Which Genes are Involved? - Statistical Planning and Analysis of Human Genetic Samples
This thesis consists of six papers related to the statistical planning and analysis of human genetic samples. The first four mainly concern statistical methodology, while the last two are applied papers in genetics and medicin. In paper A an alternative method of estimating the locus specific sibling relative risk .lambda.s(HLA) is derived, assuming a multiplicative model in coeliac disease. The method is compared to the commonly used .lambda.s(HLA)=0.25/p0. In paper B the nonparametric mean test is modified by using weighted IBD scores based on model assumptions. Selective sampling is also considered. In paper C some TDT-like statistics for detecting linkage disequilibrium with nuclear families of one of more affected children are introduced and compared. In paper D methods for multiple inference in association analysis of many loci using a case-control design are investigated. The keys to the methods are closed tests and permutation of case-control labels. In paper E the candidate gene CTLA4 is showed to be associated to coeliac disease in a Scandinavian population. In paper F the incidence of CSF abnormalities in siblings of multiple sclerosis (MS) patients and unrelated controls are examined in order to identify a subphenotype that can be utilised in the search for disease genes.