The use of seemingly unrelated regression to predict the carcass composition of lambs

The aim of this study was to develop and evaluate models for predicting the carcass composition of lambs. Forty male lambs were slaughtered and their carcasses were cooled for 24 hours. The subcutaneous fat thickness was measured between the 12th and 13th rib and breast bone tissue thickness was taken in the middle of the second sternebrae. Left side of carcasses was dissected and the proportions of lean meat (LMP), subcutaneous fat (SFP), intermuscular fat (IFP), kidney and knob channel fat (KCFP), and bone plus remainder (BP) were obtained. Models were fitted using the seemingly unrelated regression (SUR) estimator which is novel in this area, and compared to ordinary least squares (OLS) estimates. Models were validated using the PRESS statistic. Our results showed that SUR estimator performed better in predicting LMP and IFP than the OLS estimator. Although objective carcass classification systems could be improved by using the SUR estimator, it has never been used before for predicting carcass composition.     

The use of ultrasound to predict the carcass composition of live Akkaraman lambs

The aim of this study was to measure fat thickness, area and depth of the longissimus dorsi muscle using ultrasonography, to estimate carcass composition in live Akkaraman lambs. Fat thickness, area and depth of the longissimus dorsi muscle between the 12th and 13th ribs were measured in vivo and on the carcass after slaughter, using real time ultrasound in 40 Akkaraman lambs. To estimate the carcass composition, one-half of a carcass was dissected into muscle, fat and bone after slaughter. Overall, correlation coefficients between ultrasound and carcass longissimus dorsi muscle area, depth and fat thickness were 0.82, 0.60 and 0.77, respectively. Estimates of carcass composition for Akkaraman lambs based on LW explained 78%, 82%, 74%, 52%, 75%, 36% and 72% of the variations for muscle, total carcass fat, subcutaneous fat, inter-muscular fat, non-carcass fat, tail fat and bone, respectively. The introduction of UFT, ULMA and ULMD as independent variables in addition to LW in the multiple linear regression equations further improved the variations for total muscle (80%), carcass fat (84%) and bone weight (76%) whereas no improvement was observed for subcutaneous, intermuscular, non-carcass and tail fat. The results showed that in vivo ultrasound fat thickness and measurement of area and depth of the longissimus dorsi muscle in association with live weight could be used to estimate muscle, total body fat and bone weight in Akkaraman lambs.     

The use of ultrasound to predict fatness in lambs

Ultrasonic measurement of fat depth, over the eye muscle of lambs was found to be highly correlated with measurement on the carcass in two experiments with coefficients of 0·93 and 0·95.

Linear regression analysis showed that over a limited range of fatness the slope (regression coefficient) approached unity and the residual standard deviation of the regression was approximately ± 1 mm. However, the slope of the line was greater than unity over a wider range of fatness and the accuracy decreased.

The instrument used was more useful for predicting fat depth than carcass fat percentage, explaining less of the variation in the latter variable than the former.

It is apparent that further investigation of the instrument over a wide range of lambs is needed to clarify the implications for selection schemes or, alternatively, technology adopted which overcomes the underestimation found in fat animals.     

Prediction of carcass composition by impedance spectroscopy in lambs of similar weight

Previous research on impedance measurements for the prediction of carcass composition was predominantly carried out on animals that varied widely in body weight, breed, or sex. The high accuracy for the estimated lean or fat mass was mainly obtained by including the body weight in the regression equations. The objective of this study was the prediction of carcass composition in lambs of similar weight. We used 70 male German Merino Mutton lambs and 70 male German Blackheaded Mutton lambs with 35 and 45kg live weight each. Impedance measurements with different electrode placements were carried out in vivo and on carcasses 20min and 24h postmortem. The carcass composition was ascertained by dissection of the left carcass side into lean, fat, and bone. R(2)-values for prediction of lean mass by impedance and body weight ranged between 0.11 and 0.71 within breeds and weight groups and between 0.84 and 0.89in the total material. Lean percentage was estimated with R(2)=0.18-0.48 within breeds and weight groups. The corresponding values for the total material varied from 0.23 to 0.37. We conclude that the impedance method is not suitable for the prediction of lean or fat percentage, neither in lambs of similar weight nor in heterogeneous animals.     

