{"id":21495885,"url":"https://github.com/jaffar-hussein/neurone-models","last_synced_at":"2025-03-17T12:14:50.709Z","repository":{"id":135028118,"uuid":"484377371","full_name":"Jaffar-Hussein/Neurone-Models","owner":"Jaffar-Hussein","description":"These are models that are used to study the spiking of the neurones organized according to their complexity from just considering the voltage to doing the ADEX model","archived":false,"fork":false,"pushed_at":"2023-04-16T09:07:28.000Z","size":5681,"stargazers_count":0,"open_issues_count":0,"forks_count":1,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-01-23T21:51:24.337Z","etag":null,"topics":["brian","brian2","matplotlib","neuron","neuron-model","neuron-simulations","neuroscience","pyhon3"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Jaffar-Hussein.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2022-04-22T09:40:08.000Z","updated_at":"2022-05-05T09:15:05.000Z","dependencies_parsed_at":null,"dependency_job_id":"63c7f38d-0792-4ce2-accc-9d746f840c45","html_url":"https://github.com/Jaffar-Hussein/Neurone-Models","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Jaffar-Hussein%2FNeurone-Models","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Jaffar-Hussein%2FNeurone-Models/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Jaffar-Hussein%2FNeurone-Models/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Jaffar-Hussein%2FNeurone-Models/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Jaffar-Hussein","download_url":"https://codeload.github.com/Jaffar-Hussein/Neurone-Models/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":244031153,"owners_count":20386534,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["brian","brian2","matplotlib","neuron","neuron-model","neuron-simulations","neuroscience","pyhon3"],"created_at":"2024-11-23T16:13:51.927Z","updated_at":"2025-03-17T12:14:50.703Z","avatar_url":"https://github.com/Jaffar-Hussein.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"**GURA Jaffar**\n\n\n**REPORT**\n\n**Modelling of neuronal interactions at the claustrum level**\n\n**LABORATORY: UMR9197 [NEUROPSI SACLAY FRANCE](https://neuropsi.cnrs.fr)**\n\n**LABORATORY DIRECTOR: Mr. ROUYER Francois**\n\n**TEAM: Computational Neuroscience**\n\n**ACADEMIC SUPERVISOR: Prof. JACQUIR Sabir**\n\n**INTERNSHIP SUPERVISOR: Prof. PANANCEAU Marc**\n\n**ADDRESSEE: 151 Centre CEA Saclay 91400 SACLAY FRANCE**\n\n**2021/2022**\n\n# Table of Contents\n\n[Abbreviations](#abbreviations)\n\n[Abstract](#abstract)\n\n[I) INTRODUCTION](#introduction)\n\n[II) Methods](#methods)\n\n[1. Adaptive Exponential Integrate Model\n](#adaptive-exponential-integrate-model)\n\n[2. Testing the relative effects of variables to the Adex Model\n](#testing-the-relative-effects-of-variables-to-the-adex-model)\n\n[3. Modelling Spike Trains from Experimental Data\n](#modelling-spike-trains-from-experimental-data)\n\n[III) Results and Discussions\n](#results-and-discussions)\n\n[I) Interspike Intervals with a variation of the τ ω\n](#interspike-intervals-with-a-variation-of-the-τ-ω)\n\n[II) Interspike Intervals with a variation of the values of b](#interspike-intervals-with-a-variation-of-the-values-of-b)\n\n[III) Interspike Intervals with a variation of the a\n](#interspike-intervals-with-a-variation-of-the-a)\n\n[IV) Interspike Intervals with the variation of the injected current\n](#interspike-intervals-with-the-variation-of-the-injected-current)\n\n[V) Models from Experimental Data\n](#models-from-experimental-data)\n\n[VI) Conclusions ](#conclusions)\n\n[Bibliography ](#Bibliography)\n\n\n\n# Abbreviations\n\nVIP -- Vasoactive Intestinal Peptides\n\nPV - Parvalbumin\n\nSOM - Somatostatin\n\nNPY - Neuropeptide Y\n\nADEX -- Adaptive Exponential Integrate and Fire Model\n\nISI -- Interspike interval\n\ntau- Membrane time constant\n\n# Abstract\n\nSeveral studies have modelled the dynamics of neurons and not a lot\nspecifically target the neurons of the claustrum. The claustrum neurons\nare hampered by not having an agreed defined boundary. In our\ninternship, we study several models for neurons in general and see the\neffect of the modulation of their parameters. We then simulate\nexperimental data to reproduce the neuron dynamics observed.