The Effects of Sequence Variations in the Structural Dynamics and Ligand Interactions of Main Proteases in Betacoronaviruses

Abstract

The main protease (Mpro) of betacoronaviruses is an essential enzyme for viral replication and a premier target for antiviral drug discovery. The high conservation of its active site across the genus makes it an ideal candidate for developing pan-coronavirus therapeutics to combat future outbreaks. In response to the Coronavirus disease 2019 (COVID-19) pandemic, numerous large-scale screening campaigns were initiated to identify Mpro inhibitors. However, these efforts were frequently hindered by the fundamental pose classification problem, which is the inability to distinguish between correct and incorrect binding poses. Compounding this issue, the research response generated an unprecedented volume of Mpro structural data, yet existing bioinformatics platforms lack the integrated tools for its systematic, comparative analysis. These combined challenges hinder the rational design of next-generation inhibitors. In this work, we address these challenges by developing a novel, integrated computational toolkit. We first present CoviProDigy, a web-based platform designed for the comprehensive and comparative analysis of Mpro-ligand interactions, featuring specialized tools for subpocket occupancy analysis and elucidating common and unique interactions across different ligands. This functionality is designed to support scaffold hopping and medicinal chemistry optimization based on the molecular-level insights gained from known structures. To overcome the major issue of inaccurate pose selection in virtual screening, we then developed and validated a fine-tuned machine learning model that demonstrates enhanced accuracy in classifying ligand binding poses for the Betacoronavirus Mpro family. Together, these contributions provide a robust, end-to-end framework that accelerates the discovery of potent inhibitors, presenting a valuable resource for pandemic preparedness and the ongoing search for broad-spectrum antiviral agents.