J. Anim Sci. 2007. 85:1120-1125. doi:10.2527/jas.2006-694
© 2007 American Society of Animal Science
Genetic evaluation of beef carcass data using different endpoint adjustments
J. M. Rumph*,1,
W. R. Shafer
,
D. H. Crews, Jr.
,
R. M. Enns
,
R. J. Lipsey,
R. L. Quaas# and
E. J. Pollak#
* Department of Animal and Range Sciences, Montana State University, Bozeman 59717;
and
American Simmental Association, Bozeman, MT 59715;
and
Agriculture and Agri-Food Canada Research Centre, Lethbridge, Alberta T1J 4B1, Canada;
and
Department of Animal Sciences, Colorado State University, Fort Collins 80523; and
and
# Department of Animal Sciences, Cornell University, Ithaca, NY 14853
 |
Abstract
|
|---|
Carcass data from 6,795 Simmental-sired animals born from 1992 to 2001 were used to determine whether adjustment to a constant age, back-fat, HCW, or marbling score would result in differences in heritability of the carcass traits and, correspondingly, if EPD calculated using those variance components and adjustments would result in sire reranking. The endpoints were age (EPA), backfat (EPF), HCW (EPC), or marbling (EPM). The traits analyzed were 12th-rib backfat (FAT), HCW, marbling (MRB), LM area (LMA), and percentage retail cuts (PRC). The data were analyzed using an animal model, where contemporary group was included as a fixed effect and was composed of slaughter date, sex, and herd. Random effects included in the model were direct genetic and residual. Estimates of heritability ranged from 0.12 to 0.14, 0.32 to 0.34, and 0.26 to 0.27 for FAT, HCW, and LMA, respectively, for the corresponding endpoints. Heritability for MRB was estimated to be 0.27 at all endpoints. For PRC, estimates of heritability were more variable, with estimates of 0.23 ± 0.05, 0.32 ± 0.05, 0.21 ± 0.05, and 0.20 ± 0.04 for EPA, EPF, EPC, and EPM, respectively. However, because the EPF and EPC adjustments adjust for a component trait of PRC (FAT and HCW, respectively), they may be altering the trait to one different from PRC. Spearman rank correlations between EPD within a trait using EPA compared with the other endpoints were >0.90 (P < 0.01) for FAT, HCW, MRB, and LMA. For PRC, Spearman rank correlations with EPA EPD were 0.73 (P < 0.01), 0.93 (P < 0.01), and 0.95 (P < 0.01) for EPF, EPC, and EPM, respectively. For most traits and endpoints, there was little reranking among sires when alternative endpoints were used. However, adjusting PRC to EPF appears to result in a greater heritability and substantial re-ranking of sires, potentially due to the adjustment changing the trait to one other than PRC.
Key Words: beef cattle • carcass • endpoint • genetic evaluation • heritability • variance component
|
INTRODUCTION
|
|---|
Estimation of EPD for carcass traits provides cattle breeders with a selection tool for improving carcass traits in breeding animals and their commercial offspring. However, accurate estimation of these values is necessary to make them a useful tool. Currently in the US national cattle evaluation system, carcass trait EPD are calculated at a constant age endpoint, but the timing of slaughter is usually based on endpoints other than age, such as backfat thickness, marbling score, or carcass weight, or combinations thereof, in an attempt to maximize profitability by minimizing discounts and increasing premiums. Therefore, age may not be the most appropriate adjustment when calculating carcass EPD.
Studies evaluating the heritability of carcass traits at various endpoints have reported conflicting results. Most studies report that heritability does not change across endpoints (i.e., Koots et al., 1994
; Utrera and Van Vleck, 2004; Ríos-Utrera et al., 2005), but some have found the contrary (i.e., Devitt and Wilton, 2001; Shanks et al., 2001), particularly when comparing a backfat thickness endpoint with other endpoints. This may be related to the fact that adjusting for backfat is adjusting for body composition and may be changing the trait(s) being analyzed to new trait(s).
Reranking of sires is of particular concern when alternative endpoints are chosen for EPD calculations and may be indicative of the trait evaluated being changed when the endpoint is changed. Previous studies have suggested that the endpoint may affect ranking of sires (e.g., Koch et al., 1995; Shanks et al., 2001; Ríos-Utrera et al., 2005), which suggests that if the industry does not slaughter animals at a constant age, sires are potentially being ranked incorrectly using current industry standard, age-adjusted endpoints.
Therefore, the objectives of the current study were to determine 1) if genetic parameters estimated at several endpoints were significantly different, and 2) if sire reranking occurs when alternate endpoints are used.
|
MATERIALS AND METHODS
|
|---|
Animal Care and Use Committee approval was not obtained for this study because the data were obtained from an existing database. Carcass records for 6,795 Simmental-sired animals born from 1992 to 2001 were made available by the American Simmental Association (ASA), with an average age at slaughter of 456 d. The data were edited as described by Shanks (1999). However, in contrast to previous analyses using ASA data (Shanks, 1999; Shanks et al., 2001), these analyses used the most recent ASA data to better represent the current US beef industry. Summary statistics for the data are provided in Table 1.
