Date

2022-10

Author

Küçükyılmaz, Cansel Hatice
Aksoy , Selahattin
Ibrahim , Ahmed Hassan
Karabulut, Onur Can
Karpuzcu, Betül Asiye
Türk , Erdem
Süzek, Barış Ethem

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

100
views

28
downloads

Cite This

Viruses are a major cause of infectious diseases, and they lead to a substantial burden on public health and the global economy. Recently emerged outbreaks such as COVID-19 and monkeypox further emphasize the need for bioinfor- matics resources supporting research on viral diseases. Es- timation of infection capacity for viruses, identifying host shifts, and more importantly predicting virus-host protein- protein interactions (PPIs) that are a prerequisite to viral infections are main topics of interest in virus bioinformatics research. Yet, there is still a shortage of data resources to support such research when compared to alternative re- search areas in bioinformatics (e.g., cancer bioinformatics). To this end, we developed a web resource for virus bioin- formatics, namely VirBioHub. VirBioHub provides a catalog of viruses and their respec- tive hosts, selected virus-host PPI prediction tools, and a pre-computed set of viral infection predictions for various viral families and hosts. In the catalog of viruses and hosts component, for each pair of viruses and their known host - which corresponds to a known infection-, we provide a da- taset of manually curated viral attachment protein(s) and host receptor protein(s) together with their taxonomic data. Currently, VirBioHub has the interacting proteins for virus- host pairs from two viral families (89 pairs for Coronaviri- dae and 80 pairs for Adenoviridae). Virus-host PPI predic- tion component enables users to run publicly available PPI prediction tools (currently HVPPI and HOPITOR) using their own choice of virus and host protein sequences. The viral infection prediction component is designed to provide host and host shift predictions based on a previously devel- oped infection prediction pipeline originally developed for adenoviruses, namely ML-AdvInfect. Currently, VirBio- Hub has the precomputed infection prediction results for two viral families (Coronaviridae and Adenoviridae) along with the PPI data supporting these infection predictions. VirBioHub supports virus bioinformatics research by al- lowing researchers to export all its query results and da- tasets. VirBioHub is developed using React.js for the front-end, Redux Toolkit/PostgreSQL for state & data management, and Django REST for the back-end. VirBioHub acts as a central resource, providing ease of access to virus bioinformatics researchers. In the future, we plan to extend VirBioHub’s coverage in terms of viral fami- lies and PPI prediction methods. Furthermore, we intend to improve the accuracy of infection and host-shift predictions as new PPI prediction tools and data become available.

URI

https://hdl.handle.net/11511/102005

Collections

Graduate School of Informatics, Conference / Seminar