Related Experiment Video
Updated: Jul 26, 2025
07:53
A Fluorogenic Peptide Cleavage Assay to Screen for Proteolytic Activity: Applications for coronavirus spike protein activation
Published on: January 9, 2019
33.3K
Integrated data-driven and experimental approaches to accelerate lead optimization targeting SARS-CoV-2 main protease
Rohith Anand Varikoti1, Katherine J Schultz1, Chathuri J Kombala1
1Earth and Biological Sciences Directorate, Pacific Northwest National Laboratory, 902 Battelle Blvd, Richland, WA, 99352, USA.
Journal of Computer-Aided Molecular Design
|June 14, 2023
Summary
Computational methods rapidly identified novel therapeutic candidates for SARS-CoV-2 Mpro. Two compounds with low micromolar activity were validated experimentally, demonstrating an efficient drug discovery platform.
Area of Science:
- Drug discovery and development
- Computational chemistry and cheminformatics
- Machine learning in pharmacology
Background:
- Accelerating the identification of therapeutic candidates is crucial for drug development.
- Generative deep learning models can produce numerous novel compounds, but optimization of their properties remains a challenge.
- SARS-CoV-2 main protease (Mpro) is a key target for antiviral therapies.
Purpose of the Study:
- To develop and apply an integrated computational approach for the rapid identification and optimization of Mpro inhibitors.
- To generate and screen novel compound candidates using deep learning and machine learning models.
- To experimentally validate computationally identified lead compounds.
Main Methods:
- Utilized generative deep learning models to create novel compounds preserving a core scaffold.
- Applied computational tools including structural alert analysis, toxicity prediction, virtual screening, and graph neural networks.
- Employed machine learning-based 3D quantitative structure-activity relationships (QSAR) for activity and binding affinity prediction.
- Conducted experimental validation using Native Mass Spectrometry and FRET-based functional assays.
- Performed molecular dynamics simulations to analyze binding interactions and allosteric modulations.
Main Results:
- Generated tens of thousands of novel Mpro inhibitor candidates.
- Identified eight promising candidates through multi-parameter optimization and prediction.
- Two compounds, featuring quinazoline-2-thiol and acetylpiperidine cores, exhibited IC50 values in the low micromolar range (e.g., 1.7 µM and 3.41 µM).
- Molecular dynamics simulations indicated allosteric modulations upon compound binding to Mpro.
Conclusions:
- The integrated computational and experimental approach enables efficient, data-driven lead optimization for drug discovery.
- The validated compounds represent promising starting points for further development of SARS-CoV-2 Mpro inhibitors.
- This closed-loop platform can be adapted for discovering therapeutics against other protein targets.
