Simulating ion channel noise in an auditory brainstem neuron model (Schmerl & McDonnell 2013)

 Download zip file 
Help downloading and running models
" ... Here we demonstrate that biophysical models of channel noise can give rise to two kinds of recently discovered stochastic facilitation effects in a Hodgkin-Huxley-like model of auditory brainstem neurons. The first, known as slope-based stochastic resonance (SBSR), enables phasic neurons to emit action potentials that can encode the slope of inputs that vary slowly relative to key time constants in the model. The second, known as inverse stochastic resonance (ISR), occurs in tonically firing neurons when small levels of noise inhibit tonic firing and replace it with burstlike dynamics. ..." Preprint available at
1 . Schmerl BA, McDonnell MD (2013) Channel noise induced stochastic facilitation in an auditory brainstem neuron model Physical Review E 88:052722 [PubMed]
Model Information (Click on a link to find other models with that property)
Model Type: Neuron or other electrically excitable cell; Channel/Receptor;
Brain Region(s)/Organism: Auditory brainstem;
Cell Type(s): Cochlear nucleus bushy GLU cell; CN stellate cell; Ventral cochlear nucleus T stellate (chopper) neuron; Hodgkin-Huxley neuron;
Channel(s): I h; I Sodium; I Potassium;
Gap Junctions:
Gene(s): Kv1.1 KCNA1; Kv3.1 KCNC1;
Simulation Environment: MATLAB;
Model Concept(s): Bursting; Ion Channel Kinetics; Action Potentials; Methods; Noise Sensitivity; Bifurcation; Audition;
Search NeuronDB for information about:  Cochlear nucleus bushy GLU cell; I h; I Sodium; I Potassium;
clear all

%selection of deterministic or stochastic
IsDeterministic = 0;%set this to 1 for the SSE model, 0 for the deterministic model

%selection of model: only one of the following should be non-zero
IsHodgkin = 1;
IsRothmanII = 0;
IsRothmanI_II = 0;
IsRothmanI_C = 0;

%get parameters for selected model
if IsHodgkin
    %load parameters for Hodgkin-Huxley model
    Switch = 1;
    InputCurrent = 7.2; %can be modified as desired
elseif IsRothmanII
    %load parameters for Rothman-Manis Type II model
    Switch = 2;
    InputCurrent = 230; %can be modified as desired
elseif IsRothmanI_II
    %load parameters for Rothman-Manis Type I-II model
    Switch = 2;
    InputCurrent = 231; %can be modified as desired
elseif IsRothmanI_C
    %load parameters for Rothman-Manis Type I-C model
    Switch = 2;
    InputCurrent = 25.75; %can be modified as desired

%the user can decide to make some channel types stochastic and some deterministic if desired
NoiseSwitches = zeros(NumChannelTypes,1);
if IsDeterministic
    %deterministic model
    NoiseSwitches(:) = 0;
    %SSE model
    %for any channels that are to be deterministic, set the corresponding index to 0
    NoiseSwitches(:) = 1;

%simulation times, initial and input conditions
NumPoints = 1; %this is how many different channel numbers we want to run the sim for
NumRepeats = 1;
%NumChannelsToTry = round(logspace(3,8,NumPoints));
NumChannelsToTry = [10000];
dt = 0.01;
Min_t = 0;
Max_t = 400;
ts = [Min_t:dt:Max_t];
Num_t = length(ts);
OnsetTime = 0; %ms
OffsetTime = 0; %ms

%count the number of variables
StatesPerChannelType = zeros(NumChannelTypes,1);
TotalStates = 0;
for j = 1:NumChannelTypes
    NumStatesPerActivationVariable{j} = NumGatesPerActivationVariable{j} +1;
    StatesPerChannelType(j) = prod(NumStatesPerActivationVariable{j});
    TotalStates = TotalStates + StatesPerChannelType(j);
Num_Vars = 1 + NumActivationVars + TotalStates;
Params{1} = gs;
Params{2} = Es;
Params{3} = C;

%set up the input constant current, with onset and offset times
InputCurrents = zeros(Num_t,1);
for i = 2:Num_t
    if ts(i) < OnsetTime | ts(i) > Max_t-OffsetTime
        InputCurrents(i) = 0;
        InputCurrents(i) = InputCurrent;
%do the simulation for the desired number of repeats and number of channels
SpikeCount = zeros(NumPoints,NumRepeats);
InRefrac = 0;
TotalISICount = 0;
ISIs = zeros(30*NumRepeats,1);
for k = 1:NumRepeats
    for j = 1:NumPoints
        Solution = zeros(Num_Vars,Num_t);
        Conductances = zeros(NumChannelTypes,Num_t);
        Solution(1:1+NumActivationVars,1) = ICs;
        %set the number of channels of each type
        %can set arbitrarily different numbers for each type
        %could be better to do this in the params file
        NumChannelsEachType(1:NumChannelTypes) = NumChannelsToTry(j);            
        %do the simulation
        LastSpikeTime = -10^5;
        for i = 2:Num_t
            %solve the equations
            [Solution(:,i),Conductances(:,i)] = EulerMaruyama(Solution(:,i-1),dt,Switch,NoiseSwitches,Params,NumChannelTypes,ActivationVarsPerChannel,NumActivationVars,StatesPerChannelType,InputCurrents(i),NumChannelsEachType);
            %count spikes, using a rudimentary spike detector
            if InRefrac == 0 & Solution(1,i) > SpikeThreshold & Solution(1,i-1) < SpikeThreshold
                SpikeCount(j,k) = SpikeCount(j,k) + 1;
                InRefrac = 1;
                if LastSpikeTime > 0
                    TotalISICount = TotalISICount+1;
                    ISIs(TotalISICount) = ts(i)-LastSpikeTime;
                LastSpikeTime = ts(i);
            if InRefrac == 1 & Solution(1,i) < SpikeThreshold-5
                InRefrac = 0;

%plot the membrane potential as a function of time
f0= figure;hold on
box on
xlabel('Time (ms)','fontsize',14)
ylabel('Membrane potential (mV)','fontsize',14)

Loading data, please wait...