中国栽培大豆群体结构不同分类方法的比较

Comparison of various clustering methods for population structure in Chinese cultivated soybean(Glycine max(L.)Merr.)

利用215个大豆品种的135个分子标记数据,用STRUCTURE软件、PowerMarker软件和地理生态类型3种分类方法研究了大豆品种的群体遗传结构,以探索适宜的分类方法。结果表明:用STRUCTURE软件分类时,亚群间成对分化系数(Fst)的平均值最大,为0.108 3,含有相同最高频率等位基因的位点数最小,为85,说明各亚群间遗传差异最大;亚群内遗传多样度(Hs)为0.491,多态性信息含量(PIC)为0.737 6,群体内分化系数(Fis)为0.974,均为最小,说明亚群内个体间遗传相似性最高。因此,用STRUCTURE软件研究群体遗传结构最适宜。

Using geographic eco-type approach,STRUCTURE and PowerMarker softwares,molecular marker information of 135 markers in 215 soybean cultivars randomly selected from China was used to investigate genetic structure of the above population,and its purpose is to find a suitable cluster approach.Results showed that STRUCTURE software results in a maximum pair-wise Fst value of 0.108 3 among subpopulations and a minimum estimate of 85 for the number of loci with highest allelic frequency across subpopulations,suggesting the biggest genetic difference among subpopulations;and a minimum Hs estimate of 0.491,a minimum polymorphism information content value of 0.737 6 and a minimum pair-wise Fis value of 0.974 within subpopulation,indicating the largest genetic similarity within subpopulation.This means that STRUCTURE software is available for investigating population genetic structure.