Cobrapy: A Solid Foundation for a Diverse Python Metabolic Modeling Community

When the COBRApy project began, its goals were defined as providing an intuitive, object-oriented interface for constraint-based metabolic modelling in Python that 1) can represent complex biological processes, 2) meets the computational challenges of next generation stoichiometric constraint-based models and omics data sets, 3) promotes related research via freely available software, and 4) provides an interoperable alternative to the COBRA toolbox for Matlab.

Today, we can safely say that COBRApy has achieved all of those goals and has become an established tool that both academic and industry users rely on. We want to raise awareness for the maturity of COBRApy and how it has enabled productive constraint-based modeling by publishing a stable 1.0 release. In this work, we not only present the many improvements that have happened since the start of the project, but also highlight the vibrant and diverse community of Python packages built on top of COBRApy that together provide a plethora of methods related to constraint-based reconstructions and modeling.

While navigating a diverse software landscape akin to Bioconductor or the Python package index can be daunting at first, the benefits of distributed development are more publicly available tools due to the creative freedom provided, fast access to implementations of latest methods, and more direct contact between users and authors. Last but not least, a healthy open source community that encourages contributions from anyone can help mitigate the problem of abandoned academic software. Over the course of its six years of existence, COBRApy has been under the stewardship of several people and has seen contributions from almost 30 authors without whom it would not be where it is today. Here, we aim to provide a solid overview of the packages and capabilities within the community built around COBRApy.