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]
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;
import sys, pickle

sys.path.extend(["..","../networks","../simulations","../generators","../analysis"])
from networkConstants import *
from stimuliConstants import *
from simset_odor import *
from average_odor_morphs import get_filename
from sim_utils import rebin
from pylab import * # part of matplotlib that depends on numpy but not scipy

from data_utils import * # has axes_labels()

## USAGE: python movie_overlap_plots.py

RUNTIME = REALRUNTIME + SETTLETIME

## time points for the firing rate which is read from a pickled file
firingtsteps = arange(0,RUNTIME+1e-10,FIRINGFILLDT)# include the last RUNTIME point also.
lastrespstart = -int(RESPIRATION/FIRINGFILLDT)

stimseed = 844.0
odornum = 0 # 5 for odorA, 0 for odorB
numgloms = 3
## inh_options = [ (no_singles,no_joints,no_lat,no_PGs,varyRMP), ... ]
inh = (False,False,False,False,False)
resultsdir = '../results/odor_morphs'

## load in the ORN input firing rates
fname = '../generators/firerates/firerates_2sgm_'+str(stimseed)+'.pickle'
f = open(fname,'r')
frateOdorList,fratePulseList,randomPulseList, \
randomPulseStepsList,randomResponseList,kernels \
= pickle.load(f)
f.close()

## load in the mit responses
netseed = stimseed
filename, switch_strs \
    = get_filename(netseed,stimseed,inh,numgloms,\
        None,None,None,resultsdir)
f = open(filename,'r')
(mitral_responses_list,mitral_responses_binned_list) = pickle.load(f)
f.close()

fps = 30
## The simulation is played at 1/10x i.e. 0.5 s (RESPIRATION) over 5.0 s.
## It's actually played at 1/50x, but I'll put each plot for 5 frames.
NUMFRAMES = 5*fps
## We need our bindt to be: (=1/300 s)
BINDT = RESPIRATION/float(NUMFRAMES)
#BINDT = FIRINGFILLDT # 1 ms bins, same as FIRINGFILLDT
NUMBINS = int(RESPIRATION/BINDT) # num bins for 1 resp cycle of 0.5 s
## RASTER WON'T WORK -- for single trial I used mpirank=0,
## no correspondence with multi-trial data used here.
## Also should not use trial-averaging below for multi-trial when RASTER = True.
RASTER = False
## moving average window of 0.1 s or BINDT if RASTER is False/True
if RASTER: BINAVGWIDTH = BINDT
else: BINAVGWIDTH = 0.1
mitral_responses_binned_list = \
    rebin(mitral_responses_list, numbins=NUMBINS,\
        bin_width_time=BINAVGWIDTH, numresps=1)
numavgs = len(mitral_responses_list)
mitral_responses_avg = mean(mitral_responses_binned_list, axis=0)
mitral_responses_std = std(mitral_responses_binned_list, axis=0)

orntimes = arange(FIRINGFILLDT/2.0,RESPIRATION,FIRINGFILLDT)
plottimes = arange(BINDT/2.0,RESPIRATION,BINDT)

def plot_ORN_frates(axlist,lasttime):
    """ 5 for odor A, 0 for odor B.
    Take only the last respiration cycle.
    """
    for glomnum in [1,0,2]:
        ax = axlist[glomnum]
        frate = frateOdorList[glomnum][odornum]
        ornlasttimeidx = int(time/FIRINGFILLDT)+1
        ax.plot(orntimes[:ornlasttimeidx],\
            frate[lastrespstart:lastrespstart+ornlasttimeidx],\
            ['b','r','g'][glomnum],linewidth=linewidth)
        beautify_plot(ax,drawxaxis=False,drawyaxis=False,xticks=[],yticks=[])
        if glomnum==2:
            add_scalebar(ax,matchx=False,matchy=False,hidex=True,hidey=True,\
                sizex=0.2,labelx='0.2 s',sizey=2,labely='2 Hz',\
                bbox_to_anchor=[0.8,-0.2],bbox_transform=ax.transAxes)
        ax.set_xlim(0,plottimes[-1])
        ymin,ymax=ax.get_ylim()
        ax.set_ylim(0,10.5) # 10.5 Hz for all plots, to get common scale-bar

def plot_mitresponses(axlist,lasttimeidx):
    for miti,mitnum in enumerate([2,1,0,4]):
        ax = axlist[miti]
        simresponse = mitral_responses_avg[odornum,mitnum]
        ax.errorbar(x=plottimes[:lasttimeidx],y=simresponse[:lasttimeidx],\
            color=['c','m','r','g'][miti],linewidth=linewidth)
        if mitnum==1:
            add_scalebar(ax,matchx=False,matchy=False,hidex=True,hidey=True,\
                sizex=0.2,labelx='0.2 s',sizey=8,labely='8 Hz',color='w',\
                bbox_to_anchor=[1.0,-0.4],bbox_transform=ax.transAxes)
        beautify_plot(ax,x0min=True,y0min=True,\
            drawxaxis=False,drawyaxis=False,xticks=[],yticks=[])
        ax.set_xlim(0,plottimes[-1])
        if RASTER: ax.set_ylim(0,1/BINAVGWIDTH) # spike raster - 1 spike in BINAVGWIDTH
        else: ax.set_ylim(10,40) # same ylim to set common scale bar

if __name__ == "__main__":
    for timeidx,time in enumerate(plottimes):
        ## MOVIE figures: plots to overlay on simulation movie.
        ## for 1280x720 image, use 300 dpi to convert to inches needed for figsize
        fig = figure(figsize=(4.667,2.4),\
            dpi=300,facecolor='none') # none is transparent
        #inaxlist = [fig.add_subplot(4,4,i) for i in [1,2,4]]
        #outaxlist = [fig.add_subplot(4,4,i) for i in [13,14,15,16]]
        ## axisbg doesn't work, savefig overrides it
        #outaxlist = [fig.add_subplot(4,6,i) for i in [20,15,16,23]]
        ## (left,right,top,bottom) in figure coordinates for the 4 subplots for mits 2,1,0,4 above.
        ## no points of any subplot should "collide" with that of any other subplot, hence the 0.2499.
        coordslist = [(0.225,0.375,0.2499,0.0),(0.35,0.5,0.5,0.25),(0.6,0.75,0.5,0.25),(0.75,0.9,0.2499,0.0)]
        outaxlist = []
        for i in range(4):
            gs = GridSpec(1,1)
            plotcoords = coordslist[i]
            gs.update(left=plotcoords[0],right=plotcoords[1],top=plotcoords[2],bottom=plotcoords[3])
            outaxlist.append( plt.subplot(gs[:,:]) )
        ## Upi suggested that I don't plot the ORN input.
        #plot_ORN_frates(inaxlist,time)
        plot_mitresponses(outaxlist,timeidx)
        fig_clip_off(fig)
        #fig.tight_layout() # doesn't work if using GridSpec
        fig.savefig('../figures/movie/ORN_mitresponses'+str(timeidx).rjust(10,'0')+'.png',\
            dpi=fig.dpi,transparent=True)
        print "Saved plot figure number",timeidx

Loading data, please wait...