Influenza infections each year cause a major burden of sickness and death (greater than 250,000 deaths each year world wide). The most effective means of reducing illness caused by influenza is by vaccination. The influenza virus evolves by changing its surface proteins each year in response to selective pressure from the human antibody response. The World Health Organisation coordinates a massive effort to monitor the evolution of influenza around the world to try and identify the viruses that are most likely to cause infections in the coming years, and select the best candidates for manufacture into a vaccine. The methods used to do this are both new, by sequencing virus genes to monitor genetic changes, and old, by testing the antibodies that ferrets make to candidate vaccine viruses for cross-reactivity with circulating strains. Ferret antibodies are traditionally used because these animals are susceptible to influenza infections in a way similar to humans. Overall the effectiveness (reduction in risk of becoming ill with influenza) of the vaccines selected by these methods is quite poor at ~54%, so there is room for improvement. Recent technical developments have provided a new method to analyse the evolution of influenza that uses human monoclonal antibodies. It is these human antibodies that are thought to be driving the evolution of influenza in nature. We have shown that human antibodies isolated from vaccinated or infected individuals can detect changes in the virus that the ferret antibodies miss. In addition the changes that human antibodies detect frequently correlate with the changes that are appearing in the influenza viruses currently circulating in the human population. We propose to develop panels of human monoclonal antibodies that can help predict the future evolution of the virus, which will be distributed to the WHO Collaborating Centres around the world, for comparison to the standard ferret antibodies to see if this new technology can improve vaccine selection.