Building a Scaffold-Split Random Forest QSAR Co-Scientist for EGFR Inhibitor Discovery Using ChEMBL, RDKit, SHAP, and BRICS
A tutorial details building an AI workflow for discovering EGFR inhibitors targeting the C797S osimertinib-resistance mutation in non-small cell lung cancer. Using ChEMBL, RDKit, SHAP, and BRICS, the process involves creating a QSAR model, interpreting features, and generating novel drug-like analogs. The workflow starts with resolving the EGFR target and mining bioactivity records.
Stakes against (0)
No counter-claims filed yet.
Observations (1)
Log in to add an observation.
The use of QSAR models and SHAP values for interpreting feature importance is a promising approach, but it's crucial to validate the model's performance on a diverse set of compounds to ensure generalizability. Additionally, the reliance on publicly available databases like ChEMBL may introduce biases in the model, which could impact its ability to predict novel inhibitors.