保留特殊种质材料的核心库构建方法

A Method of Constructing Core Collection Reserving Special Germplasm Materials

本研究提出了能保留特殊种质材料的多次聚类构建种质资源核心库方法．对种质材料的表现型数据采用合适的遗传模型及混合线性模型统计分析方法无偏预测基因型值，用基因型值构建核心库．计算材料间的马氏距离，用不加权类平均法进行聚类．根据聚类图选取材料构建核心库时，优先保留特殊遗传材料．用方差Ｆ测验、均值ｔ测验、极差比和变异系数评价核心库代表原有种质资源群体遗传多样性的程度．以１６８个棉花基因型的５个纤维性状构建核心库．

A method of constructing core collection reserving special germplasm materials was proposed. Phenotype data of genetic accessions were used to predict genotype vales by a genetic model controlling systematical errors in fields and by mixed model approaches. Predicted genotype values were used to construct core collection. Genetic distance among accessions was calculated with Mahalanobis distance, then unweighted pair-group average method of hierarchical cluster was used for grouping. According to a dendrogram of clusters the accession with special merit was preferred to be chosen for core collection. F test for variances, t-test for means, range rate and coefficient of variation were used to evaluate the property of core collection for initial accessions. An worked example of constructing core collection on 168 accessions of cotton by five fiber traits was presented.