Dataset Card for clintox

Table of Contents

• Table of Contents
• Dataset Description
  • Dataset Summary
  • Supported Tasks and Leaderboards
  • Languages
• Dataset Structure
  • Data Instances
  • Data Fields
  • Data Splits
• Dataset Creation
  • Curation Rationale
  • Source Data
  • Annotations
  • Personal and Sensitive Information
• Considerations for Using the Data
  • Social Impact of Dataset
  • Discussion of Biases
  • Other Known Limitations
• Additional Information
  • Dataset Curators
  • Licensing Information
  • Citation Information
  • Contributions

Dataset Description

• Homepage: https://moleculenet.org/
• Repository: https://github.com/deepchem/deepchem/tree/master
• Paper: https://arxiv.org/abs/1703.00564

Dataset Summary

clintox is a dataset included in MoleculeNet. Qualitative data of drugs approved by the FDA and those that have failed clinical trials for toxicity reasons. This uses the CT_TOX task.

Note, there was one molecule in the training set that could not be converted to SELFIES (*C(=O)[C@H](CCCCNC(=O)OCCOC)NC(=O)OCCOC)

Dataset Structure

Data Fields

Each split contains

• smiles: the SMILES representation of a molecule
• selfies: the SELFIES representation of a molecule
• target: clinical trial toxicity (or absence of toxicity)

Data Splits

The dataset is split into an 80/10/10 train/valid/test split using scaffold split.

Source Data

Initial Data Collection and Normalization

Data was originially generated by the Pande Group at Standford

Licensing Information

This dataset was originally released under an MIT license

Citation Information

@misc{https://doi.org/10.48550/arxiv.1703.00564,
  doi = {10.48550/ARXIV.1703.00564},
  
  url = {https://arxiv.org/abs/1703.00564},
  
  author = {Wu, Zhenqin and Ramsundar, Bharath and Feinberg, Evan N. and Gomes, Joseph and Geniesse, Caleb and Pappu, Aneesh S. and Leswing, Karl and Pande, Vijay},
  
  keywords = {Machine Learning (cs.LG), Chemical Physics (physics.chem-ph), Machine Learning (stat.ML), FOS: Computer and information sciences, FOS: Computer and information sciences, FOS: Physical sciences, FOS: Physical sciences},
  
  title = {MoleculeNet: A Benchmark for Molecular Machine Learning},
  
  publisher = {arXiv},
  
  year = {2017},
  
  copyright = {arXiv.org perpetual, non-exclusive license}
}

Contributions

Thanks to @zanussbaum for adding this dataset.