Prédire la qualité des viandes: mythe ou réalité ?

This review was aimed at providing an overview of recent advancements obtained in the field of meat quality prediction. The different methods used in the production sectors for the development of equations and the tools used based on different types of biological markers (genomic or phenotypic) or physical (spectroscopy) are discussed. Through the different examples that are presented, it appears that the identification of biological markers is confronted with the complex determinism of quality parameters. This makes the development of generic molecular tests to be used on the field even more difficult. However, in recent years, progress has been made, benefitting from genomics, proteomics and metabolomics technological developments. The first equations that predict the sensorial quality and the technological potential of meat hint to possible applications in the near future. Concerning spectroscopy, the main results were obtained using Near Infrared Spectroscopy (NIR) with developments obtained for the prediction of the composition and nutritional value of meats. The prediction of the technological potential of meats using this method and mainly the sensorial quality is, however, more difficult. Finally, the example of the MSA (« Meat Standards Australia ») phenotypic model which predicts the sensorial quality of bovine meat based on a combination of upstream and downstream data and whose added-value for the sector has been identified in Australia, presents a genericness that has already been proven in several countries.