Ph.D. Prelim: Fathi Mubaraki

Fathi Mubaraki
Monday, November 11, 2019 - 12:00pm to 1:00pm
223 Atanasoff Hall
Event Type: 

PhyloVirus: Inferring virus reassortment and recombination, and a visualization tool for phylogenetic networks.

Objective: RNA virus diversity is the result of mutation, recombination, and reassortment. Traditional virus evolution studies rely on single-gene phylogenetic trees that do not account for all these processes. Phylogenetic network algorithms can do so but are hampered by computational limitations. We introduce an efficient software package with a graphical user interface called PhyloVirus that constructs and allows visualization of phylogenetic trees, median trees, and networks. We apply our software to influenza A virus (IAV), quantifying reassortment events in the evolution of H3N2 viruses and identify novel reassorted viruses.

Methods: We developed PhyloVirus, a multi-platform Java application which accepts a set of Newick-format gene trees and constructs a species tree using median tree methods (e.g., Strict Consensus or Deep Coalescence) or a phylogenetic network using our Robinson-Foulds Network (RF-Net) algorithm. Phylogenetic networks are visualized by displaying a phylogeny string in the extended Newick format, and drawing the network with k reticulation events. We apply RF-Net to IAV H3N2 dataset, estimating a reassortment network with over 500 strains.

Results: PhyloVirus can detect and visualize recombination, lateral gene transfer, and novel reassortment events. Our software integrates median tree phylogenetic methods and visualizations. PhyloVirus can highlight and color label taxa, rotate and reroot trees, and display gene trees and networks concurrently. Clades can be tracked between gene trees and the network, allowing identification of gene trees and clades with robust relationships and genes involved in reassortment.

Conclusion: We developed software for inferring and visualizing evolutionary trees and networks, and applied it to IAV H3N2; we recapitulated known reassortment events, detected novel events that warrant further characterization, and provide a deeper understanding of the evolutionary history of the virus. PhyloVirus expands the capabilities of prior network software and may also be applied to other viruses and bacteria.

Committee: Oliver Eulenstein (major professor), Tavis Anderson, Xiaoqiu Huang, Leslie Miller and Phillip Gauger