Grohar represents a computational tool for the visualisation of metabolic networks and their analysis using Matplotlib, networkx and cobra. Grohar was written in python together with pyqt GUI toolkit to provide a user friendly experience. Grohar is licensed under the Creative Commons Attribution license.
- Use Python 3.5+ with Spyder (get the whole package with Anaconda)
- main file = top.py
- graphical user interface is located within the file top.ui and can be modified easily with Qt Designer (included within the package pyqt-distutils)
You will need to install the following python packages - additional to the packages that are preinstalled with Anaconda:
- cobra 0.5.4 (pip install cobra==0.5.4),
- bioservices
- PyQt5
- msgpack
- python-libsbml
- bioservices
- plotly
- cobrababel
- pydot
- pydot-ng
You need to configure IPython in Spyder to open the Matplotlib graphs in a separate window. In Spyder go to Tools -- Preferences -- IPython console -- Graphics. Here you will find a setting entitled Graphics Backend, which should be set to Automatic (its default value is Inline).
Additional software that is also required:
- Gurobi
- conda config --add channels http://conda.anaconda.org/gurobi
- conda install gurobi
- You should also install the gurobi license for the solvers to work. The license and instructions are available at http://www.gurobi.com/academia/for-universities.
- Graphviz:
- http://www.graphviz.org/download/
- In Windows, add C:\Program Files (x86)\Graphviz2.38\bin to PATH
You are ready to go!
The user manual with some examples is avaible at the following link.
You can direct all your questions, comments and critiques to miha.moskon@fri.uni-lj.si.
Ivan Grohar was a Slovenian Impressionist painter. You can read more about his work on Wikipedia.
#!python