We will try to develop a surveillance-based approach for the selection of swine influenza vaccine strain candidates. Our approach will consider epidemiological, antigenic and genetic data. Using large numbers of hemagglutinin sequences from databases, we will try to identify the most common viral variants, as well as the rarest variants. We will perform vaccination experiments in pigs with selected swine influenza virus strains and perform detailed investigations of the immune response. From these experiments, we will learn whether it is possible to identify strains that best reflect the overall diversity in the HA, and whether such strains should be prioritized as vaccine strains.