Characterizing Host-Viral Interaction Interfaces for Insights into Viral Transmission and Infections 2257274

Abstract Introduction Viruses hijack human cellular machinery through protein—protein interactions (PPIs). Since interaction interfaces play a critical role in modulating PPIs, studying human-viral interaction interfaces can provide a novel perspective on viral immunomodulatory mechanisms underlying disease pathogenesis. Methods We used AlphaFold-Multimer (AFM) to structurally resolve 11,667 experimentally validated human—viral PPIs spanning 33 viral families to create comprehensive 3D interactomes. Despite the absence of paired multiple-sequence alignments, AFM outperformed other methods in accurately identifying interaction interfaces. We then systematically analyzed these 3D interaction models to reveal signatures of selection, conservation, and molecular mimicry at human—viral interfaces. Results Using 1000 Genomes data, we found that variants at interfaces exhibited population-specific selection patterns, particularly for long-coevolving viruses such as HIV. Interestingly, when the evolutionary patterns of interface and non-interface residues were compared, interface residues at human PPIs were more conserved, as expected. However, this pattern was reversed for viral residues. Systematic analyses of human interface mimicry by viral proteins revealed mutation rate-dependent differences, with higher viral mutation rates being associated with greater mimicry. Similar mutation rate-dependent differences were seen in the interface sharing between viral proteins binding to a host protein. Moreover, we identified a novel example of herpes simplex virus protein UL37 mimicking the TRAF6-binding domain on MAVS, which potentially inhibits interferon signaling by disrupting MAVS—TRAF6 interactions. Conclusion Overall, our comprehensive 3D viral interactomes provide a rich resource for exploring viral adaptation, immune evasion, and pathogenesis mechanisms, empowering researchers to generate testable hypotheses and ultimately accelerate the discovery of novel therapeutic targets and intervention strategies. Funding Source n/a Topic Categories Computational and Systems Immunology (COMP)