0 avis
Genetic analyses of mid-infrared predicted milk fat globule size and milk lactoferrin
Archive ouverte
Edité par CCSD ; Wageningen Academic Publishers -
International audience. Genetic parameters for mid-infrared (MIR) predicted milk fat globule (MFG) size and lactoferrin were estimated for first parity Canadian Holsteins. A total of 109,029 records from 22,432 cows (105,737 pedigree animals) for lactoferrin and 109,212 records from 22,424 cows (105,070 pedigree animals) for MFG size were available. The linear animal model included fixed effects of herd-test day and days in milk, fixed regressions for age-season of calving effect, random regressions for herd-year of calving, permanent environment and additive genetic effects. Regression curves (on DIM) were modelled using orthogonal Legendre polynomials of order 4 for fixed and order 5 for random effects. Co-variance components were estimated by Bayesian methods via Gibbs sampling using custom-written software and 270,000 samples after a burn-in period of 30,000 samples. Multi-trait analyses using the same model with the MIR-predicted milk components along with test-day milk, fat and protein yields, fat and protein percentage, and somatic cell score were used to estimate genetic correlations among the traits. Average daily heritability estimates over the entire lactation were 0.34 for lactoferrin and 0.50 for MFG size. Milk lactoferrin content was found to have a moderate lactation average genetic correlation with protein % (0.41) and a positive genetic correlation with fat % (0.21). Weak genetic correlations were estimated between lactoferrin and fat and protein yields, and somatic cell score at 0.07, 0.08, and 0.07, respectively. Milk fat globule size had average genetic correlations of -0.23, 0.36, and 0.54 with milk yield, fat yield, and fat %, respectively. Posterior standard deviations for all estimates were low (below 0.001). The moderate genetic correlation with fat percentage suggested that MIR prediction for MFG size may not be perfectly related to fat percentage and thus the prediction equations are not simply predicting fat percentage.
Consulter en ligne
Suggestions
Du même auteur
