Improving influenza A vaccine strain selection through deep evolutionary models
Shi, W.; Wohlwend, J.; Wu, M.; Barzilay, R.
Show abstract
Current vaccines provide limited protection against rapidly evolving viruses. For example, the flu vaccines effectiveness has averaged below 40% for the past five years. Today, clinical outcomes of vaccine effectiveness can only be assessed retrospectively. Since vaccine strains are selected at least six months ahead of flu season, prospective estimation of their effectiveness is crucial but remains under-explored. In this paper, we propose an in-silico method named VaxSeer that selects vaccine strains based on their coverage scores, which quantifies expected vaccine effectiveness in future seasons. This score considers both the future dominance of circulating viruses and antigenic profiles of vaccine candidates. Based on historical WHO data, our approach consistently selects superior strains than the annual recommendations. Finally, the prospective coverage score exhibits a strong correlation with retrospective vaccine effectiveness and reduced disease burden, highlighting the promise of this framework in driving the vaccine selection process.
Matching journals
The top 5 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Supervised fine-tuning of pre-trained antibody language models improves antigen specificity prediction 96%
- Paraplume: A fast and accurate paratope prediction method provides insights into repertoire-scale binding dynamics 96%
- Graph-Theoretical Formulation of the Generalized Epitope-based Vaccine Design Problem 95%
Similar papers in this journal
Similar papers in this journal
- The role of viral interference in shaping RSV epidemics following the 2009 H1N1 influenza pandemic 93%
- Measures of population immunity can predict the dominant clade of influenza A (H3N2) and reveal age-associated differences in susceptibility and specificity 92%
- Improved resolution of influenza vaccination responses with high-throughput live virus microneutralisation 91%
Similar papers in this journal
- Multi-strain modeling of influenza vaccine effectiveness in older adults and its dependence on antigenic distance 96%
- Language models learn to represent antigenic properties of human influenza A(H3) virus 94%
- Screening for safe opening of universities under Omicron and Delta variants of COVID-19: When less is more 94%
Similar papers in this journal
- PHIStruct: Improving phage-host interaction prediction at low sequence similarity settings using structure-aware protein embeddings 93%
- ProteinBERT: A universal deep-learning model of protein sequence and function 93%
- Topology-Driven Negative Sampling Enhances Generalizability in Protein-Protein Interaction Prediction 93%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.