STOCHASTIC PROCESSES AND DISTRIBUTION OF GENE FREQUENCIES UNDER NATURAL SELECTION
STOCHASTIC PROCESSES AND DISTRIBUTION OF GENE FREQUENCIES UNDER NATURAL SELECTION
复制标题
DOI：
10.1101/sqb.1955.020.01.006
复制
发表时间：
1955-01-01
期刊：
影响因子：
--
通讯作者：
KIMURA, M
中科院分区：
文献类型：
--
作者：
KIMURA, M
进化是自然群体中基因频率变化的随机过程。由于组成一个物种的种群是由许多个体组成的，由于进化是在很长的时间内进行的，所以支配变化过程的规律不可避免地是”统计的“。在这个意义上，进化的遗传理论，正如RA Fisher（1922）所建议的，可以与气体理论相比较。这个类比可以进一步推广：与其将种群视为基因的聚集，我们发现将种群视为基因频率（或比率）的聚集更方便。这类似于物理学中的情况，在物理学中，速度总体的规范有时比粒子总体的规范更有用（Fisher，1953）。据我所知，这个富有成果的思想最早是由Fisher在他1922年的论文中引入群体遗传学理论的，这篇论文导致了后来的阐述（Fisher，1930 a）。赖特在他的许多文件开始于1931年。基因频率的稳态分布问题在越来越一般的条件下得到了解决。在他的研究中，他提出了一个新的论点，即快速进化的最有利条件是将种群细分为许多部分孤立的局部群体。其基本思想是，这种人口结构提供了随机分化的本地群体，这是群间选择的基础。这种观点，虽然它已被许多进化论者（Dobzhansky，1951; Halfman，1949 b; Muller，1949）所接受，是一个有争议的观点，被费希尔学派所批评（例如参见Sheppard，1954）。其中一个批评来自于费希尔和福特（1947）对分离种群的Panaxia dominula蛾的研究。他们将他们对黑粉菌基因的发现概括为这样一种说法，即自然种群通常受到选择作用的影响，这种选择作用的方向和强度不时变化，而且其程度足以引起所有基因比率的波动变化。因此，他们（费舍尔和福特，1950年）认为，把一个特殊的进化优势归因于小的孤立的玉米的主张，
Evolution is a stochastic process of change in gene frequencies in natural populations. Since the populations making up a species consist of many individuals and since evolution extends over enormous periods of time, laws which govern the process of change are inevitably" statistical." In this sense the genetical theory of evolution, as RA Fisher (1922) suggests, is comparable to the theory of gases. This analogy can be pushed further: Instead of considering populations as aggregates of genes, we find it more convenient to consider populations as aggregates of gene frequencies (or ratios). This is similar to the situation in physics where the specification of the population of velocities is sometimes more useful than that of a population of particles (Fisher, 1953). As far as I know, this fruitful idea was first incorporated into the theory of population genetics by Fisher in his 1922 paper, which led to a later elaboration (Fisher, 1930a).The deductive theory of genetics of natural populations has been greatly advanced since then by S. Wright in his numerous papers starting in 1931. The problem of steady state distribution of gene frequencies has been solved under more and more general conditions. Out of his investigations he has proposed a new thesis that the most favorable condition for rapid evolution is the subdivision of the population into numerous partially isolated local groups. The essential idea is that this population structure provides stochastic differentiation of local groups which is the basis of intergroup selection. This view, though it has been accepted by many evolutionists (Dobzhansky, 1951; Haldane, 1949b; Muller, 1949) is a controversial one, having been criticized by Fisher's school (see for example Sheppard, 1954). One of the criticisms came from the study of the isolated population of the moth Panaxia dominula by Fisher and Ford (1947). They generalized their findings on the medionigra gene to the statement that natural populations in general are affected by selective action varying from time to time in direction and intensity, and of sufficient magnitude to cause fluctuating variation in all gene-ratios. Thus they (Fisher and Ford, 1950) consider that the claim for ascribing a special evolutionary advantage to small isolated com-