Articles
SELECTION OF MORPHOLOGICAL QUANTITATIVE VARIABLES IN FIG CHARACTERIZATION
Article number
798_12
Pages
103 – 108
Language
English
Abstract
The objective of this work was to select highly discriminant quantitative variables among those already proposed for fig fingerprinting.
As a first step, the analysis was carried out with 35 fig accessions from diverse geographical origins.
A total of 81 quantitative variables were studied (12 from the tree, 24 from the breba crop, 24 from the second crop and 21 from the leaves). According to the intravarietal variance data, 30 traits were selected (two from the tree, 11 from the breba crop, 11 from the second crop and 6 from the leaves). These 30 selected variables were grouped by Principal Components Analysis to eliminate redundancy resulting in five principal components that explain 93.80% of the total variability.
The results revealed that some carefully chosen highly discriminant variables are sufficient to characterize the genotypes studied.
As a first step, the analysis was carried out with 35 fig accessions from diverse geographical origins.
A total of 81 quantitative variables were studied (12 from the tree, 24 from the breba crop, 24 from the second crop and 21 from the leaves). According to the intravarietal variance data, 30 traits were selected (two from the tree, 11 from the breba crop, 11 from the second crop and 6 from the leaves). These 30 selected variables were grouped by Principal Components Analysis to eliminate redundancy resulting in five principal components that explain 93.80% of the total variability.
The results revealed that some carefully chosen highly discriminant variables are sufficient to characterize the genotypes studied.
Publication
Authors
E. Giraldo, M. López-Corrales, J.I. Hormaza
Keywords
Ficus carica, descriptors, germplasm management, characterization
Online Articles (49)