Targeting Bacterial RNA Polymerase: Harnessing Simulations and Machine Learning to Design Inhibitors for Drug-Resistant Pathogens

The increase in antimicrobial resistance presents a major challenge in treating bacterial infections, underscoring the need for innovative drug discovery approaches and novel inhibitors. Bacterial RNA polymerase (RNAP) has emerged as a crucial target for antibiotic development due to its essential role in transcription. RNAP is a molecular motor, and its function relies heavily on the dynamic shifts between multiple conformational states. While biochemical and structural experimental methods offer crucial insights into static RNAP-drug interactions, they fall short in capturing the dynamics at a molecular level. By integrating experimental data with advanced computational techniques like Markov State Models (MSMs), Generalized Master Equation (GME) Models and other machine-learning models constructed from molecular dynamics (MD) simulations, researchers can elucidate novel cryptic pockets that open transiently for antibiotic compounds and gain a more nuanced and comprehensive understanding of RNAP-drug interactions. This integrated approach not only deepens our fundamental knowledge but also enables more targeted and efficient antibiotic design strategies. In this Perspective, we highlight how this synergy between experimental and computational methods has the potential to open new pathways for innovative drug design and combination therapies that may help turn the tide in the ongoing battle against antibiotic-resistant bacteria.