Evaluation of mature cow weight: Genetic correlations with traits used in selection indices, correlated responses, and genetic trends in Nelore cattle

Transcription

1 Published December 3, 2014 Evaluation of mature cow weight: Genetic correlations with traits used in selection indices, correlated responses, and genetic trends in Nelore cattle A. A. Boligon,* 1 R. Carvalheiro, and L. G. Albuquerque* *Department of Animal Sciences, São Paulo State University, Jaboticabal, SP , Brazil; and GenSys Consultores Associados S/S Ltda. ABSTRACT: Genetic correlations of selection indices and the traits considered in these indices with mature weight (MW) of Nelore females and correlated responses were estimated to determine whether current selection practices will result in an undesired correlated response in MW. Genetic trends for weaning and yearling indices and MW were also estimated. Data from 612,244 Nelore animals born between 1984 and 2010, belonging to different beef cattle evaluation programs from Brazil and Paraguay, were used. The following traits were studied: weaning conformation (WC), weaning precocity (WP), weaning muscling (WM), yearling conformation (YC), yearling precocity (YP), yearling muscling (YM), weaning and yearling indices, BW gain from birth to weaning (BWG), postweaning BW gain (PWG), scrotal circumference (SC), and MW. The variance and covariance components were estimated by Bayesian inference in a multitrait analysis, including all traits in the same analysis, using a nonlinear (threshold) animal model for visual scores and a linear animal model for the other traits. The mean direct heritabilities were 0.21 ± (WC), 0.22 ± (WP), 0.20 ± (WM), 0.43 ± (YC), 0.40 ± (YP), 0.40 ± (YM), 0.17 ± (BWG), 0.21 ± (PWG), 0.32 ± (SC), and 0.44 ± (MW). The genetic correlations between MW and weaning and yearling indices were positive and of medium magnitude (0.30 ± 0.01 and 0.31 ± 0.01, respectively). The genetic changes in weaning index, yearling index, and MW, expressed as units of genetic SD per year, were 0.26, 0.27, and 0.01, respectively. The genetic trend for MW was nonsignificant, suggesting no negative correlated response. The selection practice based on the use of sires with high final index giving preference for those better ranked for yearling precocity and muscling than for conformation generates only a minimal correlated response in MW. Key words: Bayesian inference, beef cattle, body weight gain, scrotal circumference, visual scores 2013 American Society of Animal Science. All rights reserved. J. Anim. Sci :20 28 doi: /jas INTRODUCTION Most beef cattle breeding programs are designed to increase the production efficiency of the herd, giving priority to selection based on data obtained at the beginning of the growth phase, for example, BW or BW gains at certain ages. However, selection of young animals for greater BW may result in heavier animals at birth and at adult age due to a correlated response (Boligon et al., 2010; Pedrosa et al., 2010). Archer et al. (1998) and Forni et al. (2007) reported an increase in 1 Corresponding author: arioneboligon@yahoo.com.br Received April 1, Accepted August 31, mature weight (MW) as a correlated response to selection for greater growth rates in Angus and Nelore cattle, respectively. Mature weight is an important trait in beef cattle breeding programs because of its strong association with the maintenance costs of dams (Jenkins and Ferrell, 1994). In addition, large females may be less efficient in terms of reproductive and physiological performance (Montano-Bermudez et al., 1990; Owens et al., 1993; Silveira et al., 2004). These animals are especially undesirable in productions systems based exclusively on pasture. Genetic evaluations of MW of Nelore cows are already available (Aliança Nelore, 2011; PMGRN/Programa Nelore, 2011).

