Analytic review: some current problems of human population genetics.
Analytic review: some current problems of human population genetics.
复制标题
分析回顾：当前人类群体遗传学的一些问题。
DOI：
--
复制
发表时间：
1973
影响因子：
9.8
通讯作者：
L. Cavalli
中科院分区：
文献类型：
--
作者：
L. Cavalli
近年来，人们对人类群体遗传学的兴趣有了相当大的增加。各种动机激发了这一趋势：(1)如果不了解群体遗传学原理，就不可能对人类遗传数据进行遗传分析。由于计算机的日益普及，所涉及的数学可能很繁琐，但在很大程度上是可以商量的。(2)人类的资料可以提供一些在其他生物体中不易解决的一般进化问题的见解；已经积累或潜在可利用的物质财富是巨大的。体质人类学家日益认识到，遗传学的这一分支是他们工作的理论支柱。社会和文化人类学家可能很快就会开始意识到，有一些重要的领域是重叠的。任何与人类有关的东西往往会吸引更多的好奇心，而且在研究经费支持减少的时候，它们获得资助的机会可能更大。计算机的用处怎么强调都不过分。过去有一些统计方法，但由于涉及到大量的算术运算而没有被采用，现在这些方法在实践中是可用的。计算机还可以通过蒙特卡洛或模拟方法提供数学技术无法解决的问题的理论答案。这两个优势都应该谨慎看待。现实表明，在许多生物学估计中，甚至第一个有效数字都可能存在疑问。由于这个原因，使用需要计算机的高度复杂的方法有时可能会产生困惑，但到目前为止，使用电子计算机上已有的程序通常是最经济的。至于计算机模拟的使用，它似乎并不是（至少通常）莫顿所谴责的“唯我论”。的确，在一些情况下，可以找到一种分析技术来回答这个问题。但在模拟之前，这可能无法实现；模拟复杂现象的启发式价值，在整理基本变量或发现意想不到的结论时，不应该被最小化。此外，许多进化遗传学的数学理论本身是近似的或高度过于简化的，计算机模拟有助于显示它们的适用范围。也许有人会说，第一个动机的重要性随着……而降低
In recent years there has been a considerable increase of interest in human population genetics. Various motives have inspired this trend: (1) There can be no genetic analysis of human genetic data without an understanding of the principles of population genetics. The mathematics involved may be tedious but have largely been made negotiable, thanks to the increasing popularity of computers. (2) Human data may supply insight into some general evolutionary problems which are not so easily solved in other organisms; the wealth of material already accumulated or potentially available is enormous. (3) Physical anthropologists realize increasingly that this branch of genetics is the theoretical backbone for their work. Social and cultural anthropologists may soon start realizing that there are important areas of overlap. (4) Anything human tends to attract more curiosity and, at times of dwindling financial support for research, has perhaps a higher chance of being funded. The usefulness of computers could not be overemphasized. Statistical methods that were available but not employed because of the heavy arithmetic involved are now usable in practice. Computers can also supply theoretical answers by Montecarlo or simulation methods to problems which are out of reach for mathematical techniques. Both advantages should be -viewed with caution. Realism shows that in many biological estimations even the first significant digit may be in doubt. For this reason, the use of highly sophisticated methods requiring computers may sometimes generate perplexity, but by now it is often most economical to use programs already available for electronic computers. As to the use of computer simulation, it does not seem to be (usually at least) that exercise in "solipsism" which Morton denounces. It is true that in several cases an analytic technique can be found which answers the problem. But this may not be achievable before simulation; the heuristic value of simulating a complex phenomenon, in sorting out essential variables-or discovering unexpected conclusions-should not be minimized. Moreover, many of the mathematical theories of evolutionary genetics are themselves approximate or highly oversimplified, and computer simulation helps in showing their range of applicability. It may be argued that the first motive is likely to decrease in importance with