Predicting Future Mutations To Inform Vaccine Design

Abstract Influenza viruses constantly evolve, and mismatches between predicted and circulating strains impact vaccine effectiveness. A barrier to predicting the season-specific dominant strains is the limited ability to predict future mutations, or estimate the numerical likelihood of specific future strains. Here, we introduce a biology-aware sequence similarity metric based on deep pattern recognition of evolutionary constraints, that calculate the odds of future mutations, outperforming WHO recommended flu vaccine compositions almost consistently over the two decades.