Precise Predictions of Linear B Cell Epitopes in Protean 3D

B cell epitopes - the part of an antigen recognized by an antibody are conformational in nature; however, accurately predicting these epitopes is difficult for proteins with- out a known three-dimensional (3D) structure. In this white paper, DNASTAR presents the NeoClone method, a machine learning approach that improves the ability to predict linear B cell epitopes (a pep- tide segment) using only sequence-based information. This method provides better overall predictive accuracy than COBEpro and Epitopia, two leading epitope prediction methods.