2 Evaluation of mature cow weight 21 Studies correlating traits commonly included in selection indices and the indices themselves with MW of beef cattle are scarce (Boligon et al., 2010, 2011a; Pedrosa et al., 2010). Considering the importance of cow size for the beef cattle production system, MW should be evaluated and monitored to prevent an increase in cow size due to an indirect response to selection of animals with a greater growth potential (Boligon et al., 2010). Furthermore, the study of genetic trends over time in traits that are under direct and correlated selection permits evaluation of the results of the selection program adopted and contributes to redirect traits included in selection indices, if necessary. The objective of the present study was to estimate genetic correlations of selection indices and the traits considered in these indices with MW and correlated responses to determine whether the selection practiced among pasture-fed Nelore herds under tropical conditions would result in an undesired response in MW of beef cows. In addition, genetic trends were estimated to evaluate the adequacy of the selection criteria adopted. MATERIAL AND METHODS Animal Care and Use Committee approval was not obtained for this study because no animals were used. Data Description Data from 612,244 Nelore animals born between 1984 and 2010, belonging to different beef cattle breeding programs in Brazil and Paraguay, were used. The data are part of a single database called Aliança Nelore. The animals are kept on tropical pastures on 263 different farms. These traits were studied: visual scores of weaning conformation (WC), weaning precocity (WP), weaning muscling (WM), yearling conformation (YC), yearling precocity (YP), yearling muscling (YM), weaning and yearling indices as described below, BW gain from birth to weaning (BWG), postweaning BW gain (PWG), scrotal circumference (SC), and MW. The animals are routinely weighed at birth, weaning (about 7 mo of age), and yearling (about 18 mo of age). Females that remain on the farms are weighed at adult age (3 to 17 yr of age). In addition to BW recordings, the animals are visually evaluated at weaning and yearling using scores of conformation, precocity, and muscling. Scrotal circumference is also measured on yearling period. Repeated recordings of MW were not considered in this study and only weights obtained close to 4 yr of age were included as suggested by Boligon et al. (2010). Visual scores ranging from 1 to 5 were attributed by trained examiners, always relative to average animals within each contemporary group (CG). The greatest scores within the same CG indicate the most marked presence of the trait. The visual scores were defined as follows: conformation evaluated by the body volume of the animal and precocity that evaluates the capacity of the animal to reach a minimum level of carcass finishing in the absence of an increased BW (i.e., animals presenting earlier finishing have fat deposits mainly at the base of the tail and in the groin). Tall and lean animals with a shallow rib cage are late and should therefore receive lower finishing scores. Muscling describes the development of muscle mass as a whole and is evaluated at these sites: shoulder, loin, rump, and mainly in the hindquarter. In the beginning of its adoption, initiatives were done to implement visual scores in an absolute scale. This approach was aborted due to poor quality of data generated, where most CG presented almost no variability (i.e., examiners were capturing mainly environmental effects). In the breeding programs studied, the animals are selected using different indices. The 2 main indices used by the different programs for the selection of animals at weaning and yearling are described below: WI = [(60 sepd_bwg) + (8 sepd_wc) + (16 sepd_wp) + (16 sepd_wm)]/10 YI = [(23 sepd_bwg) + (4 sepd_wc) + (8 sepd_wp) + (8 sepd_wm) + (23 sepd_pwg) + (4 sepd_yc) + (8 sepd_yp) + (8 sepd_ym) + (14 sepd_sc)]/10 in which WI and YI = selection indices at weaning and yearling, respectively, expressed as base 10, sepd_bwg = standardized expected progeny difference (sepd) of BW gain from birth to weaning, sepd_wc = sepd of weaning conformation, sepd_wp = sepd of weaning precocity, sepd_wm = sepd of weaning muscling, sepd_ PWG = sepd of postweaning BW gain, sepd_yc = sepd of yearling conformation, sepd_yp = sepd of yearling precocity, sepd_ym = sepd of yearling muscling, and sepd_sc = sepd of scrotal circumference (at yearling). In the selection indices, the traits were weighted in such a way to obtain a balance between carcass traits (evaluated by visual scores of conformation, precocity, and muscling), growth intensity (expressed as BW gain), and sexual precocity (expressed as SC in the YI). During the selection process at weaning, on average 10% of the worst females and 50% of the worst male calves are sold based on the weaning index to adapt the number of animals to the pasture capacity during the dry season. Animals that remain in the herd are evaluated at

