ATTNSOM: Learning Cross-Isoform Attention for Cytochrome P450 Site-of-Metabolism Prediction

This repository contains the reference implementation of ATTNSOM, an isoform-aware framework for atom-level site-of-metabolism (SoM) prediction in cytochrome P450–mediated drug metabolism.

ATTNSOM integrates intrinsic molecular reactivity with cross-isoform metabolic relationships using graph neural networks and cross-attention mechanisms.

---

Table of Contents

1. Environment Setup
2. Running ATTNSOM

---

Environment Setup

1. Install Miniconda or Anaconda
   Ensure that your GPU drivers and CUDA version are compatible with the versions specified in envs/attnsom.yml.

2. Create the conda environment:

   cd envs/
   conda env create -f attnsom.yml
   conda activate attnsom
   pip install https://data.pyg.org/whl/torch-2.4.0%2Bcu118/torch_scatter-2.1.2%2Bpt24cu118-cp310-cp310-linux_x86_64.whl
   cd ..

Running ATTNSOM

To train or evaluate ATTNSOM on a dataset:

python main.py