\n\n# INTRODUCTION\n\nThe claustrum is a thin sheet of grey matter located in the forebrain,\nextending rostrocaudally along with the striatum and situated between\nthe insula and putamen (Smith, Lee and Jackson, 2020). It is the most\ndensely interconnected region of the brain (Graf *et al.*, 2020). The\nclaustrum is composed of both the spiny stellate projection neurons and\na variety of aspiny interneurons that can be differentiated by the\nexpression of various peptides and calcium-binding proteins such as PV,\nVIP, SOM, NPY and others (Smith, Lee and Jackson, 2020). Due to the high\nconnection of the claustrum and other parts of the brain, it is thought\nto take part in higher cognitive functions where it integrates\ninformation and hence supports the consciousness (Crick and Koch, 2005).\nThe claustrum is also believed to take part in attention and resilience\nfrom distraction (Atlan *et al.*, 2018). It has also been linked to\nsleep (Liu *et al.*, 2019) and impulsivity(Norimoto *et al.*, 2020).\n\nThe claustrum projection neurons and interneurons can be distinguished\nfrom each other based on their intrinsic electrical properties. The\nprojection neurons can be subdivided into 5 subclasses based on their\nintrinsic electrical properties. This makes a total of 8 subclasses of\nclaustrum neurons(Graf *et al.*, 2020).\n\nA neuron fires when it gets input from other sources. The firing of a\nneuron produces an action potential (spike), which is an abrupt\ntransient change of the membrane voltage that propagates to other\nneurons(Izhikevich, 2007). There are several models with different\nparameters allowing the simulation of neuron dynamics. In the team, we\nworked with the Adex model which is a relatively simple model with two\ndifferential equations one for the voltage with respect to time and the\nother for the adaptation variable with respect to time. The model shows\nthe evolution of the voltage in time when a current I is injected (Naud\n*et al.*, 2008).\n\nThis internship aims to reproduce the dynamics of claustrum neurons\nobserved in the experimental works (Graf *et al.*, 2020). First, we\nsimulated the ADEX model in order to understand the influence of\ndifferent parameters on the Interspike interval then we reproduced\nqualitatively some parameters of the neurons observed experimentally.\n\n# Methods\n\nAll simulations were done in the Brian2 neural simulator and Python 3\nprogramming language.\n\n### Adaptive Exponential Integrate Model\n\nThis is a model of two differential equations that model the evolution\nof membrane potential V when a certain current I is injected into the\nsystem.\n\n$$C\\frac{dV}{dt} = \\  - g_{l}\\left( V - E_{L} \\right) + g_{L}*\\ \\mathrm{\\Delta}_{T}\\exp\\left( \\frac{V - V_{T}}{\\mathrm{\\Delta}_{T}} \\right) - \\omega + Ι$$\n\nEquation Adaptation\n\n$$\\tau_{\\omega}\\frac{d\\omega}{dt} = a\\left( V - E_{L} \\right) - \\omega$$\n\nEquation Adaptation Current\n\nWhere :\n\n$C$ = Membrane Capacitance\n\nV = Membrane Potential\n\ng~L~ = Leak Conductance\n\nE~L~ = Resting Potential\n\nV~T~ = Threshold Potential\n\nω = Adaptation Variable\n\nI = Synaptic Current\n\nτ ~ω~ = Time constant\n\n∆~T\\ =~ Threshold Slope Factor\n\na = Subthreshold Adaptation\n\nWhen the current drives the potential beyond the threshold V~T~ this\nleads to the positive feedback that drives an upswing of the action\npotential. The upswing is stopped by a reset threshold that we fix, and\nthe action potential is replaced by a reset condition (Equation 1)(Naud\n*et al.*, 2008).\n\n$$if\\ V \u003e V_{T}\\ then\\ \\begin{Bmatrix}\nV \\rightarrow V_{r} \\\\\n\\omega\\  \\rightarrow {\\omega\\ }_{r} = \\omega\\  + b \\\\\n\\end{Bmatrix}$$\n\nEquation Reset Condition for the ADEX Model\n\nWhere:\n\nb = Spike triggered adaptation\n\nV~r~ = Reset Potential\n\nThe membrane potential is reset to V reset whereas the adaptation\nvariable is reset to the w~r~ which is the adaptation variable plus a\nfixed amount b. The adaptation variable accumulates during the spike\ntrain whereas the membrane potential does not.\n\nThe nine parameters can be divided into bifurcation and scaling\nparameters. The scaling parameters are involved for scaling the time\naxis. The five scaling parameters are the total capacitance (C), total\nleak conductance (g~L~ ), effective rest potential (E~L~ ), threshold\nslope factor (∆~T~ ), effective threshold potential (V~T~ ).