AlphaFold3 Covalent Ligand Discovery for Kinase BTK

Loaded with results from a prospective virtual screening campaign using AlphaFold3 cofolding predictions to discover covalent inhibitors, specifically against the model kinase BTK. It includes data demonstrating the model's performance in ranking true covalent binders over property-matched decoys and the discovery of a novel covalent small molecule. The dataset is 9.3 KB in size, indicating a limited scope.

Use Cases

• Analyze AlphaFold3 cofolding predictions and associated confidence metrics for ranking covalent binders against kinases.
• Validate the subangstrom accuracy of predicted AF3 binding modes using co-crystallography data from the BTK campaign.
• Compare the performance of AF3 predictions against state-of-the-art covalent docking tools for protein kinases.

Strengths

• Data is derived from a prospective virtual screening campaign demonstrating practical application of AlphaFold3.
• Results include validation via co-crystallography showing subangstrom accuracy for a predicted binding mode.
• The study specifically targets the kinase BTK, a prolific drug target family.

Limitations

• The dataset is very small at 9.3 KB, limiting the scope and volume of available data for analysis.
• The specific column structure and row count are unknown, complicating direct computational use.
• Focus is on a single model kinase (BTK), which may limit generalizability to other protein families.

Provenance

Source

figshare, authored by Yoav Shamir.

Collection Method

Generated from a computational study using AlphaFold3 for virtual screening and validated with experimental co-crystallography.

Freshness

Last updated on 2026-03-20.