Bioinformatics and statistical methods to study the evolution of primary HCV infection

Download files
Access & Terms of Use
open access
Copyright: Leung, Preston
Altmetric
Abstract
HCV is a rapidly mutating RNA virus, which affects 170 million people globally. The genome mutates rapidly to generate many variants (quasispecies) in every infected individual. Yet, only a few variants are successfully transmitted to infect a new host. Host immune responses are known to drive the evolution of these highly diverse viral populations. However, the patterns of co-evolving mutations on each variant within the quasispecies remain largely unknown. By employing next generation sequencing and bioinformatics analyses on longitudinal samples from 16 subjects from early acute phase of HCV infection, only one or very few transmitted/founder (T/F) viruses were shown to successfully initiate infection, and then to rapidly evolve via genomic mutations in a non-uniform distribution across the genome. By combining genomic and experimental data from HCV-specific CD8+ T-cell responses, this analysis showed that T/F viral evolution was consistent with strong immune pressures driving the onset of amino acid substitutions (fixation events) in the quasispecies of subjects who ultimately became chronically infected. By quantifying fitness of the viral populations through mathematical modelling, for the first time a direct estimate of viral fitness at the quasispecies level was determined. Viral fitness generally declined over the first 3 months of infection, with evidence of specific combinations of co-occurring mutations that conferred increased viral fitness over time, thus suggesting positive epistatic effects. Finally, to better understand the role of these co-evolving mutations in HCV evolution of virus, a network approach was developed and applied to 406 full-length HCV genotype 1a sequences. This analysis showed combinations (clusters of nodes) of genomic mutations that were unique to the acute, or to the chronic, phase of infection was identified, and others were shared between the two phases. Notably, node centrality analyses revealed that critical sites to the structural stability of the network of mutations were nodes shared between acute and chronic subjects and were likely targets of T-cells and of broadly neutralising antibodies. The bioinformatics approaches developed in this thesis to study HCV have demonstrated the importance of co-occurring mutations over the course of HCV infection. While immune escape mutations evade host immune pressure, co-occurring mutations compensate deleterious effects, thus revealing the complex strategies involved in HCV evolution. These tools can also be applied to study other viral escape dynamics in terms of immune response and potentially provide better understanding of mutational networks which can inform HCV vaccine design as well as epidemiological dynamics of other viral infections.
Persistent link to this record
Link to Publisher Version
Link to Open Access Version
Additional Link
Author(s)
Leung, Preston
Supervisor(s)
Luciani, Fabio
Bull, Rowena
Lloyd, Andrew
Creator(s)
Editor(s)
Translator(s)
Curator(s)
Designer(s)
Arranger(s)
Composer(s)
Recordist(s)
Conference Proceedings Editor(s)
Other Contributor(s)
Corporate/Industry Contributor(s)
Publication Year
2018
Resource Type
Thesis
Degree Type
PhD Doctorate
UNSW Faculty
Files
download public version.pdf 7.96 MB Adobe Portable Document Format
Related dataset(s)