MOLAN, A Machine Learning Workflow for MolecularAnalysis: Application to Melting Points

If you are using the MOLAN workflow in your research paper, please cite us as

@article{sivaraman2020machine,
  title={A machine learning workflow for molecular analysis: application to melting points},
  author={Sivaraman, Ganesh and Jackson, Nicholas E and Sanchez-Lengeling, Benjamin and V{\'a}zquez-Mayagoitia, {\'A}lvaro and Aspuru-Guzik, Al{\'a}n and Vishwanath, Venkatram and de Pablo, Juan J},
  journal={Machine Learning: Science and Technology},
  volume={1},
  number={2},
  pages={025015},
  year={2020},
  publisher={IOP Publishing}
}

Intro statement

How do I run this?

There are two options:

• Locally

This requires a standard scientific Python 3 environment with rdkit and tensorflow+pytorch and a cloned github. A simple way of getting that is installing Anaconda.

First to clone the github and then replicate a new anaconda environment using the environment.yml file:

git clone https://github.com/argonne-lcf/melting_points_ml
cd melting_points_ml
conda env create -f environment.yml

• Remotely via Google Colab

Visit google colab (requires a gdrive account) and open a colab notebook via github:

What's inside?

• data
  • 47K: Folder of json, each containing information for one molecule.
  • *csv: csv files.
• notebooks: Jupyter notebooks (run these!)
  • Exploratory_Data_Analysis.ipynb
  • semisupervised_VAE.ipynb
  • Graph_Neural_Networks.ipynb
  • Gaussian_Processes.ipynb
• code: Repo specific modules for training and creating the models.
• results: Figures and weights for models.
• media: Assorted images.

Acknowledgements

This research used resources of the Argonne Leadership Computing Facility, which is a DOE Office of Science User Facility supported under Contract DE-AC02-06CH11357. Argonne National Laboratory’s work was supported by the U.S. Department of Energy, Office of Science, under contract DE-AC02-06CH11357.

Code contributors

• Benjamin Sanchez-Lengeling beangoben::AT:::gmail.com
• Ganesh Sivaraman gsivaraman:AT::anl.gov
• Nicholas Jackson jackson.nick.e::aT:gmail.com
• Alvaro Vazquez-Mayagoitia alvaro::At:anl.gov
• Alan Aspuru-Guzik aspuru::at:utoronto.ca