|Yao, Chen - University Of Wisconsin|
|De Los Campos, Gustavo - Michigan State University|
|Vandehaar, Michael - Michigan State University|
|Spurlock, Diane - Iowa State University|
|Armentano, Lou - University Of Wisconsin|
|Coffey, Mike - Collaborator|
|De Haas, Yvette - Wageningen University|
|Veerkamp, Roel - Wageningen University|
|Staples, Charlie - University Of Florida|
|Wang, Z - University Of Alberta|
|Hanigan, Mark - Virginia Tech|
|Tempelman, Rob - Michigan State University|
|Weigel, Kent - University Of Wisconsin|
Submitted to: Journal of Dairy Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/22/2016
Publication Date: 3/1/2017
Citation: Yao, C., De Los Campos, G., Vandehaar, M.J., Spurlock, D.M., Armentano, L.E., Coffey, M., De Haas, Y., Veerkamp, R.F., Staples, C.R., Connor, E.E., Wang, Z., Hanigan, M.D., Tempelman, R.J., Weigel, K. 2017. Use of genotype x environment interaction model to accommodate genetic heterogeneity for residual feed intake, dry matter intake, net energy in milk, and metabolic body weight in dairy cattle. Journal of Dairy Science. 100(3):2007-2016.
Interpretive Summary: Improving feed efficiency in dairy cattle through genetic selection is an important goal of researchers, dairy farmers, and breeding companies world wide. Most studies on the subject to date used only a few hundred animals from a single research station because of the high cost and labor associated with measurement of daily feed intake of individual cows required for this research. However, data are now available from multiple research stations, which can be combined to estimate genetic parameters and to predict genomic estimated breeding values for feed efficiency traits. The challenge lies in considering potential interactions of genetics and environment when combining data from multiple locations. Many statistical models have been explored to address the genetic heterogeneity of data derived from several sub-populations (e.g., multi-environment or multi-breed data), with varying success. Interaction models allow effects of genetic markers to be constant and / or group-specific across environments and provide estimates of genomic correlations between subpopulations for each trait, which can be used to assess genetic similarity between subpopulations. In this study, an interaction model was assessed and compared to within-environment and across-environment models using data from multiple dairy research stations across the globe to estimate genomic variances and assess the accuracy of genomic predictions for a measure of feed efficiency called residual feed intake, or RFI, and its component traits. Overall, a novel way was developed to evaluate traits measured in multiple environments in which genetic heterogeneity may exist. Our model allows estimation of environment-specific parameters and provides genomic predictions that approach or exceed the accuracy of competing within- or across-environment models. This work benefits the dairy cattle research community and dairy industry by providing better ways to advance genetic improvement of dairy cattle for complex traits such as feed efficiency.
Technical Abstract: Feed efficiency in dairy cattle has gained much attention recently. Due to the cost-prohibitive measurement of individual feed intakes, combining data from multiple countries is often necessary to ensure an adequate reference population. It may then be essential to model genetic heterogeneity when making inferences about feed efficiency or selecting efficient cattle using genomic information. In this study, we constructed a marker × environment interaction model that decomposed marker effects into main effects and interaction components that were specific to each environment. We compared environment-specific variance component estimates and prediction accuracies from the interaction model analyses, an across-environment analyses ignoring population stratification, and a within-environment analyses using an international feed efficiency data set. Phenotypes included residual feed intake (RFI), dry matter intake (DMI), net energy in milk (MilkE), and metabolic body weight (MBW) from 3,656 cows measured in 3 broadly defined environments: North America (NAM), the Netherlands (NLD), and Scotland (SAC). Genotypic data included 57,574 single nucleotide polymorphisms per animal. The interaction model gave the highest prediction accuracy for MBW, which had the largest estimated heritabilities ranging from 0.37 to 0.55. The within-environment model performed the best when predicting RFI, which had the lowest estimated heritabilities ranging from 0.13 to 0.41. For traits (DMI and MilkE) with intermediate estimated heritabilities (0.21 to 0.50 and 0.17 to 0.53), performance of the 3 models was comparable. Genomic correlations between environments were also computed using variance component estimates from the interaction model. Averaged across all traits, genomic correlations were highest between NAM and NLD, and lowest between NAM and SAC. In conclusion, the interaction model provided a novel way to evaluate traits measured in multiple environments in which genetic heterogeneity may exist. This model allowed estimation of environment-specific parameters and provided genomic predictions that approached or exceeded the accuracy of competing within or across-environment models.