Use Of Principal Component Analysis For The Classification Of Winter Wheat.

Winter wheat is an important crop in the dry land area of eastern Washington. The objective of this study was to classify winter wheat genotypes based on the existing relations among agronomic traits. Information used regarding grain yield, percentage of protein, and test weight, were compiled for 59 soft white winter wheat genotypes for three precipitation zones: < 305 mm, 406-508 mm and > 508 mm. The information came from WSU field trials that were conducted in 5 locations of the eastern region of Washington State during 2012 season. Principal Component Analysis (PCA) and Cluster analysis were used to classify the winter wheat genotypes. The statistical analysis was performed using the Statistical Analysis System (SAS) V.9.2. PCA showed that two components explained 76% of the variance among the variables analyzed. The first component (41%) was related with the protein concentration whereas the second (35 %) represent test weight. The relation between test weight and protein as well as yield and protein were negative, there was no direct relation between yield and test weight. Cluster analysis placed the genotypes into three different groups. A comparison of the traits for each group showed that the genotypes in the first group had the highest yield followed by medium and low for the second and third group respectively. The test weight was similar for all three groups, while protein content was similar for group one and two and was less for the third group. The classification did not correspond to the three precipitation zones used. We can conclude that ACP and cluster are useful methods for the classification and description of the relations that occur between winter wheat characteristics. The obtained non-correlated variables may be used for further analysis, as well as it is important to include more locations and hard wheat genotypes.