Online Program Home
My Program

Abstract Details

Activity Number: 245
Type: Contributed
Date/Time: Monday, August 1, 2016 : 2:00 PM to 3:50 PM
Sponsor: Section on Statistics in Epidemiology
Abstract #320945
Title: Characterizing Uncertainty in Genetic Association Landscapes by Functional Bayesian Bands
Author(s): Olga A. Vsevolozhskaya* and Ilai Keren and Dmitri Zaykin
Companies: University of Kentucky and Washington Department of Fish and Wildlife and National Institute of Environmental Health Sciences
Keywords: allelic effects ; functional linear models ; Bayesian statistics ; sequencing data ; pleiotropy ; association test

In a recent article (Vsevolozhskaya et al, 2016), we proposed a novel approach based on functional linear models to estimate allelic effects of multiple variants as a smooth function varying over consecutive genetic positions. Although the approach answered lots of interesting questions, e.g., helped us estimate pleiotropic effects or the so-called 'treatment-by-trait' interaction, it has opened doors to new problems. For instance, to infer statistical significance of a genetic effect, one may construct pointwise confidence bands at each variant's positions, which are subject to multiple testing issue. Additionally, the estimated smooth functions may be affected by the "winner's curse," -- a phenomenon that may exaggerate the magnitude of the estimated effect from the initial study. To overcome th?s? issues, we extend our functional approach to Bayesian framework, which provides a statistical solution immune to both the problem of multiple testing and the winner's curse. We show the validity of our approach through an extensive simulation study and real data application.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

Copyright © American Statistical Association