Models of RNA virus evolution and their roles in vaccine design

Ojosnegros, Samuel; Beerenwinkel, Niko

Phylogenetic methods play an important role in the analysis of viral sequence data. They are used to reconstruct the evolutionary relationships between different viral strains. Three main approaches exist to phylogenetic tree reconstruction: distance-based methods, parsimony, and probabilistic methods based on maximum likelihood or Bayesian statistics [77] [78] [79] [80] . Distance-based methods start by defining an evolutionary distance between sequences and then apply hierarchical clustering algorithms to obtain a phylogenetic tree. Unweighted Pair Group Method with Arithmetic mean (UPGMA) [81] , an average linkage clustering method, and Neighbor Joining [82] are two prominent examples of this approach. Distance-based methods are computationally efficient, but the reduction of the observed data to a distance matrix presents a loss of information. Both UPGMA and Neighbor Joining can reconstruct the correct tree only under strong assumptions about the metric defined by the distance matrix and they are sensitive to violations of these assumptions.