QSAR Model for Predicting BACE1 Inhibitor Activity in Alzheimer’s Disease

This project focuses on developing a Quantitative Structure-Activity Relationship (QSAR) model to predict the inhibitory activity of potential BACE1 inhibitors, which are promising therapeutic candidates for Alzheimer’s disease. The study leverages machine learning algorithms, specifically Random Forest Regression, to analyze bioactivity data obtained from the ChEMBL database.

By using PubChem fingerprints as molecular descriptors and converting IC50 values to pIC50, the model effectively predicts the inhibitory potency of chemical compounds. The study includes exploratory data analysis, feature selection, and model deployment, with the final model enabling predictions based on molecular structure.

This work demonstrates the potential of computational drug discovery to accelerate the identification of novel BACE1 inhibitors, reducing time and cost in pharmaceutical research.