Forecasting the epidemic potential of influenza virus variants based on their molecular properties

Sequence analysis of the influenza haemagglutinin, HA (H1 and H3) suggests that many antigenic variants that are identified but which do not become predominant differ from contemporary epidemic strains in one or two amino acids, in the region 188-193. This information may assist in the optimum selection of vaccine strains when multiple variants are co-circulating. Genome analysis of H1N1 virus, from 1977 to 1983 (but not of H3N2 virus thus far) has identified two instances when large changes in total genome sequence was associated with major epidemic activity. The early detection of such gross genetic changes may provide a further indicator that can be used to forecast the likelihood of more widespread activity than normal.