Fig. 3.

Distribution of prediction accuracy of the 15 phenotypic traits in the Elite panel across 50 re-sampling runs estimated by multi-trait models. For each trait, boxplots with different colors represent models. Medians of percent change in prediction accuracy of M and G + M models relative to the G model are indicated below each box in blue if positive and in red if negative. For each model, the uppercase letters before and after the hyphen represent genetic and residual covariance structures: D = diagonal, UN = unstructured, FA = factor-analytic. The Wilcoxon Signed Rank was applied to test difference in prediction accuracy between each omics model and the GBLUP model, and significance levels are indicated above each box. *** = significant at P < 0.001, ** = significant at P < 0.01, * = significant at P < 0.05, NS = not significant