Department of Animal Production, NEIKER-BRTA Basque Institute of Agricultural Research and Development, Agrifood Campus of Arkaute s/n, E-01080 Arkaute, Spain.
Departamento de Anatomía Embriología y Genética Animal, Instituto Agroalimentario de Aragón (IA2), Universidad de Zaragoza, 50013 Zaragoza, Spain.
J Dairy Sci. 2021 Jun;104(6):6861-6872. doi: 10.3168/jds.2020-19978. Epub 2021 Mar 25.
Genomic selection has been implemented over the years in several livestock species, due to the achievable higher genetic progress. The use of genomic information in evaluations provides better prediction accuracy than do pedigree-based evaluations, and the makeup of the genotyped population is a decisive point. The aim of this work is to compare the effect of different genotyping strategies (number and type of animals) on the prediction accuracy for dairy sheep Latxa breeds. A simulation study was designed based on the real data structure of each population, and the phenotypic and genotypic data obtained were used in genetic (BLUP) and genomic (single-step genomic BLUP) evaluations of different genotyping strategies. The genotyping of males was beneficial when they were genetically connected individuals and if they had daughters with phenotypic records. Genotyping females with their own lactation records increased prediction accuracy, and the connection level has less relevance. The differences in genotyping females were independent of their estimated breeding value. The combined genotyping of males and females provided intermediate accuracy results regardless of the female selection strategy. Therefore, assuming that genotyping rams is interesting, the incorporation of genotyped females would be beneficial and worthwhile. The benefits of genotyping individuals from various generations were highlighted, although it was also possible to gain prediction accuracy when historic individuals were not considered. Greater genotyped population sizes resulted in more accuracy, even if the increase seems to reach a plateau.
基因组选择在过去几年中已在多个家畜物种中得到应用，因为它可以实现更高的遗传进展。在评估中使用基因组信息可提供比基于系谱评估更高的预测准确性，而基因分型群体的构成则是一个决定性因素。本研究旨在比较不同基因分型策略（动物数量和类型）对拉塔基亚奶绵羊品种预测准确性的影响。基于每个群体的真实数据结构设计了一项模拟研究，并使用遗传（BLUP）和基因组（单步基因组 BLUP）评估不同基因分型策略获得的表型和基因型数据。当雄性是遗传关联个体并且有具有表型记录的女儿时，对其进行基因分型是有益的。对具有自身泌乳记录的雌性进行基因分型可以提高预测准确性，而连接水平的相关性较小。雌性基因分型的差异与其估计育种值无关。无论采用哪种雌性选择策略，雄性和雌性的联合基因分型都提供了中等的准确性结果。因此，假设对公羊进行基因分型很有趣，那么加入基因分型的雌性将是有益且值得的。强调了对来自不同世代的个体进行基因分型的好处，尽管不考虑历史个体也有可能获得预测准确性。更大的基因分型群体可提高准确性，即使增加似乎达到了一个平台期。
