The genetic properties of mature weight and its relationship with early growth traits in Bonsmara breed

Abstract

This study was conducted to evaluate genetic parameters for mature weight and its relationship with early growth traits, to determine whether the recent breeding programs which are directed at improving early growth traits have resulted in an undesirable correlated response in mature weight for the Bonsmara cattle in South Africa. Traits studied were birth weight (BWT), weaning weight (WWT-205days), yearling weight (YWT-365days), 18 month's weight (EWT- 550days) and mature weight (MWT). MWT was estimated from cow weight using the Brody's growth curve and also from repeated weight measurements taken between the ages of 3 to 10 years. Contemporary groups for all traits were formed based on the herd, year and season of birth. Only contemporary groups comprising of a minimum of 8 sires were kept. Animals born as a single and reared on suckling of their own dams were considered. The final dataset consisted of a total of 105 520 pedigree identities and performance records with 69 514 progenies born between the year of 1982 and 2006 over 6 generations. The ( co )variance components for all traits were estimated by ASREML, using a single and multi-trait animal models; and in addition, the repeatedly measured MWT was analyzed using a repeatability model. The log likelihood ratio test indicated that accounting for maternal effects provided a good fit for all pre-weaning traits. Therefore, random effects fitted were; additive genetic for all traits, maternal genetic for all pre-weaning traits, direct-maternal genetic covariance for BWT, and permanent maternal environment for WWT. For analysis of MWT using the repeatability model, the direct genetic and permanent environmental effects were fitted as random effects. Direct heritability estimates from the univariate analysis for BWT, WWT, YWT, EWT and asymptotic MWT were 0.49±0.018, 0.43±0.013, 0.51±0.017, 0.55±0.02 and 0.5 1±0.064, respectively. Maternal heritability estimates were 0. 11±0.012 and 0.06±0.011 for BWT and WWT, respectively. The direct heritability estimates obtained from the multi-trait model were slightly higher than the estimates obtained from the univariate analyses. The heritability estimate for MWT from the repeatability model was 0.40±0.04. The repeatability coefficient estimated for MWT was 0.54±0.01. The direct genetic correlation estimates between MWT with BWT, WWT, YWT and EWT were 0.60±0.047, 0.54±0.051 , 0.60±0.055 and 0.66±0.053 , respectively. The realized genetic trends for BWT, WWT, YWT, EWT and MWT were -0.002±0.003 , 0.15±0.02, 0.051±0.025, 0.094±0.026 and -0.095±0.094 kg per year, respectively. All trends were positive and significant except for BWT and MWT. Using the estimated genetic correlations between early growth traits and MWT with the realized genetic trend in early growth traits, the expected correlated response in MWT was 0.22, 0.06, 0.1 kg/year for the selection emphasis given to WWT, YWT and EWT, respectively. In practical terms, these values are not far from the non-genetic trend observed for MWT. On the other hand, if the selection intensity on early growth traits would have been higher, say retaining top 5% or 10%, the predicted correlated response in MWT indicate that there would be a change ranging from 5.3 to 7.8 kg per generation. These results indicate that there is a potential to cause a significant change if intensive selection is applied on individual early growth trait without taking into consideration correlated response. Therefore, future genetic improvement programs of the Bonsmara cattle should consider a multi-trait index selection that prioritizes breeding objectives with constraints on traits that have a negative impact on production efficiencies, such as BWT and MWT.