Author: Allison FlemingPublisher:
ISBN:
Category :
Languages : en
Pages :
Book Description
Milk fat influences the economic, nutritional, flavour, and physico-chemical properties of milk and milk products. Genetic selection for fat components is therefore of interest. This would require an efficient means of phenotyping these traits routinely, on a large scale. The purpose of this study was to assess the ability to predict milk fat components using mid-infrared (MIR) spectroscopy and to examine the phenotypic and genetic variation of these traits. Individual cow milk samples from multiple breeds and herds were collected during routine milk recording multiple times throughout the lactation and the MIR spectra obtained. Milk samples were further analyzed for average milk fat globule (MFG) size and fatty acid profile using gold standard methodologies. Partial least squares regression models were employed to develop equations to predict the milk fat component traits from the MIR spectra of milk samples. Genetic parameters were estimated in Holsteins for both the measured and predicted MFG size and predicted fatty acids, along with their genetic correlations with milk production traits using multi-trait animal models. Milk fatty acid contents could be predicted with variable accuracy depending on the calibration set used and the concentration of the fatty acid in milk. In first-parity Holsteins, the predicted groups of fatty acids were found to have genetic correlations of similar magnitude with already recorded milk production traits. Differences in the genetic correlations with fat yield, fat and protein percentages, and fat to protein ratio were found for different fatty acid groups, including different trends in average daily genetic correlations at the beginning of lactation. Average MFG size was significantly affected by herd, breed, days in milk, season, and milking period. The prediction of MFG size from MIR spectra was poor, and predicted values had greater phenotypic correlations with fat percentage than their measured counterparts. A moderate heritability was found for average MFG size, but predicted values had very strong genetic correlations with fat percentage, limiting their utility. Mid-infrared technology provides an opportunity to obtain novel trait phenotypes, including fatty acids, for all milk recorded cows that could be used to help advance the Canadian dairy industry.