A new computational method that predicts antigenic clusters of the influenza A type virus H3N2 is reported this week in Nature Communications. These findings may aid in the identification of viral strains for use in vaccine production. Influenza vaccines need to be updated every 2-5 years to keep pace with the circulating viral strains that continually change their features. Previous methods have analysed the evolutionary trees of the haemagglutinin antigen gene, which is found in influenza strains. Taijiao Jiang and colleagues sequenced the haemagglutinin antigen gene in more than a 1000 strains of H3N2 that had been isolated from mainland China from 1968 to 2009. They then use a new computational method that analyses 12 structural and physiochemical features of the haemagglutinin antigen to model the antigenic clusters of the strains. Their findings allow them to follow the evolution of H3N2 in China over this period and predict strains that may be useful in future vaccination programmes.
Zoology: Mineral armour discovered in insectsNature Communications
Neuroscience: Social isolation evokes craving responses in the human brainNature Neuroscience
Ecology: Migration associated with faster pace of lifeNature Communications
Gene therapy: Concerns for the long-term safety of AAV gene therapyNature Biotechnology