Computational Analysis of Influenza A Virus Epidemics in the United States

Abstract

The influenza virus experiences strong evolutionary pressure from the human adaptive immune system. It is hypothesized that the accumulation of mutations on antigenic proteins, also known as antigenic drift, allows influenza to escape host immune responses due to poor antibody cross-reactivity. The continuous evolution of influenza strains, in conjunction with environmental and social factors, is thought the be the primary driver of seasonal influenza epidemics. However, the direct relationship between influenza mutation rate and seasonal flu severity is not well characterized. Here, I use seasonal influenza A virus (IAV) epidemic data from 2010–2023 to develop a five-metric system of scoring epidemic severity. Then, I approximated the seasonal mutation rate of IAV proteins using the extraction of branch lengths from Bayesian phylogenetic tree reconstruction. By calculating the Pearson correlation coefficient between each protein’s branch lengths and the seasonal severity score, I found no significant relationship between mutation quantity and IAV epidemic severity. My work presents insight into the complex patterns of seasonal influenza epidemics while using new perspectives to interpret phylogenetic trees and investigate genome-level changes in IAV proteins.