The accurate knowledge of true digestible amino acid (TDAA) contents of feedstuff is necessary to formulate the more accurate poultry diets for a profitable production. Several experimental approaches have been used to determine available amino acids which are highly expensive and time consuming. Prediction of nutritive value of a feed ingredient from its chemical composition via regression method has been attempted for many years. Artificial neural network (ANN) model is the powerful method which may describe the relationship between digestible amino acid contents and chemical composition. Therefore we developed multiple linear regressions (MLR) and ANN models for predicting the TDAA contents of sorghum grain based on chemical composition. The precision fed assay trials using cecectomized roosters were performed to determine the TDAA contents in 48 sorghum samples from 12 sorghum varieties differing in chemical compositions. The input variables for both MLR and ANN models were CP, ash, crude fiber (CF), ether extract (EE), and total phenols whereas the output variable was the each individual TDAA for every sample. The results of this study revealed that it is possible to satisfactorily estimate the TDAA of sorghum grain through its chemical composition. The chemical compositions of sorghum grain seems to highly influence the TDAA contents when considering components such as CP, CF, EE, ash and total phenols. It is also possible to estimate the TDAA con- tents through multiple regression equations with reasonable accuracy depending on compositions but, a more satisfactory prediction may be achieved via ANN for all amino acids. The R2 values corresponding to testing and training of the ANN model showed a higher accuracy of prediction than equation established by MLR method. In addition, the current data confirmed that the chemical composition, often considered in total amino acid prediction, could be also a useful predictor of true digestible values of selected amino acids for poultry.

Keywords