摘要
本文利用Monte Carlo模拟技术研究比较了普通选择指数、实际选择指数、约束指数和单性状选择四种方法的长期选择效果及对群体遗传结构的影响。50世代的结果表明:群体大小和遗传参数能明显影响长期指数选择的反应;实际指数选择是有效的,其总体效果好于普通指数;约束选择的早期反应虽低于普通指数,但后期效果好,而且对群体遗传结构影响较小。同时,指数选择使性状加性遗传方差和遗传力衰减变化,而遗传相关依指数公式的系数呈定向改变。建议多性状长期选择时应用实际指数法或约束指数法。
The genetic effects of long-term index selection were investigated by means of Monte Carlo Method. The response and the effect on genetic structure of population to selection using four methods, that is theoretical index, emperical index, restricted index and single-trait selection, were compared. Results obtained from 50 generations demonstrated that much greater response was achieved in large populations, and selection in small ones led to a decrease in selection limit. Genetic correlation played an important role on response in small populations. Emperical index selection was more effective than theoretical index, restricted index had less response in the early period of selection but greater response in later period and also had less effect on genetic structure of population even if the genetic correlation was intense. Meanwhile, index selection could cause decrease in additive genetic variance and beritabilities while genetic correlation changed depending on the coefficients of index formulae. The suggestion is that emperical index or restricted index should be used for long-term selection for multi-traits.
关键词
选择指数
长期选择
遗传分析
Selection index, Long-term selection, Genetic analysis Simulation
