AOPs-XGBoost: Machine learning Model for the prediction of Antioxidant Proteins properties of peptides

Abstract

Abstract Antioxidant proteins are essential for protecting cells from free radicals. The accurate identification of antioxidant proteins via biological tests is difficult because of the high time and financial investment required. The potential of peptides produced from natural proteins is demonstrated by the fact that they are generally regarded as secure and may have additional advantageous bioactivities. Antioxidative peptides are typically discovered by analyzing numerous peptides created when a variety of proteases hydrolysis proteins. The eXtreme Gradient Boosting (XGBoost) technique was used to create a novel model for the current study, which was then compared to the most popular machine learning models. We suggested a machine-learning model that we named AOPs-XGBoost, built on sequence features and Extreme Gradient Boosting (XGBoost). We used 10-fold cross-validation testing was performed on a testing dataset using the propose. AOPs-XGBoost classifier, and the results showed a sensitivity of 67.56%, specificity of 93.87%, average accuracy of 80.72%, mean cross-validation (MCC) of 66.29%), and area under the receiver operating characteristic curve (AUC) of 88.01%. The outcomes demonstrated that the XGBoost model outperformed the other models with accuracy of 80.72% and area under the receiver operating characteristic curve of 88.01% which were better than the other models. Experimental results demonstrate that AOPs-XGBoost is a useful classifier that advances the study of antioxidant proteins.