北方农业学报 ›› 2025, Vol. 53 ›› Issue (6): 43-54.doi: 10.12190/j.issn.2096-1197.2025.06.04

• 分子生物学·种质资源 • 上一篇    下一篇

基于主成分分析的谷子主要农艺性状综合评价

姚贵军1, 樊永红2, 李星星1, 苏乐平1, 牛宏伟1, 郭玮1, 韩芳1, 周雪1, 袁宏安1   

  1. 1.延安市农业科学研究院,陕西 延安 716000;
    2.宜川县农产品质量安全检测站,陕西 宜川 716200
  • 收稿日期:2025-05-20 发布日期:2026-01-26
  • 通讯作者: 周 雪(1989—),女,农艺师,硕士,主要从事谷子品种选育及栽培技术方面的研究工作。
  • 作者简介:姚贵军(1969—),男,高级农艺师,学士,主要从事小杂粮新品种选育及示范推广工作。
  • 基金资助:
    现代农业产业技术体系建设专项(CARS-06-14.5-B28); 延安市科技计划项目(2023-CYL-172)

Comprehensive evaluation of major agronomic traits of foxtail millet based on principal component analysis

YAO Guijun1, FAN Yonghong2, LI Xingxing1, SU Leping1, NIU Hongwei1, GUO Wei1, HAN Fang1, ZHOU Xue1, YUAN Hong′an1   

  1. 1. Yan′ an Academy of Agricultural Sciences, Yan′ an 716000, China;
    2. Yichuan County Agricultural Product Quality and Safety Inspection Station, Yichuan 716200, China
  • Received:2025-05-20 Published:2026-01-26

摘要: 【目的】探究谷子农艺性状之间的关系,为其新品种选育和遗传改良提供依据。【方法】对30份谷子高代品系(种)的农艺性状进行变异分析、相关性分析、主成分分析和聚类分析。【结果】参试谷子品系(种)农艺性状变异系数由高到低排序为:穗粒重>单穗重>茎粗>产量>穗粗>株高>穗长>生育期>千粒重,变异系数为8.52%~29.24%。茎粗、穗长均与单穗重、穗粒重呈极显著正相关关系(P<0.01);单穗重与穗粒重呈极显著正相关关系(P<0.01);产量与单穗重、穗粒重呈显著正相关关系(P<0.05)。主成分分析将参试谷子品系(种)的9个农艺性状简化为3个主成分因子,即产量因子、株型因子和物候期因子,累计贡献率为71.913%,20H979、22YH-122、22HB249-217品系综合评分位居前三位。聚类分析将参试谷子品系(种)农艺性状分为3大类群,第Ⅰ类群生育期较长,平均株高、穗粗、单穗重、穗粒重、产量均较低;第Ⅱ类群生育期较短,平均株高、单穗重、穗粒重、产量较高;第Ⅲ类群生育期较长,平均株高较低,穗长、穗粗、千粒重、产量较高。【结论】第Ⅲ类群中的20H979、22YH-122和22HB249-217品系兼具优异综合性状与显著产量潜力,既是优良的候选品系,又是培育突破性品种的骨干亲本,具有直接应用与遗传改良的双重价值。

关键词: 谷子, 农艺性状, 变异分析, 相关性分析, 主成分分析, 聚类分析

Abstract: 【Objective】 To explore the relationships among agronomic traits of foxtail millet and to provide a basis for new variety breeding and genetic improvement.【Methods】 Variation analysis,correlation analysis,principal component analysis,and cluster analysis were conducted on agronomic traits of 30 advanced-generation foxtail millet lines(varieties).【Results】 The coefficients of variation of agronomic traits in the tested foxtail millet lines(varieties),ranked from high to low,were:grain weight per panicle>weight per panicle>stem diameter>yield>panicle diameter>plant height>panicle length>growth period>thousand-grain weight,with the coefficients of variation ranging from 8.52% to 29.24%. Stem diameter and panicle length were both extremely significantly and positively correlated with weight per panicle and grain weight per panicle(P<0.01);weight per panicle showed a extremely significant positive correlation with grain weight per panicle(P<0.01);yield was significantly and positively correlated with weight per panicle and grain weight per panicle(P<0.05). Principal component analysis reduced the nine agronomic traits of the tested foxtail millet lines(varieties)to three principal component factors,namely the yield factor,plant architecture factor,and phenological factor,with a cumulative contribution rate of 71.913%. The comprehensive scores of lines 20H979,22YH-122,and 22HB249-217 ranked in the top three. Cluster analysis classified the tested foxtail millet lines(varieties)into three major clusters. ClusterⅠ was characterized by a longer growth period and relatively lower average plant height,panicle diameter,weight per panicle,grain weight per panicle,and yield. ClusterⅡ had a shorter growth period and higher average plant height,weight per panicle,grain weight per panicle,and yield. ClusterⅢ exhibited a longer growth period,lower average plant height,and higher panicle length,panicle diameter,thousand-grain weight,and yield.【Conclusion】 Lines 20H979,22YH-122,and 22HB249-217 in ClusterⅢ possess excellent comprehensive traits and significant yield potential. These lines are not only superior candidate lines but also backbone parents for breeding breakthrough varieties,showing dual value for direct application and genetic improvement.

Key words: Foxtail millet, Agronomic traits, Variation analysis, Correlation analysis, Principal component analysis, Cluster analysis

中图分类号: 

  • S515