Estimation of genetic parameters for production and reproductive traits in Indian Karan-Fries cattle using multi-trait Bayesian approach.

Abstract

Estimates of variance components are needed for implementing genetic selection. This study was conducted to genetic parameters for production and reproductive traits on Indian Karan-Fries cattle using multi-trait repeatability animal model. Data collected from ICAR-National Dairy Research Institute, Karnal, India (from 1988 to 2019) were used. Single-trait and multi-trait repeatability animal models were used for parameter estimation. The posterior mean of Heritability estimates for 305-day milk yield (305-DMY), lactation milk yield (LMY), lactation length (LL) were 0.20 ± 0.03, 0.19 ± 0.03 and 0.06 ± 0.02, respectively. For age at first calving (AFC), calving interval (CI), and days open (DO), the posterior mean of heritability estimates were 0.24 ± 0.08, 0.06 ± 0.01, and 0.07 ± 0.02, respectively. The repeatability estimates for 305-DMY, LMY, LL, CI, and DO were 0.37 ± 0.02, 0.34 ± 0.02, 0.15 ± 0.02, 0.09 ± 0.02, and 0.12 ± 0.02, respectively. Genetic correlation between milk production traits (305-DMY, LMY, and LL) was positive and strong (> 0.80). However, the genetic correlation between milk production trait and AFC ranges from - 0.31 to 0.12. Unfavorable strong genetic correlations were observed between production and reproductive traits (CI and DO) with values ranged from 0.5 to 0.7. Phenotypic correlations among 305-DMY, LMY, and LL were generally positive and high. The moderate heritability estimates and potential genetic variation for 305-DMY, TMY, and AFC suggested that genetic gain can be obtained for these traits through genetic selection. Low heritability estimates found for LL, CI and DO, indicating that the possibility of changing these traits through genetic selection is small. High genetic correlation observed between productive and fertility traits were unfavorable. The existed strong genetic and phenotypic correlation estimates between CI and DO indicates that recording only one of them would be sufficient in the herd. As the multi-trait model showed slight improvements in the h2 as well as r estimates for both productive and reproductive traits over univariate analysis, future selection with a multi-trait animal model applying Bayesian approach would be recommended.