Abstract
In this paper, we propose a novel hybrid binary animal migration optimization (BAMO) with k-nearest neighbor approach (KNN) to predict apoptosis protein sequences using statistical factors and dipeptide composition. Binary animal migration optimization is used for selecting a near-optimal subset of informative features that is most relevant for the classification. K-nearest neighbor approach is used as the classifier with the jackknife cross-validation. Finally, BAMOKNN is tested on a dataset including 317 proteins. Our method achieves the accuracy of 92.43%. Then, our model also tests on a testing dataset including 98 apoptosis proteins and obtains the accuracy of 94.90%. High prediction accuracy and successful prediction of apoptosis proteins suggest that BAMOKNN can be a useful approach to identify apoptosis protein locations.