|
Abstract:
|
Genome wide association studies (GWAS) are used to investigate genetic variants contributing to complex traits. Despite discovering many loci, a large proportion of "missing" heritability remains unexplained. Gene-gene interactions may help explain some of this "missing" heritability. Gene-gene dependence is an important way to measure gene-gene interaction in gene level. Traditional parametric statistical methods such as linear and logistic regression, with multi-factor dimensionality reduction (MDR) are used to address sparseness of data in high dimensions in traditional parametric methods. We propose a method for the analysis of gene-gene dependence, for unrelated SNPs on two genes. Typical methods on this problem using statistics having an asymptotic chi-squared mixture distribution, which is not easy to use. Here, we propose a Kullback-Leibler type statistic, which is an asymptotic positive normal distribution under the null hypothesis of no relationship between the two genes (SNPs), and normal under the alternative. The performance of the proposed method is evaluated by simulation studies, which show promising results. Then the method is used to analyze a real data, and identify gene-gene dependences each other among RAB3A, MADD, and PTPRN with Type 2 Diabetes (T2D) status.
|