\n\nThe remaining four parameters are bifurcation parameters and are\ndirectly related to the conductance a, the time constant τw,the spike\ntriggered adaptation b, and the rest potential Vr. The modification of\nthese four parameters results in changes in the firing patterns.\n\n### Testing the relative effects of variables to the Adex Model\n\nTo test the effect of different parameters on the evolution of the spike\ntrain we simulated the firing of a neuron while adjusting for the value\nof interest. While this was happening, we fixed the rest of the\nvariables to pre-defined control in which was our starting point for the\nexperiment as :\n\nC = 200 pF\n\ng~L~ = 10 nS\n\nE~L~ = -65mV\n\nV~T~ = -55 mV\n\nI = .120nA\n\nτ ~ω~ = 500 ms\n\ndt = 5 mV\n\na = 2 pA\n\nb = 10 pA\n\nV~r~ = -52 mV\n\nWe varied the $a,\\ b\\ ,input\\ current\\  tau$.We recorded the spike\ntrain and the ISI for the variable parameters while leaving the rest of\nthe parameters to remain the same as the control. We set the threshold\nat -40mV and the refractory period to be 5ms. The experiment ran for 4s\nfor each simulation, and we took the value from 500ms to 4000ms to avoid\nany bias at the beginning of the spike train.\n\n### Modelling Spike Trains from Experimental Data\n\nWe modelled the neurons according to the intrinsic electrical properties\nthat we fixed into the Adex model derived from the classification of the\nrat claustrum (Graf *et al.*, 2020). The Adex model (Figure 1), has 9\nvariables and we got 4 variables from the experimental data, that is the\nmembrane potential, the input current, the threshold, and the leak\nconductance.\n\nWe fixed the variables that we got from the paper into the model and\nmodulated the rest of the variables to try and simulate the experimental\nspike trains.\n\n# Results and Discussions\n\nThe Adex model has been simulated with a variation of these parameters :\n\na , b , synaptic current , and τ ~ω~ ,Then the Interspike intervals have\nbeen computed in order to look at the influence of these different\nparameters on the frequency of the spikes.\n\n### Interspike Intervals with a variation of the τ ~ω~\n\n\u003cimg width=\"718\" alt=\"image1\" src=\"https://user-images.githubusercontent.com/57854451/232287973-9c6a5e15-51ed-4813-95a4-568c21236a4f.png\"\u003e\n\n\nFigure Change of ISI for variation of tau (ms)\n\nIn Figure 3 we vary the τ ~ω~ from 0ms to 1000ms and where a is 10 nS, b\nis fixed at 10nS, and synaptic current is fixed at .120nA.\n\nAt point a in the graph the τ ~ω~ is 200ms we observe spike train with a\nhigh frequency and regular ISI. With increase to point b of τ ~ω~ is\n600ms we observe six values of ISI and hence an irregular spike train.\nThe third mark is at 1000ms we observe three distinct values of ISI and\nhence more regular than b.\n\n### Interspike Intervals with a variation of the values of b\n\nIn Figure 4 we vary the b from 0 pA to 25 pA and where a is 10 nS, τ ~ω~\nis fixed at 500ms, and synaptic current is fixed at .120nA.\n\nAt point a in the graph the b is at 7 pA we observe spike train with a\nhigh frequency and regular ISI. With increase to point b of b is 17 pA\nwe observe 4 values of ISI and hence an irregular spike train. The third\nmark is at 27 pA we observe two distinct values of ISI and hence more\nregular than b.\n\n\u003cimg width=\"648\" alt=\"image2\" src=\"https://user-images.githubusercontent.com/57854451/232287895-6f91fabe-1c68-4218-8dfc-4188fb407a59.png\"\u003e\n\n\n\n\nFigure Variation of ISI with b (pA)\n\n### Interspike Intervals with a variation of the a\n\nIn Figure 5 we vary the a from 0nS to 10nS and where b is 2nS, τ ~ω~ is\nfixed at 500ms, and synaptic current is fixed at .120nA.\n\nAt point (a) in the graph the a is at 1 nS we observe spike train with a\nhigh frequency and regular ISI. With increase to point (b) of b is 4nS\nwe observe 4 values of ISI and hence an irregular spike train. The third\nmark is at 6 nS we observe three distinct values of ISI and hence more\nregular than (b).