利用分子标记和数量性状基因型值构建作物核心种质库的研究

Study on Constructing Core Collection Based on Plant Molecular Markers and Quantitative Traits

提出了一种基于分子标记数据及数量性状基因型值构建作物种质资源核心种质库的方法.采用包括基因型与环境互作的遗传模型及相应的混合线性模型统计分析方法,无偏预测各材料的基因型值,分别用基因型值和分子标记数据计算个体间的相似系数,加权得到最终的相似距离.采用不加权类平均法(UPGMA)进行系统聚类,用多次聚类随机取样法构建核心种质库.以水稻DH 群体111个基因型8个农艺性状、175个分子标记位点的数据为实例,按四种抽样比率(25%,20%,15%,10%)构建了四个核心种质库,比较了核心种质库与整个群体的分子标记多样性及数量性状的遗传变异,评价了所用方法的有效性.

A method for constructing core collection based on plant molecular markers and quantitative traits is proposed. A genetic model with GE interaction and mixed model approaches are used for predicting genetic values of quantitative traits. Similarity coefficients between different accessions are calculated, respectively, by molecular markers and genotypic values. Equal weighted strategy on two coefficients is employed to calculate similarity distance among varieties in order to group accessions by the unweighted pair group method with arithmetic average （UPGMA） of hierarchical clustering. The core collection is developed by the stepwise clustering procedure at appropriate sampling proportion. A worked example is presented using a rice DH population consisted of 111 genetic individuals with eight quantitative traits and molecular data of 175 loci. Four core collections are constructed by random sampling at four sampling proportions （25%, 20%, 15%, 10%）, the validation of the proposed method is illustrated by evaluating the molecular marker diversity and genetic variation of quantitative traits for representing initial population by the core collections.