Using polygenic risk scores and omics to study how suboptimal sleep accelerates cognitive aging in diverse populations

PROJECT SUMMARY Suboptimal sleep, characterized by phenotypes such as insomnia, short sleep, and obstructive sleep apnea, is associated with reduced cognitive performance and increased cognitive decline and dementia risk. There is strong evidence of sleep disparities between race/ethnic and gender minorities, potentially relating to downstream disparities in cognitive aging. However, the mechanisms by which sleep phenotypes accelerate cognitive aging are not well understood. Identifying specific biological pathways and biomarkers of sleep phenotypes will enable unconfounded quantification of the effect of sleep phenotypes on cognitive aging, risk stratification, and development of potential intervention targets. We will establish a collaboration across three longitudinal cohort studies representing the diverse U.S. population: the Atherosclerosis Risk In Communities (ARIC) study, the Hispanic Community Health Study/Study of Latinos (HCHS/SOL), and the Multi-Ethnic Study of Atherosclerosis (MESA), to study how suboptimal sleep accelerates cognitive aging. We will study the associations of sleep disorders (insomnia, short and long sleep, and obstructive sleep apnea) with harmonized, global measures of cognitive aging. We will develop polygenic risk scores (PRSs) for sleep phenotypes using novel methodology, that will appropriately account for heterogeneous genetic ancestry of study individuals, and identify sleep phenotypes that are genetically strongly associated with cognitive aging based on PRSs. We will use omics (metabolomics and proteomics) to explain biological pathways underlying the sleep-cognitive aging associations. We expect that we will identify biomarkers and mediators of the sleep-related effects on cognitive aging, and using Mendelian Randomization analyses, we will be able to untangle some of the directional associations. Finally, we will study risk prediction models for cognitive aging that use sleep-related measures. We will study whether the identified associations and risk prediction models generalize to AD-specific phenotypes in datasets of two cohorts from the Rush Alzheimer's Disease Center. This work will result in the identification of omics biomarkers that measure and mediate the genetic risk of sleep phenotypes on cognitive aging, and new risk models for cognitive aging based on PRSs and omics. Importantly, our study will include individuals from diverse U.S. populations, and our methodology will ensure that the findings and models are useful across populations. This work will guide precision medicine initiatives to improve cognitive health in aging individuals from diverse genetic backgrounds.