Investigation of multi-ancestry genomic variation, telomere length, and molecular biomarkers in Alzheimer disease
Embargo Date
2027-05-12
OA Version
Citation
Abstract
Alzheimer’s disease (AD) is a complex neurodegenerative disorder characterized by amyloid plaques and neurofibrillary tau tangles. Although AD is highly heritable, much of its genetic architecture, particularly across diverse ancestral populations, remains unresolved. Telomere length (TL), a biomarker of biological aging and cellular senescence, has been implicated in age-related diseases, yet its role in AD pathogenesis and genetic risk modulation is not fully understood. To investigate ancestry-specific and shared genetic contributors to AD, I analyzed merged whole-genome sequencing (WGS) and whole-exome sequencing datasets from multi-ancestry cohorts. Using exome-wide association analyses and gene-based rare variant aggregation tests, I identified rare coding variants with protective effects against AD, including novel study-wide significant variants in BTBD8, LINGO1, and KCNG2, as well as gene-based signals in LINGO1, DYNLT4, and ADCY4. To further explore the intersection of aging biology and genetic risk, I estimated TL from large multi-ancestry WGS data and conducted a genome-wide SNP×TL interaction analysis to identify genetic variants whose effects on AD risk are modified by TL. Genome-wide SNP×TL interaction analyses revealed loci in which genetic effects on AD risk were significantly modified by TL, including variants in SEMA6A, BSN, and MST1, suggesting that telomere maintenance pathways may influence genetic susceptibility to AD. Downstream transcriptomic analysis was performed to better characterize the functional relevance of the lead loci. Finally, I examined associations between TL and AD-related molecular phenotypes, including fluid biomarkers, metabolites, and plasma proteins, to elucidate potential biological pathways linking telomere dynamics to neurodegeneration. Longer TL was associated with reduced AD risk and with several AD-related biomarkers, including plasma Aβ40 and cerebrospinal fluid neurofilament light chain. Mediation analyses further suggested that certain circulating biomarkers, including follicle-stimulating hormone in females, may partially mediate the relationship between TL and AD risk. Collectively, this thesis utilizes multi-ancestry genomic data, TL, and multi-omic biomarker data to refine our understanding of AD genetic architecture and to identify biologically plausible gene candidates and pathways contributing to disease pathology.
Description
2026