The types of genetic association studies conducted are commonly divided into candidate gene/region and genome-wide association studies. Both of these approaches involve genotyping SNPs in large collections of cases and controls (the focus of this protocol) or in large collections of individuals characterised by continuous quantitative traits. Although the general principles by which markers are selected in candidate gene vs genome-wide studies are the same, i.e. based on optimal genomic coverage, the investigator-driven methodology that is implemented in a candidate gene/region approach means that in practice they are very different. A candidate gene study is based on a prior hypothesis suggesting a potential role of the gene(s) in a particular phenotype or disease. The support for the selection of a candidate gene or region is typically based on its biological function or on its location in a region implicated in a previous linkage or association study. In a candidate gene study, the aim is to obtain the highest possible coverage of genetic variation within specified genomic boundaries, taking account of any known functional genomic characteristics. In contrast, genome-wide association studies
