calculate IPSP by IPSC

Anything that doesn't fit elsewhere.
Post Reply
lcc
Posts: 2
Joined: Sat Sep 19, 2020 3:21 am

calculate IPSP by IPSC

Post by lcc »

Now I have a model which include excitatory synapses and inhibitory synapses(gabaa and gabab),and I can record the gabaa current,but now i need to calculate the IPSP induced by gabaa current,but i donot know if it is feasible and how to achieve it.
ted
Site Admin
Posts: 6286
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: calculate IPSP by IPSC

Post by ted »

It's feasible but if you want more specific help, you'll need to provide some information about the model that you currently have. Is it a conceptual model (hypothesis) or have you actually implemented a computational model? If it is a computational model, what is your implementation (source code) of GABA-A synaptic mechanism?
lcc
Posts: 2
Joined: Sat Sep 19, 2020 3:21 am

Re: calculate IPSP by IPSC

Post by lcc »

Thank you for your reply,and I do need your help.I recreate the model based on Encoding and retrieval in a model of the hippocampal CA1 microcircuit (Cutsuridis et al. 2009)https://senselab.med.yale.edu/ModelDB/S ... 123815,but I added chloride accumulation mechanism to study the effect of the decrease of gabaa current on the depolarization of pyramidal neuron
,so now my gabaa.mod as below:
TITLE GABAergic conductance with changing Cl- concentration

NEURON {
POINT_PROCESS gaba

USEION cl READ ecl WRITE icl VALENCE -1

NONSPECIFIC_CURRENT ihco3
RANGE tau1, tau2, g
RANGE P, HCO3e, HCO3i, i

RANGE icl, ihco3, ehco3, e
GLOBAL total
}

UNITS {
(mA) = (milliamp)
(nA) = (nanoamp)
(mV) = (millivolt)
(uS) = (micromho)
(mM) = (milli/liter)
F = (faraday) (coulombs)
R = (k-mole) (joule/degC)
}

PARAMETER {
tau1 =.1 (ms) <1e-9,1e9>
tau2 = 10 (ms) <1e-9,1e9>

HCO3e = 23 (mM) : extracellular HCO3- concentration
HCO3i = 12 (mM) : intracellular HCO3- concentration
P = 0.18 : HCO3/Cl relative permeability

celsius = 37 (degC)
}

ASSIGNED {
v (mV) : postsynaptic voltage

icl (nA) : chloride current = (1-P)*g*(v - ecl)
ihco3 (nA) : bicarb current = P*g*(v - ehco3)
i (nA) : total current generated by this mechanism
: = icl + ihco3
g (uS) : total conductance, split between bicarb (P*g)
: and chloride ((1-P)*g)
factor
total (uS)

ecl (mV) : equilibrium potential for Cl-
ehco3 (mV) : equilibrium potential for HCO3-

e (mV) : reversal potential for GABAR
}

STATE {
A (uS)
B (uS)
}

INITIAL {
LOCAL tp
total = 0
if (tau1/tau2 > .9999) {
tau1 = .9999*tau2
}
A = 0
B = 0
tp = (tau1*tau2)/(tau2 - tau1) * log(tau2/tau1)
factor = -exp(-tp/tau1) + exp(-tp/tau2)
factor = 1/factor

ehco3 = log(HCO3i/HCO3e)*(1000)*(celsius + 273.15)*R/F
e = P*ehco3 + (1-P)*ecl
}

BREAKPOINT {
SOLVE state METHOD cnexp

g = B - A

icl = (1-P)*g*(v-ecl)

ihco3 = P*g*(v-ehco3)
i = icl + ihco3
e = P*ehco3 + (1-P)*ecl

}

DERIVATIVE state {
A' = -A/tau1
B' = -B/tau2
}

NET_RECEIVE(weight (uS)) {
A = A + weight*factor
B = B + weight*factor
total = total+weight
And my chloride accumulation mechanism as below:
NEURON {
SUFFIX cldifus
USEION cl READ icl WRITE cli, ecl VALENCE -1
USEION hco3 READ hco3i, hco3o VALENCE -1
GLOBAL vrat :vrat must be GLOBAL
RANGE tau, cli0, clo0, egaba, delta_egaba, init_egaba, ehco3, ecl
}

DEFINE Nannuli 11

UNITS {
(molar) = (1/liter)
(mM) = (millimolar)
(um) = (micron)
(mA) = (milliamp)
(mV) = (millivolt)
FARADAY = (faraday) (10000 coulomb)
PI = (pi) (1)
F = (faraday) (coulombs)
R = (k-mole) (joule/degC)
}