In vivo prediction of carcass composition and muscularity in purebred Texel lambs

A subjective assessment of the shape of the hind limb of purebred Texel lambs was evaluated as an in vivo predictor of carcass composition and muscularity. Lambs were taken from two flocks that were managed in a common environment, but which had either been selected for lean tissue growth rate or for improved conformation. Lambs were slaughtered at a mean age of 139 days at the end of an 11 week performance test in which they were reared indoors on a concentrate diet. Pre-slaughter measurements of live weight and ultrasonic muscle (UMD) and fat (UFD) depths at the position of the third lumbar vertebra, body length (L) and a subjective leg shape score were recorded. After slaughter, measurements were recorded for carcass side length (SL), leg length (T) and the maximum width (A) and depth (B) of the longissimus thoracis and lumborum (LTL) muscle. The side was fully dissected and various muscle weights and skeletal dimensions were used to calculate indices of muscularity as √(muscle weight/length) per unit length or as UMD/L, A/SL or B/SL. The leg shape score was positively correlated with lean weight (0.23) and proportion (0.24), lean:bone ratio (0.25), measures of LTL dimensions (0.27-0.38) and muscularity traits (0.27-0.57) but was not significantly (P>0.05) correlated with fat weights or proportions in the carcass. Live weight was the best single predictor of lean weight (RSD=0.403) and the addition of leg shape score (RSD=0.381) to prediction equations was less effective than the inclusion of UMD and UFD in combination (RSD=0.357). The addition of leg shape score to equations that included ultrasonic traits gave a significant (P<0.05) but marginal improvement in prediction (RSD=0.347). The leg shape score was the most useful in vivo predictor of carcass muscularity traits and, with R(2) in the range 0.30-0.50, had comparable predictive power to a leg muscularity score derived from muscle weight and femur length. It is concluded that the leg shape score showed potential as a predictor of carcass muscularity that was largely independent of live weight and fatness at a fixed age and was marginally associated with superior lean yield and lean:bone ratio.     

The influences of carcass weight and depot on the fatty acid composition of fats of suckling Manchego lambs

The effects of carcass weight (<5.5 kg, 5.5–6.5 kg, ⩾6.5 kg) of Manchego suckling lambs, type of fat depot of the leg (subcutaneous, intermuscular and intramuscular) and muscle (m. longissimus dorsi (LD) or m. quadriceps femoris (QF)) on the fatty acid composition were studied. Carcass fatness increased with increasing carcass weight. However, few differences in fatty acid proportion were observed between the three carcass weight groups. The lightest carcasses had lower proportions of myristic (C14:0) and palmitoleic (C16:1) acid and higher proportions of stearic (C18:0) acid, desirable fatty acids (DFA) (C18:0 + total unsaturated fatty acid (TUFA)) and a greater nutritive value ((C18:0 + C18:1)/C16:0). The fatty acid composition of subcutaneous and intermuscular fat depots of the leg were similar, whereas the intramuscular fat depot was different from both of these, since it had a lower proportion of saturated fatty acids (SFA), a greater proportion of TUFA, higher PUFA/SFA and n − 6/n − 3 ratios, and greater DFA and nutritive values. The intramuscular fat of the LD was more saturated than QF, higher SFA, and had lower value of PUFA/SFA ratio and DFA. These results indicated that the intramuscular fat quality of the leg was better than subcutaneous and intermuscular fat depots and also that intramuscular fat displayed better attributes of quality (lower SFA, higher TUFA, and greater PUFA/SFA ratio and DFA as well) than intramuscular fat of the LD from the consumer health standpoint. Despite carcass weight affected carcass fatness, it did not seem to influence the proportions of saturated or polyunsaturated fatty acids. For this reason, carcass weight probably had little effect on the organoleptic characteristics of the meat.     

