BioActivityPrediction-mTOR (Drug Discovery)

Dysregulation of the mTOR pathway has been linked to the development and progression of cancer. This app allows you to predict the bioactivity of compounds towards inhibting the mTOR enzyme to find potential leads in Drug Discovery.

Data Collection and Data Preprocessing

The data represnts mtor single proteins in Homo Sapiens. The data was extracted from ChemBL Database using chembl_webresource_client. Handled missing values, duplicate values and standardized values with varied scales.

Feature Extraction and Feature Selection

Performed feature extraction using RDKit library and PaDel-Descriptors software to extract Lipinksi Descriptors and PubChem fingerprints. Implemented Feature Selection to drop low variance features using Variance Threshold.

Model Building and Results

Having 179 features, RandomForestRegressor was modelled resulting in Rsquare of 0.68 and Mean Square Error of 0.55.