/chemicalx

A PyTorch and TorchDrug based deep learning library for drug pair scoring. (KDD 2022)

Primary LanguagePythonApache License 2.0Apache-2.0


PyPI Version Docs Status Code Coverage Build Status DOI

Documentation | External Resources | Datasets | Examples

ChemicalX is a deep learning library for drug-drug interaction, polypharmacy side effect, and synergy prediction. The library consists of data loaders and integrated benchmark datasets. It also includes state-of-the-art deep neural network architectures that solve the drug pair scoring task. Implemented methods cover traditional SMILES string based techniques and neural message passing based models.


Citing

If you find ChemicalX and the new datasets useful in your research, please consider adding the following citation:

@inproceedings{10.1145/3534678.3539023,
  author = {Rozemberczki, Benedek and Hoyt, Charles Tapley and Gogleva, Anna and Grabowski, Piotr and Karis, Klas and Lamov, Andrej and Nikolov, Andriy and Nilsson, Sebastian and Ughetto, Michael and Wang, Yu and Derr, Tyler and Gyori, Benjamin M.},
  title = {ChemicalX: A Deep Learning Library for Drug Pair Scoring},
  year = {2022},
  isbn = {9781450393850},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  url = {https://doi.org/10.1145/3534678.3539023},
  doi = {10.1145/3534678.3539023},
  booktitle = {Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining},
  pages = {3819–3828},
  numpages = {10},
  keywords = {chemistry, neural networks, deep learning},
  location = {Washington DC, USA},
  series = {KDD '22}
}

Drug Pair Scoring Explained

Our framework solves the drug pair scoring task of computational chemistry. In this task a machine learning model has to predict the outcome of administering two drugs together in a biological or chemical context. Deep learning models which solve this task have an architecture with two distinctive parts:

  1. A drug encoder layer which takes a pair of drugs as an input (blue and red drugs below).
  2. A head layer which outputs scores in the administration context - polypharmacy in our explanatory figure.

Getting Started

The API of chemicalx provides a high-level function for training and evaluating models that's heavily influenced by the PyKEEN training and evaluation pipeline:

from chemicalx import pipeline
from chemicalx.models import DeepSynergy
from chemicalx.data import DrugCombDB

model = DeepSynergy(context_channels=112, drug_channels=256)
dataset = DrugCombDB()

results = pipeline(
    dataset=dataset,
    model=model,
    # Data arguments
    batch_size=5120,
    context_features=True,
    drug_features=True,
    drug_molecules=False,
    # Training arguments
    epochs=100,
)

# Outputs information about the AUC-ROC, etc. to the console.
results.summarize()

# Save the model, losses, evaluation, and other metadata.
results.save("~/test_results/")

Case Study Tutorials

We provide in-depth case study like tutorials in the Documentation, each covers an aspect of ChemicalX’s functionality.


Methods Included

In detail, the following drug pair scoring models were implemented.

2018

2019

2020

2021


Head over to our documentation to find out more about installation, creation of datasets and a full list of implemented methods and available datasets. For a quick start, check out the examples in the examples/ directory.

If you notice anything unexpected, please open an issue. If you are missing a specific method, feel free to open a feature request.


Installation

PyTorch 1.10.0

To install for PyTorch 1.10.0, simply run

pip install torch-scatter -f https://pytorch-geometric.com/whl/torch-1.10.0+${CUDA}.html
pip install torchdrug
pip install chemicalx

where ${CUDA} should be replaced by either cpu, cu102, or cu111 depending on your PyTorch installation.

cpu cu102 cu111
Linux
Windows
macOS

Running tests

$ tox -e py

License