On Epistasis: A Methodological Review for Detecting Gene-Gene Interactions Underlying Various Types of Phenotypic Traits

Ming      Li; Xiang-Yang      Lou; Qing      Lu

Abstract

Genome-wide association study (GWAS) has become a commonly adopted approach for revealing the genetic architecture of complex diseases, with respect to uncovering the unknown genetic variants involved in the disease, their variations in the population and the magnitude of their effects. Though a substantial number of disease-susceptibility variants have been identified, the genetic architecture of complex diseases has remained elusive. It is unclear how many genetic variants in the human genome are associated with diseases, and how the genetic variants interact with one another to cause diseases. This challenge is partly due to the pervasive gene-gene interactions that underlie complex human diseases. Whereas a number of statistical methods have been developed for detecting gene-gene interactions, they are designed for various purposes, such as a particular study design, the order of the interactions being examined, and the measurement of disease phenotypes. This paper provides a survey of the currently available statistical methods and patents from the perspective of their application to various types of phenotypic traits. We also discuss the strength of each method as well as the biological interpretation of results.

Keywords: Epistasis, statistical methods, genetic association study, phenotypic traits

« Previous Next »