Discovery of Novel Potential Binders Targeting Norovirus 3CL Protease: A Machine Learning and Molecular Dynamics Approach

Norovirus, a highly transmissible virus affecting all age groups, frequently causes outbreaks in communal settings such as hospitals, schools and cruise ships. The 3CL protease, essential for viral replication, has become a primary target for antiviral drug development. This study used computational methods to identify potential inhibitors of the norovirus 3CL protease. A machine learning-based QSAR model screened 9699 bioactive compounds, selecting 431 for molecular docking. Tanimoto similarity and clustering techniques further narrowed the selection to three compounds: 10147259, 226371 and 132427679. Molecular dynamics simulations (300 ns) revealed that compound 132427679 demonstrated robust binding affinity, structural stability and consistent interactions with the 3CL protease. Root mean square deviation (RMSD), radius of gyration (Rg) and solvent-accessible surface area (SASA) analyses confirmed its stability, while principal component analysis (PCA) and free energy landscape (FEL) studies highlighted its flexibility and broad conformational exploration. Binding free energy calculations showed that 132427679(−26.15 kcal/mol) had a binding affinity comparable to the control ligand (−27.78 kcal/mol). The findings identify 132427679 as a promising candidate for antiviral development due to its stability, adaptive binding and favorable energy profile. Further experimental validation through in vitro and in vivo studies is recommended to confirm its efficacy and safety.