Setting celsius differently for two different neurons

Moderator: wwlytton

Post Reply
Krishna Chaitanya
Posts: 70
Joined: Wed Jan 18, 2012 12:25 am
Location: University of Pavia

Setting celsius differently for two different neurons

Post by Krishna Chaitanya » Wed Jan 09, 2019 4:06 pm

Hi,

I have two different neuronal cell types, each with its own v_init and celsius.

Is it possible to set these parameters independently? celsius and v_init are global variables.

I tried changing mod files by inserting a local variable for celsius but it seems like there is a difference in spike timing when global variable "celsius" is used. Although it maintains similar frequency, usage of local variable for celsius shifts timing.

Are there examples to understand this problem?

Thank you.

ted
Site Admin
Posts: 5544
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Setting celsius differently for two different neurons

Post by ted » Thu Jan 10, 2019 5:07 pm

v_init and celsius are globals. Their values will be the same everywhere.
I tried changing mod files by inserting a local variable for celsius but it seems like there is a difference in spike timing when global variable "celsius" is used.
Without seeing your code I can't begin to guess what you did, but it seems like you got an incorrect result. In order to suggest a proper approach, I will have to see the NMODL code in question.

I should mention that NEURON's global built-in celsius parameter affects not only NMODL formulas that involve this parameter, but also the equilibrium potentials for ionic species whose "ion style" results in recalculation of equilibrium potential during intialization or execution of a simulation.

Krishna Chaitanya
Posts: 70
Joined: Wed Jan 18, 2012 12:25 am
Location: University of Pavia

Re: Setting celsius differently for two different neurons

Post by Krishna Chaitanya » Tue Jan 22, 2019 11:46 am

Thank you very much Ted. I was able to solve the problem by adjusting the q10 parameter in the mod files.

ted
Site Admin
Posts: 5544
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Setting celsius differently for two different neurons

Post by ted » Tue Jan 22, 2019 12:31 pm

For the benefit of others who may read this thread--

It is possible to write custom initialization code that sets initial membrane potential or even resting potential to arbitrary values at any or all locations in one or more model cells. The only limits are (1) one's imagination and (2) how clever one might be in (a) implementing the intialization code in the first place, and (b) convincing others (reviewers especially) of the plausibility of the custom initialization. In most cases, the principle of parsimony (AKA Occam's razor, or KISS) should govern. As always, implementational details depend on the nature of the model itself and the details of the initialization that is desired. Common custom initializations that are easily achieved include
* initialization of a model cell that is not spontaneously active to a particular uniform resting potential
* initialization of a spontaneously active model to a "snapshot" of states captured at a particular time

Specifying different operating temperatures for different cells or different parts of the same cell can also be done. One way is to use code that substitutes a user-defined RANGE variable, e.g. called "mycelsius", for celsius. This must be done in all NMODL files that involve temperature-dependent rates. Also, if there is any ionic equilibrium potential ex that is not simply a constant parameter but instead is calculated at initialization and/or in the course of a simulation, then ex MUST NOT be READ. Instead, the calculation of ex must be done locally by each mechanism that uses ex, and it must employ the RANGE variable "mycelsius".

Post Reply