2D spindle morphometrics

[coralzhou]

Dear tischi, it would be very helpful for Xenopus extract folks to have 2D spindle morphometrics as a future option since the way we process our samples essentially flattens the spindles. Thanks!

[tischi]

Dear @coralzhou thank you for your request! I am a bit busy these days and am not sure when I would find the time to work on this. But could you already provide two example images where the spindles look obviously different?

[coralzhou]

Hi Christian,

Thanks for your response! It took me a few days to gather the data but I now have a folder containing laevis and tropicalis spindles, which are quite different in size and morphology.

Here is the data: https://drive.google.com/drive/folders/1rsASmty4TI7lmkV929aQCgEfVy7E9U14?usp=sharing https://drive.google.com/drive/folders/1rsASmty4TI7lmkV929aQCgEfVy7E9U14?usp=sharing

Please let me know if you need anything else!

Many thanks,

Coral Zhou Postdoctoral Fellow, Heald Lab Department of Molecular and Cell Biology University of California, Berkeley

On Nov 2, 2021, at 3:34 AM, Christian Tischer @.***> wrote:

Dear @coralzhou https://github.com/coralzhou thank you for your request! I am a bit busy these days and am not sure when I would find the time to work on this. But could you already provide two example images where the spindles look obviously different?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/tischi/spindle3d/issues/31#issuecomment-957314813, or unsubscribe https://github.com/notifications/unsubscribe-auth/AO63Y2AMLN3WBUMCX4CZJ2LUJ65BRANCNFSM5GYJEGHA. Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

[coralzhou]

Hi Christian,

Just wanted to check in about this—were you able to find time to work on it? If not, I am happy to wait.

Thanks!

Coral Zhou Postdoctoral Fellow, Heald Lab Department of Molecular and Cell Biology University of California, Berkeley

On Nov 5, 2021, at 10:33 AM, Coral Zhou @.***> wrote:

Hi Christian,

Thanks for your response! It took me a few days to gather the data but I now have a folder containing laevis and tropicalis spindles, which are quite different in size and morphology.

Here is the data: https://drive.google.com/drive/folders/1rsASmty4TI7lmkV929aQCgEfVy7E9U14?usp=sharing https://drive.google.com/drive/folders/1rsASmty4TI7lmkV929aQCgEfVy7E9U14?usp=sharing

Please let me know if you need anything else!

Many thanks,

Coral Zhou Postdoctoral Fellow, Heald Lab Department of Molecular and Cell Biology University of California, Berkeley

On Nov 2, 2021, at 3:34 AM, Christian Tischer @. @.>> wrote:

Dear @coralzhou https://github.com/coralzhou thank you for your request! I am a bit busy these days and am not sure when I would find the time to work on this. But could you already provide two example images where the spindles look obviously different?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/tischi/spindle3d/issues/31#issuecomment-957314813, or unsubscribe https://github.com/notifications/unsubscribe-auth/AO63Y2AMLN3WBUMCX4CZJ2LUJ65BRANCNFSM5GYJEGHA. Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

[tischi]

Hi @coralzhou,

To be honest I cannot predict when and whether I will find time to work on this. The issue is that the whole point of the repository was to enable 3D analysis and I am afraid this is very much backed into the code. "Extending it to 2D" may be non-trivial...

Maybe as a first step: @TobiasKletter Can you imagine/try some rick to make the 2D input data 3D such that it would produce the correct measurements? Maybe adding empty slices on the top and bottom, duplicating the slices, maybe just see what happens of you run it with the 2D data?

[TobiasKletter]

Sure! @coralzhou Thanks a lot for the image files! They are not calibrated, can you please provide us with the pixel sizes (i.e. how many px/µm)?

[coralzhou]

Hi Tobias and Christian,

Thanks for the reply!

Tobias, the images I shared all have a pixel size of 0.161 x 0.161 microns.

Hope this is helpful,

Coral Zhou Postdoctoral Fellow, Heald Lab Department of Molecular and Cell Biology University of California, Berkeley

On Nov 24, 2021, at 2:54 AM, TobiasKletter @.***> wrote:

Sure! @coralzhou https://github.com/coralzhou Thanks a lot for the image files! They are not calibrated, can you please provide us with the pixel sizes (i.e. how many px/µm)?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/tischi/spindle3d/issues/31#issuecomment-977761278, or unsubscribe https://github.com/notifications/unsubscribe-auth/AO63Y2CE3Q4TZC7655GDNLLUNS77JANCNFSM5GYJEGHA. Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

[tischi]

@TobiasKletter any luck here?

[TobiasKletter]

Actually, yes, I played around a bit but this got a little buried. What I got so far:

https://www.dropbox.com/s/ph6hkcatcp9x3ys/cycled_40x1_6-1.tif?dl=0 Using one of Coral's sample images, first I prepended and appended empty slices, and filled in the following properties: Pixel width: 0.161 microns Pixel height: 0.161 microns Voxel depth: 10 microns (For now this is pretty much arbitrary, it just needs to "stretch out" the chromatin in z such that the shortest axis is aligned with the spindle axis.)

Next, I "scaled" the image in z: run("Scale...", "x=1.0 y=1.0 z=10 width=197 height=188 depth=30 interpolation=Bilinear average create"); With the interpolation, we thus create an artificial 3D spindle + chromatin. https://www.dropbox.com/s/u60rl14ajva0lt5/cycled_40x1_6-1_INTERPOLATION.tif?dl=0

The plugin happily analyses these images and the output looks "correct", but I guess the only meaningful parameter here is spindle length. However, the chromatin and spindle volumes segmentation looks good, I think. https://www.dropbox.com/s/bp18ynw98hvrglw/cycled_40x1_6-1_INTERPOLATION-out.zip?dl=0

I next tested this strategy on files where I actually have the confocal stack. I cropped the central slice and artificially stretched it out as above and did the Spindle3D analysis. The segmentation looks quite okay compared to the original stack but there are small differences in terms of thresholding. Need to check this further on more samples, though. I will soon post an update to this.

In general, do you think this might be an avenue worth checking out?

[coralzhou]

Hi Tobias and Christian,

This seems promising! Not sure if the question at the end was directed to me but I am happy to try out any updated plugins for 2D analysis.

Best wishes,

Coral Zhou Postdoctoral Fellow, Heald Lab  http://mcb.berkeley.edu/labs/heald/ Department of Molecular and Cell Biology University of California, Berkeley

On Dec 13, 2021, at 7:06 AM, TobiasKletter @.***> wrote:

Actually, yes, I played around a bit but this got a little buried. What I got so far:

https://www.dropbox.com/s/ph6hkcatcp9x3ys/cycled_40x1_6-1.tif?dl=0 https://www.dropbox.com/s/ph6hkcatcp9x3ys/cycled_40x1_6-1.tif?dl=0 Using one of Coral's sample images, first I prepended and appended empty slices, and filled in the following properties: Pixel width: 0.161 microns Pixel height: 0.161 microns Voxel depth: 10 microns (For now this is pretty much arbitrary, it just needs to "stretch out" the chromatin in z such that the shortest axis is aligned with the spindle axis.)

Next, I "scaled" the image in z: run("Scale...", "x=1.0 y=1.0 z=10 width=197 height=188 depth=30 interpolation=Bilinear average create"); With the interpolation, we thus create an artificial 3D spindle + chromatin. https://www.dropbox.com/s/u60rl14ajva0lt5/cycled_40x1_6-1_INTERPOLATION.tif?dl=0 https://www.dropbox.com/s/u60rl14ajva0lt5/cycled_40x1_6-1_INTERPOLATION.tif?dl=0 The plugin happily analyses these images and the output looks "correct", but I guess the only meaningful parameter here is spindle length. However, the chromatin and spindle volumes segmentation looks good, I think. https://www.dropbox.com/s/bp18ynw98hvrglw/cycled_40x1_6-1_INTERPOLATION-out.zip?dl=0 https://www.dropbox.com/s/bp18ynw98hvrglw/cycled_40x1_6-1_INTERPOLATION-out.zip?dl=0 https://user-images.githubusercontent.com/45814595/145831046-68fc001c-9a79-4980-9345-c63b5191995a.png I next tested this strategy on files where I actually have the confocal stack. I cropped the central slice and artificially stretched it out as above and did the Spindle3D analysis. The segmentation looks quite okay compared to the original stack but there are small differences in terms of thresholding. Need to check this further on more samples, though. I will soon post an update to this.

In general, do you think this might be an avenue worth checking out?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/embl-cba/spindle3d/issues/31#issuecomment-992571700, or unsubscribe https://github.com/notifications/unsubscribe-auth/AO63Y2GHRYHB5QTN6QTMZH3UQYDZFANCNFSM5GYJEGHA. Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.