Epitope-Evaluator: an interactive web application to study predicted T-cell epitopes

Abstract

AbstractMultiple immunoinformatic tools have been developed to predict T-cell epitopes from protein amino acid sequences for different major histocompatibility complex (MHC) alleles. These prediction tools output hundreds of potential peptide candidates which require further processing; however, these tools are either not graphical or not friendly for non-programming users. We present Epitope-Evaluator, a web tool developed in the Shiny/R framework to interactively analyze predicted T-cell epitopes. This includes providing the distribution of epitopes across a selected set of MHC alleles, the promiscuity and conservation of epitopes, and their density and location within antigens. Epitope-Evaluator requires as input the fasta file of protein sequences and the output prediction file coming out from any predictor. By choosing different cutoffs and parameters, users can produce several interactive plots and tables that can be downloaded as JPG and text files, respectively. Epitope-Evaluator removes the programming barrier and provides intuitive tools, allowing a straightforward interpretation and graphical representations that facilitate the selection of candidate epitopes for experimental evaluation.Author SummaryWith the advent of the COVID-19 pandemic as well as past pandemics and epidemics, scientists have focused on immunological studies to develop better vaccines as well as understand immune responses. Many of the questions are centered on studying T-cell epitopes, and peptide sequences that can be presented to immune cells to elicit responses against pathogens. Although current software can produce hundreds of predictions, they are generally not user-friendly nor graphical. In order to remove the existing programming barrier, we developed a Web tool to allow scientists to analyze and filter T-cell epitopes in an easy, intuitive, interactive, and versatile way. We have included two biological cases identifying new biological insights and showing the importance of having this type of toolset, especially for nonprogrammer researchers in the immunology field.