Unraveling the Mechanism of Drug Binding to SARS-CoV-2 RNA Pseudoknot with Thermodynamics-Driven Machine Learning

2026-04-16 · Source: Machine Learning · Field: Science & Research — Life Sciences & Biology, Mathematics & Computational Sciences, Health & Medical Research · Depth: Expert, quick

Summary

A study utilized spectral map (SM), a thermodynamics-driven machine-learning method, to investigate the binding mechanism of the -1 programmed ribosomal frameshifting (-1 PRF) inhibitor merafloxacin and its analogs to the SARS-CoV-2 RNA pseudoknot. This pseudoknot, crucial for viral protein synthesis, exists in threaded and unthreaded topologies. By learning collective variables directly from molecular dynamics (MD) trajectories, researchers examined ligand-induced distortions in both pseudoknot topologies and considered neutral and ionized ligand forms at physiological pH. Free-energy landscapes revealed that ligand-induced destabilization is topology-selective: merafloxacin destabilizes the S2 stem in the threaded pseudoknot, but shifts to S1 and S3 stems in the unthreaded form. The zwitterionic form of merafloxacin uniquely imposed slow dynamics on the unthreaded pseudoknot, and neutral and zwitterionic forms showed qualitative differences in mechanism within the same RNA topology. These findings highlight the importance of pseudoknot topology, ligand type, and protonation state in shaping viral RNA conformational dynamics.

Key takeaway

For research scientists developing antiviral interventions targeting SARS-CoV-2, you should explicitly account for the RNA pseudoknot's topology and the ligand's protonation state. Your modeling efforts must incorporate physiological protonation to accurately predict drug action and identify effective small-molecule inhibitors, as these factors qualitatively alter binding mechanisms and destabilization patterns.

Key insights

Ligand binding to SARS-CoV-2 RNA pseudoknot is topology-selective and dependent on ligand protonation state.

Principles

• Physiological protonation is essential for modeling RNA-targeted drug action.
• Ligand-induced destabilization is topology-selective.

Method

Spectral map (SM), a thermodynamics-driven machine-learning method, learns collective variables directly from molecular dynamics (MD) trajectories to isolate slow dynamic modes of RNA-ligand systems.

In practice

• Model RNA-targeted drugs considering physiological protonation.
• Investigate ligand binding across different RNA topologies.

Topics

• SARS-CoV-2 RNA Pseudoknot
• -1 Programmed Ribosomal Frameshifting
• Thermodynamics-Driven Machine Learning
• Molecular Dynamics Simulation
• Ligand-RNA Binding

Best for: Research Scientist, AI Scientist

Related on AIssential

• Unraveling the Mechanism of Drug Binding to SARS-CoV-2 RNA Pseudoknot with Thermodynamics-Driven Machine Learning — Thermodynamics-driven machine learning reveals topology-selective drug binding mechanisms to SARS-CoV-2 RNA pseudoknot.
• Predicting RNA 3D structure and conformers using a pre-trained secondary structure model and structure-aware attention — trRosettaRNA2 accurately predicts RNA 3D structures and conformers by integrating pre-trained secondary structure priors and SS-aware attention.
• Reactive Flux Matching: Mechanism Discovery and Adaptive Sampling of Rare Events — Flux Matching learns reaction pathways and coordinates directly from reactive trajectory data without prior dynamic knowledge.
• RNA design across eras: from covariance models to modern generative AI — RNA's functional versatility relies on 3D folding, with generative AI building on foundational probabilistic models for structure prediction.
• Conditional Monge Gap enables generalizable single-cell perturbation modelling — CMonge is a conditional optimal transport framework that accurately predicts single-cell perturbation responses and generalizes to unseen conditions.
• Latent space mapping of interpretable structural coordinates from stochastic single-molecule signals — Latent-space mapping resolves stochastic signal warping in nanopore sensors, enabling interpretable molecular coordinate systems.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Machine Learning.