Biodemographic Approaches to Genetic Analyzes of Longevity in Longitudinal Data on Aging

• Activity Number: 531 - SPEED: Statistics in Epidemiology and Genomics and Genetics
• Type: Contributed
• Date/Time: Wednesday, August 2, 2017 : 11:35 AM to 12:20 PM
• Sponsor: Section on Statistics in Genomics and Genetics
• Author(s): Konstantin Arbeev* and Ilya Y. Zhbannikov and Olivia Bagley and Anatoliy I. Yashin
• Companies: Duke University and Duke University and Duke University and Duke University
• Keywords: longitudinal data ; genetics of longevity ; mortality ; polygenic risk score ; biodemography ; genetic-demographic model

Abstract

Modern longitudinal studies often collect genetic information in addition to follow-up data on mortality or other events. Typically, individuals are genotyped at different ages, and the demographic structure of the genotyped population provides additional information about the effect of genetic variants on the event of interest (along with follow-up data on genotyped and non-genotyped individuals). We present the general genetic-demographic approach which takes such structure into account and describe results of simulation studies which illustrate that combining information on follow-up and information on ages at biospecimen collection improves power in analyses of genetic effects on mortality compared to analyses of follow-up data alone. We also illustrate the approach in application to a genome-wide association study (GWAS) of lifespan in Cardiovascular Health Study (CHS) with genetic data from the CHS Candidate Gene Association Resource. We found that groups of individuals with different values of weighted polygenic risk scores (above/below median) constructed from the top SNPs in GWAS of lifespan (with p-value threshold 0.01) differ in chances to stay free of Alzheimer's disease thus validating further exploration of these findings in analyses of larger scale genetic data.