Prediction of retail beef yield and fat content from live animal and carcass measurements in Nellore cattle1

• Published in: Journal of animal science 2014-11, Vol.92 (11), p.5230-5238
• Main Authors: Sakamoto, L. S., Mercadante, M. E. Z., Bonilha, S. F. M., Branco, R. H., Bonilha, E. F. M., Magnani, E.
• Format: Article
• Language: English
• Subjects: Beef; Body fat; Thickness measurement
• ISSN: 0021-8812; 1525-3163
• DOI: 10.2527/jas.2012-6065

Data from 156 Nellore males were used to develop equations for the prediction of retail beef yield and carcass fat content, expressed as kilograms and as a percentage, from live animal and carcass measurements. Longissimus muscle area and backfat and rump fat thickness were measured by ultrasound up to 5 d before slaughter and fasted live weight was determined 1 d before slaughter. The same traits were obtained after slaughter. The carcass edible portion (CEP in kg and CEP% in percentage; n = 116) was calculated by the sum of the edible portions of primal cuts: hindquarter, forequarter, and spare ribs. Trimmable fat from the carcass boning process, with the standardization of about 3 mm of fat on retail beef, was considered to be representative of carcass fat content. Most of the variation in CEP was explained by fasted live weight or carcass weight (R^sup 2^ of 0.92 and 0.96); the same occurred for CEP% (R^sup 2^ of 0.15 and 0.13), and for CEP, the inclusion of LM area and fat thickness reduced the equation bias (lower value of Mallow's C^sub p^ statistics). For trimmable fat, most variation could be explained by weight or rump fat thickness. In general, the equations developed from live animal measurements showed a predictive power similar to the equations using carcass measurements. In all cases, the traits expressed as kilograms were better predicted (R^sup 2^ of 0.39 to 0.96) than traits expressed as a percentage (R^sup 2^ of 0.08 to 0.42).