Predicting Chemical Body Composition Using Body Part Composition in Boer × Saanen Goats

Abstract

Two experiments were conducted to determine which part of the empty body of Boer × Saanen intact male kids can be used to predict the chemical composition of the whole body. In the first experiment, kids were fed ad libitum and slaughtered at 5, 10, and 15 kg body weight (BW). Eighteen animals were group-fed at three intake levels (ad libitum or restricted to 30% and 60% of the ad libitum level). When the ad libitum animal in the group reached 15 kg BW, all animals in the group were slaughtered. In the second experiment, kids were fed ad libitum and slaughtered at 15, 20, and 25 kg BW. Twenty-one animals were group-fed at three intake levels and slaughtered when the ad libitum animal within the group reached 25 kg BW. Analyzed body parts included head + feet, hide, organs, neck, shoulder, ribs, loin, leg, 9–11th ribs, and half carcass. Principal component and cluster analyses showed that the neck, 9–11th ribs, and loin had the highest frequency of grouping with the empty body. These body parts were used to develop prediction models for estimating body composition. The neck, loin, and 9–11th ribs accurately and precisely predicted the dry matter, ash, fat, protein, and energy body composition of goats, with most models also incorporating BW as a predictor variable. The equations showed root mean squared error (RMSE) lower than 13.5% and a concordance correlation coefficient (CCC) greater than 0.84. Fat and protein concentrations in the loin and neck were also reliable predictors of empty body energy composition (RMSE = 2.9% of mean and concordance correlation coefficient = 0.93). Removing the loin and 9–11th ribs could reduce the carcass retail price. Using the neck to estimate body composition in growing Boer × Saanen goats provides a valuable alternative for nutrition studies, given its low commercial value.