Traits measured included 12th-rib backfat thickness (FAT, n = 6,546), HCW (n = 6,795), marbling score (MRB, n = 6,368), LM area (LMA, n = 6,728), and percentage retail cuts (PRC, n = 5,983). All measurements were collected by USDA graders, with PRC being calculated from the component traits of HCW, LMA, and FAT. Percent KPH measurements were not recorded in this data set and were assumed to be 2.5% for all carcasses. These values were used to calculate PRC, as follows: 51.34 – (2.276 x FAT, cm) – (0.0205 x HCW, kg) – (0.462 x KPH, %) + (0.1147 x LMA, cm2) (Boggs et al., 1998).
Data for each trait were adjusted to each of 4 endpoints: age (EPA), backfat (EPF), HCW (EPC), or marbling (EPM). Adjustments were made by fitting a linear and quadratic covariate for the chosen endpoint. For HCW, MRB, and FAT, heritability estimates were not undertaken when the adjustment was the trait itself. For instance, the heritability of HCW was not estimated in a model adjusting for HCW endpoint.
The data were analyzed using an animal model. There were 6,044 dams represented in the data, with an average of 1.12 and a range of 1 to 14 offspring per dam. Females with 2 or more progeny in the data made up 9.7% of the dams.
For calculation of genetic parameters and EPD, the model used was
where y is the vector of the observed phenotypes; ß is the vector of fixed effects, which included contemporary group (formed using slaughter date, sex, and herd) and the linear and quadratic covariate of adjustment; a is the vector of additive genetic effects; e is the vector of random error effects; X is the known incidence matrix associating fixed effects in ß with phenotypes in y; and Z is the known incidence matrix associating random effects in a with phenotypes in y, with zero columns associated with animals in the pedigree that do not have records.
Furthermore,
where A is the numerator relationship matrix of the 18,133 animals included in the pedigree, including those with no records, I is the identity matrix of proper order, 
is the variance due to additive genetic effects, and 
2e is the variance due to random error.
Genetic parameters were estimated using the multiple-trait, derivative-free REML program of Boldman et al. (1995), as modified by Dodenhoff et al. (1998) for calculation of SE of estimates of genetic parameters with certain models. For the first rounds of iteration, the convergence criterion (variance of the simplex) was set at 1 x 10–6. Once the convergence criterion was reached, cold restarts were continued until the –2log likelihood differed by less than 1 x 10–2 between successive restarts. The parameters obtained from the last restart were used as the final results.
Spearman rank correlations were estimated using the CORR procedure (SAS Inst. Inc., Cary, NC) to determine if the sires changed their ranking based on the EPD from alternative endpoints.
|
RESULTS AND DISCUSSION
|
|---|
Estimates of heritability for each trait adjusted to each endpoint are shown in Table 2. With the exception of PRC, estimates of heritability within a trait were similar regardless of endpoint chosen, which is in agreement with results from Bergen et al. (2006a, b) and with the review of carcass analyses by Utrera and Van Vleck (2004). However, estimates were smaller than those typically found for carcass traits using field data (i.e., Wilson et al., 1993; Hirooka et al., 1996; Pariacote et al., 1998). Estimates were 0.12 to 0.14, 0.32 to 0.34, 0.26 to 0.27, and 0.27 for FAT, HCW, LMA, and MRB, respectively. For PRC, the heritability estimate of 0.32 using EPF was significantly larger than heritabilities estimated using the other 3 endpoints (0.20 to 0.23).
Backfat
The range of EPD for sires and genetic variance for backfat thickness adjusted to an EPA, EPC, and EPM are shown in Table 3. For sires of animals in this data set, the Spearman rank correlation compared with EPA was 0.94 (P < 0.01) and 0.96 (P < 0.01) for EPC and EPM, respectively, demonstrating that there was little difference in rank among sires when endpoint was changed from the conventional EPA. Even though EPC resulted in a greater proportion of genetic variance to phenotypic variance, rankings were relatively unchanged.
HCW
Table 4 presents ranges in EPD for sires and genetic variance for HCW. Spearman rank correlations with EPA were 0.95 (P < 0.01) and 0.97 (P < 0.01) for EPF and EPM, respectively, indicating that endpoint was not reranking sires greatly.
Marbling
Results for marbling are shown in Table 5. Spearman rank correlations with EPA were 0.96 (P < 0.01) and 0.98 (P < 0.01) for EPF and EPC, respectively, which means that regardless of endpoint, sires were ranked similarly for MRB.
