Overview on computational-based B-cell epitope prediction: Dataset, method and accuracy

Vaccine development evolves to peptide-based vaccine which is effective and efficient if it is composed from epitopes of pathogens. Experimental-based epitope prediction such as X-ray crystallography is time consuming and expensive. Available data from this experiment makes a possibility to develop computer based epitope prediction method. This approach is crucial to accelerate epitope identification in wet experiment. The performance of B cell epitope prediction is still in the medium level, although the development of methods, database and software have evolved and the improvement of the earlier method is statistically significant. This state shows that the improvement of the performance is an open problem. In this study, computational based epitope prediction to support vaccine development is resumed and is studied comprehensively. Therefore, this article can be used as a reference for another researcher to identify the current state in this research area.