基于元共祖概念的猪基因组选择一步法的发展研究

Single-step method is the standard method for routine genomic selection in pigs, but it has several issues that need to be solved, including the allelic frequencies in the base population are unknown; genomic relationships and pedigree-based relationships cannot be compatible directly; and the trait-specific optimal weighting factors between the pedigree-based and marker-based relationship matrices are unknown, and single-step method cannot be effectively used in multi-breed genomic evaluation. Metafounder is assumed as the starting point of a population and it is an infinitely gametic pool. The development of metafounder provides new solutions for the issues in the single-step method, but until now, how to use single-step method with metafounders to solve the above mentioned issues has not been well investigated. Therefore, in this project, by using both simulated and real datasets, we plan to systematically study single-step method with metafounder for average daily gain and feed conversion ratio in Danish Yorkshire and American Yorkshire: first, the feasibility and reliability of using single-step method with metafounders in pigs will be explored in both univariate-trait and bivariate-trait models; then the optimal weighting factors between the pedigree-based and marker-based relationship matrices for these growth traits will be investigated; next, the performances of genomic evaluation in the two different single-step methods will be compared; finally, multi-population genomic evaluation by using a single-step method with metafounders will be studied for the first time. Its accuracy will be compared to that in within-population genomic evaluation and the significance analysis will be done for the differences between the accuracies in these two single-step methods. This project is a new development for genomic selection. The outcomes could further improve the accuracy of single-step method and also extend the scope of the application of single-step method.

一步法是猪日常基因组选择的标准方法，但存在基础群基因频率未知、基因型和系谱矩阵不直接兼容、各性状最适基因型和系谱矩阵权重比未知、且不能有效开展多群体遗传评估等问题。元共祖被假定为一个群体的开始，是一个无穷大的配子池。元共祖的创立，为解决一步法中问题提供了新思路，但如何利用元共祖解决上述问题尚未得到深入研究。本研究拟结合模拟和真实数据，以丹系和美系大白平均日增重和饲料转化率为研究对象，系统探索利用元共祖概念发展一步法的方法学问题，包括首次利用单、双性状预测模型，进行使用含元共祖的一步法进行猪遗传评估的可行性和可靠性研究；探索目标性状在两种一步法中最适基因型和系谱矩阵权重比，比较两种一步法在遗传评估中的优劣性。此外，本研究拟首次探索使用含元共祖的一步法进行多群体遗传评估，将其准确性与单群体结果做差异显著性分析。本研究是对猪基因组选择的新发展，预期成果将提高一步法的准确性，扩展一步法的适用范围。

基因组选择一步法是猪日常基因组选择的标准方法，但存在基础群基因频率未知、基因型和系谱矩阵不直接兼容、各性状最适基因型和系谱矩阵权重比未知、且不能有效开展多群体遗传评估等问题。随着元共祖概念的创立，其为解决一步法中问题提供了新思路，但如何利用元共祖解决上述问题在项目资助之前尚未得到深入研究。本研究结合了真实数据，以丹系大白猪，丹系长白猪以及美系大白猪平均日增重、饲料转化率和精液品质性状为研究对象，系统探索利用元共祖概念发展一步法的方法学问题，包括首次利用单、双性状预测模型，进行了使用含元共祖的一步法进行猪遗传评估的可行性和可靠性研究；探索目标性状在两种一步法中最适基因型和系谱矩阵权重比，比较两种一步法在遗传评估中的优劣性。此外，本研究探索了多群体猪基因组遗传评估，将其准确性与单群体结果做差异显著性分析，以探究联合遗传评估的可行性。本研究还鉴定出影响猪饲料转化率和精液品质的主效候选基因和关键信号通路。本研究是对猪基因组选择的新发展，预期成果将提高一步法的准确性，扩展一步法的适用范围。目前，本研究的核心算法和软件已成果转化至国家核心育种场，将在实践中落地，有望未来进一步提升种猪育种效率。

内容获取失败，请点击重试