model_cards/article.md

Model documentation & parameters

Algorithm Version: Which model version to use.

Primer SMILES: The SMILES string for priming the generation (has to be provided).

Maximal sequence length: The maximal number of SMILES tokens in the generated molecule.

Sampling uniquely: Generate unique sample sequences if set to true.

Number of samples: How many samples should be generated (between 1 and 50).

Model card -- REINVENT

Model Details: REINVENT is a collection of tools for de novo drug design. Here, we showcase the SMILES-based generative model. For details, see Blaschke et al. (2020); J. Chem. Inf. Model..

Developers: Thomas Blaschke and colleagues from AstraZeneca.

Distributors: Original authors' code integrated into GT4SD.

Model date: 2020, see the REINVENT 2.0 paper

Model version: N.A.

Model type: A sequence-based molecular generator from the REINVENT toolbox.

Information about training algorithms, parameters, fairness constraints or other applied approaches, and features: The code for getting unique sample sequences, randomizing scaffolds, and generation of the dataset as well as the dataloader was taken from the original implementation of Molecular Reinvent. Our implementation does not include BaseAction as a parent class for the ReinventBase where we have added all the functions of Molecular Reinvent.

Paper or other resource for more information: REINVENT 2.0 -- Blaschke et al. (2020); J. Chem. Inf. Model..

License: MIT

Where to send questions or comments about the model: Open an issue on GT4SD-REINVENT-repository.

Intended Use. Use cases that were envisioned during development: Chemical research, in particular drug discovery.

Primary intended uses/users: Researchers and computational chemists using the model for model comparison or research exploration purposes.

Out-of-scope use cases: Production-level inference, producing molecules with harmful properties.

Metrics: N.A.

Datasets: N.A.

Ethical Considerations: Unclear, please consult with original authors in case of questions.

Caveats and Recommendations: Unclear, please consult with original authors in case of questions.

Model card prototype inspired by Mitchell et al. (2019)

Citation

@article{blaschke2020reinvent,
  title={REINVENT 2.0: an AI tool for de novo drug design},
  author={Blaschke, Thomas and Ar{\'u}s-Pous, Josep and Chen, Hongming and Margreitter, Christian and Tyrchan, Christian and Engkvist, Ola and Papadopoulos, Kostas and Patronov, Atanas},
  journal={Journal of chemical information and modeling},
  volume={60},
  number={12},
  pages={5918--5922},
  year={2020},
  publisher={ACS Publications}
}