lues were set for sensitivity (0.47) and specificity (0.75) for selecting the amino acids forming discontinuous epitopes. A sensitivity of 0.47 means that 47 per cent of the epitope residues are predicted as part of the epitopes, while a specificity of 0.75 means that 25 per cent of the non-epitope residues are predicted as part of the epitopes. Outputs from both the methods were combined, and the final regions were mapped on the modelled 3D-structure as the most probable conformational epitopes. In addition, we also predicted N-linked glycosylation sites in the S protein using NetNGlyc 1.0 Server (ht