Linear B-cell epitope prediction: a performance review of currently available methods

ABSTRACTLinear B-cell epitope prediction research has received a steadily growing interest ever since the first method was developed in 1981. B-cell epitope identification with the help of an accurate prediction method can lead to an overall faster and significantly cheaper vaccine design process. Consequently, several B-cell epitope prediction methods have been developed over the past few decades, but without significant success. In this study, we review the current performance and methodology, of some the most widely used linear B-cell epitope predictors: BcePred, BepiPred, ABCpred, COBEpro, SVMTriP, LBtope and LBEEP. Additionally, we attempt to remedy performance issues of the individual methods by developing a consensus classifier, that combines the separate predictions of these methods into a single output. The performance of these methods was evaluated using a large unbiased data set. All methods performed worse than documented in the original manuscripts, since all predictors performed marginally better than random classification against the test data set. While the method comparison was performed with some necessary caveats, we hope that this update in performance can aid researchers towards the choice of a predictor, whilst conducting their research. The necessary files for the execution of the consensus method that we developed can be found at http://thalis.biol.uoa.gr/BCEconsensus/.KEY POINTSReview of the performance and methodology of currently available BCE predictorsDesign and development of consensus predictorComparison of consensus with state-of-the-art BCE prediction methodsConsensus method available at http://thalis.biol.uoa.gr/BCEconsensus/Kosmas A. Galanis has a BSc in Biology and has performed his undergrad thesis in Bioinformatics. He is interested in the development of computational methods for protein function prediction.Katerina C. Nastou is a Biologist with a PhD in Bioinformatics. Her research focuses on the study of biological networks, the computational prediction of protein function and biological database development.Nikos C. Papandreou has a PhD in Biophysics and works as Special Laboratory Teaching Staff in “Bioinformatics-Biophysics” at the Department of Biology, National & Kapodistrian University of Athens.Georgios N. Petichakis is a Computer Scientist with an MSc in Bioinformatics. His research focuses on the development of computational methods for the functional annotation of proteomes.Vassiliki A. Iconomidou is an Assistant Professor of Molecular Biophysics and the group leader of the Biophysics and Bioinformatics Lab at the Department of Biology, National and Kapodistrian University of Athens.