TY - JOUR T1 - Reliable B Cell Epitope Predictions: Impacts of Method Development and Improved Benchmarking A1 - Kringelum, Jens Vindahl A1 - Lundegaard, Claus A1 - Lund, Ole A1 - Nielsen, Morten Y1 - 2012/12/27 N2 - Author Summary The human immune system has an incredible ability to fight pathogens (bacterial, fungal and viral infections). One of the most important immune system events involved in clearing infectious organisms is the interaction between the antibodies and antigens (molecules such as proteins from the pathogenic organism). Antibodies bind to antigens at sites known as B-cell epitopes. Hence, identification of areas on the surface antigens capable of binding to antibodies (also known as B-cell epitopes) may aid the development of various immune related applications (e.g. vaccines and immunotherapeutic). However, experimental identification of B-cell epitopes is a resource intensive task, thereby making computer-aided methods an appealing complementary approach. Previously reported performances of methods for B cell epitope predictive have been moderate. Here, we present an updated version of the B-cell epitope prediction method; DiscoTope, that on the basis of a protein structure and epitope propensity scores predicts residues likely to be involved in B-cell epitopes. We demonstrate that the low performances to some extent can be explained by poorly defined benchmarks, and that inclusion of additional biological information greatly enhances the predictive performance. This suggests that, given proper benchmark definitions, state-of-the-art B cell epitope prediction methods perform significantly better than generally assumed. JF - PLOS Computational Biology JA - PLOS Computational Biology VL - 8 IS - 12 UR - https://doi.org/10.1371/journal.pcbi.1002829 SP - e1002829 EP - PB - Public Library of Science M3 - doi:10.1371/journal.pcbi.1002829 ER -