作物杂种后代基因型值和杂种优势的预测方法

METHODS OF PREDICTING GENOTYPE VALUE AND HETEROSIS FOR OFFSPRING OF HYBRIDS

本文提出了利用作物亲本和F_1预测杂种后代基因型值和杂种优势的统计分析方法.该方法运用加性-显性遗传模型,分析双列杂交试验资料,用MINQUE(1)法估算方差分量以及预测遗传效应值.由加性和显性效应预测值可进一步预测F_1,F_2,BC_1,BC_2,等不同世代的基因型值,在预测F_1群体平均优势和群体超亲优势的基础上,可以推导出其它各世代的杂种优势.提出了预测杂种后代保持超亲优势世代数的简单公式,根据杂交组合F_1群体平均优势和双亲相对遗传差异,便可预测该组合能在生产上直接利用的世代数.以棉花六个品种完全双列杂交试验资料为例,分析了各组合F_1和F_2的基因型值、超亲优势和保持5％超亲优势的世代数.

By parents and F_1 hybrids of crops, methods of predicting genotype value and heterosis are proposed for offspring of hybrids. The MINQUE (1) procedures can be used for estimating variance components and for predicting genetic effects by analyzing diallel crosses with an additivedominance model. Predicted additive and dominance effects can be further used in predicting genotype values for different generations of F_1, F_2, BC_1 and BC_2. By predicted heterosis over mid-parent value or over better-parent based on population mean, heterosis for other generations can be obtained. A simple formula is proposed for predicting the number of generations of a cross which still has heterosis over better-parent. The number of generations is a function of heterosis over mid-parent value based on population mean and genetic difference between two parents. A worked example of six-parent diallel analysis in cotton is given for predicting F_2 genotype value, heterosis and the number of generations for each cross.