Design and characterization of broadly protective influenza A(H3N2) vaccine candidates using protein language models
Seasonal influenza A viruses cause significant global morbidity each year. Although vaccination remains the primary preventive strategy, effectiveness is often reduced by antigenic drift. This challenge is particularly pronounced for influenza A(H3N2), which has required eight vaccine updates over the past decade. Here, we present a computational framework to engineer broadly reactive influenza…
Influenza A(H3N2) viruses cause substantial health issues worldwide each year, often resulting in seasonal illness. Vaccines are the key preventive measure, but their effectiveness wanes due to antigenic drift. This issue is especially severe for influenza A(H3N2), which has necessitated eight vaccine updates within the last decade.
Researchers have now developed a computational framework aimed at creating broadly protective influenza A(H3N2) vaccines. This framework utilizes protein language models to generate new hemagglutinin (HA) sequences and a machine learning model to gauge antigenic distance from current strains. In a preliminary study, seven HA candidates generated using data from 2013-2018 were tested in mice against both contemporary and future circulating viruses.
Two of these candidates produced protective levels of reactive antibodies, strong H3-specific antibody-secreting cell responses, and cross-neutralization against contemporary clades and drifted 2019-2020 strains. These results indicate that an integrated generation-selection strategy can improve vaccine coverage for both present and future A(H3N2) seasons. This approach may also have implications for designing vaccines against other influenza subtypes.
Written by urgent.news from bioRxiv's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
