CobraMod is a Python 3 open-source package for pathway-centric curation of genome-scale metabolic models (GEMs). It builds upon the COBRApy toolbox and offers a comprehensible set of functions for semi-automated network extension, curation and visualization. CobraMod supports all databases from the BioCyc collection, the KEGG database, and the BiGG Models repository. It optionally can interact with Escher for pathway and flux visualization.
This package converts pathway information into native COBRApy objects and quality-checks them while adding them to the model. This includes testing for:
- duplicate elements
- correct chemical formula
- mass balance of reactions
- reaction reversibility
- capability to carry non-zero fluxes
- adding available gene information
- MEMOTE compliance
- available cross-references
CobraMod offers user-friendly tracking of the curation process with summary output with log files and customized pathway and flux visualization with Escher.
CobraMod can easily be installed using pip :
pip install cobramod
This package offers multiple functions for modifying and extending GEMs:
- Retrieve metabolic pathway information from a database
cobramod.get_data - Transform stored data into COBRApy objects
cobramod.create_object - Add metabolites from multiple sources
cobramod.add_metabolites - Add reactions from multiple sources
cobramod.add_reactions - Test reaction capability to carry a non-zero flux
cobramod.test_non_zero_flux - Add pathway to a model
cobramod.add_pathway - Automatic cross-references
cobramod.add_crossreferences - Testing for MEMOTE compliance
Check the documentation for more information.
To cite CobraMod, please use the following paper:
CobraMod: a pathway-centric curation tool for constraint-based metabolic models
@article{10.1093/bioinformatics/btac119,
author = {Camborda La Cruz, Stefano and Weder, Jan-Niklas and Töpfer, Nadine},
title = "{CobraMod: a pathway-centric curation tool for constraint-based metabolic models}",
journal = {Bioinformatics},
volume = {38},
number = {9},
pages = {2654-2656},
year = {2022},
month = {02},
issn = {1367-4803},
doi = {10.1093/bioinformatics/btac119},
url = {https://doi.org/10.1093/bioinformatics/btac119},
eprint = {https://academic.oup.com/bioinformatics/article-pdf/38/9/2654/43481008/btac119.pdf},
}CobraMod is licensed under the GPL-3 License. Read the LICENSE for more information.
CobraMod consists of a Python and JavaScript/TypeScript part. The following briefly describes all the commands to build a local version from the source files.
Since we need Node modules to build the Jupyter integrations con CobraMod, a package.json is included in the project. All dependencies contained in it can be made available locally using yarn. The foiling command can be used for this:
yarn install
We use Vite to build the Javascript part of the Jupyter integrations. So, we can use the following command to build a bundled version of the integrations.
yarn run vite build
This creates bundled JavaScript files under '/src/cobramod/static'.
You can contribute to CobraMod by cloning the repository and installing it in
developer mode and using the dev dependency group via pip:
pip install -e ".[dev]"
Optionally, a conda environment file is supplied (environment.yml). This file contains all the dependencies to ensure reproducibility of the package. We encourage pull requests!
To report bugs and suggestions, please create an issue using the corresponding tags at https://github.com/Toepfer-Lab/cobramod/issues.