This repository contains code for the paper: Beyond Generative Models: Superfast Traversal, Optimization, Novelty, Exploration and Discovery (STONED) Algorithm for Molecules using SELFIES. By: AkshatKumar Nigam, Robert Pollice, Mario Krenn, Gabriel dos Passos Gomes, Alán Aspuru-Guzik
For cloning the repository, please have a look at the Branch Navigator section.
Before running the code, please ensure you have the following:
- SELFIES (any version) - The code was run with v1.0.1. The code is compatible with all v1+.
- RDKit
- Python 3.0 or up
- numpy
- Please have a look at our Tutorial Document. There we provide quick details on how to form local chemical subspaces, generating chemical paths & obtaining median molecules.
- Experiment B: Formation of Local Chemical Spaces
- Experiment C: Chemical Paths and Rediscovery (with a GA)
- Experiment D: Interpolation via Chemical Path formation
- Experiment E: Median Molecules for Photovoltaics (obtaining median molecules)
Make a github issue 😄. Please be as clear and descriptive as possible. Please feel free to reach out in person: (akshat[DOT]nigam[AT]mail[DOT]utoronto[DOT]ca, rob[DOT]pollice[AT]utoronto[DOT]ca, mario[DOT]krenn[AT]utoronto[DOT]ca , gabriel[DOT]gomes[AT]utoronto[DOT]ca)