Zahoránszky-Kőhalmi, G., Wan, K.K. & Godfrey, A.G. Hilbert-curve assisted structure embedding method. J Cheminform 16, 87 (2024). DOI: 10.1186/s13321-024-00850-z

G. Zahoranszky-Kohalmi, K. K. Wan, A. G. Godfrey, Hilbert-Curve Assisted Structure Embedding Method, ChemRxiv, 28 February 2020, Version 1, DOI: 10.26434/chemrxiv.11911296.v1

The publication branch of the HCASE repository contains the source code and data files in the exact same status at the time the manuscript was under review and got accepted. We will keep this branch unchanged, but we continue the development and improvement of the codebase in the main branch.

If you are interested in contributing to the HCASE repository please refer to the For Contributors section below.

To test the HCASE method, you can run the embed_uspto.py script located in the examples/EmbedUSPTO directory. This script demonstrates how to embed compounds into the HCASE space.

1. Ensure you have set up the appropriate Conda environment (hcase_examples or cpu-env) as described in the Setting Up Examples Conda Environment section.

2. Navigate to the examples/EmbedUSPTO directory:

   cd examples/EmbedUSPTO
   

3. Ensure the file 2001_Sep2016_USPTOapplications_smiles.rsmi is located in the same directory as the script.

4. Run the script:

   python embed_uspto.py
   

This script will execute the embedding process and provide the results.

We welcome contributions to the HCASE method! Here's how you can contribute:

1. Create a Feature Branch:

   • Start by creating a personal feature branch (e.g., feat_new_1) from the feature branch.

2. Develop and Test:

   • Implement your feature or fix.
   • Ensure your code is thoroughly tested and works as expected.

3. Submit a Pull Request:

   • Once your code is complete and tested, create a pull request (PR) from your feature branch to the feature branch of the HCASE repository.

4. Approval and Merge:

   • Approved features will be merged into the main branch.

Thank you for contributing to the HCASE repository!

Make sure that Git is installed along with Git Large File Support (LFS)

To install GIT LFS use:

git lfs install --skip-smudge

Conda environment manager for Python is required

See: https://conda.io/projects/conda/en/latest/user-guide/install/index.html

Clone Repository

git clone https://github.com/ncats/hcase

Pull Large Files

cd hcase

git lfs pull (this may take some time)

The primary HCASE library code is within the hcase subdirectory. See the pip installion section for imformation on how to install this code via pip.

Examples of how to use this code are found within the examples subdirectory. This directory contains a conda environment which will install the hcase code automatically.

The latest HCASE code can be installed via pip with the following command:

pip install git+https://github.com/ncats/hcase.git

Additionally certain branches can be installed with:

pip install git+https://github.com/ncats/hcase.git@BRANCH-NAME

Run the following to create the Conda environment:

For GPU support: conda env create -f examples/environment.yml (this step needs to be done only once, if you don't already have the hcase_examples conda environment)

For CPU-only support: conda env create -f environment_cpu.yml (this step needs to be done only once, if you don't already have the cpu-env conda environment)

Activate the environment: conda activate hcase_examples or conda activate cpu-env (this step is necessary every time you want to run the HCASE method and/or the experiments)

For the sake of clarity, we moved all the Jupyter Notebook (.ipynb) files into examples/publication_workflows subdirectory after our paper got published.

To view the repository code at the time of publication view the publication branch.

This repository contains source code, Jupyter notebooks, data and results files which are organized into various subdirectories.

• Source code subdirectories:

hcase, code/, and aux_code

• Data and results subdirectory:

data/

• Jupyter Notebooks:

Jupyter Notebooks can be found in the code/, examples, and data/execute_notebooks/ subdirectories (files of .ipynb extension).

• Workflows:

KNIME workflows can be found in workflows/ subdirectory.

• Source Code License of HCASE Repository

The applicable license to source code can be found under filename: code/LICENSE and aux_code/LICENSE (same license types, i.e. [https://opensource.org/licenses/MIT](MIT License)). This license is applicable to all files recursively in the source code subdirectories as defined above. The files code/NOTES and aux_code/NOTES list source code modules that were utilized and their respective licenses. These modules have their own licenses which might be different from the Source Code License of this repository, and they need to be respected accordingly.

• Data License of HCASE Repository

The applicable license to data and results can be found under: data/LICENSE and plots/LICENSE ([https://creativecommons.org/licenses/by/4.0/legalcode.txt](Creative Commons Attribution 4.0 International Public License CC-BY 4.0 International)). This license is applicable to all files recursively in the data and results subdirectory as defined above. The files listed in data/NOTES and plots/NOTES lists input files and resources utilized to perform the experiments and are considered as derivative work of those resources. These input files and resources have their own licenses which might be different from the Data License of this repository, and they need to be respected accordingly. In the same file we also list which results files can be considered as derivative works, and we also list the the respective ascendent data source(s).

• Jupyter Notebook License of HCASE Repository

Jupyter Notebboks are special in the sense that they are comprised of source code, but they can also contain embedded data and plot (graph) sections. This duality is resolved via dual-licensing as follows. The code sections of Jupter Notebooks fall under the same license as source codes, i.e. the MIT License code/LICENSEwhereas data and plot sections embedded into the Jupyter Notebooks fall under the same license as data and result files, i.e. Creative Commons Attribution 4.0 International Public License CC-BY 4.0 International License data/LICENSE. Remarks enclosed in the code/NOTES file are also valid for code section of the Jupyter Notebooks. Remarks enclosed in the data/NOTES and plots/NOTES files are also valid for the embedded data and plots of the Jupyter Notebook files.

• Workflow License of HCASE Repository

KNIME workflows utilized in this study are contained in the workflows/ directory, and are subject of the "KNIME License Terms and Conditions" see: workflows/LICENSE.

• Links to licenses

• MIT License: https://opensource.org/licenses/MIT

• Creative Commons Attribution-ShareAlike 3.0 Unported License: https://creativecommons.org/licenses/by-sa/3.0/

• Creative Commons Attribution 4.0 International Public License (CC-BY 4.0 International): https://creativecommons.org/licenses/by/4.0/legalcode.txt

• GPL-2 License: https://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html

• GPL-3 License: https://www.gnu.org/licenses/gpl-3.0.en.html

• Apache License 2.0: https://www.apache.org/licenses/LICENSE-2.0

• 3-Clause BSD license: https://opensource.org/licenses/BSD-3-Clause, https://github.com/scikit-learn/scikit-learn/blob/master/COPYING

• KNIME: KNIME License Terms and Conditions https://www.knime.com/downloads/full-license

• hilbertcurve
• networkx
• rdkit