Ensemble Modeling of Epitopes and Non-Epitopes Prediction from Protein Sequences of Zika and Dengue Viruses

Finding specific parts in a protein sequence that trigger an immune response is crucial for understanding how diseases develop, creating vaccines, and using the immune system to treat cancer. Due to many pathogen sequences being fully or partially sequenced for proteins, it's too costly and time-consuming to use experimental methods to locate these parts. In this paper, an ensemble model is proposed to predict Epitopes and Non-epitopes from protein sequences. The experimental results were determined using datasets from the ZIKA and Dengue viruses. The proposed Ensemble Model achieved the highest accuracy of 95.36% for the Zika dataset and 89.60% for the Dengue dataset. These final results can be compared with existing state-of-the-art methods. The new combination model aims to predict potential vaccines for ZIKA and Dengue, potentially saving lives globally and preventing significant outbreaks of these diseases in the future.