home
overview
graphs
examples
reference
basics
lab
logic
grouping
populations
environments
animats
synapses
cells

Cell Models

Simple integrator

Load cell1.ccm
show display
run the model

This is about the simplest cell model possible. It has a single leaky integrator compartment with a time constant of 10ms. There is a membrane conductance applied with a reversal potential of 0mV, and a recorder sampling the membrane potential.

All that happens is that the membrane charges up until the charging rate through the applied conductance equals the leak rate. Then the potential remains fixed.

Current clamp

Load cell1a.ccm
show display
run the model

The first example used a passive pipette with a fixed conductance and potential to stimulate the cell. This example shows a more usual active current clamp configuration. The CurrentClamp component (from the "lab" shelf rather than an "cell" shelf) regulates the applied potential so as to achieve the desired current input to the cell. To do this, it needs to know the current membrane potential, so there is a connection b ack from the output port on the cell to the "voltage sensor" port on the current clamp.

Spiking integrator

Load cell2.ccm
show display
run the model

In this model a ThresholdSpikeGenerator has been added. It observes the membrane potential and generates a spike each time the potential goes above its threshold. Note that there is also a connection from the output of the spike generator to the reset port on the cell click with the right button on any item to display the port names.

Without the reset connection there would be no way for the integrator compartment to know that a spike had been fired, so its potential would behave just as in the first example. You can try this by deleting the reset connection and rerunning the model.

Spiking integrator with an after hyperpolarization (AHP)

Load cell3.ccm
show display
run the model

In this model, whenever the cell fires, not only is its potential reset to -70mV after the spike, but a transient conductance is activated with a reversal potential of -90mV. The conductance rises rapidly and then decays with a time constant of 4ms giving a smooth dip after the spike. As this conductance dies away, the external drive takes over and brings the cell back up to its threshold.

You can add as many such response functions as you please to give, for example, after depolarizations as well or more complicated AHP profiles.

Synapses

Load cell4.ccm
show display
run the model

Here the constant conductance input has been removed and instead there is a population of synapses receiving input from a random spike generator. The input is too weak to make the cell fire. There are no spikes but you can see the form of the post-synaptic depolarizations. In order to show up the falling phase, the time constant of the cell has been reduced to 1ms. The role of the membrane time constant is often misunderstood: although the synaptic conductance follows a bi-exponential (fast rise slow fall) form, you only see this in the membrane potential if there is some other process in the cell (like a big leak conductance) to bring the potential back down once the depolarizing conductance shuts off.

You can try the effects of the membrane time constant on the post-synaptic potential by clicking the integrator compartment and changing the timeConstant parameter. Click run in the workbench after changing it to see the effects.

The synapse population component is somewhat different from the other items on the workbench because it does not represent a single synapse but, potentially, a whole family of synapses with similar properties. Each time a new spiking connection is made to the synapse population, another individual synapse is added just for that connection. As yet, this does not make much difference, but it is important to understand the process for synapses which saturate or where the conductance is governed by a spike timing dependent plasticity law. These are covered in more detail in the synaptic plasticity examples.

Membrane channels

Dec 01 - not ready yet. The mechanisms used so far give increasingly complex cells but still rely on explicit spike generation. Catacomb 2 allows a smooth transition from this sort of model to those based on membrane ion channels. Example 5 has the reset mechanism (the explicit connection from the spike generator back to the integrator compartment) replaced by intrinsic sodium and potassium channel kinetics which produce action potentials. This is done by attaching membraneChannel components to the integrator compartment. It refers to a kinetic scheme representation, KSChannel, of the ion channel to be used. The ion channel is attached just like any other membrane conductance. For more details on kinetic scheme channels, see the help pages on the membraneChannel component.