\n![image3](https://user-images.githubusercontent.com/57854451/232287845-6ccc4473-3da7-44f4-98aa-deb6dd18aa1d.png)\n\n\nFigure Variation of ISI for a (nS)\n\n### Interspike Intervals with the variation of the injected current\n\n![image4](https://user-images.githubusercontent.com/57854451/232287755-e1eddb00-0f2b-4973-9cce-da6ab4b39fe5.png)\n\n\n\nFigure Variation of ISI (s) with Injected current (pA)\n\nIn Figure 6 we vary the synaptic current from 0.05pA to 0.250pA and\nwhere b is 2nS, τ ~ω~ is fixed at 500ms, and a is 10 nS.\n\nAt point (a) in the graph the synaptic current is at 0.08 pA we observe\nspike train with a low frequency and an irregular spike train. With\nincrease to point (b) with synaptic current as 0.150 pA we a high\nfrequency spike train compared to (a) with a regular ISI. The third mark\nis at 0.250 pA the frequency increases from the one at (b) and the ISI\nremains regular.\n\n## Models from Experimental Data\n\nThe data divides the neurons of the claustrum into the two major groups\ninterneurons and projection neurons and further subdivides the\nprojection neurons into five subgroups according to the shape of their\nspike trains.(Graf *et al.*, 2020). The data in figure on the left are\nthe five sib-divisions of projections and one interneuron whereas the\nfigures on the right are the pre-liminary results obtained from our\nsimulations. They show a bit of difference but don't much up exactly to\nthe experimental data, this discrepancy is caused by lack of majority of\nthe variables considered in the ADEX model from the experimental data.\n\n\u003cimg width=\"1113\" alt=\"Screenshot 2023-04-16 at 10 57 30\" src=\"https://user-images.githubusercontent.com/57854451/232288070-41328f5c-84d6-4bb6-87de-dece78c436f5.png\"\u003e\n\n\n\u003cimg width=\"385\" alt=\"image6\" src=\"https://user-images.githubusercontent.com/57854451/232288027-bd3cbb14-650a-446c-a83b-fd12f9e0912a.png\"\u003e\n\n\nFigure Classification of claustrum neurons\n(a) the experimental spike trains (Graf et al., 2020) (b) the\nvalues of our simulations with data from experiments on mice (Graf et\nal., 2020)\n\n\n\n# Conclusions\n\nWe grouped the neurons of the claustrum according to their electrical\nproperties and their expression of different peptides. In total we\narrived at 8 distinct subgroups of the claustrum inter-neurons and also\nobtained preliminary results in our simulations from experimental data.\nIn addition, I learnt different neuron models with increased complexity\ntill I arrived to the Adex model that our team was using for modelling\nof the claustrum neurons.\n\nThe rest of the experimental data and all simulations can be accessed\nthrough the GitHub repository\n\u003chttps://github.com/Jaffar-Hussein/Neurone-Models\u003e\n\n# Bibliography\n\nSciences Atlan, G. *et al.* (2018) 'The Claustrum Supports Resilience to\nDistraction', *Current Biology*, 28(17), pp. 2752-2762.e7.\ndoi:10.1016/j.cub.2018.06.068.\n\nCrick, F.C. and Koch, C. (2005) 'What is the function of the\nclaustrum?', *Philosophical transactions of the Royal Society of London.\nSeries B, Biological sciences*, 360(1458), pp. 1271--1279.\ndoi:10.1098/rstb.2005.1661.\n\nGraf, M. *et al.* (2020) 'Identification of Mouse Claustral Neuron Types\nBased on Their Intrinsic Electrical Properties', *eneuro*, 7(4), p.\nENEURO.0216-20.2020. doi:10.1523/ENEURO.0216-20.2020.\n\nIzhikevich, E.M. (2007) *Dynamical systems in neuroscience: the geometry\nof excitability and bursting*. Cambridge, Mass: MIT press (Computational\nneuroscience).\n\nLiu, J. *et al.* (2019) 'The Claustrum-Prefrontal Cortex Pathway\nRegulates Impulsive-Like Behavior', *The Journal of Neuroscience*,\n39(50), pp. 10071--10080. doi:10.1523/JNEUROSCI.1005-19.2019.\n\nNaud, R. *et al.* (2008) 'Firing patterns in the adaptive exponential\nintegrate-and-fire model', *Biological Cybernetics*, 99(4--5), pp.\n335--347. doi:10.1007/s00422-008-0264-7.\n\nNorimoto, H. *et al.* (2020) 'A claustrum in reptiles and its role in\nslow-wave sleep', *Nature*, 578(7795), pp. 413--418.\ndoi:10.1038/s41586-020-1993-6.\n\nSmith, J.B., Lee, A.K. and Jackson, J. (2020) 'The claustrum', *Current\nBiology*, 30(23), pp. R1401--R1406. doi:10.1016/j.cub.2020.09.069.\n\n[^1]: The experimental spike trains obtained from (Graf *et al.*, 2020)\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjaffar-hussein%2Fneurone-models","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fjaffar-hussein%2Fneurone-models","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjaffar-hussein%2Fneurone-models/lists"}