www.neuron.yale.edu

And what happened in your case is that after you called the ion_style function with the cinit slot = 0, adding the synapse that READ cai, promoted it (for that segment only) back to 1. . . . Which suggests that the use of ion_style is surrounded by caveats, because a later change (e.g. adding of a ...

The most authoritative answers to such questions might be obtainable from the authors of one or both of these papers, at least a few of whom are probably still alive.

When using an unfamiliar feature, work through a series of toy exercises before addressing the task that is giving you difficulty. 1. Use Vector.play to drive the value of a scalar. 2. Create a single section model with nseg==1 and use Vector.play to drive the segment's v. 3. Insert extracellular in...

One other change I'd make to L5PC-db.py: get rid of this part of the cell class definition for secname in self.sections: if secname == "axon": continue h.ion_style("ca_ion",3,2,1,1,0,sec=self.sections[secname]) self.sections[secname].cao = 1.2 for seg in self.sections[secname]: seg.cai = 0.001 becau...

I see what you mean. Python statements in the L5neuron class's definition explicitly assign values to cai (1.0e-3 mM) and cao (1.2 mM) in each segment that has a mechanism with a USEION ca statement. And when you create a new L5neuron instance, those are the calcium concentrations in those segments....

Adding new 3D points that lie on the straight line between existing points and have diameters midway between those of the existing points will have no effect on L, diam, or surface area of any section and will not affect simulation results.

Before addressing fac or l2a(), note that NEURON has a concept called "currently accessed section" that allows one to write code that refers to parameters and variables that belong to a particular section without having to specifically name a particular section in the same line. forall is an iterato...

How can I specify the interspike interval/firing frequency while using a NetCon object to connect two neurons? Make the presynaptic neuron fire with the desired ISI or frequency. NetStim generates artificial presynaptic stimuli No, it generates "100% organic" events that NetCons can deliver to syna...

With regard to your class BallAndStick . . . Nice clean code, great start. Just a couple of things to resolve, and you'll be ready to extend your program so that a parameter can be used to specify how many nodes and internodes the axon has. If you're going to define an axon class, the name of the cl...

Interesting problem. How big does the network itself have to be (total # cells, total # synapses)? At what point (time) in the simulation does it occur? The first thing to do is to see if there are any peculiarities about the NMODL code itself that might cause problems, and the quickest way to do th...

Is that N-I-N-I-I-I-...-I-N-I-N (i.e. 4 nodes and a bunch of internodes? Something like that: N-I-I-I-I-......-I-I-N Two nodes separated by a lot of internodes. You'll never get a propagating action potential. but you need to know diam in order to calculate gm. And you don't know diam. . . . Diamet...

I meant 2 HH nodes at the beginning and the end and lots of internodes in the middle Is that N-I-N-I-I-I-...-I-N-I-N (i.e. 4 nodes and a bunch of internodes? or N-I-N-I-N-I-...-N-I-N (i.e. NNODES nodes and NNODES-1 internodes) "The slovenliness of our language allows us to think foolish thoughts." ...

OK, you don't need super-detailed representation of the properties of myelin. No need for extracellular. There is actually an existing literature on the effects of internode length on spike propagation. The key papers are Brill et al. Conduction velocity and spike configuration in myelinated fibres:...

I said nothing at all about spatial discretization (the value to use for nseg). That's a topic for a different discussion. Nor did I address parameters. The question that triggered this essay stipulated Params : { Myelin resistance (R_m): MOhms/mm Myelin capacitance: pF/mm longitudinal resistance: M...

This is written for general readership, so please excuse explicit discussion of much that you probably already know. The first and most important step in any modeling project is to develop a clear conceptual model. I will assume you want a model of an unbranched myelinated axon--a series of nodes of...