浙江农业学报 ›› 2024, Vol. 36 ›› Issue (4): 748-759.DOI: 10.3969/j.issn.1004-1524.20230489

• 作物科学 • 上一篇    下一篇

基于主成分分析的苦荞麦重组自交系农艺性状综合评价

薛贤滨1(), 贾琼1, 陈峥峰1, 黎瑞源2, 陈庆富1, 石桃雄1,*()   

  1. 1.贵州师范大学 生命科学学院,贵州省荞麦工程技术研究中心,贵州 贵阳 550001
    2.贵州师范大学 贵州省信息与计算科学重点实验室,贵州 贵阳 550001
  • 收稿日期:2023-04-13 出版日期:2024-04-25 发布日期:2024-04-29
  • 作者简介:薛贤滨(1998—),男,安徽合肥人,硕士研究生,研究方向为荞麦种质资源保育与创新。E-mail: xuexianbin1998@126.com
  • 通讯作者: *石桃雄,E-mail:shitaoxiong@126.com
  • 基金资助:
    国家自然科学基金(32260489);国家现代农业产业技术体系荞麦育种岗位科学家专项资金(CARS-07-A5);贵州师范大学学术新苗培养及创新探索专项(黔科合平台人才〔2017〕5726-18)

Comprehensive evaluation of agronomic characteristics of recombinant inbred lines of Tartary buckwheat based on principal component analysis

XUE Xianbin1(), JIA Qiong1, CHEN Zhengfeng1, LI Ruiyuan2, CHEN Qingfu1, SHI Taoxiong1,*()   

  1. 1. Research Center of Buckwheat Industry Technology, School of Life Sciences, Guizhou Normal University, Guiyang 550001, China
    2. Key Laboratory of Information and Computing Science of Guizhou Province, Guizhou Normal University, Guiyang 550001, China
  • Received:2023-04-13 Online:2024-04-25 Published:2024-04-29
  • Contact: SHI Taoxiong

摘要:

为筛选出综合性状优良的种质,为苦荞麦高产品种的选育推荐材料,对选取的58个具有高产潜质的重组自交系(RILs)及其双亲的10个主要农艺性状进行遗传变异分析、相关性分析、主成分分析和聚类分析。结果表明,58个RILs各性状的变异系数介于5.71%~31.55%,其中,产量、生育期和主茎分枝数的变异系数较大,粒宽和籽粒周长的变异系数较小。产量与籽粒面积、籽粒周长、株高和千粒重呈极显著(P<0.01)正相关,与粒长和粒宽呈显著(P<0.05)正相关,与生育期呈极显著负相关,产量与上述指标的相关系数绝对值从大到小依次为千粒重>株高>籽粒面积>生育期>籽粒周长>粒宽>粒长。主成分分析结果表明,前4个主成分的累计贡献率达86.987%,分别是粒形与产量因子(39.940%)、粒宽因子(24.478%)、株高因子(11.667%)、主茎分枝数与生育期因子(10.893%)。基于综合评价结果及RILs与亲本间的方差分析,共筛选出R64、R103、R164、R84、R192、R153和R214等7个综合性状优良的非米荞型RILs。这7个株系在聚类分析中均被划分在了高产、大粒、高秆和生育期短的C2类群,可将其作为示范推广品种或西南地区常规苦荞育种的优良种质资源加以利用。米荞型株系R52、R198和R101的产量极显著或显著高于米荞型亲本小米荞,可用于苦荞麦高产薄壳品种的选育。

关键词: 苦荞麦, 重组自交系, 农艺性状, 遗传变异, 主成分分析, 聚类分析

Abstract:

In order to screen accessions with better comprehensive agronomic traits and materials for breeding high yield Tartary buckwheat varieties, ten agronomic traits of 58 recombinant inbred lines (RILs) with high yield potential and the two parents were evaluated by genetic variation analysis, correlation analysis, principal component analysis and cluster analysis. The results showed that the coefficient of variation ranged from 5.71% to 31.55% for the 10 agronomic traits of 58 RILs. The coefficient of variation of seed yield, growth period and branch number of main stem were larger, and the coefficient of variation of seed width and seed perimeter were smaller. The seed yield was significantly positively correlated with seed area, seed perimeter, plant height and 1 000-seed weight at P<0.01 level, and significantly positively correlated with seed length and seed width at P<0.05 level. There was a significant negative correlation between seed yield and growth period at P<0.01 level. The absolute value of correlation coeffecient of the above indexes decreased as follows: 1 000-grain weight>plant height>seed area>growth period>seed perimeter>seed width>seed length. The results of principal component analysis showed that the cumulative contribution rate of the top four principal components was 86.987%, and the four principal components were seed shape and yield factor (39.940%), seed width factor (24.478%), plant height factor (11.667%), and branch number of main stem and growth period factor (10.893%). Based on the comprehensive evaluation results and analysis of variance between RILs and parents, seven excellent non-rice RILs were screened out, namely, R64, R103, R164, R84, R192, R153 and R214. The seven RILs were grouped into the C2 group with high yield, large seed, high stem and short growth period in cluster analysis, which could be used as promotion varieties for demonstration and high quality germplasm resources of conventional Tartary buckwheat breeding in southwest China. The yield of three rice type RILs, namely, R52, R198 and R101, was significantly higher than that of the parent Xiaomiqiao at P<0.01 or P<0.05 level, which could be used as materials for the breeding of high-yield and thin-shell Tartary buckwheat.

Key words: Tartary buckwheat, recombinant inbred lines, agronomic traits, genetic variation, principal component analysis, cluster analysis

中图分类号: