How to measure cell length and width with MorphGraphX

Dear all

Do you know how to measure cell length and width with MGX? Thanks

Sample image and/or macro code

(If this is your first post, you may not be able to attach images, in which case make a post with a link or a first post without the image attachment and a second one with)

Background

  • What is the image about? Provide some background and/or a description of the image. Try to avoid field-specific “jargon”.

This is the Arabidopsis hypocotyl epidermal cell stained with Clear See.

Analysis goals

  • What information are you interested in getting from this image?
    I want to get a heatmap with cell length and width

Challenges

  • What stops you from proceeding?
  • What have you tried already?
  • Have you found any related forum topics? If so, cross-link them.
  • What software packages and/or plugins have you tried?
    I have constructed 3D cell mesh and got cell volume and area heatmap, and I also want to get a heatmap with cell length or width. But when I run “Heat map>Measures 3D>cell atlas”, it showed an error. Is this the function to measure cell parameters? Or what should I do before running this step?

That is a pretty nice stack, 2 photon?

Do you have the mesh to go with that?

Dear Prof. Smith,

Thanks for your reply. This is a confocal image stained by ClearSee. As my work proposal, I only made epidermal cell mesh. PS: I have uploaded a new file~

confocal image:https://drive.google.com/file/d/1qW_MBuLw1pZVE2PwLAbXn-lk2s0XvoVj/view?usp=sharing
cell mesh:https://drive.google.com/file/d/1Q8wzEU6Smc-gKwz3JGLXgc9XWi9QZG92/view?usp=sharing

Hi Lei,

Thank you for the mesh. So it is a fixed sample, still very nice.

I put a Bezier line through the middle and ran 3D Cell Atlas on it, and it seems to find the lengths and widths OK. I used “Longitudinal Length” for the length and “Circumferential Length” for the width.

I tried to get a surface mesh from the original stack, but your mesh seems rotated from the stack. Although it seemed to work without it, you should probably still create one to make sure everything is calculated correctly.

You can then export the data from the Attributes. Here I have plotted the cell length as a function of the distance from the base.

Hi Lei & Richard,

The radius of the root seems fairly constant so you don’t require a surface mesh, just the Bezier line is enough to create the root coordinates.
3D CellAtlas then gives you the longitudinal (=length), radial (=depth) and circumferential (=width) coordinates & sizes of all cells
If you are interested in more details ask us or you could also check out the 3D CellAtlas user guide:

Thanks a lot. This is really helpful, I will try it.

This is really a good suggestion, I will try. Thanks :slightly_smiling_face:

Dear Richard

I have tried another sample cell mesh but the result showed a bit wired. To explain clearly I put my process below.

load 3D cell mesh;
view>ticked “draw”, “grid”, “bezier”> rest;
mesh>cell atlas 3D>tool>collapse Bezier points
move to mesh central
bezier point.bmp (4.2 MB)
select a cell on one side> A analyze cell 3D> B assign cell types (has multiple segements-No)>
x-dim>Radical cell length
y-dim>Longitudinal cell length
select one cell clusters.
Heat map> heat map new> longitudinal cell size
longitudinal cell size.bmp (4.2 MB)

this result doesn’t reply to the real parameters. Do you know where is the problem in my process?

Dear Soeren,

I have tried measured 3D cell parameters in the followed protocol
load 3D cell mesh;
view>ticked “draw”, “grid”, “bezier”> rest;
mesh>cell atlas 3D>tool>collapse Bezier points
move to mesh central
bezier point.bmp (4.2 MB)
select a cell on one side> A analyze cell 3D> B assign cell types (has multiple segments-No)>
x-dim>Radical cell length
y-dim>Longitudinal cell length
select one cell cluster.
Heat map> heat map new> longitudinal cell size
longitudinal cell size.bmp (4.2 MB)

The data seemed not to refelt the reality parameters. Do you know where is the problem with my process.

Thanks a lot.

Hi Lei,

You are right, the heatmap doesn’t look correct. I can’t see any problem in your steps (Note that you don’t need to run the “B Assign Cell Types” process, if you only have epidermis cells).
Maybe there is a bug or something less obvious went wrong.
Which version of MGX are you running? Do you see the process “Mesh/Cell Atlas 3D/Root/Tools/Display Root Cell Directions”? You could run it with only its parameter Longitudinal set to Yes, (Radial, Circumferential: “No”), which should show you the longitudinal direction as Cell Axis to confirm it’s direction (in order to see it make the mesh a bit transparent or turn “Surface” of in the Main tab.
How are the other heatmaps? or instance “Longitudinal (Absolute)” should give you the cell coordinates along the Bezier line. “Cicumeferential Angle” should give you the similar heat for the same cell files.

Dear Smith,

I suppose the MGX installed in my computer is pretty old version as it showed like below

For the other measurement function is like the image and all showed not correct data.

By the way, dose this mean that the measurement of MGX can only used for comparable, not get the real value? I mean the cell length and width.

Thanks a lot.

Best regards.

Lei Li.

Dear Soeren,

I am sorry reply to you with the other’s name, I am really sorry for this. And I have just run display root cell direction and find it really a disorganized…

Hi Lei,

Those directions don’t look good and yes, your version is pretty old (about 2-3 years old probably). My guess is that in your version it doesn’t work without surface mesh as described above, as this was implemented more recently.
I would suggest you update your MGX! I’ll send you a PM for that