Fragment-Based Hit Discovery via Unsupervised Learning of Fragment-Protein Complexes

Explore a 54-minute talk on fragment-based hit discovery using unsupervised learning of fragment-protein complexes. Delve into the challenges of finding molecules that bind to target proteins in drug discovery and learn about a novel approach using high-throughput crystallography and machine learning. Discover how the FRagment Ensemble SCOring (FRESCO) method reframes hit discovery as a denoising problem, identifying significant pharmacophore distributions from fragment ensembles. Follow the speaker's journey through retrospective validation on SARS-CoV-2 main protease and prospective discovery of novel hits. Gain insights into pharmacophore modeling, the FRESCO methodology, and its application in the COVID Moonshot project. Explore the screening workflow, results, and future outlook for this innovative approach in drug discovery.

- Intro
- Outline
- Fragment to Hit: ML Approaches and Challenges
- Introduction to FRagment Ensemble SCOring FRESCO
- Pharmacophore Modeling
- Diving Deep into FRESCO
- Retrospective Study - COVID Moonshot
- Screening Workflow
- Results
- Discussion
- Outlook and Conclusion
- Q&A