A detailed data-driven network model of prefrontal cortex (Hass et al 2016)

 Download zip file 
Help downloading and running models
Data-based PFC-like circuit with layer 2/3 and 5, synaptic clustering, four types of interneurons and cell-type specific short-term synaptic plasticity; neuron parameters fitted to in vitro data, all other parameters constrained by experimental literature. Reproduces key features of in vivo resting state activity without specific tuning.
1 . Hass J, Hertäg L, Durstewitz D (2016) A Detailed Data-Driven Network Model of Prefrontal Cortex Reproduces Key Features of In Vivo Activity. PLoS Comput Biol 12:e1004930 [PubMed]
Model Information (Click on a link to find other models with that property)
Model Type: Realistic Network;
Brain Region(s)/Organism: Prefrontal cortex (PFC);
Cell Type(s): Abstract integrate-and-fire adaptive exponential (AdEx) neuron;
Gap Junctions:
Receptor(s): GabaA; AMPA; NMDA;
Simulation Environment: C or C++ program; MATLAB;
Model Concept(s): Activity Patterns; Methods; Laminar Connectivity;
Implementer(s): Hass, Joachim [joachim.hass at zi-mannheim.de]; Hertäg, Loreen [loreen.hertaeg at tu-berlin.de]; Durstewitz, Daniel [daniel.durstewitz at plymouth.ac.uk];
Search NeuronDB for information about:  GabaA; AMPA; NMDA;
%% Worksheet runing the simulations and creating a raster plot  of the 
%% resulting spike train

% Main output: 
% - STMtx: Spike times in ms, (cell array, one cell per neuron)
% - V: Membrane potential in mV, (cell array, one cell per neuron)
% - T: Simulation time in ms (vector)

% -----------------  1) Prepare simulations  ------------------

% Set paths

% Compile MEX file (only needed at first run, and when IDNet.c is changed)
mex IDNet.c

% Set overall simulation parameters
N=1000;         % Number of neurons
M = 1;          % Number of input neurons
Str=1;          % Number of columns/stripes
SimTim=1000;    % Simulation time in ms
T_skip=500;     % Initial part of the spike train to skip for analysis in ms

sEE=1; sIE=1; sEI=1; sII=1;                         % Synaptic weight scales
pEE = 1; pEI = 1; pIE = 1; pII = 1; pE=1; pI=1;     % Connectivity scales
I = zeros(1,14);                                    % Background input 
I(1) = 250;                                         % I_ex
I(8) = 250;
I(2:7) = 200;                                       % I_inh
I(9:14) = 200;

% Compute complete simulation parameter set and construct file name
SimPar = ConfigIDNet(N,M,Str,SimTim,I,[sEE sIE sEI sII],[pEE pEI pIE pII pE pI]);
filename_1=['PFC_' num2str(I(1)) '_' num2str(I(2)) '_' num2str(N) 'N_S' num2str(Str)];
filename_2=['_s_' num2str(sEE*1) '_' num2str(sIE*1) '_' num2str(sEI*1) '_' num2str(sII*1)];
filename_3=['_p_' num2str(pEE) '_' num2str(pEI) '_' num2str(pIE) '_' num2str(pII) '_' num2str(pE) '_' num2str(pI) '_' num2str(SimTim) 'ms'];
filename=[filename_1 filename_2 filename_3];

% ---------  2)  Perform simulation or load existing data  ------------
if ~exist(filename,'file') && ~exist([filename '_all.mat'], 'file')
    SimPar.fnOut=[filename '_all'];
    SimPar.CtrPar(2) = SimTim;
    SimPar.ViewList = 1;%1:sum(SimPar.NTypes);          % Set of neurons to record the membrane potential from
    SimPar.NeuronGroupsSaveArray=[];%SimPar.ViewList;   % Set to view list to monitor additional variables
    disp(['Simulation time:' num2str(t) 's'])
    load([filename '_all'])

% --------  3) Display spike trains  ---------
STMtx=cellfun(@(x) x(x>T_skip),STMtx,'UniformOutput',false);

for ii=1:length(STMtx)
    hold on
    plot(STMtx{ii}, ii*ones(length(STMtx{ii}),1), 'b.')

% (c) 2016 J. Hass, L. Hertaeg and D. Durstewitz,
% Central Institute of Mental Health, Mannheim University of Heidelberg 
% and BCCN Heidelberg-Mannheim

Loading data, please wait...