Multi-step Correlation Test Based On The Genome Of Any Family Structure Data

Genome-wide association studies are now widely used approaches in the tradi-tional epidemiology, and genome-wide analysis is an effective way in searching for loci that affect complex traits. Now, there are many methods for association study, for example, although genetic association studies using unrelated individuals may have much more power, it is not robust to population admixture. The genome-wide family-based association studies are still necessary. Family-based genome-wide as-sociation studies usually have much more advantage during the association studies, and it is robust to population admixture especially. With the development of the sci-ence and technology, human obtain an increasing number of gene loci. The success of genome-wide association studies will depend on whether the increased number of SNPs can be translated into increased statistical power or whether the positive effects of the increased number of SNPs will be diluted by the multiple-comparison problem. Steen et al and Tao Feng et al have proposed some two-stage designs. The basic idea is that the first stage is a screening test with the whole SNPs and left part of the SNPs. The second stage is an association test with the selected SNPs. The approaches have much more power than the one-stage traditional genome-wide association studies. In this article, we proposed two two-stage approaches, we in-corporate genotypes of founders in the pedigree into both the screening test and the association test. Our simulation results show that the approaches we proposed have the correct typeâ… error rates and much more powerful.