利用基因组选择校正猕猴桃育种中的系谱错误。

Using genomic selection to correct pedigree errors in kiwiberry breeding.

作者信息

Mertten Daniel, McKenzie Catherine M, Thomson Susan, McCallum John, Andersen Dave, Baldwin Samantha, Lenhard Michael, Datson Paul M

机构信息

The New Zealand Institute for Plant and Food Research Limited (PFR), Auckland, 1142 New Zealand.

Institute for Biochemistry and Biology, University of Potsdam, 14476 Potsdam-Golm, Germany.

出版信息

Mol Breed. 2025 Mar 11;45(3):33. doi: 10.1007/s11032-025-01552-6. eCollection 2025 Mar.

DOI:10.1007/s11032-025-01552-6

PMID:40078838

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC11896956/

Abstract

UNLABELLED

In breeding programmes, accurate estimation of breeding values is crucial for selecting superior genotypes. Traditional methods rely on phenotypic observations and pedigree information to estimate variance components and heritability. However, pedigree errors can significantly affect the accuracy of these estimates, especially in long-lived perennial vines. This study evaluates the effect of pedigree errors on breeding value predictions in kiwiberry breeding and explores the benefits of using genomic selection. We applied Best Linear Unbiased Prediction (BLUP) to estimate breeding values for each genotype for a given trait. Four scenarios with varying degrees of alteration in pedigree-based relationship matrices were used to represent inaccurate relationships between genotypes. Pedigree-based breeding values were compared with genomic estimated breeding values for one vine-related and four fruit-related quantitative traits. The results showed that as the degree of altered population structure increased, the prediction accuracy of pedigree-based breeding values decreased. In contrast, genomic selection, which uses marker inheritance, maintained realised relationships between genotypes, making it a more robust method for predicting genetic merit. In kiwiberries, as in all species of the genus , only female vines bear fruit. The genotypic merit of fruit-related traits in male genotypes can only be estimated indirectly. Marker-based predictions outperformed pedigree-based predictions, especially for genotypes without phenotypic observations, such as male siblings. This study reviewed the induced population structures and introduced genomic selection into the kiwiberry breeding programme. We demonstrated that genomic selection provides more accurate breeding values by capturing true genetic relationships and reducing the effects of misidentified relationships between individuals.

SUPPLEMENTARY INFORMATION

The online version contains supplementary material available at 10.1007/s11032-025-01552-6.

摘要

未标注

在育种计划中，准确估计育种值对于选择优良基因型至关重要。传统方法依靠表型观察和系谱信息来估计方差分量和遗传力。然而，系谱错误会显著影响这些估计的准确性，尤其是在多年生藤蔓植物中。本研究评估了系谱错误对猕猴桃育种中育种值预测的影响，并探讨了使用基因组选择的益处。我们应用最佳线性无偏预测（BLUP）来估计给定性状下每个基因型的育种值。使用基于系谱的亲缘关系矩阵中不同程度改变的四种情景来表示基因型之间不准确的关系。将基于系谱的育种值与一种与藤蔓相关和四种与果实相关的数量性状的基因组估计育种值进行比较。结果表明，随着群体结构改变程度的增加，基于系谱的育种值预测准确性降低。相比之下，利用标记遗传的基因组选择维持了基因型之间的实际关系，使其成为预测遗传价值更稳健的方法。在猕猴桃中，如同该属的所有物种一样，只有雌性藤蔓结果。雄性基因型中与果实相关性状的基因型价值只能间接估计。基于标记的预测优于基于系谱的预测，尤其是对于没有表型观察的基因型，如雄性同胞。本研究回顾了诱导的群体结构，并将基因组选择引入猕猴桃育种计划。我们证明，基因组选择通过捕捉真实的遗传关系并减少个体间错误识别关系的影响，提供了更准确的育种值。