Hemagglutinin(HA) is the surface protein of influenza viruses, which is known to play important role in the infection process. It is the antigen that inducing immune responses and currently be used as the influenza vaccine potency marker. Epitope are some segments in the antigen which could be recognized by the immune system specifically and induce immune response. The T-cell epitopes of protein antigens are some peptides that presented on the cell surface by MHC molecules via the MHC class I pathway or the MHC class II pathway. The antigen presentation mechanisms are different in these two pathways. The MHC class I pathway is used to present endogeneous antigens to cytotoxic T cell, while the MHC class II pathway is used to present the extracellular antigens to helper T cell. The epitope with strong immune response is good target for vaccine design. Recently, the type of H2N2 influenza has a tendency towards popular again. To understand the immunogenicity variation among the HA proteins of H2N2 influenza for the past decade is important for deciphering the infection tendency. It is also a clue to design effective vaccine in advance.
Fuzzy measure considers a series of special classes of measures and defined by a special property, respectively. The concept of fuzzy measure theory was introduced by Choquet in 1953 and independently defined by Sugeno in 1974 in the context of fuzzy integrals. The Choquet integral is a fuzzy integral based on any fuzzy measure that provides a computational scheme for information aggregation.
In this study, we proposed a novel algorithm with high accuracy in immunogenicity prediction based on fuzzy measure and SVM classifier. Accordingly, we applied this prediction method to assess the immunogenicity variation on the HA protein of H2N2 influenza virus for the past three decades. The results show that the immunogenicity strength decreased for MHC class II response that implied the increasing tendency of influenza outbreak by H2N2 virus.
