Olfactory bulb network model of gamma oscillations (Bathellier et al. 2006; Lagier et al. 2007)

 Download zip file 
Help downloading and running models
Accession:91387
This model implements a network of 100 mitral cells connected with asynchronous inhibitory "synapses" that is meant to reproduce the GABAergic transmission of ensembles of connected granule cells. For appropriate parameters of this special synapse the model generates gamma oscillations with properties very similar to what is observed in olfactory bulb slices (See Bathellier et al. 2006, Lagier et al. 2007). Mitral cells are modeled as single compartment neurons with a small number of different voltage gated channels. Parameters were tuned to reproduce the fast subthreshold oscillation of the membrane potential observed experimentally (see Desmaisons et al. 1999).
Reference:
1 . Bathellier B, Lagier S, Faure P, Lledo PM (2006) Circuit properties generating gamma oscillations in a network model of the olfactory bulb. J Neurophysiol 95:2678-91 [PubMed]
2 . Lagier S, Panzanelli P, Russo RE, Nissant A, Bathellier B, Sassoè-Pognetto M, Fritschy JM, Lledo PM (2007) GABAergic inhibition at dendrodendritic synapses tunes gamma oscillations in the olfactory bulb. Proc Natl Acad Sci U S A 104:7259-64 [PubMed]
3 . Bathellier B, Lagier S, Faure P, Lledo PM (2006) Corrigendum for Bathellier et al., J Neurophysiol 95 (4) 2678-2691. J Neurophysiol 95:3961-3962
Citations  Citation Browser
Model Information (Click on a link to find other models with that property)
Model Type: Realistic Network;
Brain Region(s)/Organism: Olfactory bulb;
Cell Type(s): Olfactory bulb main mitral GLU cell;
Channel(s): I Na,p; I Na,t; I A; I K;
Gap Junctions:
Receptor(s): GabaA;
Gene(s):
Transmitter(s):
Simulation Environment: C or C++ program;
Model Concept(s): Oscillations; Delay; Olfaction;
Implementer(s):
Search NeuronDB for information about:  Olfactory bulb main mitral GLU cell; GabaA; I Na,p; I Na,t; I A; I K;
/*************************************************************************/
/**********             	   CmpprtmntRk4.cpp              *************/
/*************************************************************************/
/****                                                                 ****/
/****             Computing method  and Init (intialisation)          ****/
/****                   for the class Compartment                     ****/
/****                                                                 ****/
/*************************************************************************/





#include<iostream.h>
#include "CmprtmntRk4.h"
#include "CurrentRk4.h"
#include <math.h>
#include "NoiseSource.h"

/* ***************** Integrate stoch. synaptic events ********************* */
real GetX(real GI, real X, real P, real tau1, real tau2, real gMax, real MaxG, real dt){
    real Ev=0;
    if((1+Var())/2<P-floor(P)){
        Ev=1+floor(P);
    }
    X += dt*(1/tau2*GI-1/tau2*Ev*MaxG/gMax);
    return X;
}

real GetGI(real GI, real X, real tau1, real tau2, real dt){
    GI+=dt*(-(1/tau1+1/tau2)*GI-1/tau1*X);
    return GI;
}


/* ***************************   Constructor   ************************** */
Compartment::Compartment( void )
{
	// set variables to default values
	// LATER: get these from some preferences thing somewhere
	Gm  = 1;
	EGm = Gm * -0.070;
	Cm = 1;
	V[0] = V[1] = 0;
	X=0;
	GI=0;
   
}

/* ****************************  Initiation  *************************** */

void Compartment::Init( const real dt )
{
	//Sets the queue to resting potential : -70 mV
    for(int m=0; m<200; m++){Memory[m]=0;}
    dV=0;
    gMaxI=GetgMax(dt);
    X=0;
	GI=0;	
	
}


/* ****************************  General step  *************************** */

void Compartment::Step( const real dt )
{
	// This is the routine which actually calculates the new value of V.
	// Runge Kutta last step
	
	
	
	
    int curIdx = itsMaster->GetCurIdx();
	V[!curIdx] = V[curIdx] + Vk1/6 + Vk2/3 + Vk3/3 + Vk4/6; 
	dV=Vk1/6 + Vk2/3 + Vk3/3 + Vk4/6;
	
	
	for(int m=200-2; m>=0; m-- ){Memory[m+1]=Memory[m];}
	
    if (V[!curIdx]>-0.03 && V[curIdx]<=-0.03) {Memory[0]=1;}
	else {Memory[0]=0;}
	
	
}


/* ***************** Runge Kutta step function definition **************** */



                   /************ Step 1  **************/

void Compartment::Stepk1( const real dt )
{
    real Ak1 = EGm, Bk1 = Gm;
    
    for (CurrentNode *ln = itsCurrentList; ln; ln = ln->itsNext) {
		// note: this is a lot of dereferencing... can this be streamlined more?
		Ak1 += ln->itsCurrent->GetEG();
		Bk1 += ln->itsCurrent->G;
		
    }
    
    Ak1 += EI*GI;
    Bk1 += GI;
    
   int curIdx = itsMaster->GetCurIdx();
	Vk1 = -1/Cm *( V[curIdx]*Bk1 - Ak1)*dt;       
}

                   /************ Step 2  **************/

void Compartment::Stepk2( const real dt )
{
    
    real Ak2 = EGm, Bk2 = Gm;
    P=P0;
    for (CurrentNode *ln = itsCurrentList; ln; ln = ln->itsNext) {
		Ak2 += ln->itsCurrent->GetEGk1();
		Bk2 += ln->itsCurrent->Gk1;
		P+= ln->itsCurrent->GetPr();
    }
    
    GI=GetGI(GI,X, tau1, tau2, dt);
    X=GetX(GI,X,P, tau1, tau2, gMaxI, MaxGI,dt);
    Ak2 += EI*GI;
    Bk2 += GI;
    int curIdx = itsMaster->GetCurIdx();
    Vk2 = -1/Cm *((V[curIdx]+Vk1/2)*Bk2 - Ak2)*dt; 
}

                   /************ Step 3  **************/

void Compartment::Stepk3( const real dt )
{
    real Ak3 = EGm, Bk3 = Gm;
    for (CurrentNode *ln = itsCurrentList; ln; ln = ln->itsNext) {
		
		Ak3 += ln->itsCurrent->GetEGk2();
		Bk3 += ln->itsCurrent->Gk2;
    }
    Ak3 += EI*GI;
    Bk3 += GI;
     int curIdx = itsMaster->GetCurIdx();
     Vk3 = -1/Cm * ((V[curIdx]+Vk2/2)*Bk3 - Ak3)*dt;    
}

                   /************ Step 4  **************/

void Compartment::Stepk4( const real dt )
{
     real Ak4 = EGm, Bk4 = Gm;
    for (CurrentNode *ln = itsCurrentList; ln; ln = ln->itsNext) {
		
		Ak4 += ln->itsCurrent->GetEGk3();
		Bk4 += ln->itsCurrent->Gk3;
    }
    Ak4 += EI*GI;
    Bk4 += GI;
     int curIdx = itsMaster->GetCurIdx();    
     Vk4 = -1/Cm * ((V[curIdx]+Vk3)*Bk4 - Ak4)*dt;   
}