Li C T, Shi C H, Wu J G, Xu H M, Zhang H Z, Ren Y L
Department of Agronomy, College of Agriculture and Biotechnology, Zheijang University, 310029, Hangzhou, China.
Theor Appl Genet. 2004 Apr;108(6):1172-6. doi: 10.1007/s00122-003-1536-1. Epub 2004 Jan 28.
The selection of an appropriate sampling strategy and a clustering method is important in the construction of core collections based on predicted genotypic values in order to retain the greatest degree of genetic diversity of the initial collection. In this study, methods of developing rice core collections were evaluated based on the predicted genotypic values for 992 rice varieties with 13 quantitative traits. The genotypic values of the traits were predicted by the adjusted unbiased prediction (AUP) method. Based on the predicted genotypic values, Mahalanobis distances were calculated and employed to measure the genetic similarities among the rice varieties. Six hierarchical clustering methods, including the single linkage, median linkage, centroid, unweighted pair-group average, weighted pair-group average and flexible-beta methods, were combined with random, preferred and deviation sampling to develop 18 core collections of rice germplasm. The results show that the deviation sampling strategy in combination with the unweighted pair-group average method of hierarchical clustering retains the greatest degree of genetic diversities of the initial collection. The core collections sampled using predicted genotypic values had more genetic diversity than those based on phenotypic values.
为了最大程度地保留初始群体的遗传多样性，在基于预测基因型值构建核心种质库时，选择合适的抽样策略和聚类方法至关重要。在本研究中，基于992个具有13个数量性状的水稻品种的预测基因型值，对构建水稻核心种质库的方法进行了评估。性状的基因型值通过调整无偏预测（AUP）方法进行预测。基于预测的基因型值，计算马氏距离并用于衡量水稻品种之间的遗传相似性。六种层次聚类方法，包括单连锁、中位数连锁、质心、非加权配对组平均、加权配对组平均和灵活β方法，与随机抽样、优先抽样和偏差抽样相结合，构建了18个水稻种质核心种质库。结果表明，偏差抽样策略与非加权配对组平均层次聚类方法相结合，保留了初始群体最大程度的遗传多样性。使用预测基因型值抽样的核心种质库比基于表型值的核心种质库具有更多的遗传多样性。
