距离分析方法与杂种优势

Methods of Distance Analysis and Heterosis

对于75个小麦品种两年的试验结果,采用不同方法进行距离比较分析。结果表明:用表型平均值相关阵经主成分转换,计算品种间欧氏距离,再用类平均法聚类,似乎是适宜的距离分析方法。D^2值在年份间较稳定,不同年份的D^2估值,可用于预测杂种优势。根据13个性状估算的品种间距离(D^2)与株粒重杂种优势(H)之间呈曲线关系,其理论方程是:Y=-3.65+6.665X--0.1288X^2。根据该方程推算:D^2=25.87时,H出现最大值。

Using data of a two-year experiment including 75 varieties of wheat (Triticum aestivum L.) with 13 characters measured,several methods of distance analysis were studied and compared.The results showed that the best way for distance analysis is adopting phenotypic mean's correlation matrix and making transformation through principal components,then calculating Euclidean distances for each pair of varieties,and finally clustering all varieties into k clusters by group-average method.For the relationship between genetic distance (D2) and heterosis (H),the theoretical function Y=-3.65+6.665X-0.1288X22 was found,where Y = H,X=D2 The maximum heterosis could be displayed by those crosses in which the parents were moderately or more diverse.Seventy-five varieties were grouped into 7 clusters.The intra-cluster crosses would be less heterotic than inter-cluster ones.Highly significant correlation was found between D2 estimated in different years (r=0.740**),so the D2 values estimated in previous years could be used for predicting F1 generation's behavior.To get more effective prediction of heterosis from D22 those characters that have high heritability and highly significant correlations with the target character should be involved in estimating D2 values.