Using NetStim with AlphaSynapses in existing model

Moderator: wwlytton

Post Reply
JohnAnderson
Posts: 8
Joined: Sun Jun 12, 2005 10:41 am
Location: University of North Florida

Using NetStim with AlphaSynapses in existing model

Post by JohnAnderson »

Good afternoon. I'm not sure this is the right place to post this, but here goes... I have a model in which AlphaSynapses are placed in various segments of a single model neuron. I would like to have the AlphaSynapses fire in response to a spike train rather than singly, and NetStim seems like it might be a good thing to use as part of the solution. This is a single model neuron, however, not a network. Is it possible to do this with a single neuron? If so, how? If not, is there another way to do it short of inserting multiple AlphaSynapses firing in succession? Thanks for any suggestions.

ja
ted
Site Admin
Posts: 6289
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Using NetStim with AlphaSynapses in existing model

Post by ted »

JohnAnderson wrote:I'm not sure this is the right place to post this
It really belongs in the "Modeling networks" discussion area, but for the time being it can stay here. After you have read the following, I'll move the thread to its appropriate location.
This is a single model neuron, however, not a network.
Your conceptual model may best be described as a "model of synaptic integration in a single cell," but its computational implementation uses NetStims, NetCons, and biophysical synaptic mechanisms, so from that perspective it's a network.

How to proceed:

Replace each AlphaSynape with an Exp2Syn.
Make sure that each Exp2Syn's time constants are set equal to the tau parameter of the AlphaSynapse it replaces.
Drive each Exp2Syn with events generated by one or more NetStims, as appropriate for the activation pattern you have in mind. A few examples:
1. All Exp2Syns are to be activated simultaneously. Only one NetStim is needed. There will be one NetCon for each Exp2Syn; each NetCon will use the NetStim as its event source, and its target will be one of the Exp2Syns.
2. All Exp2Syns are activated at times that are completely independent of each other. For each Exp2Syn there will be a corresponding NetStim, and a NetCon that attaches the latter to the former.
3. A hybrid of cases 1 and 2: all Exp2Syns are activated synchronously by random events, and each Exp2Syn also receives its own stream of events that are independent of the events delivered to the other Exp2Syns. Set up the architecture described in case 1, and make sure that the "common" NetStim has its "random" parameter set to 1. Also create the additional NetStims and NetCons required for case 2.

If you need random event streams, you might want to read
Randomness in NEURON models
on the Documentation page at http://www.neuron.yale.edu/

If you're using precomputed event times read from a data file, instead of NetStims, you'll want to use the VecStim class, which is defined in nrn/examples/nrniv/netcon/vecevent.mod. Copy the three vecevent* files from that directory to an empty directory, compile vecevent.mod, then use NEURON to execute vecevent.hoc. Be sure to read vecevent.hoc. You may find related discussions about VecStim elsewhere in the Forum.
JohnAnderson
Posts: 8
Joined: Sun Jun 12, 2005 10:41 am
Location: University of North Florida

Re: Using NetStim with AlphaSynapses in existing model

Post by JohnAnderson »

I do want the synapses to be activated independently. Given the rather messy, nonmodular state of the hoc file (derived from Mainen & Sejnowski's model in the NEURON database) this will take some thought, but your answer points me in the right direction. Thanks very much.
Post Reply