We introduce a method for the analysis of multilocus, multitrait genetic data that provides an intuitive and precise characterization of genetic architecture. We show that it is possible to infer the magnitude and direction of causal relationships among multiple correlated phenotypes and illustrate the technique using body composition and bone density data from mouse intercross populations. Using these techniques we are able to distinguish genetic loci that affect adiposity from those that affect overall body size and thus reveal a shortcoming of standardized measures such as body mass index that are widely used in obesity research. The identification of causal networks sheds light on the nature of genetic heterogeneity and pleiotropy in complex genetic systems.

Obesity is a highly heritable and genetically complex trait with hundreds of potential loci identified. An intercross of 513 F2 progeny between the SM/J x NZB/BINJ inbred mouse strains was generated to identify quantitative trait loci (QTL) that are involved in the weight of four fat pads: mesenteric, inguinal, gonadal, and retroperitoneal. Sex and lean body weight were treated as covariates in the analysis of these fat pads. This analysis uncoupled genetic effects related to overall body size from those influencing the adiposity of a mouse. We identified multiple significant QTL. QTL alleles associated with increased lean body weight and individual fat pad weights are contributed by the NZB background. Adiposity loci are distinct from these body size QTLs and high-adiposity alleles are contributed by the SM background. An extended network of epistatic QTL is also observed. A QTL on Chr 19 is the center of a network of eight interacting QTL, Chr 4 is the center of six, and Chr 17 the center of four interacting QTL. We conclude that interacting networks of multiple genes characterize the regulation of fat pad depots and body weight. Haplotype patterns and a literature-driven approach were used to generate hypotheses regarding the identity of the genes and pathways underlying the QTL.