Acta Agriculturae Zhejiangensis ›› 2026, Vol. 38 ›› Issue (2): 248-257.DOI: 10.3969/j.issn.1004-1524.20250508

• Horticultural Science • Previous Articles     Next Articles

Indicator screening and model development for evaluating pumpkin fruit quality

LIU Jing1(), WANG Jian2, HUANG Yu1, WU Xiaohua2, GUO Xuanhe1, WANG Ying2, LI Guojing2, XU Xiaojiang1,*()   

  1. 1. Shaoxing Academy of Agricultural Sciences, Shaoxing 312003, Zhejiang, China
    2. Institute of Vegetables, Zhejiang Academy of Agricultural Sciences, Hangzhou 310021, China
  • Received:2025-07-25 Online:2026-02-25 Published:2026-03-24

Abstract:

To establish a comprehensive evaluation model for pumpkin fruit quality, 51 accessions of Cucurbita moschata and 95 accessions of Cucurbita maxima were used as materials. Eleven fruit quality indicators were measured, and correlation analysis, factor analysis, cluster analysis, and two-dimensional factor ordination were applied to construct the evaluation model. The results indicated that moisture content, cellulose, soluble sugar, β-carotene content, and sweetness of C. moschata were significantly higher than those in C. maxima. In contrast, starch content, amylose content, viscosity, dryness/wetness, and fibrous texture in C. maxima were significantly higher than those in C. moschata. The coefficients of variation for the 11 fruit quality indicators ranged from 9.00% to 69.04%. Moisture content showed the smallest coefficient of variation, while soluble sugar content exhibited the largest. Factor analysis extracted five common factors, with a cumulative variance contribution rate of 90.27%. The first common factor (F1) had the highest variance contribution rate (34.60%) and was primarily associated with moisture content, soluble sugar content, starch content, and amylose content. Based on this, an evaluation model for pumpkin fruit quality was established: Y=0.383 2F1+0.263 2F2+0.132 1F3+0.127 5F4+0.094 0F5. Furthermore, correlation analysis and cluster analysis simplified the eleven original quality indicators into three representative indicators: starch content, fibrous texture, and soluble sugar content. The scatter distribution of F1 and the second common factor (F2) confirmed high consistency between model-predicted values and actual measured values. The evaluation model developed in this study provides a methodological basis for pumpkin fruit quality assessment and serves as a reference for breeding high-quality pumpkin varieties.

Key words: pumpkin, fruit, quality, factor analysis, cluster analysis, comprehensive quality evaluation

CLC Number: