Steps to implement animal breeding for improved nutritional quality of bovine milk

Animal improvement is always directed towards a given selection objective. These objectives in dairy cattle have been changing in the last forty years passing from simple yield only potentially associated with type, towards more complex ones including also functional traits as fertility and longevity. Milk composition has been selected too in most countries, however focussing only on major components, fat and protein. During the last years interest in milk quality traits, especially nutritionally quality has been increasing. Implementing breeding for additional traits first requires that animals can be evaluated for them because data is available and they are heritable enough. The objective of our research was to develop the needed steps to implement animal breeding for improved nutritional quality. Until recently data for fatty acid (FA) composition could not be obtained easily. However calibration equations for mid-infrared (MIR) spectrometry data generated in routine milk analysis done during milk recording were developed. They are used in routine in Belgium (Walloon Region) and Luxembourg and generate MIR data used to predict FA for up to 100.000 cows for every test date, therefore up to a maximum of 1 million test-day records. Multivariate test-day models were developed that allow estimating genetic parameters and ultimately breeding values as soon as enough data is available. Daily heritabilities estimated ranged between 0.1 and 0.4 showing that sufficient genetic differences exist. Because of the lack of historical data correlated traits as milk yield, fat and protein content will be analyse together with FA. For the future research on correlations between FA and other economically important traits will be done. Results are needed to integrate FA in the selection objectives. Because of the relative lack of historical data there is a high interest to research using genomic information to provide combined breeding values.