Autoregressive model for genetic evaluation of reproductive traits in dairy cattle

• Ano de defesa: 2020
• Autor(a) principal: Silva, Hugo Teixeira
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de Viçosa; Zootecnia
• Palavras-chave em Português: Genômica; Holstein (Bovino); Melhoramento genético; Reprodução; Genética e Melhoramento dos Animais Domésticos
• Link de acesso: https://locus.ufv.br//handle/123456789/29394

In the last decades, there has been a decrease in the reproductive performance of dairy herds; an unfavorable correlated responses in consequence of the intense selection to increase milk production. To overcome this problem, breeding programs worldwide have included reproductive performance in their selection objectives. Dairy cattle breeding programs in Brazil and Portugal have also shown interest in including reproductive traits in their genetic evaluations; however, further studies are required to better understand about traits and statistical methodologies to be used in the genetic evaluation processes. Genetic analyses of longitudinal reproductive traits have been performed by using the traditional repeatability models (REP). Under these models, the environmental components between longitudinal measurements are assumed equally correlated, which is questionable since events closer in time should be more associated than distant ones. A suitable alternative would be fitting an autoregressive covariance structure to describe the different levels of associations that could exist between consecutive events over time. Therefore, the general objective in this thesis was to evaluate the autoregressive model (AR) for genetic analyses of longitudinal reproductive traits in Brazilian and Portuguese Holstein cattle. Firstly it was evaluated the performance of AR model in Portuguese Holstein herd. The reproductive traits considered were interval between calving to first service (ICF), days open (DO), calving interval (CI), and daughter pregnancy rate (DPR) measured during the first three calving orders. In general, the AR model overcome the REP model in all reproductive traits studied, corresponding with better fitting to reproductive data and higher EBV reliabilities. In the second chapter of this thesis, a similar analysis was carried out with the data of Brazilian Holstein herd. The reproductive traits evaluated were DO, CI and DPR. Once more, the AR model was superior to REP model, showing higher reliabilities and a direct response on the ranking of the best bulls of population. In addition, greater genetic variances and smaller residual variances were estimated with the AR model. The results shown in first two studies suggest that AR model is a suitable alternative for genetic evaluations of longitudinal reproductive traits in Diary cattle. In the last chapter of this thesis, we investigated the applicability of ssGBLUP methodology under the autoregressive model (H-AR) for genomic evaluation of longitudinal reproductive traits in Portuguese Holstein cattle. The use of H-AR model increased the GEBV reliabilities and reduced the prediction bias when compared to traditional analysis considering pedigree-based relationship matrix. Furthermore, when evaluating the performance of the H-AR model considering different sources of genomic information, it was observed that the inclusion of genotypic cows in the analysis of reproductive traits provided favorable responses for genomic prediction of bulls. However, improved results were obtained when bulls with no daughter and relationship information (non-contributing bulls) were disregarded in the analyses. Overall, the results suggested that the proposed AR model is a suitable alternative for genetic analyses of longitudinal reproductive traits in the Brazilian and Portuguese dairy cattle herds. In addition, they also demonstrated that ssGBLUP methodology under autoregressive model is a feasible and applicable approach to be used in the genomic analysis of reproductive traits. Keywords: Autocorrelation. Genomic evaluation. Holstein. Repeatability. Longitudinal traits. ssGBLUP