摘要
通过构建包括各种可能遗传效应的遗传模型及系数矩阵、基于EM算法实现的遗传参数和遗传率的估计、利用Akaike最大熵准则进行最佳模型的筛选等步骤,建立适合不同环境下数量性状分析的“多环境下数量性状主基因分离分析方法”,根据控制数量性状主基因数目不同,构建2大类10种遗传模型。结果表明,随着遗传率和联合世代群体样本容量的增加,各项遗传效应估计值的准确度和精确度随之提高,但比较遗传率和联合世代群体样本容量二者的功效,遗传率的作用明显高于样本容量;杂交组合驻豆03-9 ×中黄13的F2、BC1和BC2群体,主茎节数均符合两对加性–显性–上位性主基因 + 基因型 ×环境互作效应遗传模型(B-1模型)。F2代主茎节数的广义遗传率较高,说明该性状在F2分离强度较大。主茎节数的各成分分布方差不等,对于不同表现型个体选择效果有差异。
Through the construction of genetic model and coefficient matrix including various possible genetic effects, the estimation of genetic parameters and heritability based on EM algorithm, and the selection of the best model by using Akaike maximum entropy criterion, a quantitative trait analysis model suitable for different environments was established. According to the number of major genes controlling quantitative traits, 10 genetic models of 2 categories were constructed. The results showed that with the increase of heritability and sample size of combined generation population, the accuracy and precision of the estimated genetic effects were improved, but the effect of heritability was significantly higher than that of sample size in comparison of heritability and sample size of combined generation population;the number of main stem nodes of F2, BC1 and BC2 populations of Zhudou 03-9 ×Zhonghuang 13 were in accordance with the two additive pairs: The genetic model of interaction effect of sex dominant epistatic major gene and genotype ×environment (B-1 model). The broad sense heritability of the number of main stem nodes in F2 generation was higher, which indicated that the segregation intensity of this trait was higher in F2 generation. The variance of each component in the number of nodes of main stem was different, and the selection effect was different for different phenotypes.
出处
《农业科学》
2021年第4期297-305,共9页
Hans Journal of Agricultural Sciences
