摘要
控制微生物数量性状的多基因系统的诱变表达,可以用混合分布来描述,群体水平上微生物诱变的数学模型的典型情况为均匀诱变分布,即两个正态分布的混合分布.用电子计算机的蒙特卡洛模拟,对现有的三种估计方法作出了评价,发现它们都不能令人满意,提出了交互反馈最小卡方估计,这是适宜于电子计算机执行的理想的方法.参数p(即突变率)的误差估计,可以通过误差传播公式加以测定.微生物工程的诱变育种,可以分为三个阶段:预备试验,鉴定试验及验证试验.通常所说的“初筛”,实际上是鉴定试验,在此以前,应当安排由一系列实验组成的预备试验;验证试验则由复筛及车间对比组成.给出了一个实例.培养条件的优化,可以采用析因回归调控的方法.给出了一个简化的实例.
The expression of mutagenesis on polygenic system which controls a quantitative character of microbe can be represented by a mixed distribution. A mathematical model for microbial mutagenesis on populational level is a uniform-mutagenic distribution (i. e. a mixed distribution of two normal distributions).An evaluation on three estimating procedures has been made by an electronic computer with Monte Carlo Simulation. It is found that all of these procedures are not satisfactory. A procedure called Interfeedback Minimum Chi-Square Estimation is proposed which is found to be a satisfactory one and is suitable for executing at electronic computers, An estimate of error for parameter p (i. e. mutation rate) can be measured through the formula of error, propagation A muta-genie breeding in microbial engineering is divided into three stages: a preparative experiment, an appraisal experiment and a checking experiment. A primary screening, as it is generally called, is actually an appraisal experiment. Before the appraisal experiment, a preparative experiment composed of a series of trials should be taken. A checking experiment is composed of a second ary screening and a plant contrast. An illustration is given.An optimisation technique called Factorisation-Regression-Modulation-Control is recommended to optimise the circumstances of microbial culture. A simplified illustration is given.
出处
《应用科学学报》
CAS
1986年第2期95-104,共10页
Journal of Applied Sciences
