Importing TGMM Output to MaMuT

Hello,

I’ve been working with lightsheet microscopy data for a project, and we have successfully ran the most recent version of TGMM cell tracking from the Keller Lab. We were hoping to visually inspect and manually filter/correct some of our tracks using MaMuT on Fiji. However, I have been having some issues with opening the TGMM output files on MaMuT.

For background on the data: we have about 1082 time points of 960x960x599px images, all single channel and all converted to the keller lab blocktype (klb) format. I’ve successfully generated an XML for these KLBs using BigDataViewer. On the TGMM output side, we have 817,323 objects detected across the movie in 9,166 lineages.

When I attempt to import this data onto MaMuT via “Plugins>MaMuT>Import TGMM results in MaMuT”, and enter the xml directory for the image file, the “XML_finalResult_lht” folder for the TGMM output files, Fiji will start a progress bar with the title “Loading TGMM outputs”. However, even after running this overnight, there is no progress on the bar. This holds true also when I click and unclick the “crop data” option in MaMuT (as suggested in the TGMM manual). Does anybody have any idea what might be the issue in loading the data? The lack of an error message has made it tricky for us to debug it.

Thanks ahead of time for all answers!

1 Like

Hi @daksel

This looks like a complicated bug. Could you share the data - privately if you need to - so that I can investigate? I maybe need some time as I am a bit overbooked.

1 Like

Hi,

Happy to share the data in privite, although it is somewhat large even after klb compression (about 265GB). I will reach you via email about this if that is ok?

Thanks for the support!

Ok just try to send the first 10 timepoints, or something like that, with the TGMM folder for these timepoints. This should be enough.

Ok thanks for the data. But I cannot open the image in MaMuT. It crashes when I try to open it in BDV or MaMuT I get this:

[ERROR] Module threw exception
java.lang.RuntimeException: java.lang.RuntimeException: expected <SpimData> root element. wrong file?
	at bdv.ij.BigDataViewerPlugIn.run(BigDataViewerPlugIn.java:118)
	at org.scijava.command.CommandModule.run(CommandModule.java:199)
	at org.scijava.module.ModuleRunner.run(ModuleRunner.java:168)
	at org.scijava.module.ModuleRunner.call(ModuleRunner.java:127)
	at org.scijava.module.ModuleRunner.call(ModuleRunner.java:66)
	at org.scijava.thread.DefaultThreadService$3.call(DefaultThreadService.java:238)
	at java.util.concurrent.FutureTask.run(FutureTask.java:266)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
	at java.lang.Thread.run(Thread.java:745)
Caused by: java.lang.RuntimeException: expected <SpimData> root element. wrong file?
	at bdv.spimdata.XmlIoSpimDataMinimal.load(XmlIoSpimDataMinimal.java:84)
	at bdv.BigDataViewer.open(BigDataViewer.java:493)
	at bdv.ij.BigDataViewerPlugIn.run(BigDataViewerPlugIn.java:114)
	... 9 common frames omitted

It looks like the xml file is truncated. Can you have a look at it?

Hi,

I just updated the xml file - let me know if it looks ok now?

Deniz

Yep I can reproduce the bug. Thanks!
It made me realise I do not know the exact specifications of the TGMM file format, so I just asked for it.

Hi @daksel
No time so far sorry. The only progress I made was to have Kate explain to me what were the conventions for TGMM.
I still have so much to do I will still need some time sorry again.

No worries at all, I just wanted to check in! Thanks again for looking into it at all :slight_smile:

Thanks again and hope to hear from you when you get a chance to look into it,

Deniz

Just wanted to post a quick update and possibly help resolve this issue for any other user: It seems that the issue was with the indexing of timepoints. When I switched to zero-based numbering for my KLB files, and reran TGMM on those, I was able to import my result into MaMuT with no issues.

1 Like