Welcome to HSR’s documentation!
Hypershape Recognition (HSR): a Generalised Framework for Moment-Based Molecular Similarity
HSR is a versatile, moment-based similarity measure tailored for three-dimensional (3D) chemical representations annotated with atomic features. It enhances the robustness and versatility of the Ultrafast Shape Recognition (USR) method by incorporating multidimensional features for each atom, such as protons, neutrons, and formal charges.
Getting Started
Installing HSR
You can install HSR using either pip or conda:
pip install hsr
or
conda install hsr -c conda-forge
Build from source
Clone the HSR repository on your machine. Move inside it and create the conda environment:
conda env create -f environment.yml
conda activate HSR_devel
Verify the correct creation of the environment by running:
pytest
To use HSR from CLI run:
python -m hsr.hsr_cli -h
If HSR is installed with pip or conda, the above command is replaced by the simple use of hsr.
Basic Usage
Run the folowing command to get help in using HSR from CLI:
hsr -h
For more deatails of HSR’s methodology check our overview.
Licensing
HSR is licensed under the GNU Affero General Public License Version 3, 19 November 2007. For more details, see the LICENSE file in the source code repository or visit GNU AGPL v3 License.
Citing HSR
If you use HSR in your research, please cite it as follows:
[TODO: Add citation]
Contributing to HSR
We welcome contributions to HSR! If you’re interested in helping, please read our Contributing Guidelines for information on how to get started.