3 22 Boligon et al Contemporary Groups Variance and Covariance Components components were es in which y i ith trait (i n i a i m i c i is e i X i Za i Zm i W i are incidence matrices that - Table 1. Data description of weaning conformation - - n dams sires y1 X1 1 Za1 = + yn X n n Za n a1 Zm1 m1 w1 c1 e an Zmn mn wn cn en

4 Evaluation of mature cow weight 23 structure between random effects can be described as follows: a m Var c e A a A am 0 0 A am A m I Nm c I N in which a m is the maternal genetic am is the direct-maternal - c is the maternal perma- e is the residual A denotes the direct product INm is the num- N is the number Vectors a m c are location parameters of a conditional distribution y a m c tribution of distribution of y for traits that show a continuous distri- described as follows: (y amcr~ N + Z 1 a + Z 2 m + WcI N R Threshold Model which can be described as follows: UN(WI e 2 in which U e r of order s and amc- W of order r s I r r e 2 e 2 - t < t 2 K < t j t 0 t j j is the number - the categories or scores of y i animal iu i 1 if t0 Ui t 1 2 if t1ui t2 yi 3 if t2 Ui t3 in 4 if t3ui t4 5 if t4 Ui t5 in which n- t 0 to t (t to t t t 2 t t 3 t 4 Convergence Criteria

5 24 - Boligon et al. Correlated Response Δ G = r ( σ / σ ) ΔG 2,1 g g2 g1 1 r g is the genetic correlation between traits g g G is RESULTS AND DISCUSSION Posterior Distributions - 95% HPD - n of heritabilities and correlations in increasing order and discarding the (aa Genetic Trends - y i b o + b x i + e i in which y i uated in the ib o b is the x i is the ie i Figure 1. Frequency distribution of weaning conformation (WC), weaning precocity (WP), weaning muscling (WM), yearling conformation (YC), yearling precocity (YP), and yearling muscling (YM) stratified for each unit in each trait.

6 Evaluation of mature cow weight 25 Table 2. Posterior means of heritability for weaning conformation (WC), weaning precocity (WP), weaning muscling (WM), yearling conformation (YC), yearling precocity (YP), yearling muscling (YM), BW gain from birth to weaning (BWG), postweaning BW gain (PWG), scrotal circumference (SC), and mature weight (MW) Figure 2. Number of observations (bars) and means ( ) and respective SD of mature cow weight according to age. rior distributions tended to follow a normal and symmetric distribution (Table 2). The mean heritabilities for WC, WP, and WM were lower than those obtained for the same traits measured at yearling age. These results can be explained by the fact that the genetic variation in visual scores at weaning is also explained by genetic maternal effect. Cardoso et al. (2001, 2004), Koury Filho et al. (2010), and Boligon et al. (2011a) also reported a greater influence of direct additive genetic factors on visual scores obtained at yearling compared with those measured at weaning, for Angus and Nelore cattle. In the case of weaning visual scores, the contribution of the maternal genetic effect to phenotypic variance ranged from 10 to 15%, a finding justifying the inclusion of this effect in the model (Table 2). Studying Nelore animals, Koury Filho et al. (2010) and Boligon et al. (2011a) found lower maternal heritabilities for weaning visual scores, with estimates ranging from 0.04 to In contrast, Araújo et al. (2010), using data from Angus-Nelore crossbreds, reported maternal heritabilities of 0.09 ± 0.03, 0.16 ± 0.02, and 0.12 ± 0.02 for conformation, precocity, and muscling scores, respectively. The direct and maternal heritabilities for BWG (Table 2) were similar to those obtained for the other traits evaluated at weaning. The direct heritability for PWG was lower than that observed for the other traits measured at yearling age. This finding agrees with Koots et al. (1994) who demonstrated that heritability estimates are generally lower in challenging environments (i.e., the period to express PWG) compared with estimates obtained in favorable environments. Boligon et al. (2011b), using data from only 1 farm included in the present study, found direct heritabilities of 0.25 ± 0.02 and 0.30 ± 0.04 for BWG and PWG, respectively. The maternal heritability for BWG was 0.10 ± 0.01 in