Dear All, My name is Gustavo (Gus) and I am a complete newbie at NEST, so probably this will be a stupid question. However, I have thoroughly;y searched the documentation for nest-desktop, nest, nestml, and this mailing list, but couldn't find an appropriate answer. So, if this is a stupid question, sorry!
Well, I want to model a population of Izhikevich-like neurons, as described in this paper: Chen, L., Campbell, S.A. Exact mean-field models for spiking neural networks with adaptation. J Comput Neurosci 50, 445–469 (2022). https://doi.org/10.1007/s10827-022-00825-9 arXiv: https://arxiv.org/abs/2203.08341 code: https://senselab.med.yale.edu/ModelDB/ShowModel?model=267382#tabs-1 Basically, it is a light variant of the original model in Izhikevich's 2003 paper, with a standard synaptic current I = g S (e_r - u_k), with S the gating variable following its own first-order differential equation, and assuming an intrinsic current with a Lorentzian distribution (and probably something else that I do not remember right now).
I also wanted to take this opportunity to learn nest at the same time, so I decided to start with NEST-Desktop online (I have an account at EBRAINS), but couldn't figure out how to exactly configure the standard Izhikevich model to be the one in the paper, and also couldn't find a mechanism to plug a gating variable, or a way to specify the intrinsic current, and also couldn't find a Lorentzian distribution among the options for the current generators... So, I guess I should go deeper. Is that right? Can I solve this with merely nest-desktop?
Then, I started to look at some examples in nestml, including the Izhikevich one, and I think that, if I have to modify it, it is feasible. I am fluent in Python, and reading the Izhikevich tutorial seemed reasonable to me: https://nestml.readthedocs.io/en/latest/tutorials/izhikevich/nestml_izhikevi...
So, my final question... is this the way to go? Should I write my neuron model as a new variant of the original one, perhaps in nestml? Should I include the gating variable inside this model, or is there another mechanism for that? Then, I guess I need to build a current generator, right? Can I have more than one (1 for the external stimulus, 1 for the intrinsic current)? Is this plan sound? Are there any shortcuts for what I want to do? Any hints or tips you could share?
In any case, THANKS FOR THE PATIENCE!!!
cheers gus.-
Dear Gus,
As you want to create a new model based on an existing one, NESTML is certainly the way to go.
BTW, if you want to get to know the NEST community better, check out the NEST Conference 2023 (Virtual) at <%0dhttps:/nest-simulator.org/conference%0d> https://nest-simulator.org/conference<%0dhttps:/nest-simulator.org/conference%0d>!
Best, Hans Ekkehard
--
Prof. Dr. Hans Ekkehard Plesser Head, Department of Data Science
Faculty of Science and Technology Norwegian University of Life Sciences PO Box 5003, 1432 Aas, Norway
Phone +47 6723 1560 Email hans.ekkehard.plesser@nmbu.nomailto:hans.ekkehard.plesser@nmbu.no Home http://arken.nmbu.no/~plesser
From: gustavo.patow@udg.edu gustavo.patow@udg.edu Date: Thursday, 6 April 2023 at 20:37 To: users@nest-simulator.org users@nest-simulator.org Subject: [NEST Users] Population of Izhikevich-like neurons [Some people who received this message don't often get email from gustavo.patow@udg.edu. Learn why this is important at https://aka.ms/LearnAboutSenderIdentification ]
Dear All, My name is Gustavo (Gus) and I am a complete newbie at NEST, so probably this will be a stupid question. However, I have thoroughly;y searched the documentation for nest-desktop, nest, nestml, and this mailing list, but couldn't find an appropriate answer. So, if this is a stupid question, sorry!
Well, I want to model a population of Izhikevich-like neurons, as described in this paper: Chen, L., Campbell, S.A. Exact mean-field models for spiking neural networks with adaptation. J Comput Neurosci 50, 445–469 (2022). https://doi.org/10.1007/s10827-022-00825-9 arXiv: https://arxiv.org/abs/2203.08341 code: https://senselab.med.yale.edu/ModelDB/ShowModel?model=267382#tabs-1 Basically, it is a light variant of the original model in Izhikevich's 2003 paper, with a standard synaptic current I = g S (e_r - u_k), with S the gating variable following its own first-order differential equation, and assuming an intrinsic current with a Lorentzian distribution (and probably something else that I do not remember right now).
I also wanted to take this opportunity to learn nest at the same time, so I decided to start with NEST-Desktop online (I have an account at EBRAINS), but couldn't figure out how to exactly configure the standard Izhikevich model to be the one in the paper, and also couldn't find a mechanism to plug a gating variable, or a way to specify the intrinsic current, and also couldn't find a Lorentzian distribution among the options for the current generators... So, I guess I should go deeper. Is that right? Can I solve this with merely nest-desktop?
Then, I started to look at some examples in nestml, including the Izhikevich one, and I think that, if I have to modify it, it is feasible. I am fluent in Python, and reading the Izhikevich tutorial seemed reasonable to me: https://nestml.readthedocs.io/en/latest/tutorials/izhikevich/nestml_izhikevi...
So, my final question... is this the way to go? Should I write my neuron model as a new variant of the original one, perhaps in nestml? Should I include the gating variable inside this model, or is there another mechanism for that? Then, I guess I need to build a current generator, right? Can I have more than one (1 for the external stimulus, 1 for the intrinsic current)? Is this plan sound? Are there any shortcuts for what I want to do? Any hints or tips you could share?
In any case, THANKS FOR THE PATIENCE!!!
cheers gus.- _______________________________________________ NEST Users mailing list -- users@nest-simulator.org To unsubscribe send an email to users-leave@nest-simulator.org
Dear Prof. Dr. Hans Ekkehard, Thanks for your answer!!!! Sorry to insist, but regarding the part about the gating variable following its own first-order differential equation, should I include it in the NestML model itself? I guess so, but I haven't seen any example of that... Finally, what about the intrinsic current with a Lorentzian distribution... should I implement a new "current generator" (I am not sure this is the right name)??? Sounds logical, but perhaps there are other options... Thanks again!!! cheers, gus.-
-
gustavo.patow＠udg.edu -
Hans Ekkehard Plesser