抽样方式对评价小麦种质资源遗传多样性统计数的影响

摘要：以中国、澳大利亚、墨西哥、俄罗斯、荷兰和智利6个国家的733份小麦种质资源为总体，以每个国家为一个亚区，设置19种抽样方式，研究了不同抽样方式对性状平均值、变异系数和多样性指数的影响。结果表明，在均衡抽样条件下，当样本容量小于60时，样本容量大小对性状的平均值和变异系数无显著影响，但对多样性指数值具有显著影响；当样本容量大于120时，样本容量大小对性状的平均值、变异系数和多样性指数均无显著影响。在非均衡抽样条件下，当样本容量大于150时，增加样本容量可有效减少非均衡抽样对变异系数和多样性指数的影响；当样本容量在90-150之间时，缩小各亚区材料数量差异比例可有效减少非均衡抽样对变异系数和多样性指数的影响；当各亚区所取材料最小数量多于20时，可有效降低非均衡抽样对平均值、变异系数和多样性指数的影响。增加性状变异较大亚区内的资源取样数量可提高统计数的准确度。

Evaluation of different sampling methods on the genetic diversity of wheat germplasm

Abstract: The population was constructed using 733 wheat germplasm originating from 6 countries including China, Australia, Mexico, Russia, Netherlands and Chile. We evaluated the effect of 19 different sampling methods on traits means, variance coefficients and diversity index using country as subgroup. The results indicated that there were no significant effect on trait means and variance coefficient when the sample size below 60 using equilibrium sampling method, while there were significant effect on diversity index. There were no significant effect on means, variance coefficients and diversity index when the sampling size bigger than 120. Increasing sample size can reduce the effect of sampling method on variance coefficients and diversity index when the sample size above 150. Decreasing the ratio of subgroup difference can reduce the effect of disequilibrium sampling method on variance coefficients and diversity index when the sample size between 90 and 150. The effect was small to the means, variance coefficients and diversity index when the number of subgroup above 20. The accuracy can be improved by increasing the number of subgroup.