PyCoTools¶
PyCoTools is a python package that was developed as an alternative interface into COPASI, simulation software for modelling biochemical systems. The PyCoTools paper can be found here and describes in detail the intentions and functionality of PyCoTools. There are some important differences between the PyCoTools version that is described in the publication and the current version. The first is that PyCoTools is now a python 3 only package. If using Python 2.7 you should create a virtual Python 3.6 environment using conda or virtualenv. My preference is conda. The other major difference is the interface to COPASI’s parameter estimation task which is described in the tutorials and examples.
Note
I am in the process of improving this documentation. Although still valid, content in the Tutorials page is quite old and is in the process of being replaced by content in the examples.
Installation¶
First make sure you use a Python 3.6 environment.
Warning
Using Python 3.7 or 3.8 will not work at this time due to dependency issues (which are unfortunately out of my control).
Then use:
$ pip install pycotools3
Remember to source activate
your python 3.6 environment if you need to.
To install from source:
$ git clone https://github.com/CiaranWelsh/pycotools3.git
$ cd pycotools3
$ python setup.py install
The procedure is the same in linux, mac and windows.
Troubleshooting¶
Pycotools3 is only supported in Python 3 to Python 3.6. If you are using Python 2.7 or Python 3.7 please create a new conda Python3.7 environment.
$ conda create -n py36 python=3.6
$ conda activate py36
$ pip install pycotools3
The same commands should work cross platform.
Troubleshooting¶
If you get errors when trying to build a model using pycotools3.model.loada()
Make sure you have installed Copasi and added the Copasi/bin directory to
the path variable.
People¶
PyCoTools has been developed by Ciaran Welsh in Daryl Shanley’s lab at Newcastle University.
Caveats¶
- Non-ascii characters are minimally supported and can break PyCoTools
- Do not use unusual characters or naming systems (i.e. A reaction name called “A -> B” will break pycotools)
- In COPASI we can have (say) a global quantity and a metaboltie with the same name because they are different entities. This is not supported in Pycotools and you must use unique names for every model component
Citing PyCoTools¶
If you made use of PyCoTools, please cite this article using:
- Welsh, C.M., Fullard, N., Proctor, C.J., Martinez-Guimera, A., Isfort, R.J., Bascom, C.C., Tasseff, R., Przyborski, S.A. and Shanley, D.P., 2018. PyCoTools: a Python toolbox for COPASI. Bioinformatics, 34(21), pp.3702-3710.
And also please remember to cite COPASI:
- Hoops, S., Sahle, S., Gauges, R., Lee, C., Pahle, J., Simus, N., Singhal, M., Xu, L., Mendes, P. and Kummer, U., 2006. COPASI—a complex pathway simulator. Bioinformatics, 22(24), pp.3067-3074.
and tellurium:
- Medley, J.K., Choi, K., König, M., Smith, L., Gu, S., Hellerstein, J., Sealfon, S.C. and Sauro, H.M., 2018. Tellurium notebooks—An environment for reproducible dynamical modeling in systems biology. PLoS computational biology, 14(6), p.e1006220.