Combining the MAX test with methods for Family-wise error rate

Abstract

In chapter 1 the necessary biological background for this thesis is provided. Here the definition of a SNP is provided and the connection between the human genom, DNA and SNPs. Chapter 2 gives the notation and definitions that is needed to understand the basis of the Cochran-Armitage trend test which is described in detail in chapter 3. Along with the Cochran-Armitage trend test is the definition of the MAX-test given and also three different methods on how to find its distribution since it is not given asymptotically. New material here is the indirect method and using permutation to find the distribution of MAX. In chapter 4 the analysis of multiple SNPs are given by controlling the family-wise error rate. Two specific procedures were studied, the single-step procedure and the step-down procedure. These two procedures were used in the analysis of data in chapter 5, while in chapter 6 measures on linkage disequilibrium between pairs of SNPs are given. New in this chapter is the measure of linkage disequilibrium in case-control data, of which there is no documentation given anywhere have been done before. Last, in chapter 7 is the conclusion of this thesis.