Lu Y, Vandehaar M J, Spurlock D M, Weigel K A, Armentano L E, Staples C R, Connor E E, Wang Z, Bello N M, Tempelman R J
Department of Animal Science, Michigan State University, East Lansing 48824.
Department of Animal Science, Iowa State University, Ames 50011.
J Dairy Sci. 2015 Sep;98(9):6535-51. doi: 10.3168/jds.2015-9414. Epub 2015 Jul 22.
Genetic improvement of feed efficiency (FE) in dairy cattle requires greater attention given increasingly important resource constraint issues. A widely accepted yet occasionally contested measure of FE in dairy cattle is residual feed intake (RFI). The use of RFI is limiting for several reasons, including interpretation, differences in recording frequencies between the various component traits that define RFI, and potential differences in genetic versus nongenetic relationships between dry matter intake (DMI) and FE component traits. Hence, analyses focusing on DMI as the response are often preferred. We propose an alternative multiple-trait (MT) modeling strategy that exploits the Cholesky decomposition to provide a potentially more robust measure of FE. We demonstrate that our proposed FE measure is identical to RFI provided that genetic and nongenetic relationships between DMI and component traits of FE are identical. We assessed both approaches (MT and RFI) by simulation as well as by application to 26,383 weekly records from 50 to 200 d in milk on 2,470 cows from a dairy FE consortium study involving 7 institutions. Although the proposed MT model fared better than the RFI model when simulated genetic and nongenetic associations between DMI and FE component traits were substantially different from each other, no meaningful differences were found in predictive performance between the 2 models when applied to the consortium data.
鉴于资源限制问题日益重要，奶牛饲料效率（FE）的遗传改良需要更多关注。奶牛中一种广泛接受但偶尔也有争议的FE衡量指标是剩余采食量（RFI）。RFI的使用存在局限性，原因包括解释、定义RFI的各个组成性状之间记录频率的差异，以及干物质采食量（DMI）与FE组成性状之间遗传关系与非遗传关系的潜在差异。因此，通常更倾向于以DMI作为响应的分析。我们提出了一种替代的多性状（MT）建模策略，利用乔列斯基分解来提供一种可能更稳健的FE衡量指标。我们证明，只要DMI与FE组成性状之间的遗传和非遗传关系相同，我们提出的FE衡量指标就与RFI相同。我们通过模拟以及应用于来自一个涉及7个机构的奶牛FE联盟研究中2470头奶牛在产奶50至200天期间的26383条每周记录，对这两种方法（MT和RFI）进行了评估。尽管当模拟的DMI与FE组成性状之间的遗传和非遗传关联彼此有很大差异时，所提出的MT模型比RFI模型表现更好，但在应用于联盟数据时，这两种模型在预测性能上没有发现有意义的差异。
