There is another subtle, yet fundamental difference in the way that the PoPVg is calculated in GWAS and HBCGM studies, which also impacts the PoPVg estimate. A quantitative trait can be modeled by the equation “trait value ~ G + E + G*E + residual”; where G, E and G*E represents the genetic effect, the environmental effect and their interaction, and the “residual” represents the variation that can’t be explain by G or E. Many murine or human traits can be highly variable, even when repeatedly assayed in the same human subject or mouse strain. (The environmental variation in a human study can be exceedingly large.) As a result, the unexplained variation can be large, which leads to a small genetic effect (G). Multiple phenotypic measurements are made for each strain under controlled conditions, and HBCGM uses only the average value of the strain replicates. As a result, the environmental effects are eliminated (E and G*E) in a murine study, and the un-explained variation is also minimized. These factors increase the PoPVg in a HBCGM study, which increase the range of traits that can successfully evaluated by HBCGM.
