Dendrites that look not connected

Managing anatomically complex model cells with the CellBuilder. Importing morphometric data with NEURON's Import3D tool or Robert Cannon's CVAPP. Where to find detailed morphometric data.
Post Reply
tcp

Dendrites that look not connected

Post by tcp »

Dear Forum,

I have imported (import3D) a reconstruction of a neuron.
Because the reconstruction had no soma I made one and I connected the emerged dendrites.
At the shape window, not all dendrites look connected to the neuron. When I call the topology function, the diagram is as it should be.
The signal is propagated through the seemingly unconnected dendrites.

Is there any need to take further action? How can I have something that will not only be right but look right?

Kind regards,
Eleftheria
ted
Site Admin
Posts: 6289
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Dendrites that look not connected

Post by ted »

tcp wrote:Is there any need to take further action?
No. The facts that topology() shows the connections, and charge spreads throughout the cell, prove that the model is fully connected.
How can I have something that will not only be right but look right?
You might try eliminating any direct assignment to the soma's L or diam. Define its geometry with pt3d statements, and only with pt3d statements. Example: for a 10x10 um soma

Code: Select all

soma {
  pt3dadd( 0,0,0,10)
  pt3dadd(10,0,0,10)
}
Then make sure that all of the soma's child sections are connected to its 0 or 1 end, not to soma(0.5). Let me know if that helps.
tcp

Re: Dendrites that look not connected

Post by tcp »

Thanks Ted,
Geometries were already defined by pt3dadd statements and connected to soma(0) or soma(1).

Hmm, should I just carry on?

Best,
Eleftheria
ted
Site Admin
Posts: 6289
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Dendrites that look not connected

Post by ted »

It's just a cosmetic problem. However, if this is an important issue for you, I will have to be able to reproduce the problem myself--which means you should send the necessary code in a zip file to ted dot carnevale at yale dot edu.
Post Reply