This contains the models, functions and resulting data as used in: Roessert C, Dean P, Porrill J. At the Edge of Chaos: How Cerebellar Granular Layer Network Dynamics Can Provide the Basis for Temporal Filters It is based on code used for Yamazaki T, Tanaka S (2005) Neural modeling of an internal clock. Neural Comput 17:1032-58 1. Recompile if necessary. In ./cells/ifun execute: python setup.py build_ext --inplace python setup2.py build_ext --inplace python setup2b.py build_ext --inplace python setup2re.py build_ext --inplace 2. Run the simulations: Data will be saved to ./data Intermediate figure will be saved to ./figs/dump and ./figs/Pub Figures will be saved to: ./figs/Pub - To run a single instance of the one-population network execute: python Plots_Closedloop.py -o ifun -nompi - To plot FIGURE 2 python Plots_Closedloop.py -o fig1 -nompi - To plot FIGURE 3: python Plots_Closedloop.py -o fig2 -nompi - To plot FIGURE 4: python Plots_Closedloop.py -o fig3 -nompi - To plot FIGURE 5: python Plots_Closedloop.py -o fig3b -nompi - To plot FIGURE 6: python Plots_Closedloop.py -o fig4l -nompi - To plot FIGURE 7: python Plots_Closedloop.py -o fig51l -nompi - To plot FIGURE 8: python Plots_Closedloop.py -o fig52l -nompi - To plot FIGURE 9: python Plots_Closedloop.py -o figRe6 -nompi Running on a single core takes a very long time, it is suggested to use a cluster. Here is an example to run on a cluster using Sun Grid Engine: - To generate a single trace for e.g. Fig 6: qsub -v J=Plots_Closedloop.py,O=fig4lruntest -pe ompigige 64 -l rmem=32G -l mem=32G PBSinsigneo.sh - Run a trace with recurrent filter (Fig 9): qsub -v J=Plots_Closedloop.py,O=figRe6runtest -pe ompigige 64 -l rmem=32G -l mem=32G PBSinsigneo.sh All the code and data files that are used in the publication can be found here: https://github.com/croessert/ClosedLoopRoessertEtAl ** Terms of use: This code, "ClosedLoopRoessertEtAl", is a derivative of "internalclock" by Takeru Honda and Tadashi Yamazaki used under CC-BY (http://creativecommons.org/licenses/by/3.0/). The code of "internalclock" was downloaded from https://senselab.med.yale.edu/ModelDB/ShowModel.cshtml?model=115966 "ClosedLoopRoessertEtAl" is licensed under CC BY by Christian Rössert (christian.a [4t] roessert.de).