Comprehensive evaluation of the quality of different bayberry varieties in Chongqing based on principal component analysis

【Objective】In recent years, bayberry, as a functional fruit, has been increasingly favored by consumers. Fruit quality is the primary factor determining the commodity value and economic benefits of fruits, and it is also the main basis for evaluating the quality of germplasm resources. It is usually evaluated by a combination of multiple trait indicators that affect internal and external quality. At present, many studies have comprehensively applied multiple analytical methods to fruit quality evaluation, such as pear, passion fruit, citrus, etc. However, reports on the comprehensive evaluation of the quality of bayberry fruits, especially those of wild varieties and introduced varieties in Chongqing, are almost nonexistent. Therefore, this study aimed to conduct a comprehensive evaluation of the fruit quality of different bayberry resources in Chongqing to screen out the core trait indicators suitable for evaluating the fruit quality in this area and to establish a comprehensive evaluation system for bayberry fruit quality suitable for this region. At the same time, excellent germplasm would be screened out to provide some reference for the efficient and scientific evaluation, selection and breeding of new variety, and promotion of the bayberry germplasm resources in Chongqing.【Methods】29 resources of bayberry fruits were collected from the germplasm repository of the Fruit Research Institute of the Chongqing Academy of Agricultural Sciences (including 21 cultivated varieties and 8 local wild resources) as materials in 2024. A total of 33 fruit quality indicators were determined, including 8 appearance quality indicators, 10 sugar acid and nutrient indicators, 7 antioxidant capacity indicators, and 8 mineral elements. Multiple statistical analysis methods, such as difference analysis, correlation heatmap analysis, cluster analysis, and principal component analysis, were used to comprehensively evaluate and rank the quality of the bayberry fruits.【Results】The differential analysis showed that there were significant differences in the degree of variation of various quality traits of the 29 bayberry fruits. In terms of appearance quality, the coefficient of variation of the fruit shape index and edible rate was small (only 4.72% and 5.64%), and the coefficient of variation of the yellow-blue color difference b* value was large (78.93%). The coefficient of variation of the soluble solids and total sugar content was small (only 9.54% and 8.97% ), and the coefficient of the variation of other indicators was between 14.04% and 57.48%. The results showed that bayberry was sucrose accumulation fruit and citric acid dominant fruit. In terms of phenolic compounds and antioxidant capacity, the coefficient of the variation of quercetin and anthocyanins was relatively high, at 75.21% and 85.93%, respectively. Among the different resources, WM-17 and Heiruilin were the varieties with the highest total phenol and flavonoid content, respectively, Heiruilin was the variety with the strongest comprehensive antioxidant capacity, and Shuijingzhong was the variety with the lowest total phenol content. In terms of mineral nutritional quality, the coefficient of the variation of different mineral elements ranged from 15.64% to 94.39%. Among them, Fe, as the trace element with the highest content, had a relatively high coefficient of variation (76.11%), while the coefficient of the variation of Se content was as high as 94.39%, indicating extremely high variability. The correlation analysis showed that there was a significant or extremely significant correlation between the various quality indicators of different varieties of bayberry. For instance, all the color difference parameters were significantly negatively correlated with the antioxidant activity indicators, and the synthesis and function of polyphenolic substances were closely and positively correlated with the multiple elements such as Ca, Mg, Fe, etc. However, the six indicators of the soluble solids, malic acid, fructose, glucose, amino acids, and selenium content showed almost no significant correlation with other indicators, indicating that the correlation strength between the different indicators would vary. The purpose of simplifying the quality indicator evaluation system could be achieved by screening the indicators with extremely high or low correlation. Based on the correlation analysis results and coefficient of variation, 18 quality indicators with excessively high or low correlation and a small coefficient of variation were excluded, and only the remaining 15 indicators were subjected to the principal component analysis. The results classified 15 indicators into 5 principal components, with a cumulative contribution rate of 84.034% and principal component eigenvalues ranging from 1.848 to 4.183. The five principal components could be divided into two categories, reflecting the functional traits of the bayberry fruit (including the 1st, 3rd, 4th, and 5th principal components) and the appearance and flavor quality (including the 2nd principal component). At the same time, six quality indicators were selected as the core indicators for evaluating the quality of bayberry, including iron ion reduction antioxidant capacity (FRAP), average mass, sucrose content, quercetin content, Ca content, and vitamin C content. Based on the score coefficients of each principal component, 15 standardized fruit trait values were substituted into these 5 principal components to obtain the scores of each variety on the 5 principal components, namely F1-F5. Using the ratio of the variance contribution rate of each principal component to the cumulative variance contribution rate as the weight coefficient, a comprehensive score model for the bayberry fruit quality indicators was constructed as follows: F- score=0.332F1 + 0.221F2 + 0.152F3 + 0.149F4 + 0.147F5. The comprehensive score of the fruit quality for the 29 bayberry resources was ultimately calculated to vary between -0.954 and 0.897, with a higher F-score indicating better overall quality of bayberry. When the Euclidean distance was between 22 and 25, the cluster analysis grouped the 29 bayberry materials into two clusters, with the first cluster including 17 materials, which were cultivated superior varieties, and the second cluster including 12 materials, which were wild varieties. By combining the principal component comprehensive scores, it was found that the overall ranking of the two groups was Ⅰ＞Ⅱ, which was consistent with reality.【Conclusion】There were significant differences in quality indicators among different resources of the bayberry fruits. A comprehensive screening identified superior bayberry germplasm resources such as Heiruilin, HG-15, Chise, Heijing, and Zaoqimimei. Among them, Heiruilin (0.897) had the highest comprehensive score, far exceeding the second-ranked HG-15 (0.699), and performed the best in the overall evaluation of the fruit quality of the 29 bayberry resources. The research results would provide guidance for the breeding and promotion of improved bayberry varieties in Chongqing.