How to connect ARTIFICIAL_CELL with real cell

Moderator: wwlytton

Post Reply
neurite
Posts: 13
Joined: Fri Jun 22, 2007 2:06 pm
Location: Boston University

How to connect ARTIFICIAL_CELL with real cell

Post by neurite »

Hi,
I have a morphologically accurate cell which I've imported into neuron and I'd like to study the effects of synaptic noise on the cell. I need 1000 synapses on the cell and since using 1000 pre-synpatic cells as the source is very wasteful, I decided to make the source of the synapse a ARTIFICIAL_CELL. When I try to connect the ARTIFICIAL_CELL with the morphologically accurate cell with NetCon, I get the following error:

[code>]
NetCon pointer not associated with currently accessed section
Use section ... (&var(x)...) intead of ...(&section.var(x)...)
[/code]

The ARTICIFICAL_CELL is a point process, so it doesn't have a section that is accessible, so is it possible to create the network I mentioned above?
Below is my mod file and a simplified hoc file.

Code: Select all

NEURON {
	ARTIFICIAL_CELL pois_presyncell
	RANGE m
}

PARAMETER{
	freq = 1
}

ASSIGNED {
	m
	randflag
}

INITIAL {
	m = 0
}

NET_RECEIVE (w){ 
	fire()
}

PROCEDURE fire() {
    
	randflag = ceil((1000/freq)*(1/dt)*scop_random())
	if (randflag==((1000/freq)*(1/dt))){
			m = 1
		}
}

Code: Select all

create soma
soma{
	insert hh
	insert pas
}

objref syn
soma syn = new ExpSyn (0)
objref presyn
presyn = new pois_presyncell()
objref netcon
netcon = new NetCon(&presyn.m, syn, 1, 0, 1)
ted
Site Admin
Posts: 5786
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Post by ted »

An artificial spiking cell has a NET_RECEIVE block that calls net_event). Also, scop_random
samples from the uniform distribution over [0,1], so how would this PROCEDURE fire()
generate ISIs that follow a negexp distribution? Is there some reason not to use the built-in
NetStim class, which can generate events with ISIs that follow a negexp distribution?

Code: Select all

objref pre, nc
pre = new NetStim()
pre.noise = 1  // for negexp ISIs
// specify pre's start, number, interval
nc = new NetCon(pre, syn)
// specify nc's delay and weight
neurite
Posts: 13
Joined: Fri Jun 22, 2007 2:06 pm
Location: Boston University

Post by neurite »

Thanks a lot Ted! I didn't know there was a negexp ISI built in.
Post Reply