Statistics and Its Interface

Volume 7 (2014)

Number 1

We dedicate this special issue to Dr. Gang Zheng, a great colleague and dear friend to many of us.

A robust test for quantitative trait analysis with model uncertainty in genetic association studies

Pages: 61 – 68

DOI: https://dx.doi.org/10.4310/SII.2014.v7.n1.a7

Authors

Qizhai Li (Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing, China)

Wenjun Xiong (Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing, China)

Jinbo Chen (Department of Biostatistics and Epidemiology, University of Pennsylvania, Philadelphia, Penn., U.S.A.)

Gang Zheng (Office of Biostatistics Research, National Heart, Lung and Blood Institute, Bethesda, Maryland, U.S.A.)

Zhaohai Li (Department of Statistics, George Washington University, Washington, D.C., U.S.A.)

James L. Mills (Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, Maryland, U.S.A.)

Aiyi Liu (Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, Maryland, U.S.A.)

Abstract

Statistical tests that assume an additive model are commonly employed in genetic association studies. However, the true models for genetic variants are rarely known. A mis-specified genetic model may lead to loss of power in identifying the potential markers associated with a disease. In this paper, we develop a robust test based on modified $F$-test statistics for quantitative trait genetic association studies and a simple method to compute its statistical significance and power. We also study sample size calculations for designing such an association study. Numerical results, including simulation studies and a real data example, show that the proposed robust test has satisfactory performance when the model is unknown and is more robust than some existing procedures when the model is mis-specified.

Keywords

$F$-test, robust, quantitative trait, genome-wide association studies

Published 8 April 2014