Suggestion for dynamic synaptic inputs

The basics of how to develop, test, and use models.
Post Reply
ylzang
Posts: 27
Joined: Sun Mar 08, 2015 3:38 am

Suggestion for dynamic synaptic inputs

Post by ylzang »

I have built a cell model and now I want to do the in vivo simulations. When the animal is at resting condition, there are only spontaneous/background synaptic inputs. To make it simple, I assume there are 1000 synaptic contacts uniformly formed on the dendrite from another type of cells. At resting condition, then 1000 synapses will randomly activate at a very low rate. I can simulate resting condition by setting the activation time of each synapse uniformly distributed between 0 and 1000/Firing_rate_resting. Then I am confused how to do the following simulation. once the animal receives stimulus or have some movement, the activation frequency of the synapses will increase significantly in a dynamic way to code the sensory input or the movement. How should I implement this kind of synaptic inputs together with the resting state? Thanks.
ted
Site Admin
Posts: 6286
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Suggestion for dynamic synaptic inputs

Post by ted »

What do you need help with--attaching a bunch of synapses to a model cell, or making the presynaptic firing rates change in a way that represents the effect of a stimulus?
ylzang
Posts: 27
Joined: Sun Mar 08, 2015 3:38 am

Re: Suggestion for dynamic synaptic inputs

Post by ylzang »

ted wrote:What do you need help with--attaching a bunch of synapses to a model cell, or making the presynaptic firing rates change in a way that represents the effect of a stimulus?
Thanks Ted. The second one would be the question.
ted
Site Admin
Posts: 6286
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Suggestion for dynamic synaptic inputs

Post by ted »

To go any further, you will have to provide a mathematical description of the effect of a stimulus on presynaptic firing rate. It would be best if your description is based on experimental evidence.
ylzang
Posts: 27
Joined: Sun Mar 08, 2015 3:38 am

Re: Suggestion for dynamic synaptic inputs

Post by ylzang »

ted wrote:To go any further, you will have to provide a mathematical description of the effect of a stimulus on presynaptic firing rate. It would be best if your description is based on experimental evidence.
Yes, I am trying to simulate the cell spiking in vivo. It includes two stages, the resting state and the state after receiving stimulus (or locomotion).
At resting state, the cell receives spontaneous excitatory synaptic inputs (just ignore the inhibitory synaptic input to make it simple). These spontaneous synaptic inputs should be poisson process with low average firing rate.
Once I exert some sensory stimulus (or the animal has some locomotion), the cell's presynaptic input firing rate will temporarily increase significantly and linearly code the motion or the stimulus (I just make it simple here). I will send the image to your email. Sorry I don't know how to upload an image here. I guess for this case, I may have to record the presynaptic spike timing and then read it into NEURON?
Or for a first try, once the cell receives stimulus, the presynaptic input firing rate (signal) is like a sinusoidal wave FR = sin((t-stimulus_onset)/10), when t<stimulus_onset or t>stimulus_onset+10*pi. The background presynaptic firing rate is 0.2 Hz (noise).
ted
Site Admin
Posts: 6286
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Suggestion for dynamic synaptic inputs

Post by ted »

once the cell receives stimulus, the presynaptic input firing rate (signal) is like a sinusoidal wave FR = sin((t-stimulus_onset)/10), when t<stimulus_onset or t>stimulus_onset+10*pi.
sin((t-stimulus_onset)/10) is negative half of the time. Would you like to revise the expression for firing rate?
ylzang
Posts: 27
Joined: Sun Mar 08, 2015 3:38 am

Re: Suggestion for dynamic synaptic inputs

Post by ylzang »

ted wrote:
once the cell receives stimulus, the presynaptic input firing rate (signal) is like a sinusoidal wave FR = sin((t-stimulus_onset)/10), when t<stimulus_onset or t>stimulus_onset+10*pi.
sin((t-stimulus_onset)/10) is negative half of the time. Would you like to revise the expression for firing rate?
Yes, can we set it to be zero when FR<0 (this is more close to reality)? or just make FR = abs(sin((t-stimulus_onset)/10))?
Now I just want to explore how I can realise a presynaptic input as what I have described. I think I can modify it more reasonable later. Thanks again.
Post Reply