Comparison of different models for the estimation of genetic parameters in tropical goats

• Published in: Tropical animal health and production Vol. 54; no. 6; p. 381
• Main Authors: de Souza, José Ernandes Rufino, Ferreira, Josiel, Del Pilar Solar Diaz, Iara, Silveira, Robson Mateus Freitas, de Sousa, Wandrick Hauss
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Netherlands 01-12-2022; Springer Nature B.V
• Subjects: Age; Animal sciences; Animals; Bayes Theorem; Bayesian analysis; Biomedical and Life Sciences; Birth weight; Body Weight; Breeding; Covariance; Data collection; Estimates; Female; Genetic effects; Goats; Goats - genetics; Knowledge representation; Life Sciences; Males; Maternal effects; Maternal Inheritance; Mathematical models; Models, Genetic; Parameters; Parturition; Phenotype; Pregnancy; Short Communications; Software; Sorghum; Veterinary Medicine/Veterinary Science; Zoology; Anglo-Nubian goats; Improvement; Models; Performance; Genetic parameters
• ISSN: 0049-4747; 1573-7438
• DOI: 10.1007/s11250-022-03374-6

This study aimed to estimate the variance components and genetic parameters for body weight in tropical goats testing different models using Bayesian approach and investigate the effectiveness of fitting the effects of maternal genetic, permanent environmental, and covariance between direct and maternal effects. Records from 1980 to 2010 of 1453 Anglo-Nubian goats’ herd were used. Six performance growth traits: birth weight (BW, kg), at 28 (W28, kg), 56 (W56, kg), 112 (W112, kg), 140 (W140, kg), and 196 (W196; kg) days of age, were evaluated. There was a negative covariance between direct genetic effects and maternal additive for all weights. The effect of maternal permanent environment is an important source of variation for performance characteristics in goats until the 196 days, and must be considered in genetic evaluation models in order to obtain accurate predictions of breeding values of individuals. The importance of inclusion of the additive maternal effect appears to be more dependent on the structure of the data set under evaluation. Given the structure of the data, the described management and criteria for choosing the best model (deviance information criterion and the Bayes factor) should make the estimation of parameters for weights at birth and at 28 and 56 days using model IV, since that will provide more consistent results than the type I (less complex), without the need of accurate representations of knowledge prior to data collection. Over time, the breeding program will have more data and thereby increase the possibility of building a prior distribution confident that would enable the inference of parameters for more complex models. However, these are preferable components for the estimation of the characteristics and weights to 112 at 140 and at 196 days, using model I (less complex).