that study. The mean estimated heritability for SC was expressive and agrees with values reported in the literature. In Nelore animals, Boligon et al. (2011b) obtained heritabilities of 0.39 ± 0.01, 0.41 ± 0.01, and 0.44 ± 0.02 for SC at 9, 12, and 15 mo of age, respectively. Studying the Trait Mean ± SD Mode Median HPD (95%) 1 Direct heritabilities WC 0.21 ± to 0.23 WP 0.22 ± to 0.24 WM 0.20 ± to 0.21 YC 0.43 ± to 0.44 YP 0.40 ± to 0.42 YM 0.40 ± to 0.41 BWG 0.17 ± to 0.20 PWG 0.21 ± to 0.23 SC 0.32 ± to 0.34 MW 0.44 ± to 0.48 Maternal heritabilities WC 0.15 ± to 0.16 WP 0.10 ± to 0.12 WM 0.13 ± to 0.14 BWG 0.14 ± to HPD (95%) = 95% highest posterior density. same breed, Silveira et al. (2004) reported a heritability coefficient of 0.39 for SC at 18 mo of age. Heritabilities of both indices were equal to 1 as expected. With respect to MW, the mean estimated heritability (0.44 ± 0.018) suggests the possibility of rapid responses to selection, indicating that this trait can be used for the evaluation and identification of biotypes that are compatible with the production system. Although MW is not commonly evaluated in beef cattle programs, it is Table 3. Posterior means of genetic correlations between mature weight and the other traits studied Trait 1 Mean ± SD Mode Median HPD (95%) 2 WC 0.43 ± to 0.48 WP 0.17 ± to 0.24 WM 0.19 ± to 0.24 YC 0.55 ± to 0.58 YP 0.21 ± to 0.25 YM 0.22 ± to 0.26 WI 0.30 ± to 0.33 YI 0.31 ± to 0.32 BWG 0.16 ± to 0.20 PWG 0.34 ± to 0.39 SC 0.21 ± to WC = weaning conformation; WP = weaning precocity; WM = weaning muscling; YC = yearling conformation; YP = yearling precocity; YM = yearling muscling; WI = weaning index; YI = yearling index; BWG = BW gain from birth to weaning; PWG = postweaning BW gain; SC = scrotal circumference. 2 HPD (95%) = 95% highest posterior density.

7 26 Boligon et al. Figure 3. Genetic trends for weaning index, yearling index, and mature weight in genetic SD units per birth year and for mature cow weight in genetic SD units per birth year of calf in Nelore cattle. easily measured, a fact that could encourage breeders to weigh the animals that remain in the herd at this age. It should be emphasized that the choice of a selection criterion does not only depend on how much the trait is subject to transmission to offspring but also on its correlation with other traits and its economic importance. Table 3 shows the genetic correlations of MW with the traits considered in the selection indices and with the indices themselves. Positive genetic correlations of low to medium magnitude were observed between MW and the visual scores (Table 3). At the 2 ages (weaning and yearling), the greatest genetic associations were obtained between MW and body conformation score when compared with the other scores, a finding also reported by Pedrosa et al. (2010). In Nelore animals, Boligon et al. (2011a) estimated genetic correlations of 0.30 ± 0.06, 0.27 ± 0.06, and 0.32 ± 0.05 between MW and conformation, precocity, and muscling scores measured at weaning and of 0.34 ± 0.05, 0.30 ± 0.04, and 0.36 ± 0.04 for the same scores measured at yearling age. The posterior means of genetic correlations between MW and weaning and yearling indices were 0.30 ± 0.01 and 0.31 ± 0.01, respectively. Considering only these results is not possible to determine whether the selection for higher indices or traits alone may result in undesirable response in MW of cows. Therefore, is important to check the correlated responses and genetic trends. Favorable genetic trends were obtained for the weaning and yearling indices (Fig. 3), with the observation of a decrease of genetic trends between 2000 and 2004, a period characterized by an expressive increase in the number of controlled animals. The correlated responses expected for MW as a function of the genetic correlations and direct responses of the indices were 0.27 and 0.35 kg/yr based on parameter estimates for weaning and yearling indices, respectively. As illustrated in Fig. 3, the estimated genetic trend for MW confirmed the expected correlated response, which was 0.35 kg/yr. However, the genetic trend for MW was nonsignificant at the end of studied period (the last 10 yr), suggesting no negative correlated response. One possible explana- Figure 4. Dispersal of breeding values (BV) predicted for mature weight (MW) using 896 sires (top 20%) for yearling index (YI), considering sires with better BV for yearling scores of precocity and muscling than conformation (precocious) and sires with better BV for yearling scores of conformation than precocity or muscling (late).

