PROTAC-Splitter: A Machine Learning Framework for Automated Identification of PROTAC Substructures

This repository contains the program code to split PROTAC molecules into their constituent substructures.

A Gradio app is available to split PROTAC molecules and visualize the results at this link: https://huggingface.co/spaces/ailab-bio/PROTAC-Splitter-App.

Table of Contents 📜

• Installation
• Usage
• Data Availability
• Contributing
• License
• Reference

Installation 🛠️

The package was developed and tested in Python 3.10.8. Always use a virtual environment to install the package.

For using the code under the scripts directory in this repository, run the following commands:

git clone https://github.com/ribesstefano/PROTAC-Splitter.git
cd PROTAC-Splitter
pip install -r requirements.txt
pip install -r scripts/requirements.txt

# Add the package to the PYTHONPATH
export PYTHONPATH=$PYTHONPATH:`pwd`/protac_splitter

Alternatively, you can install the package using pip (again, in a virtual environment):

pip install git+https://github.com/ribesstefano/PROTAC-Splitter.git

Usage 🚀

To use the package, please refer to the function split_protac in the protac_splitter/protac_splitter module.

Here is an example of how to use the function:

import pandas as pd
from protac_splitter import split_protac

# Split a PROTAC molecule reported as a SMILES string
protac_smiles = "CC(C)(C)S(=O)(=O)c1cc2c(Nc3ccc4scnc4c3)ccnc2cc1OCCOCCOCCOCCOCC(=O)Nc1cccc2c1CN(C1CCC(=O)NC1=O)C2=O"
ligands = split_protac(protac_smiles)
print(ligands)

# One can also feed a DataFrame to the function
df = pd.read_csv("my/local/file.csv")
split_df = split_protac(df, protac_smiles_col="PROTAC SMILES")
print(split_df.head())

Alternatively, you can use the Gradio app at scripts/protac_splitter_app.py have a GUI to split PROTAC molecules.

gradio scripts/protac_splitter_app.py

Data Availability 📥

Curated public data, the synthetic PROTACs dataset, and trained models are available for download from Zenodo at: https://doi.org/10.5281/zenodo.15797309.

Contributing 🤝

We welcome contributions to this project! If you have suggestions for improvements, bug fixes, or new features, please open an issue or submit a pull request.

License 📄

This project is licensed under the MIT License - see the LICENSE file for details.

Reference 📝

If you find this work useful, please consider citing it via:

@article{Ribes2025PROTACSplitter,
  title   = {PROTAC‐Splitter: A Machine Learning Framework for Automated Identification of PROTAC Substructures},
  author  = {Stefano Ribes and Ranxuan Zhang and Télio Cropsal and Anders Källberg and Christian Tyrchan and Eva Nittinger and Rocío Mercado},
  journal = {ChemRxiv},
  year    = {2025},
  month   = {Jul},
  day     = {08},
  doi     = {10.26434/chemrxiv-2025-bn1nv},
  url     = {https://chemrxiv.org/engage/chemrxiv/article-details/686670983ba0887c33677fc8},
  license = {CC BY 4.0}
}