Resumen de Genetic characterization based on the best linear unbiased predictor of traits related to cluster architecture of vitis vinifera l.
The genotypic effect and the interaction between genotype and gibberellic acid (GA3) treatment (g×t) of 23 traits associated with cluster architecture was characterized with a progeny ?Ruby Seedless? × ?Sultanina? (n = 140). The genotypic effects were significant for all the traits and g×t effect was significant only for rachis length (rl), fresh weight (rw), lateral shoulder length (sl) and number of berries per cluster (rb). Best linear unbiased predictors (BLUPs) of these effects were calculated. Multivariate factorial analyses of BLUPs showed highly significant correlations and a latent correlation structure among the traits. The BLUPs of rl, rw, number of internodes (ri), and sl were responsible for 27.8% of the total variance (first factor). A quantitative trait loci (QTLs) analysis showed that the linkage group 18 (LG18) and LG5 harbored several significant QTLs for the first factor. These QTLs were supported mainly by a paternal additive effect. Considering g×t, six significant QTLs were mapped in LG17, LG10, LG14 and LG18. These QTLs had mainly maternal and dominant effects. The correlation structure together with QTL analyses revealed possible pleiotropic effects. In conclusion the QTLs detected indicate that rl, rw, ri and sl have a clear genetic basis and due to their importance in the total variance they are good determinants of the genetic diversity of cluster architecture.