8 Evaluation of mature cow weight 27 tion for this finding is related to the fact that breeders are recommended to use sires with high final index giving preference for those better ranked for yearling precocity and muscling than for conformation. Moreover, the selection practice in which females are discarded when they have lower reproduction rates can also be associated to nonsignificant changes in mature weight, because large females may be less efficient in terms of reproductive performance (Kaps et al., 1999; Meyer et al., 2004). As can be seen in Fig. 4, when considering the top 20% of sires in terms of breeding values for the final index, precocious sires (i.e., greater breeding values for YP and YM compared with YC) presented reduced average breeding values for MW (7.70 kg) than late sires (19.01 kg; i.e., greater average breeding values for YC compared with YP or YM). In addition, precocious sires left 68% more sons in the population than late sires. Among the traits considered in the selection indices, the conformation score showed the greatest genetic correlation with MW (Table 3), with this trait presenting the least weight in both the weaning index (8% for conformation at weaning) and yearling index (4% for conformation at weaning and 4% for conformation at yearling). The genetic trend for the yearling index was equivalent to an average genetic progress of 0.16 to 0.68%/ year of the phenotypic mean of the traits included in the index. For MW, the genetic gain corresponded to 0.08% of the phenotypic mean. Taken together, the present results suggest that selection based on the indices used in this study, which comprise not only the BW but also the weight composition of the animals, has a favorable effect on the growth curve, restricting the response to selection for increased MW of cows, when compared with the separate use of the traits considered in these indices. The genetic trend for MW of cows per birth year of their calves was not biologically important (Fig. 3), a finding demonstrating that the average genetic merit of MW of productive cows was stable over time. Rumph et al. (2004) studied genetic trends in MW of Hereford cows selected based on an index that comprised an increase in yearling weight and a reduction in birth weight and observed an increase in MW of 0.38 kg/yr. In cases in which indirect selection for MW is above the expected, selection based on traits of economic interest or indices that comprise such traits, combined with EPD for MW, may yield desired results in terms of improving the production performance of beef cattle. However, further studies are needed because investigations in this area are scarce. LITERATURE CITED Aliança Nelore: Sumário de touros Gensys Consultores Associados S/C Ltda, Sumário Aliança Nelore (In Portuguese.) (Accessed September 8, 2011.) Araújo, R. O., P. R. Rorato, T. Weber, D. M. Everling, J. S. Lopes, and M. A. Dornelles Genetic parameters and phenotypic and genetic trends for weight at weaning and visual scores during this phase estimated for Angus-Nellore crossbred young bulls. Rev. Bras. Zootec. 39: Archer, J. A., R. M. Herd, P. F. Arthur, and P. F. Parnell Correlated responses in rate of maturation and mature size of cows and steers to divergent selection for yearling growth rate in Angus cattle. Livest. Prod. Sci. 54: Boligon, A. A., L. G. Albuquerque, M. E. Z. Mercadante, and R. B. Lobo Models for genetic evaluation of Nelore cattle mature body weight. J. Anim. Sci. 86: Boligon, A. A., F. Baldi, and L. G. Albuquerque. 2011b. Genetic parameters and relationships between growth traits and scrotal circumference measured at different ages in Nellore cattle. Genet. Mol. Biol. 34: Boligon, A. A., M. E. Z. Mercadante, L. G. Albuquerque. 