Investigating the Multivariate Genetic Architecture of Aging-Related Traits

Aging encompasses a range of phenotypes, including lifespan, healthspan, aging-related diseases, and other
aging-related outcomes. Aging-related phenotypes are correlated with each other but not perfectly. We will
develop a multivariate genomic model of aging-related outcomes and diseases in order to 1) understand whether
aging-related phenotypes are best described individually, clustered in small groups, or as a single group and 2)
identify genetic variants that underly processes shared across multiple aging-related phenotypes. First, we will
examine genetic correlations among aging-related phenotypes such as Alzheimer's disease, longevity, epigenetic
aging, telomere length, parental lifespan, healthspan, frailty, and other aging-related diseases (e.g. Parkinson's
disease). We will then examine patterns of genomic covariation among aging-related phenotypes using Genomic
Structural Equation Modeling. Finally, we will perform multivariate genome-wide association studies and derive
polygenic scores from these GWAS.

A wide range of characteristics and behaviors are related to aging, including lifespan, aging-related diseases such
as Alzheimer's disease and Parkinson's disease, and frailty. Aging-related characteristics are genetically
correlated, meaning that some genetic variants affect more than one aging-related phenotype (e.g. some genetic
variants may confer risk for Alzheimer's disease and may be related to shorter lifespan). We aim to identify how
different aspects of aging are genetically related to each other, and whether we can use broader groups of
correlated phenotypes to identify genetic variants related to certain aspects of aging.