Detecting the Common and Individual Effects of Rare Variants on Quantitative Traits by Using Extreme Phenotype Sampling.

Ya-Jing Zhou,Li-Li Chen,Yong Wang

doi:10.3390/genes7010002

Abstract

Next-generation sequencing technology has made it possible to detect rare genetic variants associated with complex human traits. In recent literature, various methods specifically designed for rare variants are proposed. These tests can be broadly classified into burden and nonburden tests. In this paper, we take advantage of the burden and nonburden tests, and consider the common effect and the individual deviations from the common effect. To achieve robustness, we use two methods of combining p-values, Fisher’s method and the minimum-p method. In rare variant association studies, to improve the power of the tests, we explore the advantage of the extreme phenotype sampling. At first, we dichotomize the continuous phenotypes before analysis, and the two extremes are treated as two different groups representing a dichotomous phenotype. We next compare the powers of several methods based on extreme phenotype sampling and random sampling. Extensive simulation studies show that our proposed methods by using extreme phenotype sampling are the most powerful or very close to the most powerful one in various settings of true models when the same sample size is used.

Highlights

Hundreds of common genetic variants associated with many complex diseases and human traits have been successfully identified by the genome-wide association studies (GWAS)
These common genetic variants with minor allele frequencies (MAF ) >3% have small to moderate effects, and explain only a small fraction of disease heritability for common disease [1,2,3,4]
We explore the advantage of the extreme phenotype sampling in rare variant analysis and refine this design framework for future large-scale association studies on quantitative traits

Summary

Introduction

Hundreds of common genetic variants associated with many complex diseases and human traits have been successfully identified by the genome-wide association studies (GWAS). These common genetic variants with minor allele frequencies (MAF ) >3% have small to moderate effects, and explain only a small fraction of disease heritability for common disease [1,2,3,4]. Next-generation sequencing technology will soon sequence the whole genome of large groups of individuals and will make testing rare variants possible. Rare variants are difficult to detect even with large sample size.

Methods

Results

Discussion

Conclusion