README for model associated with the paper

Moss BL, Fuller AD, Sahley CL, Burrell BD (2005) Serotonin 
modulates axo-axonal coupling between neurons critical for learning
in the leech. J Neurophys, accepted with minor revisions

Abstract
S cells form a chain of electrically coupled neurons that extends
the length of the leech CNS and plays a critical role in 
sensitization during whole-body shortening.  This process requires
serotonin, which acts in part by altering the pattern of activity
in the S cell network.  Lent (1982) observed serotonin-containing
axons and varicosities in Faivre's nerve where the S-to-S cell 
electrical synapses are located.  To determine whether serotonin
modulates these synapses, S cell action potential (AP) propagation
was studied in a two-ganglion chain containing one electrical 
synapse.  Suction electrodes were placed on the cut ends of the
connectives to stimulate one S cell while recording the other, 
coupled S cell's APs.  A third electrode, placed en passant, 
recorded the APs near the electrical synapse before they propagated
through it.  Low concentrations of the gap junction inhibitor 
octanol increased AP latency across the two-ganglion chain and this
effect was localized to the region of axon containing the electrical
synapse.  At higher concentrations, APs failed to propagate across
the synapse.  Serotonin also increased AP latency across the
electrical synapse, suggesting that serotonin reduced coupling
between S cells.  This effect was independent of the direction of
propagation and increased with the number of electrical synapses in
progressively longer chains.  Furthermore, serotonin modulated 
instantaneous AP frequency when APs were initiated in separate S 
cells and in a computational model of S cell activity following 
mechanosensory input.  Thus, serotonergic modulation of S cell 
electrical synapses may contribute to changes in the pattern of
activity in the S cell network.

Model summary:

It is a linear network of 5 electrically-coupled neurons.  The idea
is to see whether a decrease in the gap junctional conductance 
alters the pattern of action potentials in the network following
simulated sensory input.  In order to generate the simulated data in 
the paper (Fig. 10), one has to run the model with all of the 
electrical synapses set to a "control" conductance value, and then 
run it again with the "modulated" conductance values. The user can
also choose whether to display the intracellular recording from the
anterior or posterior end of the network.

More details about the model and how to use it:

The model is a linear network of five S cells connected by
non-rectifying electrical synapses.  The five S cells are designated
10-10a-10b-10c-10d.  Each cell consists of ten compartments (each
compartment is designated cell #_compartment #) so the whole model looks
like this:

10_1-10_2-10_3-10_4-10_5-10_6-10_7-10_8-10_9-10_10---10a_1-10a_2-10a_3-
10a_4-10a_5-10a_6-10a_7-10a_8-10a_9-10a_10---10b_1-1-0b_2-10b_3-10b_4-10
b_5-10b_6-10b_7-10b_8-10b_9-10b_10---10c_1-1-0c_2-10c_3-10c_4-10c_5-10c_
6-10c_7-10c_8-10c_9-10c_10---10d_1-1-0d_2-10d_3-10d_4-10d_5-10d_6-10d_7-
10d_8-10d_9-10d_10

-        Connections between compartments with each cell

axon.es (g = 0.196)

---     Electrical synapses between S cells

     "Control" value:  elsyn.es (g = 0.098)

"Modulated" value:  serelsyn.es (g = 0.064)


In the default simulation, the model displays the intracellular
recording from the anterior end of the network (compartment 10_1) and
all four electrical synapses are set to the "control" conductance value
(elsyn.es).

To display the intracellular recording from the posterior end, the user
must modify the simulation so that it displays the voltage from
compartment 10d_10.  To do this: load the simulation, go to Graphic
Output and select Modify (this opens cable_10.ous), click on the single
entry in the Variables list on the right, select Modify, select
V[10d_10...]<ivr from the pull down list.   The simulation must then be
re-loaded.

To change the conductance of the electrical synapses to the "modulated"
values, one can modify each electrical synapse in the network so that it
is uses serelsyn.es rather than elsyn.es.  This is fairly tedious, so an
easier way is to simply open elsyn.es and change the values of g1 and g2
from 0.098 to 0.064.  To do this, go to Edit Formula, click on the es
box, open elsyn.es, click on the shaded g1 and g2 variables to change
the values.  Again, the simulation must be re-loaded.

Note: recall that SNNAP requires that parent directories have no spaces
in the folder names, e.g. on windows install this zip archive in a 
directory like C:\nrnmodels rather than under 
c:\Documents and Settings\.
(has spaces in the name)

Thanks to the people who made SNNAP possible (see contributors at):
http://snnap.uth.tmc.edu/