玉米自交系抗旱相关性状的主成分分析与模糊聚类  

Principal component analysis and fuzzy clustering on drought tolerance related traits of maize inbred lines

作　　者：胡树平[1] 苏治军[1] 于晓芳[1] 陈春梅 高聚林[1] 吕佳雯[3] 王鹏飞[1] 许子清 

机构地区：[1]内蒙古农业大学,内蒙古呼和浩特010019 [2]内蒙古种子管理站,内蒙古呼和浩特010010 [3]包头市农业科学研究所,内蒙古包头市014013 [4]包头市种子管理站,内蒙古包头市014010

出　　处：《干旱地区农业研究》2016年第6期81-88,176,共9页Agricultural Research in the Arid Areas

基　　金：内蒙古杰出青年基金(2011JQ02);国家玉米产业技术体系(CARS-02-63);国家粮食丰产科技工程(2011BAD16B13;2012BAD04B04;2013BAD07B04;2011BAD16B14)

摘　　要：采用主成分分析(Principal Components Analysis,PCA)和模糊聚类(Fuzzy Clustering,FC)分析,对51个玉米自交系在正常灌溉和干旱胁迫条件下的主要农艺形状及部分生理指标进行分析并对其抗旱性进行综合评价。主成分分析结果表明,4个综合主成分(2012TMC:丰产性因子-雄穗发育性因子,光合调控性因子,生长发育性因子,繁茂性因子;2012HT:丰产性因子-生长发育性因子,光合调控因子,繁茂性因子,雄穗发育因子)可代表两个地区(2012TMC和2012HT)玉米自交系抗旱性91.05%的原始数据信息量。利用抗旱性度量值进行模糊聚类,可将51个玉米自交系的抗旱性划分为4类,即耐旱、中度耐旱、中度干旱敏感和干旱敏感。其中H201、H21、英64、吉842、早49、吉8415、东46和沈137抗旱性强且稳定性好。利用主成分分析,结合模糊聚类的方法进行玉米自交系的抗旱性综合评价,不仅可以避免单一指标的片面性,而且可以较好地揭示玉米自交系抗旱相关性状与其抗旱性的关系。The purpose of this paper is to explore the methods of drought tolerance evaluation and to make overall evaluation on the relationship between some agronomic traits as well as physiological indicators,such as photosynthesis,and the drought tolerance of maize inbred lines cultivars.Principal Component Analysis（PCA） and Fuzzy Clustering（FC）are used in the paper aimed at assessing the drought tolerance of 51 maize inbred lines cultivars based on some related performance under the conditions of normal irrigation and drought stress.The result of PCA shows that 4 principal components can represent 91.05%（the average between 2012 TMC and 2012HT） information of the raw data of the traits related to drought tolerance of maize inbred lines.According to FC,the drought tolerance of those 51 maize inbred lines cultivars can be divided into 4 parts,drought tolerance,moderate drought,moderate drought sensitive and drought sensitive.Which could reflect the drought characteristics of the origins and planting areas of those cultivars.Among those cultivars,the drought tolerance of H201、H21 and Ying 64 are better.Based on the above contents,it comes to a conclusion that it is an effective way to comprehensively evaluate the drought tolerance of foxtail millet by Principal Component Analysis together with Fuzzy Clustering,which not only can avoid the bias and the instability of single factor analysis,but also better reflect the relationship between different drought tolerance related traits and drought tolerance in maize inbred lines.

关 键 词：玉米自交系 农艺性状 生理指标 抗旱性 主成分分析 模糊聚类 

分 类 号：S513[农业科学—作物学] S503.4