Abstract
Background: Patients with COVID-19 often have an impact on populations with underlying co-infection. METTL3 plays a crucial role in numerous infectious diseases, making the research on broad-spectrum anti-infective drugs targeting METTL3 both captivating and compelling. Objective: The study aims to identify anti-infection potential compounds targeting METTL3 using a comprehensive virtual screening approach. Method: The approach involves ensemble docking and molecular dynamics simulations to predict and validate the potential of natural compounds. This was complemented by the application of deep learning, specifically CNN binary classification and regression models, to predict the anti-infection capabilities of the compounds identified through docking. Experimental validation through Surface Plasmon Resonance (SPR) further confirmed the computational predictions. Network and gene ontology analyses provided insights into the compounds' biological targets and pathways. Results: Ensemble docking approach achieved high prediction accuracy. We identified 17 natural compounds as potential METTL3 inhibitors, with Rosavin and Eriocitrin showing strong antiinfection potential against HIV-1, SARS-CoV-2, Mycobacterium tuberculosis, and Influenza A. Rosavin's low toxicity and strong METTL3 binding affinity were confirmed by SPR experiments and MD simulations. Network and gene ontology analyses suggest Rosavin may enhance immune responses by disrupting interferon degradation. Conclusion: The multidimensional analysis identified Rosavin as a potent METTL3 inhibitor with significant anti-COVID-19 co-infection diseases potential.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call