Robust Linear Regression Methods in Association Studies

• Activity Number: 301
• Type: Contributed
• Date/Time: Tuesday, August 2, 2011 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Vanda Milheiro Lourenço*+ and Ana Maria Pires and Matias Kirst
• Companies: Technical University and Technical University and University of Florida
• Keywords: Non-normality ; Outlier ; SNP ; Power ; M-regression ; Least squares

Abstract

Data normality is a mathematical convenience. In practice, experiments usually yield data with nonconforming observations. In the presence of this type of data, classical least squares statistical methods perform poorly, giving biased estimates, raising the number of spurious associations and often failing to detect true ones. Robust statistical methods are designed to accommodate certain types of data deficiencies, allowing for reliable results under various conditions. We analyze the case of statistical tests to detect associations between genomic individual variations (SNPs) and quantitative traits when deviations from the normality assumption are observed. We consider the classical ANOVA tests for the parameters of the appropriate linear model and a robust version of those tests based on M-regression. We show through a simulation study and a real data example, that the robust methodology can be more powerful and thus more adequate for association studies than the classical approach.