Because of their simplicity, population-based association studies of candidate genes have been widely used and perhaps abused. There are several major problems with this approach in alcoholism research. The first problem is the choice of candidate gene. Numerous biochemical pathways are likely to be involved in addictive behavior, including pathways related to alcohol metabolism. These pathways involve numerous enzymes and other molecules, and the genes encoding all of these molecules are therefore potential candidate genes. Thus, the number of candidate genes is large and increases daily with the application of new technologies to alcohol research. The second problem is that with a disorder that involves multiple genes, such as alcoholism, the effects of each gene are probably small. Therefore, large sample sizes in multiple populations are often required to detect such genes. Third, spurious associations between certain alleles and the disorder are likely to show up if the two samples (i.e., alcoholics and nonalcoholics) are not well matched with respect to important characteristics such as ethnicity.
