Fig. 5

Results from 100 iterations for simulation scenarios 1–3. The boxplots represent the distributions of estimated variances (posterior modes) from a model with genetic groups and heterogeneous additive genetic variances \(\sigma _{A_1}^2\) and \(\sigma _{A_2}^2\) (left panel), compared to the results from a model that only allowed for a single homogeneous variance \(\sigma _{A}^2\) (right panel). Dashed lines indicate the reference values that were used to generate the data (black: \(\sigma _{A_1}^2\), red: \(\sigma _{A_2}^2\), green: \(\sigma _{E}^2\))