On the Analysis of Virus Phenotypes

Closely related viral species have the capacity to cause drastically different clinical symptoms in humans. The 2013 - 2016 west African Ebolavirus outbreak which resulted in the death of 11,000 people was caused by the species Zaire ebolavirus, yet the near identical Reston ebolavirus is asymptomatic in humans. Similarly the SARS-CoV-2 virus instigated widespread disruption owing to its ability to spread easily between individuals, while the SARS-CoV virus caused far fewer cases and spread much less effectively, despite the two viruses being members of the same species.

Recent advances in next generation sequencing have allowed for the collection and processing of vast quantities of biological data. Using the extensive assortment of viral genomes openly available we can identify and analyse differentially conserved positions (DCPs), residues that are highly conserved amongst a species but differ between closely related species. Using advanced modelling and structural analysis we can determine which DCPs are likely to have an impact on the differing levels of pathogenicity between species.

Here we report the VP24 Ebolavirus protein is key to pathogenicity, and that few key residue differences in the VP24/human karyopherin binding site are responsible for the lack of pathogenicity of Reston virus. Additionally we use this data to propose that the newly discovered Bombali virus is also not pathogenic in humans using sequence comparison between the pathogenic and non-pathogenic species. To further this approach we also consider the requirement to adapt filoviruses to a novel species, to determine key amino acid changes that are responsible for pathogenicity.