The assessment of carcass composition of Awassi male lambs by real-time ultrasound at two different live weights

The accuracy of ultrasonography to measure fat thickness (FT) and longissimus dorsi muscle (LM) traits (area, depth and width) in live Awassi male lambs and predict carcass FT and LM traits was studied.Twenty six Awassi male lambs were randomly divided into light (L) (n = 13) and heavy (H) (n = 13) finishing lambs. Slaughter weight of lambs in L and H groups were 40 and 45 kg, respectively. FT and LM traits, cross-sectional area between the 12th and 13th rib were measured using real-time ultrasound in vivo and on the carcass after slaughter. All ultrasound and carcass measurements were the same except live weight (LW) (P < 0.001), cold carcass weight (CCW) (P < 0.001) and carcass LM width (P < 0.05). Overall, correlation coefficients between ultrasound and carcass FT, LM depth, width and area were 0.79, 0.82 (P < 0.001); 0.60, 0.58 (P < 0.05); 0.48, −0.17 (P > 0.05) and 0.89, 0.87 (P < 0.001), respectively, for lambs in L and H groups. The introduction of ultrasound FT and ultrasound LM area as independent variables in addition to LW in the multiple regression equations further improved the variations for carcass FT (84%, 71%), carcass LM area (79%, 79%), CCW (72%, 65%) for lambs in light and heavy groups whereas no improvement was observed for carcass yield.These results indicate that in vivo ultrasound FT and measurement of the LM area in association with LW could be used to estimate carcass FT, carcass LM area and CCW in different LW Awassi lambs.     

The assessment of carcass composition of Awassi male lambs by real-time ultrasound at two different live weights

The accuracy of ultrasonography to measure fat thickness (FT) and longissimus dorsi muscle (LM) traits (area, depth and width) in live Awassi male lambs and predict carcass FT and LM traits was studied.Twenty six Awassi male lambs were randomly divided into light (L) (n = 13) and heavy (H) (n = 13) finishing lambs. Slaughter weight of lambs in L and H groups were 40 and 45 kg, respectively. FT and LM traits, cross-sectional area between the 12th and 13th rib were measured using real-time ultrasound in vivo and on the carcass after slaughter. All ultrasound and carcass measurements were the same except live weight (LW) (P < 0.001), cold carcass weight (CCW) (P < 0.001) and carcass LM width (P < 0.05). Overall, correlation coefficients between ultrasound and carcass FT, LM depth, width and area were 0.79, 0.82 (P < 0.001); 0.60, 0.58 (P < 0.05); 0.48, −0.17 (P > 0.05) and 0.89, 0.87 (P < 0.001), respectively, for lambs in L and H groups. The introduction of ultrasound FT and ultrasound LM area as independent variables in addition to LW in the multiple regression equations further improved the variations for carcass FT (84%, 71%), carcass LM area (79%, 79%), CCW (72%, 65%) for lambs in light and heavy groups whereas no improvement was observed for carcass yield.These results indicate that in vivo ultrasound FT and measurement of the LM area in association with LW could be used to estimate carcass FT, carcass LM area and CCW in different LW Awassi lambs.     

The use of various live animal measurements to predict carcass and meat quality in two divergent lamb breeds

Live weight, subjective scores of condition and conformation, live animal video image analysis (LVIA), ultrasound and X-ray computed tomography (CT) scanning were used to investigate the best method or combination of methods for predicting carcass and meat quality traits in live Texel and Scottish Blackface lambs. Predictors derived from CT alone accounted for a high proportion of the variance in dissected fat and muscle weight in Texel lambs (adjusted R² = 0.8), as well as intra-muscular fat content in the loin (0.6), but lower proportions in Blackface lambs (0.7 for fat, 0.4–0.5 for muscle and intra-muscular fat), after adjusting for sire and fixed effects. Adding traits measured by other in vivo methods increased prediction accuracies (adjusted R²) by up to 0.26, depending on trait and data set. Shear force and ultimate pH could not be accurately predicted using the traits considered here (adjusted R² < 0.4). Although the same methods tended to be best for predicting product quality traits between breeds, prediction accuracies differed.     

The relationship of carcass measurements to carcass composition and intramuscular fat in Spanish beef

Forty beef carcasses were classified for conformation and fatness. Besides, carcass weight, fat thickness (FT), carcass dimension, marbling by computer image analysis and ultrasound readings was recorded to complement grading. For predicting intramuscular fat (IMF) content, FT, number of intramuscular flecks and conformation increased R²-value from 0.19 to 0.64 compared to conformation alone. For visual marbling, ultrasound readings and thoracic depth (TD) increased the R²-value from 0.24 to 0.57 compared to fatness score (FS). The best variables for predicting weight of fabricated subprimals were carcass weight or compactness which is a function of carcass weight (R² between 0.94 and 0.63). Fatness score was poorer than FT for predicting yield of subprimals cuts from round (R² = 0.16 vs. 0.50) and ultrasound readings for less valuable subprimals (R² = 0.31 vs. 0.39). These results showed that other variables could be used in combination with carcass fatness or conformation to achieve a more accurate estimation of fat and carcass yield.     