Longissimus Muscle Area
Results for LMA are shown in Table 6. Spearman rank correlations with EPA were 0.99 (P < 0.01), 0.90 (P < 0.01), and 0.99 (P < 0.01) for EPF, EPC, and EPM, respectively. The lower correlation between EPA and EPC is likely due to the fact that these are positively correlated traits (Crews and Kemp, 2001) and adjusting LMA for HCW decreases the genetic variability in LMA, which is in agreement with results shown by Lee et al. (2000). It appears that this adjustment results in an altered definition of LMA EPD that likely does not reflect industry practices.
In Figure 1a, the ranking of the 100 highest accuracy sires for LMA using EPA and EPF is shown. Ranking is very similar between these 2 endpoints. However, as shown in Figure 1b, there is a greater amount of reranking when comparing EPA and EPC. As discussed previously, EPC does result in more reranking than other endpoints and likely should not be used for calculation of national cattle evaluations. In Figure 1c, the comparison of EPA and EPM is shown to result in little reranking among these high accuracy sires.
View larger version (9K):
[in this window]
[in a new window]
|
Figure 1. Rank of high accuracy bulls for LM area at an age endpoint (EPA) compared with alternative endpoints: A) backfat thickness endpoint (EPF); B) HCW endpoint (EPC); and C) marbling score endpoint (EPM).
|
|
Percentage Retail Cuts
Results for PRC are shown in Table 7. Because FAT, HCW, and LMA are all component traits of PRC, adjusting for FAT essentially eliminated the effect of that trait in the PRC calculation. This adjustment, in turn, changes the PRC from a trait partially influenced by backfat to a trait solely dependent on HCW and LMA and whose meaning is different than the original concept of PRC. This is supported by the fact that the estimate of heritability was similar to those obtained for HCW at all endpoints.
View this table:
[in this window]
[in a new window]
|
Table 7. Summary statistics for sire EPD and genetic variance () for percentage retail cuts adjusted to 4 end-points1
|
|
Unlike the other traits, PRC was sensitive to endpoint adjustment. Compared with EPA, estimates of genetic variance were 65, 84, and 82% for EPF, EPC, and EPM, respectively. Although all adjustments relative to EPA resulted in reduced genetic variance, the adjustment for EPF was of most concern with the adjustment removing more than one-third of the genetic variation. This is similar, but more extreme than the reduction by 16% found by Devitt and Wilton (2001). The reduction is partially due to FAT being a component trait of PRC. The reduction in genetic variation using EPC is also due to the fact that HCW is a component trait of PRC, although the reduction seen using this adjustment is not as extreme as for EPF.
Spearman rank correlations were 0.73 (P < 0.01), 0.93 (P < 0.01), and 0.95 (P < 0.01) for EPF, EPC, and EPM, respectively. Although all 3 endpoints result in decreased genetic variance relative to EPA, rankings were similar for EPC and EPM. The much lower correlation of 0.73 indicates that adjustment for FAT produces a change in the defined trait likely due to the fact that FAT is a component trait for PRC and the EPF is altering the trait so that it can no longer be considered PRC. Although HCW and FAT are component traits of PRC, the difference when adjusting to EPC is not as extreme as when adjusting to EPF, compared with the traditional EPA. This difference may be explained by the increased coefficient of variation seen in the FAT vs. HCW phenotypes used to calculate PRC. The coefficient of variation for FAT is 41.3% and for HCW is 11.6%. Therefore there is greater chance of change in FAT than in HCW within the PRC equation.
Figure 2 depicts the ranking of high accuracy sires for PRC using EPA compared with EPF, EPC, and EPM. Reranking among these high accuracy sires is the greatest in PRC, particularly using EPF in support of the Spearman rank correlations calculated using all animals in the pedigree.
View larger version (9K):
[in this window]
[in a new window]
|
Figure 2. Rank of high accuracy bulls for percentage retail cuts at an age endpoint (EPA) compared with alternative endpoints: A) backfat thickness endpoint (EPF); B) HCW endpoint (EPC); and C) marbling score endpoint (EPM).
|
|
For FAT, HCW, LMA, and MRB, endpoint does not appear to influence sire rank, so EPD calculated to EPA, EPF, EPC, and EPM should essentially result in similar outcomes regardless of the endpoint used to decide the slaughter date. Choice of endpoint would be a concern for PRC because the EPF significantly reranks sires relative to the current EPA adjustment. Further investigation is necessary to determine which adjustment is most predictive of PRC based on the way cattle are currently slaughtered in the United States.
In conclusion, for most traits, there is little reranking of sires when evaluated at alternate endpoints. However, endpoint has a large effect on the ranking of sires for percentage retail cuts and LM area. Adjusting percentage retail cuts for backfat and LM for carcass weight appears to change the definition of these traits. It has been shown that these traits rerank sires across varying endpoints, but it is unclear as to which endpoint is the most predictive of future progeny performance. Further investigation is needed to determine whether these alternative endpoints result in a more predictive estimate of EPD than the traditional age endpoint.
