Knowledge-based virtual screening of HLA-A*0201-restricted CD8+ T-cell epitope peptides from herpes simplex virus genome.

A novel knowledge-based method is developed to virtually screen potential HLA-A*0201 binders from large-scale peptide candidates. This method utilizes the information from both the crystal structures and experimental affinities of various peptides bound with HLA-A*0201 to construct a single-position mutation free energy profile for accurately characterizing HLA-A*0201-peptide interaction and for effectively predicting the binding affinities of peptides to HLA-A*0201. We employ this method to analyze physicochemical properties and structural implication underlying the specific recognition and association between the HLA-A*0201 and a large panel of peptide segments generated from the herpes simplex virus type 1 (HSV-1) genome, and to evaluate the binding potencies of these peptide candidates to HLA-A*0201. As a result, 288 out of 38,020 candidates are predicted as the potential high-affinity binders of HLA-A*0201, from which three most promising peptides are picked out for further development of potent vaccines against HSV-1. In addition, we also demonstrate that this newly proposed method can successfully identify 8 known binders and 3 known nonbinders from the glycoproteins D and K of HSV-1.