摘要
理想株型可以显著提高作物产量,但谷子中株型性状与产量性状之间的关系还不清楚。旨为谷子株型育种提供理论基础和种质资源,以矮宁黄×晋谷21号杂交自交构建的126个F 6重组自杂交系(RIL)群体为材料,在长治、榆次和大同3个生态环境条件下,对10个株型性状(株高、主茎长度、穗长、穗下节长、穗粗、分蘖数、节数、旗叶长、旗叶宽和旗叶面积)和3个产量性状(穗质量、穗粒质量和千粒质量)进行表型鉴定,基于最佳线性无偏估计(BLUE)进行相关性分析、偏相关分析、主成分分析、逐步回归分析和聚类分析。表型变异分析结果表明,双亲的株高和主茎长度在3个生态环境下均呈极显著差异,穗下节长和节数在2个生态环境下呈显著或极显著差异,穗长、穗粗、穗质量、穗粒质量和分蘖数在单一生态环境下呈显著或极显著差异。RIL群体中,株型和产量性状变异丰富,频率分布近似正态分布,13个性状的变异系数为6.86%~31.71%,除榆次的主茎长度外,其他性状在各生态环境下均存在双向超亲分离现象。相关性分析和偏相关性分析结果发现,穗质量、穗粒质量与株高、主茎长度、穗长、穗下节长和节数呈极显著正相关,而与分蘖数呈极显著负相关,穗质量与旗叶长呈显著正相关。主成分分析将13个性状简化为4个主成分,累计贡献率可达93.938%。多元回归分析结果显示,拟合度R 2为0.614,主茎长度、穗长、分蘖数是影响穗质量的主要因素。聚类分析将RIL群体划分为7类,其中第Ⅴ类群包括3份材料,产量最高,株高中等,综合性状好,可作为今后谷子株型育种的骨干亲本。
Ideal plant architecture can significantly improve crop yield,but the relationship between plant architecture traits and yield traits is still unclear in foxtail millet.To provide a theoretical basis and germplasm resource for plant architecture breeding in foxtail millet,10 plant architectures traits(plant height,main stem length,panicle length,panicle neck length,panicle diameter,tiller number,node number,length of flag leaf,width of flag leaf and area of flag leaf)and 3 yield traits(panicle weight,panicle grain weight and 1000-grain weight)were analyzed via 126 F 6 recombinant inbred lines(RIL)from a F 1 arrived from a cross between Aininghuang and Jingu 21 under three ecological environment consisting of Changzhi,Yuci and Datong.Based on the best linear unbiased estimation(BLUE),variance analysis,correlation analysis,partial correlation analysis,principal component analysis,multiple regression analysis and cluster analysis were conducted.The phenotype variation analysis showed there were an extremely significant difference between plant height and main stem length in three ecological environments,a significantly or extremely significant difference between panicle neck length and node number in two ecological environments,and the significant or extremely significant differences among panicle length,panicle diameter,panicle weight,panicle grain weight and tiller number in single ecological environment.In RIL population,an abundant variation was observed for 13 traits with the approximately normal frequency distribution,and the variation coefficients ranged from 6.86%to 31.71%.Except for the main stem length in Yuci,other traits showed a transgressive separation in three ecological environments.Correlation analysis and partial correlation analysis indicated that panicle weight and panicle grain weight were extremely significant positive correlated with plant height,main stem length,panicle length,panicle neck length and node number,while they were extremely significant negative correlated with tiller number,panicle weight was significantly positive correlated with flag leaf length.Principal component analysis simplified 13 traits into 4 principal components,and the cumulative contribution rate was up to 93.938%.The fitting degree R 2 of multiple regression analysis was 0.614,and main stem length,panicle length and tiller number were the main factors affecting panicle weight.The RIL population was divided into 7 groups via cluster analysis,among which GroupⅤconsisted of 3 materials with the highest yield,medium plant height and good comprehensive traits,which could be used as elite parents for ideal plant architecture breeding.
作者
何妙玲
王智兰
杜晓芬
韩康妮
连世超
李禹欣
成锴
李颜方
王军
HE Miaoling;WANG Zhilan;DU Xiaofen;HAN Kangni;LIAN Shichao;LI Yuxin;CHENG Kai;LI Yanfang;WANG Jun(School of Life Science,Shanxi University,Taiyuan 030006,China;Millet Research Institute,Shanxi Agricultural University,Shanxi Key Laboratory of Minor Crop Germplasm Innovation and Molecular Breeding,Shanxi Key Laboratory of Genetic Resources and Breeding in Minor Crops,Changzhi 046011,China)
出处
《华北农学报》
CSCD
北大核心
2023年第4期91-100,共10页
Acta Agriculturae Boreali-Sinica
基金
国家重点研发计划项目(2018YFD1000700)
山西省回国留学人员科研资助项目(HGKY2019101)
山西省农业科学院杂粮分子育种平台专项(YGC2019FZ3)
山西省青年基金项目(20210302124233)
中央引导地方科技发展资金项目(YDZJSX2022A043)。
关键词
谷子
重组自交系组自交系
株型
产量性状
相关分析
主成分分析
聚类分析
Foxtail millet
Recombination inbred line
Plant architecture
Yield traits
Correlation analysis
Principal component analysis
Cluster analysis