Genetic diversity evaluation and selection methods of sweet potato hybrid F1 population based on SSR markers and phenotypic detection

Abstract

Sweet potato (Ipomoea batatas (L.) Lam.) is a vital global crop, with breeding focused on both high starch and high yield. Hybrid populations are crucial for genetic improvement, but research on sweet potato hybrid F₁ populations remains limited. To explore the genetic laws of important traits in hybrid progenies, this study investigates the genetic diversity and efficient selection methods of the hybrid F₁ population from crossing between Yushu No.12 (high starch content) and Luoxushu No.9 (high yield) using phenotypic detection and SSR markers. Coefficients of variation, genetic distances, and similarity coefficients results showed that the F₁ population has rich genetic diversity. The parents and F₁ progenies could be clustered into 4 and 6 categories based on phenotypic detection and SSR markers, respectively. The results of transgressive inheritance analysis and cluster analysis showed that the hybrid F₁ population of sweet potato was closer to the female parent and might exhibit matroclinous inheritance. Based on the principal component analysis (PCA) results, a comprehensive scoring model was developed to select superior progeny. Correlation analysis revealed a strong link (r = 0.6420) between the hardness and starch content of storage root, suggesting hardness could be used for rapid screening high-starch materials. Mantel test showed SSR markers as more reliable for evaluating genetic diversity than phenotypic analysis. These findings uncover the genetic diversity information of sweet potato F₁ generation, and provide strategies for the rapid and accurate selection of hybrid progenies, and lay theoretical foundation for deciphering the genetic mechanisms of important traits in sweet potato.

Keywords: Genetic diversity; Hybrid F1 population; Phenotypic trait; SSR; Sweet potato.

Fig. 1

Frequency distribution of 11 phenotypic traits Kur, kurtosis; Ske, skewness; Sig., one-sample Kolmogorov-Smirnov significance value, which is equivalent to the p-value in this study. If Sig. (i.e., p) > 0.05, it means the trait is normally distributed

Fig. 2

Boxplot analysis of phenotypic traits of the top 20 and the rest in the F₁ population SG, a group composed of 20 plants selected from the F₁ population; UG, a group composed of unselected plants from the F₁ population

Fig. 3

The correlation analysis of 11 phenotypic traits The correlation coefficient threshold values were α = 0.01, r = 0.1478 and α = 0.05, r = 0.1127

Fig. 4

Genetic distance matrix comparison of SSR markers and phenotypic detection D1, Genetic distance based on SSR maker; D2, Genetic distance based on phenotypic detection

Fig. 5

Cluster analysis based on phenotypic detection (A) and SSR markers (B) The different colors represent Category I-IV (SG1-SG4) identified by cluster analysis based on phenotypic detection in Fig. 2A, and the different colors represent Category I-VI (SG1-SG6) identified by cluster analysis based on SSR markers in Fig. 2B. The female parent (304) and the male parent (305) are highlighted in red in the figure. The same applies to Fig. 7

Fig. 6

Frequency distribution of 11 phenotypic traits in various clusters based on phenotype detection clustering analysis results

Fig. 7

Comparison of cluster analysis based on phenotypic detection and SSR markers The dendrogram on the left represents the results of the cluster analysis based on SSR molecular markers, while the dendrogram on the right is based on phenotypic detection. Identical plant numbers are connected by lines in the middle. For clarity, the color of the connecting lines follows the dendrogram based on SSR makers