Research Project
10301113
SUMMARY Alzheimer's disorder (AD) is a devastating neurodegenerative disease and the most common cause of dementia. There are approximately six million Americans with AD and 29.8 million worldwide, making AD one of the most pressing public health issues as the world's population continues to age. Presently, there is no known effective prevention or cure exists, and current AD medications only alleviate symptoms or slow decline rates. The landscape of AD drug trials is gloomy. One possible reason is that AD is a heterogeneous disorder but trials are designed treating it as a monolithic disease. Although lifestyle and environmental risk factors clearly affect AD, the primacy of genetic influences suggests that categorization by genetic basis should be prioritized in developing effective interventions. Genetics can offer insights on risk prediction, disease mechanism, and new therapeutic targets. Heritability of AD estimates range from 49-79%, but the conventional single nucleotide variants (SNVs) identified to date only account for <50% of AD heritability. Multiple studies have highlighted the roles of copy number variants (CNVs) in AD. We hypothesize that a systematic investigation of genome-wide CNVs at the full spectrum (i.e. small and large in size, common and rare in frequency, and coding and no-coding in genomic regions) from whole-genome sequencing (WGS) can further enhance the knowledge of AD etiology and risk. Leveraging the rich resources from the Alzheimer's Disease Sequencing Project (ADSP), we propose to focus on a large multi-ethnic WGS sample (n>17,000) composed of AD cases and normal healthy elderly controls, and to (1) detect and genotype CNVs from WGS for ADSP case-control samples; (2) perform association analysis to identify genome regions of CNVs contributing to AD; and (3) conduct cross-ethnic association studies to find ethnic-shared or ethnic-unique AD-associated CNVs. Successful completion of our aims will provide (i) the first large-scale CNV investigation of AD genetics using WGS data; (ii) new CNV calling method for WGS based on the current best practices; (iii) new CNV association strategies to address issue of breakpoint non-alignment and enhance association power; (iv) multi-ethnic characterization of shared and unique CNV risk factors for AD; and (v) optimized computational pipelines with open- source code and released standardized images (e.g., Docker images and Bioconductor packages) that are easily deployable in other large-scale WGS association projects.
