231份烤烟种质资源SSR标记遗传多样性及其与农艺性状和化学成分的关联分析

本文旨在分析烤烟种质资源遗传多样性，通过关联分析寻找与烤烟农艺性状和化学成分相关联的分子标记，为分子标记辅助选择育种和提高烤烟育种效率奠定基础。利用均匀分布于烟草24个连锁群上的120个SSR标记分析了231份烤烟种质资源的遗传多样性及群体遗传结构，在此基础上采用TASSEL 5.0 GLM（general linear model）和MLM （mixed linear model）方法进行标记与表型性状变异（8个农艺性状和5个化学成分）的关联分析。通过遗传多样性分析，共检测出403个等位变异，变异范围2-21个，平均每个标记3.36个；引物的多态性信息含量（PIC）为0.036~0.785，平均值为0.408；供试材料间遗传距离为0.003~0.371，平均值为0.148。通过非加权组平均法（UPGMA）进行聚类分析，在遗传距离（GD）值为0.153水平上可将231份烤烟材料聚为4大类。通过群体遗传结构分析将供试材料划分成3个亚群，且每个亚群分别包含117、38和76份烤烟材料。通过GLM_Q和MLM_Q+K两种关联分析模型，分别检测出17个和9个与13个表型性状显著相关，其中9个标记在两个年份和两种模型中均被检测到与同一性状在P < 0.001水平上极显著相关，且MLM_Q+K未能检测到新的关联标记；3个株高相关标记（TM10481、PT54964和TM20580）、2个叶数相关标记（TM10846和PT60728）、1个茎围相关标记（TM25276）和1个叶长相关标记（TM11215）与已报道的QTL定位结果一致，可应用于烤烟分子标记辅助选择育种。 

SSR marker-based analyses on genetic diversity and relevant variations of agronomic traits and chemical composition of 231 flue-cured tobacco germplasm resources

The aim of this study was to investigate genetic diversity of flue-cured tobacco (Nicotiana tabacum L.) germplasm and to screen molecular markers for agronomic and chemical traits. A total of 231 flue-cured tobacco germplasms were genotyped with 120 simple sequence repeats (SSRs) markers uniformly distributed on 24 linkage groups. Associations of SSR variation with eight agronomic traits and five chemical components were investigated by two contrasting statistical models, i.e. general linear model (GLM) and mixed linear model (MLM). Overall, 403 alleles were identified in all accessions, ranging from 2 to 21 per locus, with an average of 3.36 alleles per locus. Mean polymorphic information content was 0.408, ranging from 0.036 to 0.785. Genetic distance between varieties ranged from 0.003 to 0.371, with a mean of 0.148. Using unweighted pair-group method with arithmetic mean (UPGMA), all accessions were clustered into four groups at a genetic distance of 0.153. Using population genetic structure analysis, the test materials were divided into three subpopulations. Association analysis by GLM_Q and MLM_Q+K models revealed that seventeen and nine markers were significantly associated with a total of 13 phenotypic traits, respectively. Moreover, nine markers were consistently associated with the same traits at the p < 0.001 level in two years using both models. Specifically, three markers associated with plant height (TM10481, PT54964, and TM20580), two markers with leaf number (TM10846 and PT60728), one marker with stem circumference (TM25276), and one marker with leaf length (TM11215) supported published quantitative trait locus (QTL) mapping results. These seven markers could be used in future molecular marker-assisted selection to improve breeding efficiency.