Potential of CIELAB colour scores to gauge the quality of sorghum as a feed grain for chicken-meat production

Context Cereal grains used by the poultry industry in Australia vary widely in available energy and protein content, which is often reflected as variation in bird performance. Rapid or real-time techniques for measuring the apparent metabolisable energy (AME) content of cereal grains for birds include near infrared spectroscopy, rapid visco-analysis starch pasting profiles and colour analysis. Aims This study involved retrospective colour analysis of Australian sorghum samples reported in recent publications, and sorghum samples used in commercial production of chicken meat in Australia. The main objective was to develop regression models as tools to predict AME values for sorghum from colour analysis of the grain for timely assistance to nutritionists formulating commercial diets and purchasing sorghum grain. Methods Stepwise regression analysis was used to correlate AME values for 18 samples of red, yellow and white sorghum with their CIELAB colour variables L*, a* and b*, which indicate lightness (from black to white), green-red component and blue-yellow component, respectively. The model was then used to predict AME values for sorghum in previously reported studies. Key results The multivariate model AMEsorghum (MJ/kg DM) = 31.139 – 0.189 L* – 0.604 a* + 0.189 b* (P = 0.0021, R2 = 0.638) was shown to predict AME of red sorghum samples to within an average difference of 0.67 MJ/kg DM in one published study. The sorghum sample showing the largest difference contained kafirin 61.5 g/kg. Data from another published study indicated larger differences (0.93 MJ/kg DM) between predicted and measured values for sorghum. The largest difference of 1.41 MJ/kg DM was observed for a sample of white sorghum containing the lowest concentrations of kafirin (41.4 g/kg), phytate (4.93 g/kg) and total phenolics (3.00 mg GAE/g). Conclusions CIELAB colour analysis has potential as a rapid, inexpensive indicator of AME values for sorghum as a feed grain for chicken-meat production, but high concentrations of antinutritive components, such as kafirin, detract from this potential. Implications A rapid, inexpensive indicator of kafirin, such as near infrared, is required to complement CIELAB colour analysis.