质量—数量性状的遗传分析 Ⅳ.极大似然法的应用

Genetic Analysis for Qualitative-Quantitative Traits IV. Application of the Maximum Likelihood Method

应用极大似然法分析质量—数量性状的遗传。假定分离世代中的同一主基因基因型因多基因和环境的变异而呈正态分布,不同主基因基因型则为具不同平均数和相同方差的混合正态分布。提出以EM算法对主基因的分离比率、基因效应、微基因的遗传方差等进行极大似然估计以及以似然比方法测验有关遗传假设的程序。以大麦一个杂交组合的F_2代株高资料为例演示了分析过程。结果表明该组合的株高遗传涉及一个主基因和一组微基因;主基因杂合体为完全显性,显性效应达 14.97cm;微基因的遗传方差为65.14cm^2;环境方差为 14.80cm^2。本文的原理和方法可以推广于多个分离世代的联合分析以及利用分子标记构建数量性状基因图谱等。

Maximum likelihood method is applied to the genetic analysis of qualitative-quantitative traits. Assume that individuals with the same major gene genotype in a segregating generation are normally distributed due to the variations of polygenes and environment, and individuals with different major gene genotypes follow a mixture of normals with different means and the same variance. The maximum likelihood estimation of segregating ratios and the effects of major gene as well as the polygenic variation is proposed through the EM algorithm, and the test of corresponding genetic hypotheses by the likelihood ratio. An example of the plant height for an F2 generation in a cross of barley is used for the demonstration. The results indicate that the genetic difference of plant height in this cross relates to one major gene and a group of polygenes; heterozygote of the major gene shows complete dominance with dominance effect 14.97cm; polygenic variance is estimated as 65.14cm2 and environmental variance as 10cm2. Principles and methods in this study can be extended to the joint analysis of multiple generations and the mapping of quantitative trait locus with the aid of genetic markers such as RFLP.