Machine learning prediction and experimental validation of antigenic drift in H3 influenza A viruses in swine
The antigenic diversity of influenza A virus (IAV) circulating in swine challenges the development of effective vaccines, increasing zoonotic threat and pandemic potential. High throughput sequencing technologies are able to quantify IAV genetic diversity, but there are no accurate approaches to adequately describe antigenic phenotypes. This study evaluated an ensemble of non-linear regression models to estimate virus phenotype from genotype. Regression models were trained with a phenotypic dataset of pairwise hemagglutination inhibition (HI) assays, using genetic sequence identity and pairwise amino acid mutations as predictor features. The model identified amino acid identity, ranked the relative importance of mutations in the hemagglutinin (HA) protein, and demonstrated good prediction accuracy. Four previously untested IAV strains were selected to experimentally validate model predictions by HI assays. Error between predicted and measured distances of uncharacterized strains were 0.34, 0.70, 2.19, and 0.17 antigenic units. These empirically trained regression models can be used to estimate antigenic distances between different strains of IAV in swine using sequence data. By ranking the importance of mutations in the HA, we provide criteria for identifying antigenically advanced IAV strains that may not be controlled by existing vaccines and can inform strain updates to vaccines to better control this pathogen.