To view the publications and organisms associated with each model, please see the key.
A collection of M models downloaded from published studies, along with notebooks which parse and solve them. The notebooks can be viewed at nbviewer.
To parse these reconstructions into a computing M-models, install
cobrapy
version 0.4.0b1 or later, along with the optional dependcies libsbml and scipy.
To load the reconstructions from the Microsoft Excel format, the
pandas library is also necessary. Then run
the load_models.py script.
The resulting all_models.mat file can be used with both cobrapy and the
COBRA toolbox. This script also generates SBML files which use the SBML level 3
with the fbc package (version 2), which are placed in the "sbml3" directory.
As shown in the included notebooks, these models can all be used to compute rational solutions as well as loopless solutions. They also can be solved by the MATLAB cobra toolbox.
Some of the models, however, could not be parsed correctly into solving models. Still, they give consistent results between floating point and exact solvers
Please note that these models are not the official versions. Many of the models in here have been updated at the original source. This is merely a snapshot of models which were used in the analysis for this study.
All code in this repository is released into the public domain under the unlicense. Please see http://unlicense.org for details. The work may also be used under the terms of CC0.
The models themselves, however, are provided only for reproducing the results in the included scripts. All other uses must be consistent with the initial license the model was released under.