Application of cluster analysis and principal component analysis in ampelographic research of seedless vine cultivars

Venelin Roychev, Neli Keranova

Abstract: The phenotypic similarity and remoteness of 18 table seedless vine cultivars have been studied by means of Cluster Analysis and Principal Component Analysis (PCA). It has been found that, according to their phenological indices, they can be divided into three clusters; according to their agrobiological characteristics – into four clusters; according to their technological parameters – into four clusters. The researched seedless cultivars are arranged in four clusters depending on the three groups of indices – phenological, agrobiological and technological. The first cluster includes those of the cultivars which are characterized by small berry length and width and average weight of 100 berries, moderate sugar contents, low resistance to picking and pressure; the second one – the cultivars with maximum average cluster weight, high average weight of 100 berries and high average yield per decare; the third one – the cultivars with high percentage of mesocarp in the berry; the fourth one – the cultivars having the shortest budding phenophase and the highest average yield per vine and per decare. The three groups of indices are transformed into seven factors, explaining 87,015% of the total dispersion.

Keywords: ampelographic indices; Cluster Analysis; Principal Component Analysis (PCA); seedless vine cultivars

Date published: 2021-08-20