1 Corresponding author: janice{at}montana.edu
Received for publication October 18, 2006.
Accepted for publication January 9, 2007.
|
LITERATURE CITED
|
|---|
Bergen, R., S. P. Miller, J. W. Wilton, D. H. Crews Jr., and I. B. Mandell. 2006a. Genetic correlations between live yearling bull and steer carcass traits adjusted to different slaughter endpoints. 1. Carcass lean percentage. J. Anim. Sci. 84:546–557.[Abstract/Free Full Text]
Bergen, R., S. P. Miller, J. W. Wilton, and I. B. Mandell. 2006b. Genetic correlations between live yearling bull and steer carcass traits adjusted to different slaughter end-points. 2. Carcass fat partitioning. J. Anim. Sci. 84:558–566.[Abstract/Free Full Text]
Boggs, D. L., R. A. Merkel, and M. E. Doumit. 1998. Livestock and carcasses: An integrated approach to evaluation, grading, and selection. Kendall/Hunt Publishing Company, Dubuque, IA.
Boldman, K. G., L. A. Kriese, L. D. Van Vleck, C. P. Van Tassell, and S. D. Kachman. 1995. A manual for use of MTDFREML. A set of programs to obtain estimates of variance and covariances [Draft]. USDA, ARS, Washington, DC.
Crews, D. H., Jr., and R. A. Kemp. 2001. Genetic parameters for ultrasound and carcass measures of yield and quality among replacement and slaughter beef cattle. J. Anim. Sci. 79:3008–3020.[Abstract/Free Full Text]
Devitt, C. J. B., and J. W. Wilton. 2001. Genetic correlation estimates between ultrasound measurements on yearling bulls and carcass measurements on finished steers. J. Anim. Sci. 79:2790–2797.[Abstract/Free Full Text]
Dodenhoff, J., L. D. Van Vleck, S. D. Kachman, and R. M. Koch. 1998. Parameter estimates for direct, maternal, and grandmaternal genetic effects for birth weight and weaning weight in Hereford cattle. J. Anim. Sci. 76:2521–2527.[Abstract/Free Full Text]
Hirooka, H., A. F. Groen, and M. Matsumoto. 1996. Genetic parameters for growth and carcass traits in Japanese Brown Cattle estimated from field records. J. Anim. Sci. 74:2112–2116.[Abstract]
Koch, R. M., K. E. Gregory, and L. V. Cundiff. 1995. Genetic aspects of beef carcass growth and development. Page 9 in Proc. 5th Genetic Prediction Workshop, Kansas City, MO.
Koots, K. R., J. P. Gibson, C. Smith, and J. W. Wilton. 1994. Analyses of published genetic parameter estimates for beef production traits. 1. Heritability. Anim. Breed. Abstr. 62:309–338.
Lee, J. W., S. B. Choi, J. S. Kim, J. F. Keown, and L. D. Van Vleck. 2000. Parameter estimates for genetic effects on carcass traits of Korean Native Cattle. J. Anim. Sci. 78:1181–1190.[Abstract/Free Full Text]
Pariacote, F., L. D. Van Vleck, and R. E. Hunsley. 1998. Genetic and phenotypic parameters for carcass traits of American Shorthorn beef cattle. J. Anim. Sci. 76:2584–2588.[Abstract/Free Full Text]
Ríos-Utrera, A., L. V. Cundiff, K. E. Gregory, R. M. Koch, M. E. Dikeman, M. Koohmaraie, and L. D. Van Vleck. 2005. Genetic analysis of carcass traits of steers adjusted to age, weight, or backfat thickness slaughter end-points. J. Anim. Sci. 83:764–776.[Abstract/Free Full Text]
Shanks, B. C. 1999. Genetic evaluation of carcass traits in Simmental cattle. MS Thesis Montana State University, Bozeman.
Shanks, B. C., M. W. Tess, D. D. Kress, and B. E. Cunningham. 2001. Genetic evaluation of carcass traits in Simmental-sired cattle at different slaughter end-points. J. Anim. Sci. 79:595–604.[Abstract/Free Full Text]
Utrera, A. R., and L. D. Van Vleck. 2004. Heritability estimates for carcass traits of cattle: A review. Genet. Mol. Res. 3:380–394. http://www.funpecrp.com.br/gmr Accessed Oct. 10, 2006.[Medline]
Wilson, D. E., R. L. Willham, S. L. Northcutt, and G. H. Rouse. 1993. Genetic parameters for carcass traits estimated from Angus field records. J. Anim. Sci. 71:2365–2370.[Abstract]
Top
Abstract
INTRODUCTION
