The ability of single genes vs full genomes to resolve time and space in outbreak analysis
Gytis Dudas1,2, Trevor Bedford1
1 Fred Hutchinson Cancer Research Center, Seattle, Washington, United States.
2 Gothenburg Global Biodiversity Centre, Gothenburg, Sweden.
Inexpensive pathogen genome sequencing has had a transformative effect on the field of phylodynamics, where ever increasing volumes of data have promised real-time insight into outbreaks of infectious disease. As well as the sheer volume of pathogen isolates being sequenced, the sequencing of whole pathogen genomes, rather than select loci, has allowed phylogenetic analyses to be carried out at finer time scales, often approaching serial intervals for infections caused by rapidly evolving RNA viruses. Despite its utility, whole genome sequencing of pathogens has not been adopted universally and targeted sequencing of loci is common in some pathogen-specific fields.
In this study we aim to highlight the utility of sequencing whole genomes of pathogens by re-analysing a well-characterised collection of Ebola virus sequences in the form of complete viral genomes (~19kb long) or the rapidly evolving glycoprotein (GP, ~2kb long) gene. We quantify changes in phylogenetic, temporal, and spatial inference resolution as a result of this reduction in data and compare these to theoretical expectations. We propose a simple intuitive metric for quantifying temporal resolution, i.e. the time scale over which sequence data might be informative of various processes as a quick back-of-the-envelope calculation of statistical power available to molecular clock analyses.