Sites of Metabolism (SoMs) Data Analysis

Code used in Metabolite Identification Data in Drug Discovery, Part 2: Site-of-Metabolism Annotation, Analysis, and Exploration for Machine Learning

Examplified by AstraZeneca site-of-metabolism (SoM) data for 120 compounds and approved drugs from DrugBank.

Environment Setup

Create the environment from environment.yml

conda env create -f environment.yml
conda activate az_som_py310

Alternative: Create a virtual environment and install dependencies

conda create -n som_env_py310 python=3.10
conda activate som_env_py310
conda install -c conda-forge rdkit
pip install -r requirements.txt

Notes

• Data files are expected in the data/ directory.
• Outputs are written to the output/ directory.
• Active learning results generated using code from FAME.AL