cuntzlab / treestoolbox Goto Github PK

When Trying to import a Neurolucida file using the "load Neurolucida one soma" Function. The software always gets stuck on Please...Wait...Yeah... Make somata.

When uploading using just load Neurolucida, The structure looks great but when exporting to NEURON, the soma is broken up into multiple soma's despite it starting as one.
Is there a solution to this?
Kind Regards,
Laura

Hi,

First of all thank you for a great toolbox!!!

I have trees which were reconstructed with a "square" to represent the size of the soma. I wanted to delete this square, but still have a soma in my tree. For this I first used the function quaddiameter_tree, which removed the square, and then implemented the soma_tree function to create the soma; I used for maxD the value at tree.D(1), since this is the diammeter of the soma. After doing so, if I compare my original tree with the final one, soma_tree.D(1) is half the value of tree.D(1). Shouldn't they be the same value? Am I doing something incorrectly or could this be a small bug?

My code is something as:

start_trees
tree = load_tree('females.mtr');
q_tree = quaddiameter_tree(tree, 0.01, [], []);
s_tree = soma_tree(q_tree, tree.D(1), [], []);

And output is:

tree.D(1)
ans = 18.8000

s_tree.D(1)
ans = 9.4000

I appreciate any input you could give me!
Nat.

(The edit was not to overwrite the function with a variable)

Is there a way to break up an imported neuron (from neurolucida) into proximal, middle and distal branches?
I was trying to use the path length function on the gui but am unsure what the correct procedure is.

Hi! I was looking at the paper "Dissecting Sholl Analysis into its functional components" and wanted to use the rootangle_tree function with the show option, but I get the following error message:

rootangle_tree(sample_tree, '-s')

Error using trapz (line 58)
LENGTH(X) must equal the length of Y in dim 2.

Error in rootangle_tree (line 80)
plot(mAngV,pdf/trapz(AngV,pdf),'black')

I went to look into the code for root angle and it seems the variable AngV is a 1x25 double, while mAngV and pdf are both 1x24 doubles, which is what causes this error. Could this be a code error or maybe is it due to the matlab version I'm using? (R2016b)

To try to "go around" the generated error, I did the plot by using:

plot(mAngV, pdf/trapz(AngV(1:24), pdf), 'black')

which does generate a plot, but I am not sure if it is correct, since it does not start in x=0 like the plots in Figure 3C from the paper.

The plot is also exactly the same if instead of using AngV(1:24), I use AngV(2:25).

I would very much appreciate your input about the plot, if I can do it like this or if it needs to be another way! :)

Hi
I am plotting neurons whose axons initiate at dendrite branches. The toolbox automatically connects the initiate segment back to soma with a straight line. Is there a way to turn off this feature.
Thanks

Hey there,

on your main page you list a contributor that has written a Python wrapper for the Trees Toolbox. Is it available already? How would I go about, if I wanted to run your code from a Python script?

Thanks a lot!

`[tree,path,~] = load_tree('C:\Users\Experiment\cameraLucida\Example_data\FR130\FR130_1_tracing.swc','r');
tree = resample_tree(tree, 5, '-s'); % resample tree every 5 microns

%% Step 2: group nodes to branches
[rtree, order] = sort_tree(tree,'-LO');
[sect, vec] = dissect_tree(rtree);`

It is only segmenting half the dendrites that are there. Do you know what the issue is?

Many Thanks!

Hi Hermann, this is Sumit.

ROOTANGLE_TREE function assumes that tree begins at [0,0,0], which is not generally the case [X2, Y2 and Z2 values in the function are not pure displacement values for trees that are not rooted at 0,0,0].

Line 64 to 67 needs to updated or the tree needs to be translated (default to 0,0,0) as part of the ROOTANGLE_TREE function.

Otherwise angles generated for trees not beginning at [0,0,0] are nonsensical. This makes all the downstream results such as BF, K etc false.

%% just add the following after resampling in line 50

% if root is not at [0,0,0] translate the tree, so that root lies in [0,0,0]

if ~(rtree.X(1)==0 &&  rtree.Y(1) == 0 && rtree.Z(1) ==0 )
       rtree = tran_tree(rtree);
end