Statistical Methods Based On Correlation Analysis Of Haploid

Now, candidate genes association tests are performed by using several intragenicSNPs, simultaneously, by testing SNP haplotype or genotype effects in multi-factorialdiseases or traits. It is widely recognized that multi-marker haplotypes are moreinformative than a single marker, and there is a strong evidence that several mutationswithin genes can interact each other that has a large effect on the observed phenotype.But, the number of haplotypes increases with the increase in the number of typedSNPs,. As a result, large numbers of haplotypes will introduce large degrees of freedomin haplotype-based tests, and thus limit the power of the tests.In this article, we propose using the principal component method of statistics toreduce the dimension of tests, and then construct association tests on thelower-dimensional space to test the association between haplotypes and a quantitativetrait using population-based samples. The proposed method allows ambiguoushaplotype. We use simulation studies to evaluate the typelerror rate of the tests, andcompare the power of the proposed tests with that of the tests without dimensionreduction, and the tests with dimension reduction by merging rare haplotypes. Oursimulation results show that the proposed tests have correct type I error rates and aremore powerful than other tests in most cases considered in our simulation studies.