WRN activity model
WRN is a RecQ-family DNA helicase, and its inherited loss causes Werner syndrome, a premature-aging disorder. Cancers with high microsatellite instability (MSI-H), common in colorectal, endometrial and gastric cancer, depend on WRN to survive, so WRN inhibitors are in clinical trials as a synthetic-lethal treatment for them. The reference structure 9S18 has the clinical inhibitor HRO761 bound between the two lobes of the helicase domain, locking the enzyme in an inactive shape.
A model trained on measured activity data for WRN (Oncology). Give it compounds and it predicts their IC50 against WRN, so you can rank candidates by it.
Predict with this model
- UniProt
- Q14191
- Reference structure
- PDB 9S18, ligand YHC
- Training data
- TSAP (AI-QSAR): 65 compounds, IC50
- Validation
- Predicted and measured values are compared on compounds held out of training.
- Cost
- From 402 credits
- Model version
- TSAP (AI-QSAR): Ridge regression on docking features

Validation results
Predicted and measured values compared on compounds held out of training.
TSAP (AI-QSAR)
Pearson r
0.90
R² (CoD)
0.72
Holdout compounds
16
How to use it
Under Predict with this model, press TSAP (AI-QSAR). Its wizard opens with the WRN model already selected.
Upload the SMILES of the compounds to predict as a .smi, .csv or .txt file.
Check the estimated credits and time, then submit.
The result is a table of compounds ranked by predicted IC50.
Limitations
- Predictions can be less accurate for compounds whose structures differ a lot from the training data.
- Validation used only about 16 compounds, so performance on new compounds can differ from these metrics.
- Predictions do not replace experiments. Use them to decide what to synthesize and measure first.
Run your own predictions with the WRN model
Sign up and get 500 credits. No card required. You can also choose to get news about new models and workflows.
