| We have studied the agronomic, yield and quality traits of151soybean accessions for2years. SCoT, CDDP and ISSR molecular markers were utilized in polymorphic amplification for studying the genetic relations and population structures of soybean varieties from different geographical origins, further analyzing the relation between molecular markers and traits in order to identify the highly associated loci to these traits, and to sequence the elite loci. The results were as follows:1. The151accessions of soybean cultivars used in this study have broad genetic diversity, which shows a normal distribution and can be used in further association analysis. Yield is a complicated quantitative trait, and there is significant positive correlation between yield and valid pods per plant, hundred seeds weight and growth stage. The more the weight of hundred seeds weight is, the fewer the valid pod per plant and seeds per pod is. It is required for soybean breeders to put more weight on valid pods per plant and seed number per pod than on hundred seeds weight. There is excessive significant negative correlation between protein and fat oil.2. We have established an optimized SCoT-PCR system of20μL, including2.0mmol·L-1Mg2+,0.2mmol·L-1dNTPs,0.250μmol·L-1primer,1.5U Taq polymerase and30ng DNA template. A total of41polymorphic primers were selected from82primers, and there were182polymorphic bands, average primer PIC being0.656, which best implied the genetic diversity of soybean cultivars. Genetic clustering analysis basically coincided with population structure analysis, both dividing all the tested cultivars into three groups. Genetic Diversity between domestic cultivars and foreign ones is larger than that between foreign soybean and the wild soybean germplasm. Landrace and released cultivars were grouped into different clusters. Small genetic diversity existed among cultivars from Heilongjiang, Inner Mongolia, jilin due to the frequent regional germplasm exchange. Closely related cultivars from the same ancesters were clustered together.3. We have established an optimized CDDP-PCR system of20μL, including2.0mmol·L-1Mg2+,1.5U Taq polymerase,0.375μmol·L-1primer,0.3mmol·L-1dNTPs, and40ng DNA template. A total of17polymorphic primers, selected from21primers, produced82polymorphic bands, average primer PIC being0.644, which best implied the genetic diversity of soybean cultivars. All the tested cultivars were divided into three groups. Genetic distance of cultivars from Heilongjiang, Inner Mongolia, jilin were minimal. Most of the landrace was divided into the same group, so was cultivars of the same breeding institute. Cluster analysis based on genetic distance and model were broadly consistent and different geographical distribution of cultivars could be seen from the analysis,4. A total of37polymorphic primers were selected from100ISSR primers, a total of150polymorphic bands were generated, average primer PIC being0.546, which basically implied the genetic diversity of soybean cultivars. However, the clustering was not as effective as those done with SCoT and CDDP marker. The151accessions of soybean germplasm were divided into three groups:there was diverse genetic background among some of the domestic and foreign cultivars, whereas, the genetic relation was closer between the rest of the foreign cultivars and landraces; most of the cultivars with similar descend and middle and late mature varieties were clustered into the same groups; model analysis of population structure implies simple genetic background among tested cultivars, which coincided well with genetic distance analysis.5. The efficiency of markers were ranked as CDDP>SCoT>ISSR according to their polymorphic loci identifying ability and genetic diversity coefficient. There was correlation between any two markers, and all these three kinds of marker were efficient and dependable in genetic identification of soybean germplasm, SCoT and CDDP being superior to ISSR. From the cluster analysis with the three markers, it could be suggested that foreign cultivars can be used to expand the genetic basis of domestic cultivars; cultivars from Heilongjiang, Inner Mongolia, Jilin were closely related genetically; most landraces were clustered together into the same group; close genetic relation existed among the cultivars with same ancestor or the same pedigrees. Sub-cluster analysis based on model indicated three different pedigrees:there was great variance between domestic cultivars and foreign ones; there is correlation between soybean sub group distribution and geographical origins. This study suggested that cluster analysis based on genetic distance and that based on model were highly coincided with each other, whether the single marker system or combined marker system was used.6. There were different LD among27028locus combinations induced from the233marker loci; correlation analysis between233polymorphic marker loci and mean phenotype (two years) of151accessions of soybean, using General linear model (GLM) and Mixed linear model (MLM) programs of TASSEL4.0software, indicated that there was significant association between41loci and14phenotype traits for141times (P<0.005). In two year repetitions,22correlation loci were detected for32times; the two model analysis system identified21correlation loci repeatedly for32times, and the correlation loci detected with MLM is less than that with GLM.7. The loci associated with such morphological phenotype as the bottom pod height, valid branch, plant height and main stem number were9,5,6,2respectively and detected for37times. The loci associated with flowering and growth period were7and5respectively, and detected for30times. The loci associated with valid pod, seeds per plant, seed weight per plant, hundred seeds weight, seeds per pod, yield per plot were6,8,5,5,4and6respectively, and detected for54times. The loci associated with oil content, protein content were5and4respectively, detected for20times. The explanation rate of phenotypic variation of correlated loci detected was0.0483to0.1622.8. SCoT loci were cloned and sequenced. The results showed that, SC06-2was862bp in length, and highly similar with small nuclear ribonucleoprotein Sm D2-like; full-length of SC21-1was437bp, and highly similar with UDP-glycosyltransferase89A2-like subunits; full-length of SC21-6was243bp, and highly similar with the auxin-responsive protein IAA11; SC28-3was392bp in length, and highly similar with GDP-L-galactose phosphorylase; full-length of SC60-2was900bp, and highly similar with the disease protein At4g27190; full-length of SC71-2was703bp, highly similar with the E3ubiquitin ligase protein CIP8; full-length of SC82-1was900bp, highly similar with exocytosis members Exo70B1.The others were highly similar with soybean genome sequence, but the function is unknown. |
PDF Full Text Request