Figure 7

Accuracy of selection in breeds A and B in five scenarios. (a) and (b) Results for a low correlation of LD phase between breeds A and B (r = 0.2 for markers 1 cM apart) (c) and (d) Results for a high correlation of LD phase between breeds A and B (r = 0.7 for markers 1 cM apart). The plotted accuracies are means from 30 replicates. Sc. Ref: Selection criteria in both breed A and B was for purebred performance (P) and both breeds had Separate training sets. Sc.1: Selection criteria in breed A was for crossbred performance (C) and selection criteria in breed B was for purebred performance and both breeds had separate training sets. Sc.2: Selection criteria in both breed A and B was for crossbred performance and both breeds had separate training sets. Sc.3: Selection criteria in breed A was for crossbred performance and selection criteria in breed B was for purebred performance and both breeds had a Common training set. Sc.4: Selection criteria in both breed A and B was for crossbred performance and both breeds had a common training set. It should be noted that accuracies in this Figure are correlations between the selection criterion and the EBV of interest. Thus, when selection is for purebred performance, accuracy is the correlation between GEBVP and TBVP, while when selection is for crossbred performance, accuracy is the correlation between GEBVC and TBVC.