Computational Identification of Mitochondrial Binding Sites for General Anesthetics

Overview

Drug discovery relies on an accurate assessment of how a compound binds to a target and how that may influence biological function. The recent revolution in AI has now significantly expedited modeling those interactions. Notably, the Nobel Prize in 2024 was awarded to AlphaFold of Google DeepMind for its landmark improvements in predicting protein geometry-one of the main components for calculating binding affinity. In this study, we aim to assess the capabilities and limitations of these AI-based prediction methodologies and explore how they can be combined with more traditional geometry-prediction techniques. More specifically, we are focusing on testing the following: the robustness of predictions to perturbations; the ability of predictions to model changes in protein conformation; and the advantages of the predictions when combined with molecular docking techniques to screen for binding sites. Our goal is to use these AI tools to further understand binding interactions, identify drug targets, and enhance molecular-docking strategies. This study may ultimately facilitate successful binding predictions and improve the screening of molecular targets for new or existing drugs for more effective therapies.