Epperson Bryan K, Li Tian-Quan
Department of Forestry, Michigan State University, East Lansing, Michigan, 48824.
Evolution. 1997 Jun;51(3):672-681. doi: 10.1111/j.1558-5646.1997.tb03651.x.
Spatial autocorrelation statistics have been studied in theoretical population genetic models and widely used in experimental studies of spatial structure in many plant and animal populations. However, the statistical properties of spatial autocorrelation statistics have remained uncharacterized. Little is known about how values of spatial autocorrelation statistics in population samples depend on the level of dispersal and scheme of sampling. In this paper, we characterize the statistical properties of join-count spatial autocorrelation statistics for population genetic surveys under various conditions of dispersal and sampling. The results indicate generally high statistical power. These results can provide a method to estimate gene dispersal based on standing spatial patterns of genetic variation observed within populations.
空间自相关统计已在理论群体遗传模型中得到研究，并广泛应用于许多动植物群体空间结构的实验研究。然而，空间自相关统计的统计特性仍未得到描述。关于群体样本中空间自相关统计的值如何依赖于扩散水平和抽样方案，人们知之甚少。在本文中，我们描述了在不同扩散和抽样条件下群体遗传调查的联合计数空间自相关统计的统计特性。结果表明其统计功效总体较高。这些结果可以提供一种基于群体内观察到的遗传变异的现有空间模式来估计基因扩散的方法。
