NLM DIR Seminar Schedule

Abstract:

Metatranscriptomics as well as targeted approaches uncover more and more diverse RNA virus families, yet more to be expected. Thorough protein annotation and comparison is essential to get insights into function and evolution of the viruses and their proteins. In addition to sequence and protein profile based methods, protein structure comparison adds a powerful tool to uncover protein function and relationships. In this study, we used protein structure modeling and subsequent structure comparison searches to illuminate the remaining ‘dark matter’ in hundreds of thousands of previously discovered RNA viruses. Only a few domains and small proteins within this ‘dark matter’ could be confidently assigned a distinct fold and function. The vast majority of the domains showed either ‘generic’ folds (e.g. single alpha-helices) or no high confidence structure prediction. Thus, it appears that notwithstanding the continuing discovery of new RNA viruses by metatranscriptomics, all the protein domains shared by large groups of these viruses have already been identified. The rest of the viral proteome appears to consist of poorly structured domains including intrinsically disordered ones that likely mediate interactions between viral and host proteins. In the course of this work, a Riboviria ‘structurome’ was compiled from already annotated and initially non-annotated (‘dark matter’) proteins and domains encoded in viral genomes. Comparing structures within this ’structurome’ helps to understand protein relationship across virus families.