metworkpy 0.9.0

Tools for creating and working with network models of metabolism.

Welcome to MetworkPy

• Welcome to MetworkPy
  • Usage
  • Issues and Pull Requests
  • Licensing
  • References
    • iMAT References
    • Kullback-Leibler Divergence
    • Mutual Information
    • Permutation Testing

MetworkPy is a Python library containing tools for working with and analyzing metabolic networks. This functionality includes:

• Generating network representations of Genome Scale Metabolic Networks (GSMMs)
• Integrating gene expression data with GSMMs
• Evaluating where the metabolism is most perturbed using divergence metrics

Documentation

Documentation can be found at https://metworkpy.readthedocs.io

Usage

MetworkPy can be installed with pip:

pip install metworkpy

The documentation includes some instructions for getting started. For an example of the usage of MetworkPy see the examples directory . For a more advanced example associated associated with the application note for MetworkPy (including some test data) see https://github.com/Ma-Lab-Seattle-Childrens-CGIDR/metworkpy_application_note.

Issues and Pull Requests

If you experience any problems while using MetworkPy (including the documentation), please create a GitHub issue in this repository. When creating an issue, a minimal reproducible example of the issue will make getting you help much easier. You can also create issues for any enhancements you would like to see in MetworkPy. Contributions are welcome! Please see the CONTRIBUTING.md for more information.

Licensing

This project makes use of the following external libraries:

• COBRApy licensed under the LGPL-2.1
• NetworkX licensed under the BSD-3-Clause
• NumPy licensed under the BSD-3-Clause
• Optlang licensed under Apace-2.0
• Pandas licensed under the BSD-3-Clause
• SciPy licensed under the BSD-3-Clause
• SymPy licensed under the BSD-3-Clause

The mutual information implementation where partially inspired by those found in the feature_selection module of scikit-learn, and the tests for those methods were adapted from those in scikit-learn, which is licensed under the BSD-3-Clause. Additionally, the implementation of the iMAT functionality was inspired by gembox (which uses a GPL-3.0-only license), and dexom-python (which uses the GPL-3.0-only license).

The permutation test implementation uses modified code from SciPy's stats module (licensed under the BSD-3-Clause, see above) for estimating the empirical p-value in line with Phipson, B., & Smyth, G. K. (2010).

The code for the centrality submodule is based on code from NetworkX (licensed under the BSD-3-Clause, see above) function closeness_centrality. Specifically,

• closeness_centrality_subset is based on networkx.algorithms.centrality.closeness_centrality, modified to calculate the closeness only to a susbet of nodes
• betweenness_centrality_subset is based on networkx.algorithms.centrality.betweenness_centrality_subset, modified to change the rescaling of the nodes to only be relative to the number of paths between the subset of nodes, not all nodes in the graph.
• _rescale is modified from network.algorithms.centrality.betweenness_centrality_subset._rescale, modified for the changed rescaling of the betweenness_centrality_subset
• _accumulate_subset is essentially unmodified from network.algorithms.centrality.betweenness_centrality_subset._accumulate_subset (included to avoid issues with later implementation changes in NetworkX).

References

iMAT References

1. Shlomi T, et al. Network-based prediction of human tissue-specific metabolism, Nat. Biotechnol., 2008, vol. 26 (pg. 1003-1010)
2. Hadas Zur, Eytan Ruppin, Tomer Shlomi, iMAT: an integrative metabolic analysis tool, Bioinformatics, Volume 26, Issue 24, December 2010, Pages 3140–3142, https://doi.org/10.1093/bioinformatics/btq602

Kullback-Leibler Divergence

1. Q. Wang, S. R. Kulkarni and S. Verdu, "Divergence Estimation for Multidimensional Densities Via k-Nearest-Neighbor Distances," in IEEE Transactions on Information Theory, vol. 55, no. 5, pp. 2392-2405, May 2009, doi: 10.1109/TIT.2009.2016060.

Mutual Information

1. Kraskov, A., Stögbauer, H., & Grassberger, P. (2004). Estimating mutual information. Physical Review E, 69(6), 066138.
2. Ross, B. C. (2014). Mutual Information between Discrete and Continuous Data Sets. PLoS ONE, 9(2), e87357

Permutation Testing

1. Phipson, B., & Smyth, G. K. (2010). Permutation p-values should never be zero: Calculating exact p-values when permutations are randomly drawn. Statistical Applications in Genetics and Molecular Biology, 9(1). https://doi.org/10.2202/1544-6115.1585

Sampling

1. Galuzzi, B. G., Milazzo, L., & Damiani, C. (2024). Adjusting for false discoveries in constraint-based differential metabolic flux analysis. Journal of Biomedical Informatics, 150, 104597. https://doi.org/10.1016/j.jbi.2024.104597