Olfactory bulb microcircuits model with dual-layer inhibition (Gilra & Bhalla 2015)

 Download zip file 
Help downloading and running models
Accession:153574
A detailed network model of the dual-layer dendro-dendritic inhibitory microcircuits in the rat olfactory bulb comprising compartmental mitral, granule and PG cells developed by Aditya Gilra, Upinder S. Bhalla (2015). All cell morphologies and network connections are in NeuroML v1.8.0. PG and granule cell channels and synapses are also in NeuroML v1.8.0. Mitral cell channels and synapses are in native python.
Reference:
1 . Gilra A, Bhalla US (2015) Bulbar microcircuit model predicts connectivity and roles of interneurons in odor coding. PLoS One 10:e0098045 [PubMed]
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; Olfactory bulb main interneuron periglomerular GABA cell; Olfactory bulb main interneuron granule MC GABA cell;
Channel(s): I A; I h; I K,Ca; I Sodium; I Calcium; I Potassium;
Gap Junctions:
Receptor(s): AMPA; NMDA; Gaba;
Gene(s):
Transmitter(s): Gaba; Glutamate;
Simulation Environment: Python; MOOSE/PyMOOSE;
Model Concept(s): Sensory processing; Sensory coding; Markov-type model; Olfaction;
Implementer(s): Bhalla, Upinder S [bhalla at ncbs.res.in]; Gilra, Aditya [aditya_gilra -at- yahoo -period- com];
Search NeuronDB for information about:  Olfactory bulb main mitral GLU cell; Olfactory bulb main interneuron periglomerular GABA cell; Olfactory bulb main interneuron granule MC GABA cell; AMPA; NMDA; Gaba; I A; I h; I K,Ca; I Sodium; I Calcium; I Potassium; Gaba; Glutamate;
## USAGE:
## python2.6 generate_firefiles_gran_baseline.py <invitro|noresp>
## no extra arg for invivo, use extra arg for invitro and noresp options

import sys
sys.path.extend(['..','../networks'])

from networkConstants import * # for gran_spines
from stimuliConstants import *
from neuro_utils import *
    
from pylab import * # part of matplotlib that depends on numpy but not scipy

from generate_firerates_odors import *
from generate_firefiles_odors import *

## granule baseline firing - write files
def gran_files(mit_base_rate, filebase, invitro_str):
    ### Seed only if called directly, else do not seed.
    ### Also seeding this way ensures seeding after importing other files that may set seeds.
    ### Thus this seed overrides other seeds.
    seed([100.0]) ##### Seed numpy's random number generator.

    for i in range(MAXNUMAVG_GRANS):
        firefilename = filebase+\
            '/firetimes_gran_baseline'+invitro_str+'_'+str(i)+'.txt'
        firefile = open(firefilename,'w')
        for num in range(num_gran_baseline_files):            
            #### for each spine on the granule,
            #### make a mitral fire at a baseline rate (slightly varied randomly).
            #ornstimvector_merged = []
            ## there will be at least one non-baseline synapse connected already.
            #for i in range(gran_spines-1):
            #    mit_base_f_rand = uniform(mit_base_rate*0.75,mit_base_rate*1.25)
            #    ornstimvector = poissonTrain(MAXRUNTIME+SETTLETIME,\
            #        mit_base_f_rand,REFRACTORY) # from moose_utils.py
            #    ornstimvector_merged.extend(ornstimvector)
            #ornstimvector_merged.sort()
            #firefile.write(' '.join([str(t) for t in ornstimvector_merged])+'\n')
            
            ### instead of merging and sorting large number of granules,
            ### (smaller variance due to averaging);
            ### just generate the full baseline excitation to granule
            ### with frate as gaussian/normal with variance ~= mean
            ### roughly matches Carleton (in vitro) & Cang&Isaacson (in vivo)
            gran_base_rate = mit_base_rate*(gran_spines-1)
            gran_base_f_rand = normal(gran_base_rate,sqrt(gran_base_rate))
            if gran_base_f_rand<0: gran_base_f_rand = 0
            ornstimvector = poissonTrain(MAXRUNTIME+SETTLETIME,\
                gran_base_f_rand,REFRACTORY) # from moose_utils.py
            firefile.write(' '.join([str(t) for t in ornstimvector])+'\n')
            
        firefile.close()
        print "wrote ", firefilename

def gran_files_resp(filebase, extrastr, weight, showfig):
    ### Seed only if called directly, else do not seed.
    ### Also seeding this way ensures seeding after importing other files that may set seeds.
    ### Thus this seed overrides other seeds.
    seed([100.0]) ##### Seed numpy's random number generator.

