Abstract: Methods of constructing core collections for rice were used to evaluate 992 rice varieties with 13 quantitative traits. The genotypic values of the varieties for the traits studied were predicted by using the adjusted unbiased prediction (AUP) method. Based on the genotypic values, the Mahalanobis distances were calculated and employed to measure the genetic similarities among rice varieties. Three hierarchical clustering methods, the complete linkage method, the Ward’s method and the unweighted pair-group average method, combining with three sampling strategies including the random sampling, the preferred sampling and the deviation sampling, were proposed to develop nine core collections of rice germplasm. Results showed that larger genetic variation of the initial collection could be retained by using the deviation sampling method along with the unweighted pair-group average method of hierarchical clustering. The core collections based on genotypic values had more genetic diversities than those of phenotypic values.

Key words: rice, core collection, genotypic value, germplasm

摘要： 以992份水稻品种为材料，通过13个性状的测定和评价，进行水稻核心种质的构建方法研究。采用调整无偏预测法（AUP）无偏预测水稻性状的基因型值，用基因型值计算不同遗传材料间的马氏距离，分别用3种聚类方法(最长距离法、不加权类平均以及离差平方和)结合随机取样法、优先取样和变异度取样3种抽样方法构建出9个不同的水稻核心种质。水稻核心种质和原种质资源群体的方差差异百分率、均值差异百分率、极差符合率和变异系数变化率的比较结果表明，采用不加权类平均法进行多次聚类、结合变异度取样的方法可以较好地构建出水稻核心种质。基于基因型值构建的水稻核心种质比利用表现型值构建的核心种质更能代表原种质资源群体的遗传多样性。

关键词: 水稻, 核心种质, 基因型值, 种质资源