Viral genomics and pangenome graphs

How do we represent viral diversity without losing the variation that matters?

Graph genomes have changed human genetics, but their tools do not transfer directly to viruses, where within-host diversity, recombination and incomplete references are far greater. Our hepatitis B sequence variation graph improves alignment and sample-specific consensus construction (PLOS ONE, 2024), and we are extending graph-based pangenome and quasispecies methods to hepatitis B and HIV for cure-directed research.

In parallel, we study the human plasma virome, the community of viruses circulating in blood, as a sensitive bioindicator of HIV risk. Rolling-circle amplification and nanopore sequencing read circular DNA viruses such as anelloviruses end to end, and we have characterized the read-level errors this approach introduces (Viruses, 2026).