Comparação entre metodologias para avaliar a idade à primeira concepção em éguas Puro-Sangue Inglês

There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

Abstract

No Brasil, devido à existência do ano hípico para a raça Puro-Sangue Inglês, os dados reprodutivos são geralmente truncados, já que os criadores tentam obter produtos nascidos no começo do ano hípico, a fim de aproveitarem a vantagem competitiva (animais mais desenvolvidos, maduros e treinados) destes em relação aos animais nascidos mais tardiamente no mesmo ano hípico. Para análise destes dados deve-se utilizar metodologias adequadas. Sendo assim, este trabalho objetivou comparar três metodologias: método da máxima verossimilhança restrita, com a utilização do aplicativo MTDFREML, análise bayesiana sem dados censurados por meio do programa MTGSAM e análise bayesiana com dados censurados pelo programa LMCD, para avaliação da idade à primeira concepção em éguas da raça Puro-Sangue Inglês, a fim de verificar seu impacto na escolha de reprodutores durante a seleção. O arquivo de dados continha 3509 registros de idade à primeira concepção (meses) de éguas da raça Puro-Sangue Inglês. As estimativas de herdabilidades obtidas foram 0,23, 0,30 e 0,0926 (escala log), para o MTDF, MTGSAM e LMCD, respectivamente. Considerando todos os animais do pedigree (6713), as correlações de ranking variaram de 0,91 a 0,99. Quando considerado apenas os garanhões (656), variaram de 0,48 a 0,99 (considerando diferentes porcentagens de machos selecionados) entre as metodologias de avaliação. As maiores mudanças na classificação geral foram observadas quando comparado o LMCD com as outras duas metodologias. Considerando que o modelo linear censurado é o mais adequado para análise de características com dados censurados, observouse que as informações de censura levariam a escolha de diferentes animais durante o processo de seleção, quando comparado com as outras duas metodologias de análise.

Translated abstract

In Brazil, due to the breeding season for Thoroughbred, the reproductive data are normally truncate, since the breeders try to get animals that were born at the beginning of the breeding season in order to take their competitive advantages (more developed, mature and trained animals) compared to animals born later in the same breeding season. To analyze these data suitable methods should be used. Then, this paper aims to compare three methodologies: the method of maximum restricted likelihood, using MTDFREML, bayesian analysis without censured data by software MTGSAM and bayesian analysis with censured data by software LMCD, to evaluate age at first conception in thoroughbred mares, in order to verify its impact on the choice of stallions during selection. The database contained 3509 records for age at first conception (months) for thoroughbred mares. The heritability estimates were 0.23, 0.30 and 0.0926 (log scale), for MTDF, MTGSAM and LMCD, respectively. Considering all animals in the pedigree (6713), ranking correlations varied from 0.91 to 0.99. When only stallions were considered (656), those varied from 0.48 to 0.99 (considering different percentages of selected males) between evaluation methods. The highest changes in the general classification were observed when LMCD was compared to the other two methods. As the linear censured model is the most suitable for trait analysis with censured data, it was observed that censure information would lead to the choice of different animals during the selection process, when compared to the two other methodologies.

Related collections

Most cited references 26

Record: found
Abstract: found
Article: not found

The Bayesian controversy in animal breeding.

A J Blasco (2001)

Frequentist and Bayesian approaches to scientific inference in animal breeding are discussed. Routine methods in animal breeding (selection index, BLUP, ML, REML) are presented under the hypotheses of both schools of inference, and their properties are examined in both cases. The Bayesian approach is discussed in cases in which prior information is available, prior information is available under certain hypotheses, prior information is vague, and there is no prior information. Bayesian prediction of genetic values and genetic parameters are presented. Finally, the frequentist and Bayesian approaches are compared from a theoretical and a practical point of view. Some problems for which Bayesian methods can be particularly useful are discussed. Both Bayesian and frequentist schools of inference are established, and now neither of them has operational difficulties, with the exception of some complex cases. There is software available to analyze a large variety of problems from either point of view. The choice of one school or the other should be related to whether there are solutions in one school that the other does not offer, to how easily the problems are solved, and to how comfortable scientists feel with the way they convey their results.