Differential Performance Between Structural and Functional B-Cell Epitopic Residue Prediction

We compared the performance between the structural and functional B-cell epitope data set with sequence profiling method based on three propensity scales. One of the scales is called relative connectivity, which is newly derived from the topological parameters of residue networks. The other two are Parker´s hydrophilicity and Levitt´s index, known as the best indices so far for B-cell epitope prediction. Among all the scales and data sets tested, the best performance is always achieved with the relative connectivity, suggesting it might be the best scale for both structural and functional B-cell epitopic residue prediction. Furthermore, all the scales perform better on the structural rather than functional data sets. This indicates that these scales based sequence profiling prediction might be more appropriate for the structural B-cell epitopic residue prediction. Besides, we found that the different performance between data sets is not related to the epitopic residue density of the data set.