浙江农业学报 ›› 2018, Vol. 30 ›› Issue (1): 50-57.DOI: 10.3969/j.issn.1004-1524.2018.01.07

• 园艺科学 • 上一篇    下一篇

菜用大豆主要农艺性状的相关性、聚类及主成分分析

贺礼英, 尹成杰, 黄守程, 何庆元, 舒英杰*   

  1. 安徽科技学院 农学院, 安徽 凤阳 233100
  • 收稿日期:2017-09-14 出版日期:2018-01-20 发布日期:2018-02-09
  • 通讯作者: 舒英杰,E-mail: shuyj@ahstu.edu.cn
  • 作者简介:贺礼英(1993—),女,山西汾阳人,硕士研究生,主要从事大豆栽培生理研究。E-mail: 1876396278@qq.com
  • 基金资助:
    安徽省自然科学基金项目(1608085MC69); 安徽省高校优秀青年人才支持计划重点项目(gxyqZD2016216); 安徽科技学院稳定人才项目; 安徽省教育厅自然科学研究重点项目(KJ 2016A172)

Correlation, clustering and principal component analysis of primary agronomic traits of vegetable soybean

HE Liying, YIN Chengjie, HUANG Shoucheng, HE Qingyuan, SHU Yingjie*   

  1. College of Agronomy, Anhui Science and Technology University, Fengyang 233100, China
  • Received:2017-09-14 Online:2018-01-20 Published:2018-02-09

摘要: 对沿淮地区广泛种植的41个菜用大豆品种的生育期、株高、茎粗、倒伏性、主茎节数、结荚高度、单株荚数、单株有效荚数、荚长、荚宽、单荚粒数、鲜豆百粒质量等12个主要农艺性状进行了测定,并进行了变异、相关、聚类及主成分分析。结果表明,41个菜用大豆品种的12个农艺性状均存在较大的变异,其中结荚高度变异最大(变异系数31.41%),荚宽的变异最小(变异系数为6.45%)。相关性分析表明,除生育期与株高、单株荚数、荚长,荚长与荚宽间的相关系数大于0.5外,其他指标间的相关系数的绝对值小于0.5。通过聚类分析,利用类平均法在欧氏距离5.597 3处可将41个菜用大豆品种划分为4大类群, 各类群的农艺性状差异明显。主成分分析表明,生育期、株高、茎粗、倒伏性、主茎节数、结荚高度及单株荚数等7个农艺性状累计贡献率为87.028 1%,可以反映出菜用大豆主要农艺性状的基本特征。

关键词: 菜用大豆, 农艺性状, 相关分析, 聚类分析, 主成分分析

Abstract: Variation, correlation analysis, cluster analysis and principal component analysis of primary agronomic traits of 41 vegetable soybean varieties that were planted around Huaihe River, including growth period, plant height, stem diameter, lodging, nodes on main stem, pod height, pods per plant, effective pods per plant, pod length, pod width, grains per pod and fresh weight of 100 grains were studied. The results showed that 12 agronomic traits of 41 vegetable soybean varieties had large variation, and the variation of pod height was the highest (variation coefficient was 31.41%), the least variation coefficient was 6.45% of pod width. Correlation analysis showed that, the correlation coefficients between growth period and plant height, pods per plant, pod length, as well as that between pod length and pod width were larger than 0.5, the absolute value of correlation coefficient between the other traits was less than 0.5. By cluster analysis, 41 species of vegetable soybean varieties were divided into four groups at 5.597 3 of the Euclidean distance, and the agronomic traits of each type were different. The principal component analysis showed that the total contribution rate of the seven agronomic traits that included growth period, plant height, stem diameter, lodging, nodes on main stem, pod height and pods of per plant was 87.028 1%, which could reflect the main agronomic traits of vegetable soybean.

Key words: vegetable soybean, agronomic traits, correlation analysis, cluster analysis, principal component analysis

中图分类号: