www.neuron.yale.edu

Hello again Sir!
I am trying to simulate a second messenger system having differential equations with time-delay present in some equations of system but not others. (Specifically I am trying to model: Dynamic properties of regulatory motifs associated with induction of three temporal domains of memory in aplysia. J Comput Neurosci. 2005 Mar-Apr;18(2):163-81. Pettigrew DB, Smolen P, Baxter DA, Byrne JH. )
So in my mod file I am using derivimplicit method to solve these differential equations and to simulate timedelay I am trying to use If...else statememnt inside my DERIVATIVE block. But I am getting an error that no. of equations are more than no. of states. Is there anyway arround this?
Thanks!

There is a DELAY keyword in the model description
language but it works only with the fixed step
method because it builds a ring buffer of length
tdelay/dt
You generally should not do any state manipulation in a BREAKPOINT block because that is called many times per time step---twice
per time step for the fixed step method in order
to calculate di/dv.
If the state delays are actually just delayted events
in response to trigger thresholds, then it is best
to use NET_RECEIVE blocks with WATCH
statements.