Predictive tools
Convene a group of stakeholders to determine how best to use new tools, such as computational approaches, machine learning, and systems biology, to enhance understanding of influenza virus evolution and to improve capabilities to predict circulating influenza virus strains, including emergence of novel viruses.
Progress Highlights
(Note: Although stakeholders have not been convened, recent research provides insight on use of the new tools outlined in the milestone; therefore, this milestone is categorized as in progress.)
Huddleston 2024 quantified the effects of reducing forecast horizons and submission lags on the accuracy of forecasts for A/H3N2 populations to determine whether technologies for more rapid vaccine development could improve long-term forecasts for seasonal influenza. Results showed the potential to improve the accuracy of existing influenza forecasting models by using new influenza vaccine platforms such as mRNA and increasing global sequencing capacity.
Kim 2024 used a cross-sectional antibody-derived estimate of population susceptibility to different clades of influenza A (H3N2) could predict the success of clades in the following season and found that representative measures of population immunity might improve evolutionary forecasts and inform selective pressures on influenza.
Parino 2024 developed a multiscale epidemiological model calibrated on worldwide genetic data through phylogeographic inference, simulating migration fluxes between epidemics occurring in different countries and identifying model parameterizations to predict global influenza virus circulation.
Welsh 2024 used use deep mutational scanning to map how mutations to HA proteins of two H3N2 strains affect neutralization by serum from individuals of a variety of ages, to investigate how antigenic effects of viral mutations vary across the human population and how this heterogeneity affects virus evolution.