Source: Ontario Ministry of Agriculture, Food and Rural Affairs
The beef industry, just like any other, wants to sustain profitability while decreasing costs. Beef producers are provided with genetic selection tools such as expected progeny differences (EPDs) and estimated breeding values (EBVs), but it is ultimately up to them to determine the economic importance of each trait. Although economic breeding objectives and selection indexes have long been developed, the widespread use of these tools in the beef industry has recently gained momentum . A study done by Enns and Nicolli (2008) observed the long-term genetic change in a commercial beef breeding program due to a selection index designed for a specific economic breeding objective.
A nucleus Angus seedstock herd located in New Zealand was used for the study. This particular herd was ideal since data was available for the entire length of the breeding program and the breeding objectives did not change over a 17 year period. The goal of this breeding program was to produce bulls for distribution and to use in the company’s commercial breeding herds in order to genetically improve profitability.
Figure 1. Stabilizer seedstock herd from Southern Ontario.
In 1976, a breeding objective was developed for the herd and was used up until 1993. This economic breeding objective took into account harvest weight of surplus progeny at 30 months of age, dressing percentages of harvest progeny and culled cows, the net fertility of the cow measured by calves weaned/cow exposed, and the body weight of the cow at disposal. Feed intake was not included because the developers felt there was not enough information available on the correlations between feed intake of cattle on pasture and the traits likely to be included in the selection index. Therefore, they adjusted gross income to account for the extra feed cost associated with larger animals, using a standard relationship between body size and feed intake.
The traits used in the selection index were:
- fertility of the dam
- average lifetime body weight of the dam
- weaning weight
- yearling weight
These traits were weighted by the number of progeny the cow had produced at the time the index was being calculated for an individual. Unlike previous selection indexes, this index includes traits that do not directly affect profitability such as, weaning weight and yearling weight. The modern analysis which were used by Enns and Nicoll to reinterpret the data included additional techniques such as calculating direct, maternal, permanent environment and residual effects separately. The current computational technology also allowed the use of procedures such as best linear unbiased predictions (BLUP) to calculate more accurate estimations.
The study found that direct weaning weight, post-weaning gain, harvest weight, and yearling weight all have average EBVs that increased significantly over the 1976-1993 time period. No significant difference in the EBV for maternal weaning weight (MWW) was observed. This is most likely due to the fact that MWW was found to be negatively correlated to direct weaning weight (DWW), and since DWW was more heavily weighted within the selection index there would not be much of a change in the average EBV for MWW.. The average EBV for mature body weight also did not change significantly. This was most likely due to the fact that the developers were accounting for feed intake cost by adjusting the gross income associated with larger carcass weights. This adjustment meant that selection pressures were indirectly resulting in smaller mature body weights to avoid the cost penalty associated with a larger carcass weight.
The realized genetic response for number of calves weaned almost doubled what was predictedback in 1976. This is most likely due to the fact that the developers were using an assumed heritability of 0.05 while the actual estimated heritability obtained from the recent data analysis was 0.20. The estimates obtained from the recent analysis would cause an increase in the response to selection due to an increase in accuracy.
Figure 2. Cow-calf pair from a Stabilizer seedstock herd.
Previously, selection indexes were created using least squares adjustments and deviations of adjusted records from contemporary group means. These records included the individuals own records, the dams records, and the repeated records of the progeny of the dam. Presently, selection indexes are created using all the available data including the different components associated with the traits such as its direct, maternal, permanent environment and residual components. These, along with pedigree information, were fully accounted for in the 2008 analysis. It would therefore be reasonable to say that the response to selection calculated in 1976 would be less accurate compared to the response calculated at the time of this study.
The main limitation of the breeding objective developed in 1976 was the lack of computational technology. The technology present today allows us to work with larger data sets in order to get more statistically significant results. In addition, the lack of information on genetic parameters would also limit the breeding objective. As previously mentioned, the developers did not directly incorporate feed intake in the breeding objective. Now we have more estimations for genetic parameters, and with present technology these estimations are even more accurate, which will in turn result in faster genetic change. There are more estimates and fewer assumptions which, as seen in the results, can drastically change the outcome of a breeding program. The breeding objective can be more precisely defined by directly accounting for costs such as feed intake. This allows traits such as feed intake and body weight to be incorporated independently, which then allows selection for moderate to high body weight while maintaining or decreasing the cost of feed.
Currently, the beef industry has “generalized” breeding objectives and selection indexes developed on a national production system. Future research should be done on the rate of genetic change of systems using these generalized breeding objectives and selection indexes in comparison to the rate of genetic change from specific breeding objectives and selection indexes designed for specific production of marketing sectors.
A more recent study was conducted in 2012 by Åbyii et al. looking at the inclusion of functional traits in a bio-economic model in addition to regular production traits. Examples of such functional traits would be:
- herd life of cow
- age at first calving
- calving interval
- twinning frequency
- calving difficulty
- limb and/or claw disorders
The results were that functional traits were just as economically important as production traits within the model. For this reason, it is equally important to include functional traits in a bio-economic model in addition to production traits.
In conclusion, an economic breeding objective developed in 1976 along with a multi-trait selection index resulted in long-term genetic gains in the beef cattle industry. The simple economic breeding objectives and selection index such as the ones developed in 1976 built a solid foundation on which to improve. This was apparent when looking at the 2008 reanalysis of the 1976-1993 data using modern techniques, as more accurate estimations were calculated due to increased knowledge, power of technology, and BLUP estimations. Improvements in genetic selection for economic breeding objectives continues, as in 2012 it was shown that functional traits would be as economically important as production traits. As research continues moving forward, new ideas will come to light providing a stronger foundation in order to make more effective selection decisions.
i Enns, R. M., & Nicoll, G. B. (2008). Genetic change results from selection on an economic breeding objective in beef cattle. Journal of animal science, 86(12), 3348-3357.
ii Åby, B. A., Aass, L., Sehested, E., & Vangen, O. (2012). A bio-economic model for calculating economic values of traits for intensive and extensive beef cattle breeds. Livestock Science, 143(2), 259-269.
Author: Jennifer Proulx, Msc Candidate, Department of Animal and Poultry Science/University of Guelph