INTRODUCTIONThe goal of association studies is to discover genetic variation that differs in frequency between cases and controls or between individuals with different phenotypic values. Until a few years ago, the only method available for such studies was low-throughput analysis in which a single gene was selected and either genotyped for known genetic variants or sequenced to identify such variants. With the completion of the Human Genome Mapping Project, we have learned that single-nucleotide polymorphisms (SNPs) are frequent in the genome and that variants in physical proximity tend to correlate in genotype. Therefore, a major international effort was started to map this correlation in the form of the International HapMap Project. The concurrent advances in genetic laboratory techniques, statistical methods, and computing power, coupled with the information from the HapMap, have allowed large-scale microchip-based technologies to be used to assay large numbers of SNPs quickly and easily. Thus, truly genome-wide association studies (GWAS) can now be performed, analogous to linkage studies of Mendelian diseases in having no prior hypothesis of the chromosomal location responsible for disease. In this article, we will only discuss case-control studies in which family members are not analyzed, but the principles apply to large-scale family-based association studies as well.