1 Comparison of Linear Mixed Model and Random Regression Model for Weaning Weight in Hereford Cattle over a United States Longitudinal Gradient. (1st March 2018)

• Record Type: Journal Article
• Authors: Langdon, J M; Delgadillo, J S; Herring, A D; Blackburn, H D; Speidel, S E; Sanders, S; Krehbiel, B C; Riley, D G
• Is Part Of: Journal of animal science. Volume 96(2018)Supplement 1
• Page Start: 1
• Page End: 1
• Publication Date: 2018-03-01
• Subjects: Weaning weight -- Genetic parameters -- Random regression; Livestock -- Periodicals; Livestock; Electronic journals; Periodicals; 636.005
• DOI: 10.1093/jas/sky027.001
• Languages: English
• ISSNs: 0021-8812

Abstract

Abstract: It is of interest to the beef cattle industry to determine accurate estimates of heritability for economically important traits such as weaning weight (WW) for improved genetic merit prediction, accuracy of selection, and overall beef cattle production and profitability across the United States. Therefore, the objective of this study was to compare the use of either a linear mixed model or a linear random regression model to identify the best approach to accurately fit the data. Both models were evaluated to determine heritability estimates for WW using American Hereford Association records (n = 226, 845) in the US across different longitudinal coordinates. Both models utilized contemporary group and eco-region as fixed effects. The random regression model used an additive genetic component, additive maternal component, and a linear random regression, which modeled the interaction between the additive genetic and additive maternal components with longitudinal coordinates. A log-likelihood ratio test was conducted to determine the most informative model, where the random regression model was the best fit for the data ( P < 0.001). Heritability estimates from the random regression model ranged from 0.14 to 0.32 from 124.3º (west) to 69.18º (east), respectively. These estimates increase steadily from west to east; however, proportion of phenotypic variance explained by the additive maternal component does not change substantially with longitude. This result implies genetic-environment interactions may need to be considered for improved genetic predictions and interpretations. However, further research should be conducted to determine for what traits and other cattle populations this phenomenon exists.