Related Experiment Videos
Using simulation models to predict feed intake: phenotypic and genetic relationships between observed and predicted
C B Williams1, G L Bennett, T G Jenkins
1USDA, ARS, US Meat Animal Research Center, Clay Center, NE, USA. williams@email.marc.usda.gov
Journal of Animal Science
|May 16, 2006
Summary
The Decision Evaluator for the Cattle Industry (DECI) and Cornell Value Discovery System (CVDS) accurately predict individual dry matter intake (DMI) in cattle. Predicted DMI shows strong genetic correlations, supporting its use in genetic evaluations for improved feed efficiency.
Area of Science:
- Animal Science
- Genetics
- Agricultural Engineering
Background:
- Accurate prediction of individual dry matter intake (DMI) is crucial for efficient cattle production and genetic improvement.
- Existing systems like DECI and CVDS offer potential for predicting DMI, but their accuracy and utility in genetic evaluations require validation.
Purpose of the Study:
- To evaluate the predictive accuracy of DECI and CVDS for individual DMI in cattle.
- To assess the feasibility of incorporating predicted DMI data into genetic evaluations for cattle breeding programs.
Main Methods:
- Collected postweaning data on 504 steers, including observed DMI (OFI), average daily gain (ADG), and carcass measurements.
- Utilized environmental data (temperature, wind speed) and animal data as inputs for CVDS and DECI to predict feed required (FR) under various conditions.
- Estimated genetic parameters using a REML animal model, analyzing phenotypic and genetic correlations between observed and predicted DMI.
Main Results:
- Both DECI and CVDS demonstrated good predictive accuracy for DMI, with DECI showing a slightly higher R-squared value (0.53) compared to CVDS (0.44).
- High phenotypic and genetic correlations were observed between observed DMI (OFI) and predicted DMI from both systems, particularly when considering maintenance and ADG (e.g., DFRmg, CFRmg).
- Heritability estimates for predicted DMI were comparable to or higher than that of observed DMI, suggesting potential for genetic selection.
Conclusions:
- Predicted DMI from DECI and CVDS, especially models incorporating maintenance and ADG, are accurate enough for use in genetic evaluations.
- The strong genetic relationships indicate that predicted DMI can be a valuable tool for improving feed efficiency in cattle breeding.
- Further research is needed to validate these findings in diverse populations, particularly those with existing genetic variation in feed efficiency.