We have utilized the mouse as a model system to study the genetics of metabolic and vascular diseases [3–5]. Rather than focus initially on natural or induced mutants of single genes, we utilize the complex endogenous genetic variation between strains in genetic crosses to identify causative genetic loci and ultimately the underlying variations responsible for trait differences [6]. This design more closely reflects the situation faced when studying human populations. The genetic composition of each individual mouse is restricted to that of the two parental strains and is defined at every locus across the genome. The application of quantitative trait locus (QTL) analysis allows the identification of those chromosomal regions that contain a genetic variation that influences trait expression (for a comprehensive review on QTLs see [7]). We [5,8], and others [9–17], have recently extended the power of this approach by incorporating genome-wide gene expression array analysis, which allows us to model the “genetics of gene expression” using similar methods. An immediate extension of this approach is toward dissecting the genetic regulation of complex phenotypes, which would greatly improve the progression from candidate locus to candidate gene.