霍尔巴绵羊形态结构主成分分析

Principal Component Analysis of Morphological Structure of Huoerba Sheep

为描述霍尔巴绵羊的体型外貌,于2021年4月在西藏日喀则市某公司随机选取131只成年母羊进行表型测量,并对相关变量进行分析、拟合以及预测。结果表明:除臀长、头长、头宽外,体长、体高、胸围、臀高、胸深、胸宽、臀宽、管围、耳长与体重呈极显著正相关(P<0.01),胸围与体重相关系数最高。基于14个表型指标进行主成分分析,提取的4个成分占总方差65.52%,其中PC1与体型相关指标有较高关联,PC2与体型形状指标相关,PC3与头部指标相关,耳长对因子构成的贡献最小。利用主成分分析提取因子拟合线性回归方程(R^(2)=0.843),较直接运用测量指标拟合(R^(2)=0.742)所得到的模型准确度更高。综上,主成分提取的因子可用于霍尔巴绵羊育种过程定向性状指标选择。

To explore the body shape and appearance of Huoerba sheep,131 adult ewes were randomly selected from a company at Shigatse City in Tibet in mid of April 2021 for phenotypic measurement,and the related vari⁃ables were further analyzed and predicted.The results demonstrated that except hip length,head length and head width,the body length,body height,abdominal circumference,chest circumference,hip height,chest depth,chest width,hip width,tube circumference and ear length were positively correlated with body weight sig⁃nificantly(P<0.01),and the chest circumference is strongly correlated with the body weight.Based on the princi⁃pal component analysis of 14 phenotypic indicators,the extracted four components accounted for 65.52%of the total variance,.Among them,PC1 was highly correlated with somatotype related indicators,PC2 was correlated with somatotype shape indicators,PC3 was correlated with head related indicators,and ear length had the least contribution to factor composition.The accuracy of linear regression equation(R^(2)=0.843)fitted by principal com⁃ponent analysis is higher than that of measurement index(R^(2)=0.742).Taken together,these factors extracted from principal component analysis can be useful predictor for directional trait selection in the breeding process of Huoerba sheep to determine its body shape.