Genetic Diversity of Sour Passion Fruit Revealed by Predicted Genetic Values

Abstract

Core Ideas The sporophytic homomorphic self‐incompatibility system advanced the greatest diversity within populations. The phenotypic correlation network facilitates rapid observation of the relationships among groups of variables related to vegetative vigor, incidence of leaf diseases, and physicochemical properties of passion fruit. The correlation networks associated with the relative contribution of the variable can help in the selection of important variables. Sour passion fruit (Passiflora edulis Sims) is a highly suitable crop for small farmers because of its high market value and short time to maturity. However, passion fruit breeding programs need to provide cultivars to producers with quality, disease resistance, and production improvements. The objective of this study was to assess the genetic diversity among and within improved populations of sour passion fruit (SPF) using predicted genetic values, select the most divergent accessions, and evaluate the relationships among groups of traits of vegetative vigor, incidence of leaf diseases, and physicochemical properties of fruits. Twenty‐three genotypes were selected and divided into five populations with variable numbers of accessions. Phenotypic data were transformed into genotypic values using mixed model restricted maximum likelihood‐best linear unbiased prediction. We tested the hypothesis that there is differentiation among populations and between their components and the accessions. Graphic dispersion of the genotypes around the centroids of five populations was performed. The Tocher method was used to group the accessions. The relative importance of the variables and correlation networks was evaluated. Populations showed genetic variability; the hybrid combinations involving the population I accessions 1 (B1 29 PL1), 2 (B1 41 PL3), 4 (B2 9 PL 3), 5 (B2 37 PL3), 10 (G1 B1 9), and 19 (BRS GA1) can be exploited in breeding programs to improve the quality of SPF. The correlation network facilitates the observation of the relationships among groups of traits and helps in the selection of variables when associated with the relative contribution.