PARAMETER {
DCl = 2 (um2/ms) : Kuner & Augustine, Neuron 27: 447
tau = 3000 (ms)
cli0 = 4.25 (mM)
clo0 = 135 (mM)
hco3i0 = 12 (mM)
hco3o0 = 23(mM)
P = 0.18
celsius = 37 (degC)

}

ASSIGNED {
diam (um)
icl (mA/cm2)
cli (mM)
hco3i (mM)
hco3o (mM)
vrat[Nannuli] : numeric value of vrat equals the volume
: of annulus i of a 1um diameter cylinder
: multiply by diam^2 to get volume per um length
egaba (mV)
ehco3 (mV)
ecl (mV)
init_egaba (mV)
delta_egaba (mV)
}

STATE {
: cl[0] is equivalent to cli
: cl[] are very small, so specify absolute tolerance
cl[Nannuli] (mM) <1e-10>
}


BREAKPOINT {
SOLVE state METHOD sparse
ecl = log(cli/clo0)*(1000)*(celsius + 273.15)*R/F
egaba = P*ehco3 + (1-P)*ecl
delta_egaba = egaba - init_egaba
}

LOCAL factors_done

INITIAL {
if (factors_done == 0) { : flag becomes 1 in the first segment
factors_done = 1 : all subsequent segments will have
factors() : vrat = 0 unless vrat is GLOBAL
}
cli = cli0
hco3i = hco3i0
hco3o = hco3o0
FROM i=0 TO Nannuli-1 {
cl = cli
}
ehco3 = log(hco3i/hco3o)*(1000)*(celsius + 273.15)*R/F
ecl = log(cli/clo0)*(1000)*(celsius + 273.15)*R/F
egaba = P*ehco3 + (1-P)*ecl
init_egaba = egaba
delta_egaba = egaba - init_egaba
}

LOCAL frat[Nannuli] : scales the rate constants for model geometry

PROCEDURE factors() {
LOCAL r, dr2
r = 1/2 : starts at edge (half diam), diam = 1, length = 1
dr2 = r/(Nannuli-1)/2 : full thickness of outermost annulus,
: half thickness of all other annuli
vrat[0] = 0
frat[0] = 2*r : = diam
FROM i=0 TO Nannuli-2 {
vrat = vrat + PI*(r-dr2/2)*2*dr2 : interior half
r = r - dr2
frat[i+1] = 2*PI*r/(2*dr2) : outer radius of annulus Ai+1/delta_r=2PI*r*1/delta_r
: div by distance between centers
r = r - dr2
vrat[i+1] = PI*(r+dr2/2)*2*dr2 : outer half of annulus
}
}

KINETIC state {
COMPARTMENT i, diam*diam*vrat {cl}
LONGITUDINAL_DIFFUSION i, DCl*diam*diam*vrat {cl}

~ cl[0] << ((icl*PI*diam/FARADAY) + (diam*diam*vrat[0]*(cli0 - cl[0])/tau)) : icl is Cl- influx
FROM i=0 TO Nannuli-2 {
~ cl <-> cl[i+1] (DCl*frat[i+1], DCl*frat[i+1])
}
cli = cl[0]
}
ted
Site Admin
Posts: 6286
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: calculate IPSP by IPSC

Post by ted »

Something to think about before you go any further:
D_Cl is 2 um2/ms. That means radial concentration gradients vanish rapidly in neurites with diameters on the order of 5 um or smaller. Much faster than the time course of GABA-A synaptic conductance change or transmembrane current. This means chloride accumulation mechanisms can treat cli as independent of distance from the surface of the cell, at least in dendrites.

Spatial discretization poses a special challenge in models that involve longitudinal diffusion of solutes. Electrical signals propagate much faster than chemical signals do, so values of nseg that would be fine for membrane currents and electrical potential will be way too coarse to capture the spatial variation of concentrations along the length of sections. You'll have to experiment with nseg to determine what values give sufficient accuracy without imposing excess computational burden.

A minor comment:
If
chloride current = (1-P)*g*(v - ecl)
bicarb current = P*g*(v - ehco3)
then "HCO3/Cl relative permeability" is equal to P/(1-P), not P.
i need to calculate the IPSP induced by gabaa current
First you need to be sure that your NMODL code is correct and has no inconsistencies of units (have you checked it with modlunit?). Then you have to come up with a model cell to which you can attach your GABA-A synapse. The selection of model cell and where to attach the synapse are important choices--you should have a good reason for selecting a particular model cell and synaptic location. You'll need to drive the synapse with an event using a Netcon whose weight[0] is equal to the experimentally determined peak conductance of a single GABA-A synapse. And you'll need to decide on the resting potential of the model cell and the reversal potential of your synaptic mechanism.
Post Reply