Revealing deep evolutionary relationships between RNA viruses using predicted structural models of viral RNA...

Area of Science:

• Virology
• Structural Biology
• Bioinformatics
• Evolutionary Biology

Background:

• The RNA-dependent RNA polymerase (RdRP) is the sole homologous gene in the kingdom Orthornavirae, crucial for inferring viral evolutionary relationships and taxonomy.
• Limited sequence similarity and alignment challenges in RNA viruses hinder deep evolutionary analyses.
• Experimental structures of viral RdRPs are scarce, restricting structure-based phylogenetic studies.

Purpose of the Study:

• To overcome limitations in experimental RdRP structures by employing protein structure prediction.
• To construct a comprehensive, structure-based phylogenetic tree for viral RdRPs.
• To evaluate current RNA virus taxonomy and explore deep evolutionary histories.

Main Methods:

• Utilized AlphaFold to predict structure models for 989 viral RdRPs.
• Employed Homologous Structure Finder for structural alignment, identifying 211 structurally equivalent residues across 96 virus genera.
• Constructed a structure-based phylogenetic tree using equivalent residues and validated it with a jackknifing approach.

Main Results:

• The generated phylogenetic trees largely support current RNA virus classifications at the class rank.
• The study's findings do not support the monophyly of the phyla Pisuviricota and Duplornaviricota.
• Flaviruses were observed to group separately from other Kitrinoviricota members in the phylogenetic analysis.

Conclusions:

• Protein structure conservation provides a robust method for inferring deep evolutionary histories of RNA viruses, even when sequence homology is lost.
• Structure-based phylogenetics offers valuable insights into viral evolution and taxonomy, complementing sequence-based methods.
• The study highlights potential revisions needed for current RNA virus classification, particularly concerning specific phyla and viral groups.