Bayesian evaluation of the effect of non-genetic factors on the phenomics for quality-related milk nutrients and yield in Murciano-Granadina goats.

Abstract

The aim of this study was to evaluate the effect of non-genetic factors on the variability of milk production and composition using Bayesian linear regression. We analyzed 2594 milk records from 159 dairy goats from the breeding nucleus of the Murciano-Granadina breed. Bayesian linear regression was used to determine the effects of non-genetic factors on the phenomics for quality-related milk nutrients and yield. Multivariate regression model significantly explained 21.5%, 40.0%, 41.5%, 44.3%, 44.6%, and 47.5% of the variability in somatic cell count (SCC, sc/mL), lactose (%), protein (%), milk yield (kg), fat (%), and dry matter (%), respectively. Although the aforementioned factor combination significantly conditions milk production and composition, SCC may be particularly affected by collateral factors. Milking routine and drying period factors are reference predictors to be considered in the evaluation of milk production and composition progression. Drying period extensions positively repercussed on milk yield and lactose content, but negatively affected fat, protein, dry matter contents, and somatic cell count. Variability across drying years may depend on the drying season rather than the drying month course, except for milk yield, for which an increasing trend was reported from winter to summer. Including drying period-related non-genetic factors in genetic evaluations improves the accuracy of the regression models and permits to boost the commercial possibilities and profitability of local breeds.