Thank you to our colleagues @edsherer and Mikhail Reibarkh

Also, you don't need massive numbers of decoys to estimate screening enrichment! DUD-E+ data set available at jainlab.org/downloads

researchsquare.com/article/rs-258… Good news! Colleagues at the INSERM in France have used our ForceGen and Surflex-Dock methods to identify an already approved drug to potentially inhibit SARS-CoV-2 cell entry. #Covid19

Introducing our #ModuleOfTheMonth for April... Surflex eSim3D! Surflex eSim3D supports 3D #ligand-based design efforts, helping you to understand binding conformations to identify & optimise novel active compounds More info 👉 optibrium.com/project/surfle… BioPharmics LLC #DrugDesign

I'm thrilled to announce a new preprint describing collaborative work with Ajay Jain and Ann Cleves Jain, "Deep-Learning Based Docking Methods: Fair Comparisons to Conventional Docking Workflows". arxiv.org/abs/2412.02889

Patrick Walters Diego del Alamo Time-splits with the PDB can work fine, but they need to be more like 25/75% Known/Unknown. Scaffold-type splits also can make sense. See our (Ann Jain Optibrium) recent paper on non-cognate ligand docking: link.springer.com/article/10.100…