The use of various live animal measurements to predict carcass and meat quality in two divergent lamb breeds

Live weight, subjective scores of condition and conformation, live animal video image analysis (LVIA), ultrasound and X-ray computed tomography (CT) scanning were used to investigate the best method or combination of methods for predicting carcass and meat quality traits in live Texel and Scottish Blackface lambs. Predictors derived from CT alone accounted for a high proportion of the variance in dissected fat and muscle weight in Texel lambs (adjusted R² = ∼0.8), as well as intra-muscular fat content in the loin (∼0.6), but lower proportions in Blackface lambs (∼0.7 for fat, 0.4–0.5 for muscle and intra-muscular fat), after adjusting for sire and fixed effects. Adding traits measured by other in vivo methods increased prediction accuracies (adjusted R²) by up to 0.26, depending on trait and data set. Shear force and ultimate pH could not be accurately predicted using the traits considered here (adjusted R² < 0.4). Although the same methods tended to be best for predicting product quality traits between breeds, prediction accuracies differed.     

Relevance of carcass palpation in lambs to protecting public health

We recently reviewed the Food Safety and Inspection Service's (FSIS's) inspection procedures for lambs. As a result, FSIS published a Federal Register notice informing the public of its intent to change from an inspection system that requires extensive carcass palpation to an inspection system that requires no carcass palpation for lambs. This decision was based on the following three points. (i) Extensive carcass palpation in lambs does not routinely aid in the detection of food safety hazards that result in meat-borne illnesses. (ii) Hands are capable of spreading or adding contamination to the carcasses. (iii) FSIS inspection systems must reflect science-based decisions as they pertain to meat-borne illnesses consistent with a Pathogen Reduction/Hazard Analysis and Critical Control Point environment.     

Fatty acid composition and carcass characteristics of growing lambs fed diets containing palm oil supplements

Thirty 45-day-old male Ojalada lambs were used to investigate the effect of palm oil (as such or in the form of calcium soap) on carcass characteristics and fat composition. Dietary palm oil did not significantly affect dorsal fat thickness, carcass fatness or carcass conformation. The palm oil supplement did not affect C16:0 levels in intramuscular and subcutaneous fat, but did increase the percentage of C16:0 and reduce PUFA and MUFA content in internal fat depots. Supplementation with palm oil in the form of calcium soap did not cause significant differences in the composition of intramuscular fat. Subcutaneous and perirenal fat of lambs fed treatments including palm oil as such contained higher levels of some saturated fatty acids than that of those fed fat in the form of calcium soap. Mesenteric fat of lambs that received calcium soap had less % C18:1 and more % C18:3 than that of those given palm oil.     

Plasma leptin in growing lambs as a potential predictor for carcass composition and daily gain

Leptin is an adipocyte derived hormone and correlates highly to the extent of body fat tissue. The aim of this study was to determine if leptin could serve as an early predictor for carcass composition and final growth rate in lambs with special emphasis on size and cellularity of the different body fat depots. Thirty intact male ad libitum fed lambs were blood sampled at 20, 25, 30, 35, and 40kg live weight. After slaughtering at 40kg, lean and the visceral, subcutaneous and intermuscular fat were measured by dissection. The fat cell diameter was determined in subcutaneous and perirenal fat. Average daily gain from birth to slaughter correlated to leptin only at 30 and 35kg live weight (r=-0.56 and -0.61, P<0.01) and thus leptin cannot be regarded as a suitable early predictor for growth rate. That goes for the prediction of subcutaneous and intermuscular fat, too; because no relationships were detected between early leptin concentrations and the amount of these tissues. Leptin concentrations measured just before slaughter were related to all fat tissues except the pelvic and intramuscular fat. Among the visceral fat depots, omental fat expressed the highest correlations to leptin (r=0.60, P<0.001). Additionally, leptin concentrations at 35 and 40kg live weight increased with increasing fat cell diameters (r=0.38, P<0.05 to r=0.59, P<0.001). This study indicates that leptin concentration measured in the slaughter weight range has the greatest potential to assess body fat content, whereas an earlier prediction does not seem to be feasible. Further studies should clarify if these results are reproducible for other breeds or species.     

