ABSTRACT
Although great progress in GWAS has been made, the significant SNP associations identified by GWAS account for only a few percent of the genetic variance. Searching for remaining genetic variance is a great challenge. One way to discover remaining genetic variance or “missing heritability” is to study gene–gene and gene–environment interaction. The interactions hold a key for dissecting the genetic structure of complex diseases and elucidating the biological and biochemical pathway underlying the diseases.
This chapter first reviews the odds ratio, disequilibrium, and information measure of gene–gene and gene–environment interaction and introduces the relative risk, odds-ratio, disequilibrium, and information measure-based statistics for testing the gene–gene and gene–environment interaction. The classical statistical methods for gene–gene and gene–environment interaction detection are designed for common variants; they are not suitable for rare variants and next-generation sequencing data. To deal with next-generation sequencing data, this chapter also covers the current development of gene-based gene–gene and gene–environment interaction analysis: function regression models for single quantitative trait, multivariate function regression models for multiple quantitative traits, canonical correlation analysis for both single and multiple quantitative traits, and functional logistic regression for qualitative traits. All gene-based methods are designed for interaction analysis with next-generation sequencing data.
