the distance() function

The basics of how to develop, test, and use models.
ted
Site Admin
Posts: 5647
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: the distance() function

Post by ted » Mon Jan 20, 2014 5:44 pm

By any chance did you install one of these "Debian Science" packages--the "Debian Science packages for modeling of neural systems", or maybe "Simulator-independent specification of neuronal network models" ?

rlindroos
Posts: 17
Joined: Thu Jan 02, 2014 2:23 pm

Re: the distance() function

Post by rlindroos » Tue Jan 21, 2014 6:44 am

By any chance did you install one of these "Debian Science" packages--the "Debian Science packages for modeling of neural systems", or maybe "Simulator-independent specification of neuronal network models" ?
I don't remember unfortunatelly. I will reinstall NEURON using the link you posted above and try again. However your reply in the previous post makes me belive that this for some reason is the intended behaviour of import3D.
Cells typically have somas that are electrically compact, so the proximal ends of all neurites that arise from the soma are at the same potential. However, the most proximal 3d measurement (x,y,z, and diameter, or in the case of swc files, x,y,z, and radius) on any such neurite will be made at some nonzero distance from the soma's center, and should be drawn at the proper location in any graphical rendering of the cell's structure.
Is this the standard way of doing these measurments? I'm pretty sure that in GENESIS each line of a morphology file (except the first one defining the soma) is interpretted as the end point of a compartment - no mather if it is the first point on a dendritic stem or not. I agree that the length of the first compartments are better defined your way but the electrical vs anatomical surface area of the soma might differ a lot. For example the morphology I use have a stem where the first point is located much further away from the soma than the somatic diameter. There will hence be a big difference between the anatomical and electrical surface area of the soma.

Anyhow since it is more important for me that I can compare my NEURON model under construction with its GENESIS equivalent, than the accuall anatomical resemblance with the medium spiny neuron it is based upon, I will go ahead and try to fix this manually. I do know that the surface area will also differ between the two models but that is a later problem.

rlindroos
Posts: 17
Joined: Thu Jan 02, 2014 2:23 pm

Re: the distance() function

Post by rlindroos » Tue Jan 21, 2014 7:25 am

I'm still puzzled about what it was you got from Ubuntu's Software Center; will look into this.
I have now reinstalled NEURON using the .deb file. The reason why I thought I got it trough the Software Center the first time is because when clicking on the link to the deb package you are asked if you want to open with the Software Center or save it. Hence by specifying open with the sofware center will install it for you.

This time I saved it and installed it using the terminal instead and have so far noticed no difference.

ted
Site Admin
Posts: 5647
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: the distance() function

Post by ted » Tue Jan 21, 2014 10:03 am

Cells typically have somas that are electrically compact, so the proximal ends of all neurites that arise from the soma are at the same potential. However, the most proximal 3d measurement (x,y,z, and diameter, or in the case of swc files, x,y,z, and radius) on any such neurite will be made at some nonzero distance from the soma's center, and should be drawn at the proper location in any graphical rendering of the cell's structure.
Is this the standard way of doing these measurments? I'm pretty sure that in GENESIS each line of a morphology file (except the first one defining the soma) is interpretted as the end point of a compartment
Neither GENESIS nor NEURON nor any other simulation software dictates what anatomists do when they collect anatomical data from cells. The general policy for measuring neurites is to collect the xyzdiam at its origin and end and at every intervening point where one judges there to be a significant change of diameter or direction. There is no uniform policy with regard to somas; common practices are:
--guess what diameter sphere would have a similar surface area or roughly the same mean diameter (based on a 2d projection of a complex 3d object!)
--make a series of xyzdiam measurements starting at one end and working toward the other
--trace a single outline of the cell ("silhouette")
--trace a stack of outlines at different z levels
the electrical vs anatomical surface area of the soma
Exactly what is the difference? Defined in the literature somewhere?

Where exactly to draw the line between soma and dendrite is a matter of judgement, and that can be rather arbitrary for structures that taper gradually, such as the apical end of a pyramidal cell. That said, in many cells the soma amounts to only a tiny fraction of total cell surface area--a few percent--and omitting it entirely or replacing it with a ~5 um diameter sphere would have little effect on simulation results (as long as whatever channels and synapses attached to it were redistributed over the proximal 50 um or so of each primary neurite).
the morphology I use have a stem where the first point is located much further away from the soma than the somatic diameter.
One must wonder what the original cell looked like; wouldn't assume that there is a physical gap between soma and the "first point." Presumably diameter tapers gradually over the distance between the "soma measurement point" and the "first point."

Post Reply