A comparison of alternative methods for predicting the carcass composition of crossbred lambs of different breeds and crosses

Carcass data for a total of 2808 lambs from three breed trials were used to evaluate a series of linear measurements, visual scores and the proportions of tissues in joints as predictors of carcass composition. The trials involved crossbred lambs out of different dam types by sires of the main British meat breeds and by Ile de France, Oldenburg and Texel sires. Each trial was carried out over a number of years and involved approximately equal numbers of female and castrated male lambs. Potential predictors were examined in terms of the precision of prediction and in terms of the robustness of prediction equations to differences in sire breed and sex. The overall standard deviation (s.d.) of carcass lean percentage averaged over trials was 4·3. Combinations of simple measurements, including visual fat scores, percentage perinephric and retroperitoneal fat in carcass and M. longissimus dimensions achieved a residual s.d. of 2·5 for carcass lean percentage. The application of an overall prediction equation to individual sire breed means resulted in considerable bias (predicted-actual lean percentage): the mean square deviation was 0·75. In comparison, the sex differences were relatively unimportant. The precision of sample joints was examined in relation to their cost of dissection. The best end neck and shoulder joints offered a high level in relation to cost: typical residual s.d. were 1·5 for carcass lean percentage. Joints and combination of joints with high predictive precision tended to have equations that were robust to differences between sire breeds.     

On the use of milk composition measures to predict the energy balance of dairy cows

Milk composition varies with energy status and was proposed for measuring energy balance on-farm, but the accuracy of prediction using monthly samples is not high. With automated sampling and inline milk analysis, a much higher measurement frequency is possible, and thus improved accuracy of energy balance determination may be expected. Energy balance was evaluated using data in which milk composition was measured at each milking. Three breeds (Danish Holstein, Danish Red, and Jerseys) of cows (623 lactations from 299 cows) in parities 1, 2, and 3+ were used. Data were smoothed using a rolling local regression. Energy balance (EBal) was calculated from changes in body reserves (body weight and body condition score). The relationship between EBal and milk measures was quantified by partial least squares regression (PLS) using group means data. For each day in lactation, the within-breed and parity mean EBal and mean milk measures were used. Further PLS was done using the individual cow data. The initial PLS models included 25 combinations of milk measures allowing a range of nonlinear effects. These combinations were as follows: days in milk (DIM); DIM raised to the powers 2, 3, and 4; milk yield; fat content; protein content; lactose content; fat yield; protein yield; lactose yield; fat:protein ratio; fat:lactose ratio; protein:lactose ratio; and milk yield:lactose ratio, together with 10 “diff()” variables. These variables are the current minus the previous value of the milk measure in question. Using group means data, a very high proportion (96%) of the variability in EBal was explained by the PLS model. A reduced model with only 6 variables explained 94% of the variation in EBal. This model had a prediction error of 3.82 MJ/d; the 25-variable model had a prediction error of 3.11 MJ/d. When using individual rather than group means data, the PLS prediction error was 17.3 MJ/d. In conclusion, the mean Ebal of different parities of Holstein, Danish Red, and Jersey cows can be predicted throughout lactation using 1 common equation based on DIM, milk yield, milk fat, and milk protein measures.     

The effect of fasting on liveweight and carcass characteristics in lambs

In two experiments, involving 250 and 300 lambs, respectively, lambs were fasted (with access to water) and slaughtered after 0, 1, 2, 3 and 4 days. The effects of initial fat score, initial liveweight and sex (wethers and ewes) on the patterns of loss in liveweight, hot carcass weight and carcass characters (fat depth at the GR and C sites, kidney fat weight and the weights of chemical carcass components) were examined. In experiment 2 the effects of weaning and transport on fasting losses were also examined.

The rate of liveweight loss (expressed as a percentage of initial liveweight) was greater in the initial stages of the fast and was affected by fat score, with a greater loss in lambs with a lower fat score. The pattern of loss in hot carcass weight with fasting was curvilinear and in experiment 2 was affected by both fat score and liveweight, with leaner, heavier lambs having a greater loss. In experiment 2 the decrease in fat depth at the GR site interacted with fat score, with a greater loss in lambs with a lower fat score. Transport per se had no effect on hot carcass weight or carcass characteristics.     