    delay = delay_mean
    risetime = risetime_mean
    duration = duration_mean
    ######### get double sigmoid params
    spread2_factor = 4.0 # not used presently
    spread1, center2, spread2 = \
        compute_dblsigmoid_params(risetime, duration, spread2_factor)
    #spread2 = spread2_factor*spread1
    ########## shift curve to include latency
    ## I had taken center1 = 0.0, now shift the curve
    ## so that t=0 is actually where first sigmoid is 0.05*peak
    ## and then add the delay / latency to it!
    offset = - invert_dblsigmoid(0.0, spread1, center2, spread2, 0.05) + delay
    odorparamsR_e = [ gran_baseline_for_resp_tuning ]
    odorparamsR_e.extend( [offset, spread1, center2+offset, spread2] )

    ## inhibition kicks in 200ms later, delay_mean above is 154ms
    delay = 200e-3+delay_mean # ms
    risetime = risetime_mean
    duration = duration_mean/1.5
    ######### get double sigmoid params
    spread2_factor = 4.0 # not used presently
    spread1, center2, spread2 = \
        compute_dblsigmoid_params(risetime, duration, spread2_factor)
    #spread2 = spread2_factor*spread1
    ########## shift curve to include latency
    ## I had taken center1 = 0.0, now shift the curve
    ## so that t=0 is actually where first sigmoid is 0.05*peak
    ## and then add the delay / latency to it!
    offset = - invert_dblsigmoid(0.0, spread1, center2, spread2, 0.05) + delay
    odorparamsR_i = [ gran_baseline_for_resp_tuning ]
    odorparamsR_i.extend( [offset, spread1, center2+offset, spread2] )

    kernelR = getkernel(odorparamsR_e,odorparamsR_i)
    frate = float(weight) * (
        receptorFiringRate(0, 1.0, 0, \
            kernelR, kernelR, kernelR) \
            + gran_baseline_apart_from_resp_tuning
        )
    figure()
    plot(firingtsteps, frate, color=(1,0,0), marker=',')
    ylim(0,60)
    ## Take the last respiration period from the end and integrate
    lastfrate = frate[-int(RESPIRATION/FIRINGFILLDT):]
    frateavg = sum([fratei*FIRINGFILLDT for fratei in lastfrate]) / RESPIRATION
    print "Average firing rate for respiratory tuned response is",frateavg
    firefilename = filebase+'/firetimes_gran_baseline'+extrastr+'_'
    ## RUNTIME is defined in simset_odor.py imported via generate_firefiles_odors.py
    for i in range(MAXNUMAVG_GRANS):
        ## write_odor_files() is in generate_firefiles.py
        ornstimvector_merged = write_odor_files(num_gran_baseline_files,
            frate, firefilename+str(i), RUNTIME, firingtsteps, 
            vary=(frateavg,sqrt(frateavg)) )
    if showfig:
        figure()
        ratebins = [rate/float(num_gran_baseline_files)\
            for rate in plotBins(ornstimvector_merged, respbins, RUNTIME, SETTLETIME)]
        plot(tlist, ratebins, marker=',')
        show()

if __name__ == "__main__":
    ## seed for every function separately

    filebase = '../firefiles/firefiles_baseline'
    #filebase = '../firefiles/firefiles_whitenoise'
    #filebase = '../firefiles/firefiles_variedinh'
    
    if len(sys.argv)>1:
        arg1 = sys.argv[1]
        if arg1=='invitro':
            ### granule baseline firing in vitro
            ### gran_files has a loop of gran_spines iterations
            ### that effectively makes firing rate = mit_base_f_invitro*gran_spines
            gran_files(mit_base_f_invitro, filebase, '_invitro')
        elif arg1=='noresp':
            ### granule baseline firing in vivo :
            ### for activity-dep inhibition & random pulses for tracheotomized rat
            ### constant rate
            ### gran_files has a loop of gran_spines iterations
            ### that effectively makes firing rate = mit_base_f*gran_spines
            gran_files(mit_base_f, filebase, '_noresp')
            gran_files(extraexc_factor*mit_base_f, filebase, '_noresp_extra')
        else:
            print "Unrecognized param",arg1
    else:
        ### In vivo, we have a constant baseline firing and a respiratory tuned firing on top of it,
        ### since average mitral response is respiration tuned - see network_constants.py
        ### mean is roughly 34-35Hz when the weight (2nd param below) is 1.0
        gran_files_resp(filebase, '', 1.0, showfig=False)
        gran_files_resp(filebase, '_extra', extraexc_factor, showfig=False)