BAYESIAN INFERENCE OF GENETIC PARAMETERS FOR 305 DAY MILK YIELD IN TURKISH HOLSTEINS VIA GIBBS SAMPLING


GEVREKÇİ Y. , MESTAV B., TAKMA Ç.

FRESENIUS ENVIRONMENTAL BULLETIN, cilt.27, ss.6388-6393, 2018 (SCI İndekslerine Giren Dergi) identifier

  • Cilt numarası: 27 Konu: 9
  • Basım Tarihi: 2018
  • Dergi Adı: FRESENIUS ENVIRONMENTAL BULLETIN
  • Sayfa Sayıları: ss.6388-6393

Özet

In this study, variance components of 305 day milk yield (dMY) are estimated. Milk yield records from the Cattle Breeders' Association of Turkey were collected between 2001 and 2011. The data include 4395 first-calving records of animals with 3227 dams and 550 sires of them. The fixed effects in the model are the calving year, calving season, herd and covariate effect which is the age at first calving. The analysis was implemented by Gibbs Sampling methodology with a single run of the Monte Carlo Markov Chains (MCMC). Estimates of marginal posterior densities of all unknown parameters were obtained by MCMC GLMM (Generalized Linear Mixed Model) package in R Project software. The convergence was verified through graphical inspection as trace plots. The posterior mean of direct heritability for 305 dMY was calculated as 0.04 +/- 0.0012. Unlike conventional methods such as Analysis of Variance (ANOVA), Gibbs Sampling gives point estimates within the parameter space. The estimated results of this study suggest that Gibbs Sampling method seems to be a flexible and reliable procedure for the genetic evaluation of 305 dMY and it can be useful in the breeding programs for Holsteins.