Genetic Diversity And Association Analysis Of Loci For Yield And Quality Traits In Wheat

Wheat (Triticum aestivum L.) is one of the first crops purposely cultivated by human beings and most widely distributed in the world which is feeding about40%world population. Wheat is the third most-produced cereal crop after maize and rice in China. In China wheat occupied more than one ninth of global planting area and accounted for16.8%of total wheat output at2011, in which93.1%of the area grew winter wheat and approximately94.5%of total wheat production were comprised by winter wheat. Hence, Understanding the genetic diversity of Chinese winter wheat and identifying SSR markers associated with yield and quality characters will provide useful information support for enhancing the efficiency of further breeding.Genetic diversity, population structure and linkage disequilibrium (LD) of91wheat (Triticum aestivum L.) elite cultivars and breeding lines were investigated by exploring88wheat genomic simple sequence repeat (SSR) markers which were evenly distributed throughout the wheat genome, and then association analysis between SSR markers and yield as well as quality characteristics were performed using TASSEL GLM (General Linear Model) and MLM (Mixed Linear Model) programs. The major results are as follows:1. A total of883alleles were detected, with2to30alleles per locus and a mean genetic richness of10.03. The average genetic diversity index (GD) was0.703, with values ranging from0.160to0.932. The mean polymorphism information content (PIC) value was0.667, ranging from0.148to0.929. Further analysis showed that B genome had the highest genetic diversity and D genome the lowest.2. The91Chinese winter wheat could be divided into three、four and three subgroups based on UPGMA、STRUCTURE and PCA analyses, respectively. The result manifested there was no obvious relationship amongst geographic eco-type, but a partial correlation with pedigrees. Nei’s and AMOVA analysis both indicated that the majority genetic variation occurred within different individuals. Furthermore, gene flow proved that there was some high frequency of gene intercommunication among different subpopulations.3. Linkage disequilibrium (LD) analysis showed that the different degrees of LD were detected not only among syntenic markers but also among nonsyntenic ones, suggesting that historical recombination may exist among linkage groups. The proportion of LD loci pairs when p less than0.01in whole genome、A、B and D genome were17.29%、18.77%、11.08%、25.57%, respectively. The baseline of LD decay for genome was r2=0.0287and mean LD decay distance at the whole genome was4.3cM, while A genome was3.7cM and1.0cM as well as4.1cM for B and D genome. The final analysis indicated that D genome had the highest degree of LD and B genome the lowest. The results of this study were expected to provide valuable information for future association mapping by using this Chinese winter wheat.4. Six models were used to evaluate the effects of population structure (Q), relative kinship (K) and principal components analysis (PCA) on three selected traits for marker-trait associations. The results demonstrated that the model controlling relative kinship (K), population structure and relative kinship (Q+K), principal components analysis and relative kinship (PCA+K) outperformed other models in terms of population structure adjustment and pseudo association reducing.5. Association analysis by K model found seven markers (xcfd8、xcfd10、xwmc580、 xwmc719、xwmc749、xgwm194、xbarcl34) might be significantly (p<0.01) associated with agronomic traits including iron content in grain、test weight、stability time、first internode length below spike、plant height and thousand kernel weight, ranging from0.10to0.28of variation was able to be explained by those markers. And seven markers (xcfd8、xcfd10、 xwmc219、xwmc256、xwmc580、xwmc749、xbarc126) found by Q+K model might be highly related to the iron content in graiin、zinc content in grain、test weight、stability time、 development time、grain length、grain width and thousand kernel weight, which could account for about0.10to0.22of variance. In addition, PCA+K model closely connected seven markers (xcfd8、xcfd201、xwmc537、xgwm194、xbarc96、xbarc134、xbarc187) with the iron content in grain、stability time、stretch area、first internode length below spike、second internode length below spike、uppermost internode diameter、plant height、flag leaf width and thousand kernel weight, and which satisfied to explain0.08to0.28of variation.