the same marginal penetrance and would not be found associated with the disease in a one-at-a-time search but only when examined simultaneously. Multivariate statistical models, such as linear or logistic regression, can circumvent these limitations by examining the overall dependency structure between genotypes, phenotype, environmental, and clinical variables. However, traditional regression models require large sample sizes and/or experimental and control samples that are sufficiently different in terms of the phenotype of interest to confer significant power [46]. The amount of data produced by the new genotyping technology requires novel techniques that go beyond “traditional statistical thinking” in order to accommodate the potential complexity of genetic models.
