Development and evaluation of empirical equations to interconvert between twelfth-rib fat and kidney, pelvic, and heart fat respective fat weights and to predict initial conditions of fat deposition models for beef cattle1

Abstract

The Davis growth model (DGM) simulates growth and body composition of beef cattle and predicts development of 4 fat depots. Model development and evaluation require quantitative data on fat weights, but sometimes it is necessary to use carcass data that are more commonly reported. Regression equations were developed based on published data to interconvert between carcass characteristics and kilograms of fat in various depots and to predict the initial conditions for the DGM. Equations include those evaluating the relationship between the following: subcutaneous fat (SUB, kg) and 12th-rib fat thickness (mm); visceral fat (VIS, kg) and KPH (kg); DNA (g) in intermuscular, intramuscular, subcutaneous, and visceral fat depots and empty body weight; and contributions of fat (kg) in intramuscular (INTRA), SUB, and VIS fat depots and total body fat (kg). The intermuscular fat (INTER, kg) contribution was found by difference. The linear regression equations were as follows: SUB vs. 12th-rib fat thickness (n = 75; P < 0.01) with R(2) = 0.88 and SE = 10.00; VIS vs. KPH (kg; n = 78; P < 0.01) with R(2) = 0.95 and SE = 2.82; the DNA (g) equations for INTER, INTRA, SUB, and VIS fat depots vs. empty body weight (n = 6, 5, 6, and 6; P = 0.08, P < 0.01, P < 0.01, and P = 0.05) with R(2) = 0.57, 0.93, 0.93, and 0.66, and SE = 0.03, 0.003, 0.02, and 0.03, respectively; and initial contribution of INTRA, SUB, and VIS fat depots vs. total body fat (n = 23; P < 0.01) for each depot, with R(2) = 0.97, 0.99, and 0.97, and SE = 0.61, 0.93, and 1.41, respectively. All empirical equations except for DNA were challenged with independent data sets (n = 12 and 10 for SUB and VIS equations and n = 9 for the initial INTER, INTRA, SUB, and VIS fat depots). The mean biases were -2.21 (P = 0.12) and 2.11 (P < 0.01) kg for the SUB and VIS equations, respectively, and 0.05 (P = 0.97), -0.37 (P = 0.27), 1.82 (P = 0.08), and -1.50 (P = 0.06) kg for the initial contributions of INTER, INTRA, SUB, and VIS fat depots, respectively. The random components of the mean square error of prediction were 73 and 26% for the SUB and VIS equations, respectively, and similarly were 99, 85, 62, and 61% for the initial contributions of INTER, INTRA, SUB, and VIS fat depots, respectively. Both the SUB and VIS equations predicted accurately within the bounds of experimental error. The equations to predict initial fat contribution (kg) were considered adequate for initializing the fat depot differential equations for the DGM and other beef cattle simulation models.