2011a. Genetic associations of conformation, finishing precocity and muscling visual scores with mature weight in Nelore cattle. Livest. Sci. 135: Boligon, A. A., M. E. Z. Mercadante, S. Forni, R. B. Lobo, and L. G. Albuquerque Covariance functions for weights from birth to maturity in Nelore cows. J. Anim. Sci. 88: Cardoso, F. F., R. A. Cardellino, and L. T. Campos Componentes de (co)variância e parâmetros genéticos para caracteres produtivos à desmama de bezerros Angus criados no estado do Rio Grande do Sul. (In Portuguese.) Rev. Bras. Zootec. 30: Cardoso, F. F., R. A. Cardellino, and L. T. Campos Componentes de (co)variância e parâmetros genéticos de caracteres pós-desmama em bovinos da raça Angus. (In Portuguese.) Rev. Bras. Zootec. 33: Faria, C. U., C. U. Magnabosco, L. G. Albuquerque, A. De Los Reyes, L. A. F. Bezerra, and R. B. Lôbo Análise genética de escores de avaliação visual de bovinos com modelos bayesianos de limiar e linear. (In Portuguese.) Pesqui. Agropecu. Bras. 43: Forni, S., M. Piles, A. Blasco, L. Varona, H. N. Oliveira, R. B. Lobo, and L. G. Albuquerque Analysis of beef cattle longitudinal data applying a non-linear model. J. Anim. Sci. 85: Geweke, J Evaluating the accuracy of sampling-based approaches to the calculation of posterior moments. In: J. M. Bernardo, J. O. Berger, A. P. Dawid, and A. F. M. Smith, editors, Bayesian statistics 4. Oxford Univ. Press, New York, NY. p Gianola, D., and J. L. Foulley Sire evaluation for ordered categorical data with a threshold model. Genet., Sel., Evol. 15: Gianola, D., and D. Sorensen Likelihood, Bayesian, and MCMC methods in quantitative genetics. Springer-Verlag, New York, NY. Heidelberger, P., and P. D. Welch Simulation run length control in the presence of an initial transient. Oper. Res. 31: Jenkins, T. G., and C. L. Ferrell Productivity through weaning of nine breeds of cattle under varying feed availabilities. I. Initial evaluation. J. Anim. Sci. 72: Kaps, M., W. O. Herring, and W. R. Lamberson Genetic and environmental parameters for mature weight in Angus cattle. J. Anim. Sci. 77: Koots, K. R., J. P. Gibson, C. Smith, and J. W. Wilton Analyses of published genetic parameter estimates for beef production traits. 1. Heritability. J. Anim. Breed. Genet. 62: Koury Filho, W., L. G. Albuquerque, S. Forni, J. A. II V. Silva, M. J. I. Yokoo, and M. M. Alencar Estimativas de parâmetros genéticos para os escores visuais e suas associações com peso corporal em bovinos de corte. (In Portuguese.) Rev. Bras. Zootec. 39:

9 28 Boligon et al. Meyer, K., D. Johnston, and H. Graser Estimates of the complete genetic covariance matrix for traits in multi-trait genetic evaluation of Australian Hereford cattle. Aust. J. Agric. Res. 55: Misztal, I., S. Tsuruta, T. Strabel, B. Auvray, T. Druet, and D. H. Lee BLUPF90 and related programs (BGF90). In: Proc. 7th World Congress on Genetics Applied to Livestock Production, Montpellier, France. Communication no Montano-Bermudez, M., M. K. Nielsen, and G. H. Deutscher Energy requirements for maintenance of crossbred beef cattle with different genetic potential for milk. J. Anim. Sci. 68: Owens, F. N., P. Dubeski, and C. F. Hanson Factors that alter the growth and development of ruminants. J. Anim. Sci. 71: Pedrosa, V. B., J. P. Eler, J. B. S. Ferraz, J. A. II V. Silva, S. Ribeiro, M. R. Silva, and L. F. B. Pinto Parâmetros genéticos do peso adulto e características de desenvolvimento ponderal na raça Nelore. (In Portuguese.) Rev. Bras. Saude Prod. Anim. 11: PMGRN: Programa de Melhoramento Genético da Raça Nelore. Portuguese. Rumph, J. M., D. D. Kress, K. C. Davis, D. C. Anderson, and D. L. Boss Genetic analysis of mature cow weights in a population of inbred Hereford cattle. Proc. - Am. Soc. Anim. Sci., West. Sect. 55: R Development Core Team R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Silveira, J. C., C. McManus, A. S. Mascioli, L. O. C. Silva, A. C. Silveira, J. A. S. Garcia, and H. Louvandini Fatores ambientais e parâmetros genéticos para características produtivas e reprodutivas em um rebanho Nelore no estado do Mato Grosso do Sul. (In Portuguese.) Rev. Bras. Zootec. 33: Smith, B. J Bayesian output analysis program (BOA) version 1.1 user s manual. (Accessed June 6, 2011.)