|Host institution:||Swiss Federal Institute of Technology Zurich (ETH), Switzerland|
|Local supervisor:||Prof. Dr. Niko Beerenwinkel (ETH Zurich, Computational Biology Group)|
|Local co-supervisor:||Prof. Dr. Karin Metzner (University Hospital Zurich, Dept. of Infectious Diseases and Hospital Epidemiology)|
|Project partner:||ESR 4, ESR 9|
Viruses exist in their hosts as populations of genetically heterogeneous particles often referred to as viral quasispecies. Intra- host genomic diversity can result in phenotypic heterogeneity, and it has been linked to viral pathogenesis and virulence. NGS can be used to assess the genomic heterogeneity of virus populations in a cost-efficient manner, but this approach is challenging due to sequencing errors and short read length. In this work package, we will develop improved tools for the reconstruction of viral genomic diversity from both short-read (Illumina) and long-read (PacBio, MinION) NGS data, and we will leverage the power of the inferred population structure to inform models of virus micro-evolution, to detect selection, and to correlate population features to viral and host phenotypes.
Specifically, the objectives are:
- Reconstruction of viral haplotypes (full-length strain) from NGS data (BC) while removing sequencing errors, including application to Drosophila C virus (ESR 9) and Deformed wing virus (ESR 4) quasispecies data.
- Extension by fitting parametric representations of virus populations (e.g. mutation-selection balance).
- Detection of patterns of positive or negative selection after viral passaging.
- Investigation of the correlation between viral quasispecies and (viral and host) phenotypes such as viral load, host immune response and survival, and specific readouts of virus-induced pathology.