Breeds and muscle types modulate performance of near-infrared reflectance spectroscopy to predict the fatty acid composition of bovine meat.

This study aims to assess near-infrared reflectance spectroscopy feasibility for predicting beef fatty acid (FA) composition. Experimental scheme included four breeds (Angus, Blond d'Aquitaine, Charolais, Limousin) and three muscles, Longissimus thoracis (LT), Rectus abdominis (RA), Semitendinosus (ST). The results showed that 1) increasing FA content variability with several breeds increased calibration model reliability (R(2)CV>0.86) for the major individual and groups of FA unless polyunsaturated FAs, 2) Longissimus thoracis FAs were better predicted than RA FAs while no ST FAs were correctly predicted (R(2)CV<0.71). This difference could be explained by FA content, FA variability or specific muscle physico-chemical characteristics.