KNPsim

KNPsim is a implementation of the KNP framework (see Halnes et al., https://www.ncbi.nlm.nih.gov/pubmed/27820827), using the FEniCS software for finite element problems. KNP is a framework for solving ion concentration dynamics by assuming the Nernst-Planck equation and electroneutrality. Neuronal input is provided as point sources to the system. This code was used to produce the results presented in Solbrå et al. 2018: https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1006510

Installation

KNPsim is currently only supported on Linux (but it should work on macOS as well). Make sure that you have Anaconda installed on your system. First clone or download the repo, then navigate to the top directory of the repo in a terminal and enter the following commands:

conda create -n knpsim -c conda-forge fenics ipython h5py matplotlib scipy python=3
source activate knpsim
python setup.py install

Testing

Run the command pytest from the top directory. Make sure you have pytest installed (conda install pytest).

Reproducing the results of the paper

In order to reproduce the results of (TODO: future paper), run the script run_all_examples_and_gather_data.sh. Be advised that this might run for a few days on a normal computer. Afterwards, go to KNPsim/make_the_figures, open the notebooks and run all the cells in each notebook. Note that the figures have been formatted in Inkscape for publication purposes.

How to set up an electrodiffusion problem in knpsim

A script using knpsim will usually go through the following steps:

1. Create an instance of Geometry (will involve making or loading a mesh).
2. Create an instance of Simulator.
3. Create instances of the ion species.
4. (optional) Set up current point sources
5. Create an instance of Time_solver
6. Initialize the simulation by calling simulator.initialize_simulator()
7. (optional) Create an instance of State_saver, to save the result.
8. Run the simulation by calling time_solver.solve()