plotting a network

Moderator: wwlytton

Post Reply
R_Jarvis

plotting a network

Post by R_Jarvis »

I wonder if it is possible to plot a whole network of neurons in shape plot?

I get the sense that you can plot only one soma which you access in the code.

Some times when I open shape plot from after executing a hoc file that includes many neurons I can see a trail of parallel lines. The parallel lines resemble a train track extending out from the first or last neuron that was made. I wonder if these parallel lines are the soma of other neurons in the network?

If it is possible to view a whole network at once I wonder what is the best way to do it?

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

Re: plotting a network

Post by ted »

1. Shape plots show all sections that exist. If sections appear to be clipped or lie outside the frame of a shape plot, use "View = plot" to rescale the image.

2. If there is more than one cell, chances are they will all be drawn on top of each other, unless special care has been taken to prevent visual overlap. Templates that define cell classes can be written in such a way as to provide a method for specifying the xyz coordinates of the "origin" of each cell (0 node of the root section) individually. This controls the apparent placement of the entire cell in a space plot. This feature is built into templates that are written by the Network Builder or Cell Builder; the method is called position() and is used thusly (assuming that the name of the cell class is MyCell):

Code: Select all

objref cel
cel = new MyCell()
cel.position(10, 5, -25) // move origin of cel to (10, 5, -25)
Although position() controls the appearance of shape plots by affecting the screen placement of renderings of cells, it has no effect on network connectivity.

3. Artificial spiking neurons do not appear in shape plots because they don't have sections.

4. NetCons are not displayed, so shape plots of most networks will look like the neural equivalent of a collection of moths.
R_Jarvis

Re: plotting a network

Post by R_Jarvis »

Okay thanks for that.

It seems like if I put some thought into positioning the cells I can graphically convey the number of cells in a network. Even if the positions that they are plotted in says nothing about their underlying connectivity.
Post Reply