Related Experiment Video
Updated: Jul 5, 2025
08:53
Using Scaffold Liposomes to Reconstitute Lipid-proximal Protein-protein Interactions In Vitro
Published on: January 11, 2017
8.9K
Local Scaffold Diversity-Contributed Generator for Discovering Potential NLRP3 Inhibitors
Weichen Bo1, Yangqin Duan1, Yurong Zou1
1Department of Biotherapy, Cancer Center and State Key Laboratory of Biotherapy, West China Hospital, Sichuan University, Chengdu 610041, China.
Journal of Chemical Information and Modeling
|January 23, 2024
Summary
We developed a novel AI method, the local scaffold diversity-contributed generator (LSDC), to create diverse drug candidates. This approach successfully identified potent NLRP3 inhibitors with novel scaffolds, advancing AI-driven drug discovery.
Area of Science:
- Medicinal Chemistry
- Computational Chemistry
- Artificial Intelligence
Background:
- Deep generative models are vital for de novo drug design.
- Existing models struggle with scaffold diversity under multiple constraints.
- Scaffold diversity is crucial for identifying novel drug leads.
Purpose of the Study:
- To introduce a novel AI model, the local scaffold diversity-contributed generator (LSDC), for enhanced molecular generation.
- To improve scaffold diversity in generating drug candidates that meet multiple objectives.
- To discover novel inhibitors targeting the NLRP3 inflammasome.
Main Methods:
- Developed the local scaffold diversity-contributed generator (LSDC) model.
- Applied LSDC for generating diverse molecules targeting NLRP3.
- Utilized molecular docking and bioactivity assays for validation.
Main Results:
- LSDC generated molecules with significantly greater scaffold diversity compared to state-of-the-art methods.
- Identified 12 novel molecules, including those with previously unreported scaffolds.
- Discovered two potent NLRP3 inhibitors, A22 (IC50=38.1 nM) and A14 (IC50=44.43 nM).
- Compound A14 demonstrated high oral bioavailability (83.09% in mice).
Conclusions:
- The LSDC model effectively enhances scaffold diversity in AI-driven drug design.
- Successfully identified novel, potent NLRP3 inhibitors through integrated AI generation and experimental validation.
- This work provides a valuable framework for combining AI with wet lab experiments in drug discovery.
