BayesPharma · Real Docking
Actual AutoDock Vina searchPose output + provenance
Dock molecules against a real binding-site hypothesis.
Provide a prepared receptor, an explicit 3D search box and one or more ligand SMILES. BayesPharma prepares ligands with RDKit + Meeko, executes AutoDock Vina, stores poses and ranks completed ligands by Vina score.
Docking is not proof of binding. A more negative Vina score is a computational ranking signal, not measured affinity. Validate receptor state, pocket definition, protonation/tautomers and benchmark the protocol against known active/inactive controls before trusting prospective rankings.
1 · Docking experiment
checking runtime…Prepared PDBQT is preferred because receptor protonation, cofactors, metals, waters and alternate locations should be reviewed deliberately.
One ligand runs synchronously. Multiple ligands launch an asynchronous campaign and reuse the receptor maps.
2 · Real docking results
not startedNo docking run yet. Results shown here will come from the `/docking/v1` AutoDock Vina service, not the Generative Chemistry affinity heuristic.
Progress0%
Completed0
Successful0
Failed0
| Rank | Ligand | SMILES | Vina kcal/mol | Runtime | Pose | Status |
|---|---|---|---|---|---|---|
| Run a docking experiment to populate this table. | ||||||
Interpretation checklist
Before advancing a hit: inspect the pose, compare known actives/inactives, test alternative receptor conformations where justified, examine key interactions/clashes, then combine docking with medicinal chemistry, selectivity, ADMET, synthesis and experimental evidence. Do not select wet-lab compounds from score alone.