I have a Biologically realistic cell which i want to convert in Izhekevich type one. I looked at the code posted in one of the models (http://senselab.med.yale.edu/modeldb/s ... h\izh.mod). It says that we have to create a dummy variable vv for voltage instead of using NEURONS default v. I wanted to use NEURONS default v along with the compartment properties used in the biological model ( Length, diam, Cm etc). I have come up with the following code which gives the correct result for the Izhikevich model from the book "Dynamical Systems in Neuroscience" pg 282-283.
Code: Select all
NEURON {
POINT_PROCESS Izhi
NONSPECIFIC_CURRENT Iizhi
RANGE Iizhi
RANGE k, vr, vt, a, b, vpeak, c, d, uinit
RANGE Ifast
}
UNITS {
(mV) = (millivolt)
(nA) = (nanoamp)
(pA) = (picoamp)
(uS) = (microsiemens)
(nS) = (nanosiemens)
}
PARAMETER {
k = 0.7 (pA/mV2)
vr = -60 (mV)
vt = -40 (mV)
a = 0.03 (1/ms)
b = -2 (nS)
vpeak = 35 (mV)
c = -50 (mV)
d = 100 (pA)
uinit = 0 (pA)
}
ASSIGNED {
v (mV)
Iizhi (nA)
Ifast (nA)
}
STATE {
u (nA)
}
INITIAL {
net_send(0,1)
}
BREAKPOINT {
SOLVE recvars METHOD cnexp
Ifast = k*(v-vr)*(v-vt)
Iizhi = (-(Ifast - u))/1000
}
DERIVATIVE recvars {
u' = a*(b*(v-vr)-u)
}
NET_RECEIVE (void) {
if (flag == 1) {
WATCH (v > vpeak) 2
} else if (flag == 2) {
net_event(t)
v = c
u = u+d
}
}
My question is that will resetting NEURONS default v in NET_RECEIVE block cause problems for this model. It seems to work correctly but just wanted to know as we will be using this implementation to convert Biologically realistic multi-compartment models to this type.