ModelDB is moving. Check out our new site at The corresponding page is

Salamander retinal ganglion cell: ion channels (Fohlmeister, Miller 1997)

 Download zip file 
Help downloading and running models
A realistic five (5) channel spiking model reproduces the bursting behavior of tiger salamander ganglion cells in the retina. Please see the readme for more information.
1 . Fohlmeister JF, Miller RF (1997) Impulse encoding mechanisms of ganglion cells in the tiger salamander retina. J Neurophysiol 78:1935-47 [PubMed]
Model Information (Click on a link to find other models with that property)
Model Type: Neuron or other electrically excitable cell;
Brain Region(s)/Organism:
Cell Type(s): Retina ganglion GLU cell;
Channel(s): I Na,t; I A; I K; I K,Ca; I Calcium;
Gap Junctions:
Simulation Environment: NEURON;
Model Concept(s): Activity Patterns; Ion Channel Kinetics; Influence of Dendritic Geometry; Detailed Neuronal Models; Calcium dynamics;
Implementer(s): Hines, Michael [Michael.Hines at];
Search NeuronDB for information about:  Retina ganglion GLU cell; I Na,t; I A; I K; I K,Ca; I Calcium;
Fohlmeister JF, Miller RF.
Impulse encoding mechanisms of ganglion cells in the
tiger salamander retina.
J Neurophysiol 1997 Oct;78(4):1935-47

A reprint of this article can be obtained from

This model is initially setup to produce figure 1 automatically.
After viewing the currents, you may simulate figure 3a in the
followng way.
1) close the 6 graph windows.
2) Destroy the SEClamp object by executing the statement
	objref VoltageClamp
3) Select the
menuitem and double click on the
4) press the Init&Run button.

The following parameter changes to the current working code supplied
by Bob Miller's lab were made in collaboration with
Michael Hines in order to
semi-quantitatively reproduce figures 1 and 3a.

1) Table 1 indicates that gnabar_spike = .05 S/cm2. However the curves
for the Na-current portion of Fig 1 use the default mod file value of
gnabar_spike = .04 .

2) The initial Ca Rev. Potential of Fig 1 requires that the Table 1 value
of celsius = 22 degC be used instead of NEURON's default value of 6.3.
This statement was added to the current.hoc file.

3) To get the fig 1 behavior of slowly increasing calcium concentration
with constant calcium current over 10 ms we modified current.hoc so that
depth_cad = 3 (microns)
taur_cad = 10 (ms)
instead of the working code values of 2.5 and 1.5 respectively.

4) The working code used a spherical soma of 5um in diameter. This
was changed to 25um as indicated in the paper and gives the results
of fig 3 when a current of .02 nA is injected which is the amount
specified in the paper.

The largest discrepancy is
that the amplitude of Ca-activated K-current is lower
in this example model than is indicated in fig 1 of the paper.

Questions about how to execute the model should be addressed to Questions about parameters should
be addressed to
J. F. Fohlmeister,
Physiology Dept.,
6-255 Millard Hall
University of Minnesota
435 Delaware St. SE.
Minneapolis, MN  55455

20120109 capump.mod updated from euler to derivimplicit: see for more: - TMM
20130826 Update from Ted Carnevale: capump.mod cleaned up to 1)
reflect this mechanism is just a simple calcium accumulation mechanism
that approximates pump action by a first order decay of cai and 2)
since neither drive_pump nor kt were used for anything, every mention
of them was deleted for clarity's sake.

Loading data, please wait...