Patterns of amino acid substitutions on the hemagglutinin molecules of antigenic variants of H3N2 influenza A viruses.

The rapid development of molecular biology during the last two decades has accelerated the accumulation of results from genetic and experimental studies on zoonotic pathogens. Massive biological data sets enhance the chance to find important factors that are essential for zoonosis control and have not been addressed so far. At the same time, however, the dramatic increase in data volume causes difficulty in analysis by hand, or even by existing computer programs. In this regard, the larger the data sets on individual zoonotic pathogens that are accumulated in the databases, the more important role bioinformatics may well play in comprehensive studies to develop effective strategies for control of zoonoses. Here we present bioinformatics technologies developed for influenza control. Influenza A viruses are zoonotic pathogens that have been isolated from various animals, including humans. All subtypes of influenza A viruses are maintained in aquatic birds and they are known to be the source of these viruses. Hemagglutinin (HA) is the major target of antibodies that neutralize viral infectivity. HA undergoes antigenic changes with the accumulation of amino acid substitutions. The structural changes in antigenic sites of HA are responsible for the viral escape from neutralizing antibodies induced by previous infection or vaccination. Thus influenza virus strains that acquire novel antigenic structures cause annual epidemics worldwide, and it is believed that eradication of influenza is difficult to achieve. Moreover, the high rate of mutation in the HA gene causes substantial difficulties to select an effective vaccine strain prior to each influenza season. To predict future antigenic changes, it is important to understand the evolution of the virus associated with antigenic changes caused by amino acid substitutions in the past. We introduce bioinformatics technologies developed at the author’s laboratory, aiming to predict the future evolution of influenza viruses. The techniques include sequence data analysis employing information theory to find patterns of evolution of viruses, and molecular modeling using homology modeling and molecular dynamics simulation of viral proteins. Through these technologies, we investigate the past, current and future evolution of influenza viruses.