Genetic parameters for carcass composition and performance data in crossbred lambs measured by Video Image Analysis

A total of 7074 crossbred lambs, produced by mating crossbred Mule ewes with terminal sire rams were used in this study. Of these, 630 were scanned using a Video Image Analysis (VIA) to estimate carcass quality traits. Genetic parameters for average daily gain (ADG), scanning live weight (SW), ultrasonic measures of muscle (UMD) and fat (UFD) depths, cold carcass weight (CCW) and VIA measurements of primal carcass joint weights (LEG, CHUMP, LOIN, BREAST and SHOULDER) were estimated using multivariate animal models. Additionally, VIA traits were evaluated under a repeatability model, considering the primal joints as repeated measures of the same trait. Direct heritability estimates were low to moderate (0.08–0.26) for VIA measurements of primal joints. Repeatability estimates for VIA traits were high (>0.90). Moderate to high heritability estimates (0.25–0.55) were found for performance traits (ADG, SW, UMD and UFD) and CCW. Genetic correlations between VIA traits and ADG were strong (0.75–0.93). Most of the VIA traits were highly correlated to SW (0.60–0.97). UFD was significantly negatively correlated with UMD (−0.22), ADG (−0.18) and CCW (−0.18). The results of this study suggest that selection on performance and carcass traits, measured by VIA, could possibly improve primal meat yield of carcass cuts without increasing the overall carcass fatness. High repeatability estimates of VIA traits and moderate heritabilities of the most valuable carcass joints suggests that including VIA information in breeding programs would be useful in order to improve carcass quality.     

Effects of previous diet and duration of soybean oil supplementation on light lambs carcass composition, meat quality and fatty acid composition

Forty Merino Branco ram lambs were used to study the effects of initial diet and duration of supplementation with a conjugated linoleic acid (CLA) promoting diet, on carcass composition, meat quality and fatty acid composition of intramuscular fat. The experimental period was 6 weeks. The experimental design involved 2 initial diets (commercial concentrate (C); dehydrated lucerne (L)), and 2 finishing periods (2 and 4 weeks) on dehydrated lucerne plus 10% soybean oil (O). Data were analysed as a 2 × 2 factorial arrangement with initial diet and time on finishing (CLA promoting) diet as the main factors. The lambs were randomly assigned to four groups: CCO; COO; LLO; LOO according to the lamb’s diet fed in each period.Lambs initially fed with concentrate showed higher hot carcass weights (11.2 vs 9.6 kg) than lambs fed initially with lucerne. The increase of the duration of finishing period reduced the carcass muscle percentage (57.4% vs 55.5%) and increased the subcutaneous fat percentage (5.67% vs 7.03%). Meat colour was affected by initial diet. Lambs initially fed with concentrate showed a lower proportion of CLA (18:2cis-9, trans-11 isomer) (0.98% vs 1.38% of total fatty acids) and most of n−3 polyunsaturated fatty acids than lambs initially fed with lucerne. Initial diet did not compromise the response to the CLA-promoting diet and the proportion of 18:2cis-9, trans-11 in intramuscular fat increased with the duration of time on the CLA-promoting diet (1.02% vs 1.34% of total fatty acids).     

Prediction of carcass composition in the rabbit

Carcass composition of two synthetic rabbit breeds was predicted from retail cuts and external measurements by using regression equations. Breed R has a higher adult weight and reaches slaughter weight 1 week before breed V. Sixty rabbits of each breed were slaughtered when they (approximately) reached the Spanish commercial liveweight of 2kg. The carcasses were measured and retailed according to the norms of the World Rabbit Scientific Association. Rabbit carcass composition is well defined by meat percentage of the commercial carcass and ratio meat bone . External measurements on the carcass, retail cuts and meat of retail cuts or muscular masses are all bad predictors of carcass meat percentage or ratio meat bone (R(2) < 0.53). The ratio meat bone of the hind leg can give reasonable predictions for carcass meat percentage and meat bone ratio (R(2) = 0.60 and 0.69). Dissectible carcass fat weight and dissectible carcass fat percentage can be predicted by the perirenal fat weight (R(2) = 0.77 and 0.69). Fat depots had a low predictive power for fat percentage of the dissected meat in the half carcass.