Automated recommendation of the seasonal influenza vaccine strain with PREDAC

Abstract Vaccination is currently the most effective way to protect against influenza viruses. This study presents PREDAC (PREDict Antigenic Cluster), a sequence-based computational method to model antigenic clusters of influenza viruses. PREDAC could improve vaccine recommendations for influenza viruses and has been used to support the influenza vaccine strain recommendations at the Chinese National Influenza Center. Moreover, a PREDAC server which including tools for the prediction of antigenic variants and antigenic clusters of both influenza A and B viruses was built and made publicly available at http://www.computationalbiology.cn/home/ . PREDAC could not only facilitate systematic studies of the antigenic evolution of influenza viruses but also provide user-friendly tools for vaccine strain recommendations for influenza viruses.