Unmasking Heterogeneity in Longitudinal Data: Biodemographic Genetic Analyses

Anatoliy I. Yashin, Duke University
Svetlana V. Ukraintseva, Duke University

Genetic variants associated with lifespan may have pleiotropic associations with other traits measured in longitudinal data. In this paper we evaluated how such variants influence age patterns of physiological indices, incidence rates associated with major human diseases, cause-specific and all-cause mortality rates. We performed genome-wide association study of human lifespan using data collected in the Original cohort of the Framingham Heart Study. The key feature of this approach is the proper treatment of missing data, information about age structure of study participants at the time of blood collection, their familial structure, right censoring and possible population stratification. We found that a number of genetic variants associated with lifespan shows pleiotropic associations with other traits: they are also significantly associated with age at onset of cancer and CVD. We conclude that studying genes’ functioning together with aging related changes in phenotypes results in important insights concerning biological mechanisms of lifespan regulation.