcriptomic Mechanisms of Contributing to Alzheimer's Disease Risk in Diverse Ancestral Populations

PROJECT SUMMARY Alzheimer disease (AD) is the leading cause of dementia in the elderly in the United States and occurs in all ethnic and racial groups. Although >20 susceptibility loci have been associated with AD, much of the genetic influence on AD remains unknown. This is particularly true for diverse ethnic populations such as Hispanics (HI) and African Americans (AA) that are underrepresented in most genetic studies compared to non-Hispanic Whites (NHW). Though emerging studies have begun to unravel some of the ancestry associated genetic risk of AD, understanding the role of DNA variation is only one critical step in understanding the complex underlying biology variants do not always provide direct information on the genes and mechanisms influenced. To expand the context of ongoing large-scale genomic genotyping and whole genome sequencing efforts to individuals of diverse ancestries, here we propose a transcriptomic, epigenomic, and genomic association study of NHW, AA, and HI from Puerto Rico (PR) and Peru (PER) addressing diverse admixed populations. Completion of this project will provide powerful functional genomic resources that could be applied to large existing and emerging AD datasets to better understand the biological impact of genetic variation on AD risk. Specifically, we will 1) generate whole-blood RNAseq, single-cell RNAseq, and DNA methylation datasets on existing collections of NHW, AA, PR, and PER AD cases and elderly cognitively normal controls; 2) identify transcriptomic and methylation differences including gene expression, alternative splicing, allele specific expression, and site and regional methylation differences between NHW, AA, and HI cases and controls that may drive disease risk differences between the ancestries; 3) combine these data with available DNA genotyping/whole genome sequencing data to identify population and disease-specific expression/methylation QTLs; 4) generate the first multiethnic imputation panel of expression/methylation QTL effects for estimating gene expression traits in diverse (NWH, AA, HI) AD datasets; and 5) apply these panels to a broader set of thousands of multi-ethnic genotyped samples. The multi-omics approach outlined here will provide much needed functional context to the ongoing DNA variant discovery efforts in these populations while addressing the important problem of disparities in AD research. These studies will broaden the spectrum of AD risk to gene expression and methylation and expand existing genomic efforts to a broader AD community.