Machine-Learning\nApproach to Identify Potential Dengue\nVirus Protease Inhibitors: A Computational Perspective

The global prevalence of dengue virus\n(DENV), a widespread flavivirus,\nhas led to varied epidemiological impacts, economic burdens, and health\nconsequences. The alarming increase in infections is exacerbated by\nthe absence of approved antiviral agents against the DENV. Within\nflaviviruses, the NS3/NS2B serine protease plays a pivotal role in\nprocessing the viral polyprotein into distinct components, making\nit an attractive target for antiviral drug development. In this study,\nmachine-learning (ML) techniques were employed to build predictive\nmodels for the screening of a library containing 32,000 protease inhibitors.\nUtilizing GNINA for structure-based virtual screening, the top potential\ncandidates underwent a subsequent evaluation of their absorption,\ndistribution, metabolism, excretion, and toxicity properties. Selected\ncompounds were subjected to molecular dynamics simulations and binding\nfree energy calculations via MM/GBSA. The results suggest that comp530\npossesses binding potential to DENV protease as a noncovalent inhibitor\nwith multiple positions for chemical substitutions, presenting opportunities\nfor optimizing their selectivity and specificity. However, other compounds\npredicted via ML models may still provide a promising start for covalent\ninhibitors.