Why Demographic Analysis Is Crucial in Studying Genetics of Exceptional Longevity

Anatoliy I. Yashin, Duke University
Deqing Wu, Duke University
Alexander Kulminski, Duke University
Liubov Arbeeva, Duke University

The genome-wide association studies (GWAS) of human longevity are often used to identify genetic determinants of exceptional lifespan. An important problem in such analyses is possible genetic difference in ancestry in subpopulations of study participants. To control for such population stratification the use of Principal Component Analyses is recommended. In this approach several first principal components of genetic covariance matrix (which presumably capture genetic variability induced by population stratification) are included as additional covariates in the statistical procedure of selecting genetic variants in GWAS. In this paper we performed GWAS of human lifespan using data on the Original cohort of the Framingham Heart Study and show that such recommendation has to be used with care. This is because population stratification may be generated by demographic process of mortality selection in genetically heterogeneous population. Controlling for population stratification in this case may remove genes we are looking for from the analyses

  See extended abstract

Presented in Session 92: The Oldest Old: Determinants of Very Long Lifespans