High-performance virtual screening by targeting a high-resolution RNA dynamic ensemble.

Area of Science:

• Structural Biology
• Computational Chemistry
• Drug Discovery

Background:

• Dynamic ensembles of RNA structures offer potential for advancing RNA-targeted drug discovery.
• High-throughput screening (HTS) is crucial for identifying novel therapeutic compounds.

Purpose of the Study:

• To evaluate the efficacy of using experimentally determined dynamic RNA ensembles in virtual screening for drug discovery.
• To assess the impact of ensemble accuracy on hit enrichment during virtual screening.

Main Methods:

• Experimental high-throughput screening of ~100,000 small molecules against the transactivation response element (TAR) RNA.
• Integration of NMR spectroscopy data and molecular dynamics simulations to generate dynamic TAR RNA ensembles.
• Ensemble-based virtual screening of generated sublibraries.

Main Results:

• Virtual screening using accurate dynamic ensembles achieved high performance (AUC ~0.85-0.94).
• ~40-75% of all identified hits were found within the top 2% of scored molecules.
• Enrichment significantly decreased with control ensembles lacking NMR data input.

Conclusions:

• Experimentally validated dynamic RNA ensembles significantly enhance hit identification in virtual screening campaigns.
• The accuracy of the RNA ensemble is critical for achieving effective enrichment of true drug candidates.
• This approach holds promise for accelerating RNA-targeted drug discovery.