Association Mapping For Plant Height, Branch Number,Node Number On Main Stem, Stem Diameter And Pod Number Per Plant In Soybean(Glycine Max L. Merr.)

Soybean (Glycine max L. Merr.), with abundant protein and oil, is an important economic crop in China. Soybean yield is related to its agronomic traits, i.e., plant height, branch number, node number on main stem, stem diameter and pod number per plant. Thus, it is necessary to investigate the genetic foundation of agronomic traits in soybean. However, there are few studies about association mapping for agronomic traits in soybean under multiple environments using multiple approaches.In this study there were three aspects. First,257soybean cultivars obtained, by stratified random sampling, from6geographic ecotypes in China were used to scan SSR markers on the genome. The molecular information was used to describe the genetic structure, genetic diversity and linkage disequibrium between markers. Then, the phenotypic observations for the above-mentioned five agronomic traits were measured in Jiangpu experimental station in2009and2010. These values along with SSR marker information were used to detect quantitative trait loci (QTL) using generalized linear model (GLM), enriched compression mixed linear model (E-cMLM) and epistatic association mapping (EAM) approaches. Finally, novel alleles of the identified QTL for the above five traits were mined according to the mapped QTL and elite cross combinations for the above five traits were predicted. The main results are as follows.1) PowerMarker3.25software was used to analyze the genetic diversity in the soybean population. Results showed that there was abundant allelic variation in the soybean population.135SSR markers generated a total of891alleles, with the range from2to24alleles per locus and the average of6.6±3.39alleles per locus. The polymorphic information content was0.56±0.20, with the range from0.0153to0.9146. The gene diversity was0.60±0.20, with the range from0.0154to0.9201.The above population under study was partitioned into4subpopulations using the STRUCTURE2.2software. The linkage disequilibrium (LD) beween two markers was calculated using TASSEL2.1software. Results showed that there was LD between two markers on the same chromosome or on the different chromosomes. The LD degree decreased with the increase of genetic distance between two linked markers. When the distance was less than20cM, the decrease is very quick. However, there was no obvious decrease with the increase of marker distance while the distance was more than20cM.2) The GLM, E-cMLM and EAM apprpaches were used to identify the linked markers for plant height, node number on main stem, branch number, pod number per plant and stem diameter. As a result, forty-seven, ten and fifty-eight main-effect QTL were respectively detected by the above-mentioned three approaches. Fifteen common main-effect QTL between EAM and GLM, one common main-effect QTL between E-cMLM and GLM, one common main-effect QTL among the three approaches were identified. In addition, one epistatic QTL for pod number per plant and32QTL-by-envronment interactions were found by the EAM method.3) In the stable QTL detected, novel alleles for each QTL were mined, for example, satt669(145bp) and satt534(152bp) for node number on main stem, satt102(154bp) and satt102(157bp) for branch number, and satt382(395bp) and satt534(152bp) for stem diameter.Based on the results from the EAM approach, superior parent cross combinations for the above-mentioned traits were predicted, for example, deqingsharendou and hefeiliangtangjiaoshuangqingdou for branch number, nanchengqingpidadou and hefeiliangtangjiaoshuangqingdou for node number on main stem, yixingwuhuangdou and jidou12for stem diameter, nanchengqingpidadou and hefeiliangtangjiaoshuang-qingdou for pod number per plant. Note that hefeiliangtangjiaoshuangqingdou may be used to simultaneously improve branch number, node number on main stem and pod number per plant.