Use of multivariate analysis in evaluation of oil flax variety samples in control nursery
2014
Ivanova,E.V. | Andronik,E.L. | Maslinskaya,M.E.
Literature review showed the increasing role of statistical data processing using multivariate analysis in breeding. The research was realized in the conditions of Republic of Belarus in 2012-2013 in the Institute of Flax, Ustie (Orsha district, Belarus) on sod-podzolic soil. Factor analysis was conducted, and such informative characters as seed yield, resistance to lodging, oil content, oil yield were found. As a result of clustering of samples in the control oil flax nursery by the informative characters, forms for breeding purposes contrasting by phenotype were isolated. The results of the factor analysis made it possible to raise the profile descriptions of the control samples nursery linseed by determining the optimal set of features. For the studied genotypes such signs had appeared: lodging resistance (load factor = 0.75), seed yield (-0.82), the oil content in the seeds (-0.73) and the collection of oil (-0.93). Cluster analysis of informative features allowed to estimate the similarity between samples of nursery and more accurately pick contrasting phenotypic patterns: C-6, C-7, C-8, C-13, C-16, C-20, C-22 for the selection and further study.
显示更多 [+] 显示较少 [-]