GraphCompass (Graph Comparison Tools for Differential Analyses in Spatial Systems) is a Python-based framework that brings together a robust suite of graph analysis and visualization methods, specifically tailored for the differential analysis of cell spatial organization using spatial omics data.
It is developed on top on Squidpy and AnnData.
GraphCompass provides differential analysis methods to study spatial organization across conditions at three levels of abstraction:
-
Cell-type-specific subgraphs:
i. Portrait method,
ii. Diffusion method.
-
Cellular neighborhoods:
i. GLMs for neighborhood enrichment analysis.
-
Entire graphs:
i. Wasserstein WL kernel,
ii. Filtration curves.
Tutorials for the different methods can be found in the notebooks folder.
You will find all the necessary dependencies in the pyproject.toml file. Dependencies are automatically installed when you install GraphCompass.
You can install GraphCompass via pip from PyPI.
$ pip install graphcompassor you can install GraphCompass by cloning the repository and running:
$ pip install -e .Mayar Ali, Merel Kuijs, Soroor Hediyeh-zadeh, Tim Treis, Karin Hrovatin, Giovanni Palla, Anna C Schaar, Fabian J Theis, GraphCompass: spatial metrics for differential analyses of cell organization across conditions, Bioinformatics, Volume 40, Issue Supplement_1, July 2024, Pages i548–i557, https://doi.org/10.1093/bioinformatics/btae242
[COMING SOON] Contributions are very welcome. To learn more, see the [Contributor Guide].
Distributed under the terms of the MIT license, GraphCompass is free and open source software.
If you encounter any problems, please file an issue along with a detailed description.
This project was generated from @cjolowicz's Hypermodern Python Cookiecutter template.
[COMING SOON] [contributor guide]: https://github.com/theislab/graphcompass/blob/main/CONTRIBUTING.md