Machine learning-driven docking of diverse DDAs as promising cysteine protease inhibitors targeting Mpox virus.

Abstract

The rise of zoonotic viruses like Monkeypox (mpox) presents significant challenges to public health, the economy, and modern medical practices. These pathogens, which can transfer from animals to humans, have the potential to cause both localized outbreaks and global pandemics. Monkeypox, recently recognized as a zoonotic virus, is particularly concerning due to its severe impact, especially on children and those with weakened immune systems. In light of the pressing need for effective treatments, repurposing existing drugs and utilizing computational modeling have emerged as vital strategies for discovering potential therapeutic agents. Research has demonstrated the promise of Direct Acting Antivirals (DAAs) against various viral infections. By employing computational tools and existing data, we can quickly identify potential treatments to combat the current mpox outbreak. Given that the cysteine protease of mpox bears similarities to proteases found in viruses such as HCV and HIV, it is plausible that DAAs could inhibit mpox protease. We applied machine learning techniques, including Support Vector Machines (SVM), Reinforcement Learning (RL), and K-Nearest Neighbors (KNN), to analyze a set of 86 DAAs. The compounds predicted to be effective inhibitors were then assessed using structural modeling methods. Our docking simulations identified four DAAs-Paritaprevir (DB09297), Ledipasvir (DB09027), Lenacapavir (DB15673), and Bictegravir (DB11799)-as having particularly strong binding affinities for mpox protease. Key interacting residues, such as Cys328, Tyr270, His241, and Gly329, were found to be critical in the binding process. These results indicate that FDA-approved DAAs might provide new treatment avenues for mpox. Nevertheless, additional validation through experimental studies is necessary to confirm the biological effectiveness of these drug candidates. This research provides a foundational basis for exploring DAAs as potential new treatments for mpox, with future investigations required to fully determine their therapeutic value.

Supplementary information: The online version contains supplementary material available at 10.1007/s40203-025-00374-w.