中文摘要：

【目的】从分子水平了解中国大蒜种质资源的群体遗传结构和遗传背景。【方法】利用AFLP、SSR和InDel这3种分子标记对国家无性繁殖蔬菜资源圃保存的212份大蒜资源进行检测，通过Mega软件进行最大相似性聚类分析，structure2．1软件进行群体遗传结构分析，SSPS软件进行分子标记与大蒜辣素含量和21个数量性状进行一元线性模型检测，考察两者之间的关联性及群体遗传结构的影响。【结果】3种分子标记在212份种质中扩增出502个位点，多态性位点为492个。群体遗传结构与聚类分析均将所有资源划分为5个群体，划分的类别基本一致。然而，群体遗传结构分析划分的5个群体，群内遗传信息多样性指数较小。对212份种质的22个数量性状与分子标记的线型模型分析表明，包括大蒜辣素含量在内的多个数量性状受群体遗传结构的影响较小。【结论】中国大蒜种质资源遗传背景丰富，群体遗传结构对数量性状的分布影响较小，适合进一步进行性状与分子标记之间关联分析研究。

英文摘要：

[Objective] The objective of the experiment is to understand the diversity and genetic structure of garlic germplasm resources in China. [Method] Two-hundred and twelve accessions of garlic germplasm preserved in China were detected and analyzed based on AFLP, SSR and InDel. Neighbor-joined clustering was analyzed by soft＇ware Mega, and the population genetic structure was studied by software Structure 2.1. The effect of genetic structure on association analysis between the molecular markers and 21 quantitative traits including allicin content was evaluated by general linear model （GLM） by software S SPS 13.0. [ Result ] Totally, 502 allels were amplified by AFLP, SSR and InDel primers, and 492 of them were polymorphic among 212 accessions of garlic. All accessions were divided into 5 groups by both structure analysis and neighbor-joining clustering. However, the Shannon＇s index of each group assumed by genetic structure analysis was smaller than that assumed by neighbor-joining clustering, which indicated that the genetic structure analysis could interpret genetic relationship among the individual accessions in more details. Most of traits including allicin content were slightly affected by population structure, which indicated that the germplasm in this study was acceptable to be the populations for association mapping. [ Conclusion] The garlic germplasm resources in China possess a great diversity. The population structure of the garlic germplasm resources slightly affect the quantitative traits including the allicin content, which means that the population of the garlic germplasm is a potential candidatenatural population for association analysis between the molecular markers and the quantitative traits.