There can be variation between animals in how stable their genetic merit is across different environments due to genotype-by-environment (G×E) interactions. This variation could be used in breeding programs to select robust genotypes that combine high overall performance with stable genetic ranking across environments. There have been few attempts to validate breeding values for robustness in livestock, although this is a necessary step towards their implementation in selection decisions. The objective of this study was to validate breeding values for the robustness of body weight across different growth environments that were estimated using reaction norm models in sheep data. Using threefold cross-validation for the progeny of 337 sires, the average correlation between single-step breeding values for the reaction norm slope and the realised robustness of progeny across different growth environments was 0.21. The correlation between breeding values for the reaction slope estimated independently in two different datasets linked by common sires was close to the expected correlation based on theory. Slope estimated breeding values (EBV) obtained using reaction norm models were predictive of the phenotypic robustness of progeny across different environments and were consistent for sires with progeny in two different datasets. Selection based on reaction norm EBV could be used to increase the robustness of a population to environmental variation.

中文翻译：

---

澳大利亚绵羊稳健性反应标准育种值的验证


由于基因型与环境 (G×E) 的相互作用，动物之间的遗传价值在不同环境中的稳定性可能存在差异。这种变异可用于育种计划，以选择稳健的基因型，将高整体性能与跨环境的稳定遗传排名结合起来。尽管这是在选择决策中实施的必要步骤，但很少有人尝试验证牲畜稳健性的育种值。本研究的目的是验证不同生长环境下体重鲁棒性的育种值，这些值是使用绵羊数据中的反应规范模型估计的。对 337 头公牛的后代进行三重交叉验证，反应规范斜率的单步育种值与不同生长环境下后代实现的稳健性之间的平均相关性为 0.21。在由共同公牛链接的两个不同数据集中独立估计的反应斜率育种值之间的相关性接近基于理论的预期相关性。使用反应范数模型获得的斜率估计育种值（EBV）可以预测不同环境中后代的表型鲁棒性，并且对于具有两个不同数据集中的后代的公牛来说是一致的。基于反应范数 EBV 的选择可用于提高群体对环境变化的鲁棒性。