% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@INPROCEEDINGS{Jiang:890909,
author = {Jiang, Han-Jia and van Albada, Sacha},
title = {{A} {M}ulti-layer {M}icrocircuit {M}odel of {S}omatosensory
{C}ortex with {M}ultiple {I}nterneuron {C}lasses},
school = {University of Cologne},
reportid = {FZJ-2021-01240},
year = {2021},
abstract = {Three major classes of GABAergic inhibitory interneurons
play critical and distinct roles in the regulation of
cortical network dynamics and signal processing [1, 2]. The
mechanisms of regulation are linked to the anatomical and
physiological diversity of these interneurons. While
experimental studies provide realistic observations,
neuronal network simulations with parameters from
experimental data can aid in direct exploration of the
mechanisms and conveniently test hypotheses. Therefore, we
develop a microcircuit model of mouse somatosensory (barrel)
cortex incorporating three major classes of interneurons as
a tool to study cortical network dynamics and sensory signal
processing.The simulation software NEST 2.16.0 [3] is used
to create a multi-layer (L2/3, L4, L5, and L6) cortical
microcircuit model adapted from [4]. The neuron model is the
leaky-integrate-and-fire neuron model with exponentially
decaying post-synaptic currents (PSC). The network includes
populations of excitatory (Exc) cells and three classes of
interneurons: parvalbumin (PV), somatostatin (SOM) and
vasoactive intestinal peptide (VIP) cells. The
layer-specific numbers of each cell type are based on the
data of a mouse barrel column [5, 6]. Cell-type-specific
membrane parameters are according to the measurements in
[7]. Probabilities of recurrent connections are determined
from experimental data of paired recordings or algorithmic
estimates by [8]. Each neuron receives optimized background
inputs with cell-type-specific numbers of connections. Two
hundred thalamic cells are created and connected to Exc and
PV cells with layer-specific probabilities from [9] to test
the network responses to a 10-ms transient thalamic input of
100 spikes/s. The weights of excitatory recurrent
connections are set according to local intracortical unitary
excitatory postsynaptic potentials (uEPSPs) [5], while those
of inhibitory connections are multiplied by a relative
inhibitory strength (g). The background and thalamic inputs
are only excitatory, with weights according to
thalamocortical uEPSPs [10]. Synaptic short-term
plasticities (STPs) are included in the recurrent
connections according to the Tsodyks model [11]. The STP
parameters of different connections are individually fitted
to match the experimentally measured STPs.The ground-state
firing rates of the populations in the optimized model are
comparable to those of in vivo data from [12], although some
deviations remain (Figure 1A). The lower firing rates of VIP
cells and Exc cells in L5 and L6 compared to the in vivo
data are possibly due to the incomplete STP and connectivity
data (data not shown), respectively. The changes of firing
rates in L2/3 in response to activation of PV, SOM, and VIP
cells show their respective roles of inhibition and
disinhibition (Figure 2A), as observed in most experimental
studies. Over a range of external input and relative
inhibitory strengths, the model is able to fulfill the
criteria on firing rates, spiking irregularity, and pairwise
correlations of spike counts derived in [13] (Figure 1B). In
addition, the model with STPs shows clearer multi-layer
spiking responses to simulated transient thalamic input as
compared to a version without STPs (Figure 2B). Further
mechanistic analysis of the model may reveal the mechanisms
behind these results and explore the specific roles of
different interneuron types in state-dependent modulation of
sensory signal processing. Specifically, with the
experimental data of the thalamic input responsible for the
whisker sensation [9], this model is useful in simulating
somatosensory inputs and provides a tool to analyze the
roles of interneurons in its regulation and signal
processing.},
month = {Feb},
date = {2021-02-01},
organization = {The 5th HBP Student Conference,
virtual (virtual), 1 Feb 2021 - 4 Feb
2021},
subtyp = {After Call},
cin = {INM-6 / IAS-6 / INM-10},
cid = {I:(DE-Juel1)INM-6-20090406 / I:(DE-Juel1)IAS-6-20130828 /
I:(DE-Juel1)INM-10-20170113},
pnm = {5231 - Neuroscientific Foundations (POF4-523) / HBP SGA3 -
Human Brain Project Specific Grant Agreement 3 (945539)},
pid = {G:(DE-HGF)POF4-5231 / G:(EU-Grant)945539},
typ = {PUB:(DE-HGF)24},
url = {https://juser.fz-juelich.de/record/890909},
}