Linear B-cell epitopes prediction using bagging based proposed ensemble model

Vaccine design through experimental analysis is too costly and time taking process. By using computational intelligence approaches, cost and time can be reduced. Identification of linear B-cell epitopes is the main concern of peptide vaccine designs, immunodiagnosis, and antibody productions. It can be performed by developing a suitable machine learning model. In this paper, prediction of linear B-cell epitopes has been performed by using a bagging-based proposed ensemble model. The goal of using a bagging-based ensemble approach is to improve the prediction performance of various base classifiers. Here, five existing base classifiers are combined for ensembling to predict the B-cell epitopes or non-epitopes. This prediction is based on a bagging-based voting system, which is performed under binary class classification. The proposed ensemble model achieved 82.06% accuracy, which is better than to some existing models. Finally, the proposed ensemble model has been tested using sevenfold cross-validation, and where it provided almost consistent performance. © 2022, The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of Computer Applications and Management.