论文部分内容阅读
运用数字计算机模拟研究了长期人工选择的反应。考虑了不同的基因作用模型(加性、显性与超显性)、群体含量(小群体N_f=Nm=10,Ne=20;大群体N_f=Nm=100,Ne=200)、遗传力水平以及位点间连锁强度等因素共54个组合,270个群体各49代的选择反应。结果表明,对于三种基因作用模型,只有强烈的连锁(相邻位点之间交换率为r=0.01)在小群体中对选择反应有显著影响,它可以阻碍有利基因的固定,加速它们的丢失,从而降低选择反应。中等强度的连锁(r=0.1)或大群体中的紧密连锁都没有显著的作用。大群体中的选择反应要明显地高于小群体中的,特别是遗传力低的性状。本文还讨论了不同选择系杂交以获致进一步进展的问题,讨论了长期选择理论研究结果与实际模拟的偏差。
Using digital computer simulation to study long-term artificial choice of response. Considering different gene action models (additive, dominant and super dominant), population content (small population N_f = Nm = 10, Ne = 20; large population N_f = Nm = 100, Ne = 200) As well as the linkage strength between sites and other factors a total of 54 combinations, 270 groups of 49 generations of selective response. The results showed that for the three gene action models, only a strong linkage (exchange rate between adjacent loci was r = 0.01) had a significant effect on selection responses in small groups, which could hinder the fixation of favorable genes and speed up their Lost, thus reducing the choice of response. Moderate-strength catenaries (r = 0.1) or close clumps in large groups had no significant effect. The response to selection among large groups was significantly higher than that among small groups, especially those with low heritability. This paper also discusses the problem of different selection lines crossing for further progress, and discusses the deviation of long-term selection theory from actual simulation.