畜牧兽医学报 ›› 2023, Vol. 54 ›› Issue (9): 3677-3688.doi: 10.11843/j.issn.0366-6964.2023.09.009

• 遗传育种 • 上一篇    下一篇

基于主成分分析和因子分析估计新疆褐牛体型性状的遗传参数

胥磊1, 张梦华1, 张涛1, 耿娟2, 范守民3, 杨光维3, 郭杨2, 邓强2, 李金芝4, 刘江卫5, 黄锡霞1*, 王雅春6*   

  1. 1. 新疆农业大学动物科学学院, 乌鲁木齐 830052;
    2. 新疆维吾尔自治区畜牧总站, 乌鲁木齐 830004;
    3. 伊犁哈萨克自治州畜牧总站, 伊宁 835099;
    4. 塔城农牧科技有限公司, 塔城 834799;
    5. 新疆燕犇褐牛繁育发展有限责任公司, 昌吉 831200;
    6. 中国农业大学动物科学技术学院, 北京 100193
  • 收稿日期:2023-03-22 发布日期:2023-09-22
  • 通讯作者: 黄锡霞,主要从事动物遗传育种研究,E-mail:au-huangxixia@163.com;王雅春,主要从事分子及数量遗传学研究,E-mail:wang-yachun@cau.edu.cn
  • 作者简介:胥磊(1994-),男,新疆伊宁人,博士生,主要从事动物遗传育种研究,E-mail:q609468041@sina.com
  • 基金资助:
    国家重点研发计划项目(2021YFD1200903-8);国家自然科学基金(31360543);国家现代农业产业技术体系(CARS-36)

Genetic Parameter Estimates for Body Conformation in Xinjiang Brown Cattle Based on Principal Component Analysis and Factor Analysis

XU Lei1, ZHANG Menghua1, ZHANG Tao1, GENG Juan2, FAN Shoumin3, YANG Guangwei3, GUO Yang2, DENG Qiang2, LI Jinzhi4, LIU Jiangwei5, HUANG Xixia1*, WANG Yachun6*   

  1. 1. College of Animal Science, Xinjiang Agricultural University, Urumqi 830052, China;
    2. Xinjiang Uygur Autonomous Region Animal Husbandry Station, Urumqi 830004, China;
    3. Animal Husbandry Station of Yili State, Yining 835099, China;
    4. Tacheng Agricultural and Animal Husbandry Technology Co., Tacheng 834799, China;
    5. Xinjiang Yan Ben Brown Cattle Breeding Development Co., Changji 831200, China;
    6. College of Animal Science and Technology, China Agricultural University, Beijing 100193, China
  • Received:2023-03-22 Published:2023-09-22

摘要: 旨在估计新疆褐牛体型性状的遗传参数,同时应用主成分分析和因子分析进行降维后探索各主成分和因子之间的遗传规律,为育种方案制定提供参考。本研究采用SAS软件进行主成分和因子分析,利用BLUPF90软件采用平均信息约束最大似然法配合动物模型对9个新疆褐牛核心育种场的1 016头新疆褐牛泌乳牛的27个体型性状及主成分和因子得分表型进行遗传参数估计。研究结果表明,除蹄踵深度、蹄角度和乳房平衡性外,其它体型性状均属于中高遗传力性状。主成分(1~8;特征值 ≥ 1)和潜在因子(1~8;特征值 ≥ 1)都能够解释总方差的57%,各主成分和因子能够综合反映牛只体躯结构、肌肉度、尻部、肢蹄和泌乳系统的信息,各主成分和各因子的遗传力估计值范围分别为0.13(PC6)~0.46(PC1)和0.08(F3)~0.48(F1),基于主成分分析估计遗传相关的范围为-0.76(PC3-PC6)~0.74(PC3-PC4),因子分析估计遗传相关的范围为-0.75(F3-F6)~0.86(F1-F6)。结果提示,新疆褐牛体型性状中除蹄踵深度、蹄角度和乳房平衡性外,其它性状均属于中高遗传力性状,在体型性状的评估中可以考虑使用主成分和潜在因子,将多维度的体型性状简化为新的变量,这种方法可以避免分析大量相关性较高的性状,从而降低分析大量数据而造成的计算负担。

关键词: 新疆褐牛, 体型性状, 主成分分析, 因子分析, 遗传参数

Abstract: The objective of this study was to estimate the genetic parameters of body conformation traits in Xinjiang brown cattle. In addition, principal component analysis and factor analysis were applied to examine the genetic patterns among the main components and factors after reducing the dimensions. The aim was to provide valuable insights for the development of breeding programs. SAS software was utilized for principal component and factor analysis. BLUPF90 software was employed to estimate the genetic parameters for 27 body conformation traits and the phenotypes of principal component and factor scores. The data consisted of 1 016 lactating cows from 9 core breeding farms of Xinjiang brown cattle. The average information restricted maximum likelihood method with animal models was used for estimation. The findings of the study revealed that most body conformation traits exhibited medium to high heritability, with the exception of heel depth, foot angle, and udder balance. When analyzing the data, it was determined that the principal components (1 to 8; eigenvalues ≥ 1) and latent factors (1 to 8; eigenvalues ≥ 1) collectively accounted for 57% of the total variance. These principal components and factors provided valuable information regarding the body frame, muscularity, rump, feet&legs, as well as the mammary system of the cattle. The heritability estimates for each principal component and factor ranged from 0.13 (PC6) to 0.46 (PC1) and from 0.08 (F3) to 0.48 (F1), respectively. Furthermore, the genetic correlations estimated through principal component analysis varied from -0.76 (PC3-PC6) to 0.74 (PC3-PC4), while the factor analysis yielded correlations ranging from -0.75 (F3-F6) to 0.86 (F1-F6). These results indicate that most body conformation traits in Xinjiang brown cattle exhibit medium to high heritability, except for heel depth, foot angle, and udder balance. Additionally, both principal components and latent factors can be employed to evaluate body conformation traits, effectively reducing the multidimensional traits into new variables. This approach eliminates the need to analyze numerous highly correlated traits, thereby reducing the computational burden associated with handling extensive datasets.

Key words: Xinjiang brown cattle, body conformation traits, principal component analysis, factor analysis, genetic parameters

中图分类号: