Identifying New TACE Inhibitors Using Deep Learning and Drug Screening
Author Information
Author(s): Yasir Muhammad, Park Jinyoung, Han Eun-Taek, Han Jin-Hee, Park Won Sun, Hassan Mubashir, Kloczkowski Andrzej, Chun Wanjoo
Primary Institution: Kangwon National University School of Medicine
Hypothesis
Can deep learning models effectively identify novel inhibitors targeting the TNF-α converting enzyme (TACE) from FDA-approved drugs?
Conclusion
The study successfully identified Vorinostat as a potential TACE inhibitor, demonstrating the effectiveness of deep learning in drug discovery.
Supporting Evidence
- Vorinostat was identified as a potential TACE inhibitor through deep learning predictions.
- Molecular docking and dynamics simulations confirmed Vorinostat's strong binding interactions with TACE.
- Biological evaluation demonstrated Vorinostat's inhibitory activity against TACE in RAW 264.7 cells.
- The study highlights the potential of repurposing FDA-approved drugs for new therapeutic uses.
- Deep learning models achieved high accuracy in predicting TACE inhibitory potential.
- Statistical analysis showed significant differences in binding affinities among the screened compounds.
- Vorinostat's binding energy was among the lowest, indicating strong potential as a TACE inhibitor.
- The integration of computational methods streamlined the drug discovery process.
Takeaway
Researchers used computer models to find new uses for existing medicines, discovering that Vorinostat might help treat inflammation by blocking a specific enzyme.
Methodology
The study combined deep learning, molecular docking, and molecular dynamics simulations to screen FDA-approved drugs for TACE inhibition.
Potential Biases
Potential bias in the selection of compounds based on existing FDA approvals.
Limitations
The study primarily focused on FDA-approved drugs, which may limit the exploration of other potential inhibitors.
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website