Model files from the paper:

F. Zeldenrust, P. Chameau, W. J. Wadman, 'Spike and Burst Coding in Thalamocortical Relay Cells', PLoS Computational Biology, 2018

Questions on how to use this model should be directed to f.zeldenrust at neurophysiology.nl

Synopsis

Mammalian thalamocortical relay (TCR) neurons switch their firing activity between a tonic spiking and a bursting regime. In a combined experimental and computational study, we investigated the features in the input signal that single spikes and bursts in the output spike train represent and how this code is influenced by the membrane voltage state of the neuron. Identical frozen Gaussian noise current traces were injected into TCR neurons in rat brain slices to adjust, fine-tune and validate a three-compartment TCR model cell (Destexhe et al. 1998, accession number 279). Three currents were added: an h-current (Destexhe et al. 1993,1996, accession number 3343), a high-threshold calcium current and a calcium- activated potassium current (Huguenard & McCormick 1994, accession number 3808). The information content carried by the various types of events in the signal as well as by the whole signal was calculated. Bursts phase-lock to and transfer information at lower frequencies than single spikes. On depolarization the neuron transits smoothly from the predominantly bursting regime to a spiking regime, in which it is more sensitive to high-frequency fluctuations. Finally, the model was used to in the more realistic "high-conductance state" (Destexhe et al. 2001, accession number 8115), while being stimulated with a Poisson input (Brette et al. 2007, Vogels & Abbott 2005, accession number 83319), where fluctuations are caused by (synaptic) conductance changes instead of current injection. Under "standard" conditions bursts are difficult to initiate, given the high degree of inactivation of the T-type calcium current. Strong and/or precisely timed inhibitory currents were able to remove this inactivation.

Use
---

There are three folders. For each folder: extract the archive, run nrnivmodl in the channels directory (linux/unix) or mknrndll (mswin or mac os x) (see
http://senselab.med.yale.edu/ModelDB/NEURON_DwnldGuide.html for more help) to compile the channels, and then run (use nrngui in linux/unix) the tc....hoc file.

  1. The 'current clamp' folder, simulates the experiments. It will result in the voltage traces (and resulting spike trains) used in figures 9-11. The used 'holding currents' are set as 'El1.stim.amp' (i.e. the amplitude of the current injected via electrode 1). The data can be saved with procedures 'fileopen()' and 'slaop()'. Currently, the membrane potential of the soma, as well as the current and gating variables of the h-current and T-type calcium current are recorded and saved. A sample excerpt from the generated voltage trajectory for El1.stim.amp = -0.7 ('holding potential' -80 mV):

    screenshot 1

  2. In the 'influence_ih_iT' folder, the current-clamp experiments are repeated, but now the h-current is reduced by a factor (see figure 12 of the paper), that can be set at the top of the .hoc-file (as well as the 'holding potential' or 'membrane state' Vhold). This will then automatically adjust the mean input current ('El1.stim.amp'), the h-current conductivity and change the relevant filenames for saving (procedures 'fileopen()' and 'slaop()'). A sample excerpt from the generated voltage trajectory for 'holding potential' vhold= -80 mV and factor=0.1:

    screenshot 2

  3. In the 'high_conductance_state', the experiment as shown in figure 13 of the paper is simulated. A 'high-conductance-state ion channel' (fl) is used to either simulate a 'classic', no or an inhibitory high-conductance-state. Moreover, two (exponential) synapses that receive (the same) Poisson spike trains, 'Esynapse' and 'Esynapse_i' are added to the dendrite. Again, 'fileopen()' and 'slaop()' can be used to save the results. A sample excerpt from the generated voltage trajectory and a couple of other quantities for the 'classic' high conductance state:

    screenshot 3

  4. Additional note: the data related to the model, the experiments with which it was verified, are also public (an account must be generated, however it is free) and can be found here: https://data.donders.ru.nl/collections/shared/di.dcn.DSC_626840_0002_144/versions/1?0. This is hosted by Radboud University, The Netherlands.