Fig. 5
Errors on estimates of genetic parameters (a) and breeding values (b) under sire modeling scenarios for open mating. Black diamond-shaped points indicate the relative bias on genetic variances, and black lines the regression lines of relative errors of variance estimates of worker effects on that of queen effects. The controlled mating strategy in the simulation always used single sire mating (SS). DPQs were open-mated with a heterogeneous drone population. From left to right, the results were obtained by excluding DPQs colony records (O_NoPheno); accommodating for the effect of subpopulations of open-mating drones by: assigning a distinct open-mating pseudo sire in the pedigree for initial BQs and each half of the DPQs (O_TwoPSP); or by adding a fixed (O_FixedGroup) or random (O_RandGroup) non-genetic effect in the statistical model describing the phenotypes, with one level per drone subpopulation