遗传 ›› 2021, Vol. 43 ›› Issue (4): 340-349.doi: 10.16288/j.yczz.20-351

• 研究报告 • 上一篇    下一篇

基于高密度SNP标记估计群体间遗传关联

周子文, 王雪, 丁向东()   

  1. 中国农业大学动物科技学院,畜禽育种国家工程实验室,农业农村部动物遗传育种与繁殖重点实验室,北京 100193
  • 收稿日期:2020-10-19 修回日期:2021-02-17 出版日期:2021-04-16 发布日期:2021-03-26
  • 通讯作者: 丁向东 E-mail:xding@cau.edu.cn
  • 作者简介:周子文,在读硕士研究生,专业方向:动物遗传育种。E-mail: zhouzw834@163.com
  • 基金资助:
    国家现代农业产业技术体系项目编号(CARS-35);国家重点研发计划项目编号(2019YFE0106800);河北省重点研发计划项目资助编号(19226376D)

Measuring genetic connectedness between herds based on high density SNP markers

Zhou Ziwen, Wang Xue, Ding Xiangdong()   

  1. National Engineering Laboratory for Animal Breeding, Key Laboratory of Animal Genetics, Breeding and Reproduction of Ministry of Agriculture and Rural Affairs, College of Animal Science and Technology, China Agricultural University,Beijing 100193, China
  • Received:2020-10-19 Revised:2021-02-17 Online:2021-04-16 Published:2021-03-26
  • Contact: Ding Xiangdong E-mail:xding@cau.edu.cn
  • Supported by:
    Supported by China Agriculture Research System No(CARS-35);the National Key Research and Development Project No(2019YFE0106800);Modern Agriculture Science and Technology Key Project of Hebei Province No(19226376D)

摘要:

联合育种的准确性受到群体间遗传关联程度的影响。本研究通过比较基于系谱数据和基因组数据计算的群体遗传关联,探究高密度SNP标记在遗传关联估计中的应用前景。本研究同时使用了模拟数据和真实数据,采用6种不同的遗传关联计算方法,包括PEVD(prediction error variance of differences)、PEVD(x)、VED(variance of estimated difference)、CD(generalized coefficient of determination)、r(prediction error correlation)和CR(connectedness rating),比较基于构建不同的关系矩阵(A、G、Gs、G0.5和H矩阵)的群体间遗传关联。模拟数据和实际数据结果表明,除PEVD(x)和VED方法外,PEVD、CD、r和CR基于基因组信息的G、Gs和G0.5阵计算的遗传关联程度均高于基于系谱信息的A阵,基于同时利用系谱和基因组信息的H阵遗传关联结果一般介于A阵与G阵之间。当CR和r为0时,CD都较高,高估了群体遗传关联。用r度量3个遗传分化程度不同的猪场间遗传关联时,基于G阵的r值均为0.01,不能准确反映群体真实遗传关联。随着遗传力的提高,所有群体遗传关联评估方法都有所改善,但遗传力为0.1时,PEVD基于A阵结果优于G阵,中高遗传力性状用于估计遗传关联优于低遗传力性状。本研究证明高密度SNP标记比系谱信息估计群体间遗传关联更有优势,CR是衡量遗传关联稳健而可靠的评价指标,计算简单,受性状遗传力影响较小。PEVD可以作为补充,量化具体群体遗传关联下的育种值预测误差情况。G矩阵比Gs、G0.5阵能更好反映群体遗传关联。

关键词: 猪, 遗传关联, 系谱, 基因组, 关系矩阵

Abstract:

The accuracy of genetic evaluations in different herds is affected by the degree of genetic connectedness among herds. In this study, we explored the application of high density SNP markers in the assessment of genetic connectedness by comparing the genetic connectedness based on pedigree data and genomic data. Six methods, including PEVD (prediction error variance of differences between estimated breeding values), PEVD (x), VED (variance of estimated difference between the herd effects), CD (generalized coefficient of determination), r (prediction error correlation) and CR (connectedness rating), were implemented to measure the genetic connectedness based on different relationship matrices (A, G, Gs, G0. 5 and H). Our results from both simulated data and SNP chip data indicated that, except for the PEVD (x) and VED methods, the genetic connectedness obtained by PEVD, CD, r and CR based on G. Gs and G0.5 matrices (using genome information only) were superior to those based on A matrix (using pedigree information only). Generally, for most approaches, the genetic connectedness based on H matrix (using both pedigree and genome information) was somewhere between A matrix and G matrices. CD could overestimate the degree of genetic connectedness as it was still very high when CR and r were close to 0. The method r could not accurately reflect the true genetic connectedness of the populations. It generated 0.01 of genetic connectedness for all three pig breeding farms, which were actually genetically different with each other. With increasing of heritability, the degree of genetic connectedness obtained by all methods were increased as well. However, in the case of heritability 0.1, PEVD based on A matrix performed better than based on G matrix, suggesting that traits with medium and high heritability are more suitable for the assessment of genetic connectedness compared to traits with low heritability. Our findings indicated that high-density SNP markers have advantages over pedigree analysis for the measurement of genetic connectedness, and CR is a robust and reliable method to assess genetic connectedness. Further, CR is easily calculated and less affected by heritability of trait. PEVD is good supplement to quantify the prediction errors of estimated breeding values under the specific genetic connectedness. In comparison, G matrix can reflect genetic connectedness better than its extensions Gs and G0.5 matrix.

Key words: swine, genetic connectedness, pedigree, genome, relationship matrix