Regression Models for Haplotype-Based Association Studies.

Oh,  So Hee; Namkung,  Jung Hyun; Park,  Tae Sung

Genomics Inform > Volume 5(1); 2007 > Article

Original Article

Genomics & Informatics 2007;5(1): 19-23.

Regression Models for Haplotype-Based Association Studies.

So Hee Oh, Jung Hyun Namkung, Tae Sung Park

¹Department of Statistics, Seoul National University, Seoul 151-742, Korea. tspark@stats.snu.ac.kr
²Interdisciplinary Program in Bioinformatics, Seoul National University, Seoul 151-742, Korea.

Abstract

In this paper, we provide an overview of statistical models for haplotype-based association studies, and summarize their features based on the design matrix. We classify the design matrix into the two types: direct and indirect. For these two kinds of matrices, we present and compare characteristics using a simple hypothetical example, and a real data set. The motivation behind this study was to provide practitioners with an improved understanding, to facilitate the informed selection of the appropriate haplotype-based model and to improve the interpretability of the models.

Keywords: Association; Case-control Study; Design Matrix; Generalized Linear Model; Haplotype