Study On Genetic Diversity Of Some Wheat Core Collections

In recent years, the speed of genetic improvement of Wheat Varieties in China has become slow, one important reason is that the level of genetic diversity of parental resources is reduced. In this research, some wheat core collections were analyzed to achieve two goals:one is to understand the genetic basis, and to provide reference for proper utilization of wheat materials, the other is to prevent the further loss of wheat germplasm.When crop varieties are analyzed, more than two variables that have a certain relationship need analyze in many cases. To fully reveal the relationship between these variables, a multivariate statistical analysis method is should be employed. There are two commonly methods for multivariate statistical analysis. one that contains correlation analysis and principal component analysis is to understand the relationship between variables. The other is to determine the attribution of a sample and contains clustering analysis and discrimination analysis. In this study, seven agronomic traits of 105 wheat core collections were measured and the data were analyzed by multivariate statistical analysis. The main conclusions were as follows:(1) The results of the analysis of Agronomic traitsThe variation coefficient of each agronomic trait was in turn. Spikelet number > spikelet density > spike length > plant height > 1000-grain weight > grain per spike > grains per spike.These results indicated that the most abundant variation was spikelet number that had the widest range to choose. Grains per spike had the smallest variation so that it had the narrowest range to select.The genetic diversity index of the tested varieties was between 0.14 and 2.07, and the average value was 1.74, which showed that the genetic diversity of the tested materials was high.(2) The results of correlation analysis and partial correlation analysisCorrelation analysis showed that eight pairs of traits were extremely significant correlation. Among them, two pairs were negative correlation at extremely significant level,six pairs were positive correlation at extremely significant level. Additionally, two pairs of traits showed significant positive correlation. Partial correlation analysis showed that 9 pairsof traits were extremely significant partial correlation, and 3 pairs had significant partial correlation.(3) Principal component analysisBased on the analysis for three principal components of seven traits, we can draw a conclusion: a good variety should have a moderate or high value of the first principal component, a moderate value of the second principal component, and a high value of the third principal component.(4) results of Cluster analysisThe similarity of two clustering analysis was 80%. The tested cultivars were divided into four groups with corresponding relationship. The first group had abundant variation; the second group might not be suitable for production. The third group which was the most valuable type might be not only suitable for production but also had high collection value. the fourth type was multi-grain type, which can be used to breed with other varieties(with reference to the results of system cluster analysis).