Genotype by environment interaction on growth and carcass traits in beef cattle in the tropics

Abstract

Miller, S.P., Thallman, R.M. et al., unpublished results). These results support published results from laboratory species (Fry et al., 1998) and broiler chickens (Long et al., 2008) across different environments. Hence GxE interactions will be important considerations in SNP association analyses of complex beef production traits in future. Introduction A wide variety of beef cattle production systems may be found in the tropics, which could be classified as pasture based systems, mixed farming systems, and industrial systems, following the classification by Seré and Steinfeld (1995). In tropical regions, pasture based systems and mixed faming systems are mainly practiced, with different levels of intensification in management. In that context, one major goal for cattle breeders is to match the type of animal they use to the management conditions they apply. In this study, our objective was to investigate the genotype 3 environment interaction (GEI) that may affect growth traits (live weight, average daily gain, yearling weight) and carcass traits (slaughter weight, hot carcass weight, muscular tissue weight, adipose tissue weight and digestive tract weight) expressed across two contrasted systems. Material and methods The analysis was made in a database of 718 young Creole calves, born between 1999 and 2008, and distributed between two management conditions during their post weaning growth. The calves were born in different suckling herds, and weaned at about 211 days, at a weaning weight of 157 kg, on average. After weaning and an adaptation period of 40 days on average, 346 steers and 372 heifers were separated into two groups, each on a different management system. The penned animals (176 steers and 200 heifers) were fattened indoor, and fed with cut grass and concentrate (maize: 68%; wheat middling: 22%; soya meal: 8%; UF 5 1.05; PDIN 5 113 g), distributed up to 60% of their voluntary intake. The grazed animals (196 steers and 146 heifers), were fed outdoor at pastures, only with grazed tropical forages. During the entire experiment, the animals were regularly weighted every 2 weeks. At the end of the growing period, 298 steers were sacrificed. Slaughter ages were defined in order to obtain equivalent live weight range at the end of the fattening period: 14–17 months for penned steers, and 17–21 months for grazing steers. As soon as they arrived at the slaughter house, the steers were kept fasting for 24 hours, and then weighted just before slaughter. The calves were issued from 32 sires and 163 dams; the pedigree was 6 generations deep and a total of 961 animals are included in the pedigree file. Descriptive statistics were obtained with the SAS & software, and the genetic analysis was performed with the ASREML and MATLAB software. The longitudinal dataset collecting the observations on live weight of each animal at different ages was analyzed by a random regression with Legendre polynomials of order two fitted for age, animal effect and permanent individual effect. The fixed part of the model included a fixed effect coding for contemporary groups (by sex, management system, group of weighting), and the covariables age and weight at weaning; the random part included the genetic animal effect and the permanent environment individual effect. The datasets collecting observations on the other traits were analyzed by several bivariate models. The genetic correlations between traits expressed within the same environment were computed by implementing bivariate models where the response variables were two different traits, expressed within the same environment. The GEI affecting each trait was measured by implementing