Stochastic Processes Are Key Determinants of Short-Term Evolution in Influenza A Virus

Abstract

Understanding the evolutionary dynamics of influenza A virus is central to its surveillance and control. While immune-driven antigenic drift is a key determinant of viral evolution across epidemic seasons, the evolutionary processes shaping influenza virus diversity within seasons are less clear. Here we show with a phylogenetic analysis of 413 complete genomes of human H3N2 influenza A viruses collected between 1997 and 2005 from New York State, United States, that genetic diversity is both abundant and largely generated through the seasonal importation of multiple divergent clades of the same subtype. These clades cocirculated within New York State, allowing frequent reassortment and generating genome-wide diversity. However, relatively low levels of positive selection and genetic diversity were observed at amino acid sites considered important in antigenic drift. These results indicate that adaptive evolution occurs only sporadically in influenza A virus; rather, the stochastic processes of viral migration and clade reassortment play a vital role in shaping short-term evolutionary dynamics. Thus, predicting future patterns of influenza virus evolution for vaccine strain selection is inherently complex and requires intensive surveillance, whole-genome sequencing, and phenotypic analysis. A comparative analysis of 413 complete genomes of the H3N2 subtype of human influenza A virus sampled from New York State, United States, the largest and first population-based study of its kind, reveals that viral evolution within epidemic seasons is dominated by the random importation of genetically different viral strains from other geographic areas, rather than by natural selection favoring local strains able to escape the human immune response. Multiple clades of genetically distinct viral strains cocirculate across the entire state within any season and occasionally exchange genes through reassortment. Both genetic diversity and geographic viral “traffic” are extensive within epidemics. Therefore, the evolution of influenza A virus is strongly shaped by random migration and reassortment and is far harder to predict than previously realized. Consequently, intensive sampling and whole-genome sequencing are essential for selecting viral strains for future vaccine production.