VirHostNet 2.0: surfing on the web of virus/host molecular interactions data

Abstract

VirHostNet release 2.0 (http://virhostnet.prabi.fr) is a knowledgebase dedicated to the network-based exploration of virus-host protein-protein interactions. Since the previous VirhostNet release (2009), a second run of manual curation was performed to annotate the new torrent of high-throughput protein-protein interactions data from the literature. This resource is shared publicly, in PSI-MI TAB 2.5 format, using a PSICQUIC web service. The new interface of VirHostNet 2.0 is based on Cytoscape web library and provides a user-friendly access to the most complete and accurate resource of virus-virus and virus-host protein-protein interactions as well as their projection onto their corresponding host cell protein interaction networks. We hope that the VirHostNet 2.0 system will facilitate systems biology and gene-centered analysis of infectious diseases and will help to identify new molecular targets for antiviral drugs design. This resource will also continue to help worldwide scientists to improve our knowledge on molecular mechanisms involved in the antiviral response mediated by the cell and in the viral strategies selected by viruses to hijack the host immune system.

Figure 1.

VirHostNet 2.0 annotations statistics. (a) Growth statistics. The number of non-redundant (i.e. unique) VH ppi annotated between 2009 and 2014 is given for VirHostNet and VirusMint. This figure highlights a 5-fold enrichment of the VirHostNet 2.0 annotation between 2009 and 2014. (b) Experiments summary. The pie chart represents the proportion of annotated experimental methods used to characterize VH and VV ppi (other—other methods). (c) Independent experiments support. This pie chart represents the proportion of VH and VV non-redundant ppi supported by one or more than two independent experiments.

Figure 2.

A VirHostNet 2.0 web interface short tour. The purpose of our case study is to visualize molecular cross-talks between two important pathways involved in the innate immune response of the host cell. The result of such complex question is obtained step-by-step in only four consecutive queries. Step S1. From the ‘home’ page (a), enter the sentence ‘Regulation of autophagy’ in the ‘pathway’ quick-search area (b). Alternatively enter a personalized set of proteins by using the multiple proteins search (c). After few seconds of search, the loader image disappears (d) and ppi (blue edges) between the set of human proteins (blue nodes)—annotated in KEGG database as participating to autophagy function—are represented as a graph (e). A summary table (f) provides the list of proteins and the list of non-redundant and redundant ppi corresponding to the graph or its sub-selection. Step S2. After selecting all the proteins, add viral partners (red nodes) of the autophagy-related proteins by using the button panel at the top of the Cytoscape web page (g). Step S3. Then add a pin representation of the jak-STAT pathway by entering ‘Jak-STAT signaling pathway’ in the quick-search area (h). Step S4. Finally, search for molecular interactions (i.e. the putative molecular cross talks) between the autophagy, the jak-STAT signaling pathway and the set of viral proteins, by using the ‘add interactions’ button in the panel at the top of the Cytoscape page (g).