Estimação bayesiana dos parâmetros genéticos da curva de crescimento de lhamas

Abstract

This research refers to Bayesian modeling applied to the estimation of genetic parameters associated with growth curves in llamas. For this purpose, the work is structured into two parts: the first comprises the theoretical framework, presenting the foundations of nonlinear growth models and Bayesian inference; and the second consists of two scientific articles. In the first article, parameters of nonlinear growth models were estimated using both frequentist and Bayesian procedures. The results highlight the potential of applying nonlinear functions to model the weight–age relationship in llamas under a Bayesian approach. The second article addresses the fitting of growth curves and the joint estimation of genetic parameters, as well as the comparison among different nonlinear functions through a hierarchical Bayesian model. The findings indicate that the use of hierarchical Bayesian models with the NUTS/brms algorithm is computationally efficient and statistically effective to analyze llamas growth curves, evidencing high genetic potential for adult weight selection, in contrast to the low heritability of curve shape parameters. Moreover, this study represents the first application of the NUTS algorithm in hierarchical growth models for South American camelids. For this purpose, monthly weight records from birth to one year of age were analyzed in both studies, including male llamas and individuals from the K’ara and Ch’accu varieties from the Quimsachata Experimental Station (Peru). Overall, the findings reinforce the potential of hierarchical Bayesian modeling as a useful methodological tool for understanding llamas growth dynamics and improving genetic selection strategies in llamas, contributing to both scientific advancement and the strengthening of production systems based on this species. Keywords: Bayesian hierarchical models; Genetic improvement; Growth curves; Heritability; South-american camels.