Integrate and fire with voltage dependent conductances

The basics of how to develop, test, and use models.
Post Reply
rchaudhuri

Integrate and fire with voltage dependent conductances

Post by rchaudhuri »

Hey

I'm new to Neuron, and I'm trying to build a network of integrate and fire neurons. None of the provided artificial cell types are what I need. I have AMPA, NMDA and GABA synapses in my model. The NMDA synapses have a voltage dependent conductance which seems to mean that I'll need to update every time step.

Should I try to write a single POINT_PROCESS that updates every time step, use this for my cells and then connect them together using NetCon?

Thanks,
Rishi
ted
Site Admin
Posts: 5810
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Integrate and fire with voltage dependent conductances

Post by ted »

rchaudhuri wrote:The NMDA synapses have a voltage dependent conductance which seems to mean that I'll need to update every time step.
If synaptic mechanisms involve a conductance change, you are forced to use numerical integration.

For a good starting point, see the NEURON implementation of the Vogels & Abbott model with COBA (COnductance BAsed) integrate & fire cells, which is available from ModelDB:
http://senselab.med.yale.edu/modeldb/Sh ... odel=83319
Be sure to read the implementation notes in the README file
http://senselab.med.yale.edu/modeldb/Sh ... RON\README

AMPA and GABA-A synapse properties are easy to replicate with ExpSyn and Exp2Syn. For NMDA synapses you can start by copying the source for Exp2Syn (exp2syn.mod), then in the copy change the name of the mechanism, define a FUNCTION f that specifies the voltage-dependence of the conductance, and change this statement
g = B - A
in the BREAKPOINT block to
g = f(v) * (B - A)
Depending on how much mecahnistic detail you want to include, you might want f to include magnesium effects. Unless there is a good reason to have a formal state variable that represents extracellular Mg concentration, it is most expedient to declare a PARAMETER called mg and include mg in the formula for f.
rchaudhuri

Re: Integrate and fire with voltage dependent conductances

Post by rchaudhuri »

Thank you. That makes a lot more sense than the way I was trying to do it.

Rishi
Post Reply