Problems with script in Anaconda Navigator (jupyer)

[StiThor]

Hi,

Am getting this error message:

Do you have a solution for the problem?

[nilseling]

Hi @StiThor, could you please upgrade to the newest version of the pipeline? This should fix the error.

[StiThor]

Hi @nilseling,

thanks for the quick response! I can try to use the newest version of the pipeline. Is the version of Anaconda Navigator important? I have one computer running with version 2.1.1 with the old ImcSegmentationPipeline which works as it should. While the other computer is running with the newest version of Navigator (2.1.4) which doesn't work.

[nilseling]

I'm not too familiar with Anaconda Navigator but what I'm assuming is that you are using a newer version of the tifffile library (see #90 and #100). The problem should be resolved when you create a conda environment following the documentation.

[StiThor]

Following the Usage protocol with the new version of the pipeline gives me this error:

Any idea on how to solve this one?

[nilseling]

Please put a # in front of the third line in the script.

[StiThor]

That works! :) Now another error occurs. Any suggestions ?

[StiThor]

I ran the program as an administrator

[nilseling]

Hmm, this one is not trivial for me to solve as I'm not familiar with handling file permissions in Windows. You don't have write access to the folder. This is not an ImcSegmentationPipeline issue but rather a problem of the computational setup.

[StiThor]

Good morning @nilseling,

Thanks for all the help you gave me! I was able to get it working :) The Errno13 error is not a computational setup problem as we first thought, but rather a problem with the path to the panel.csv file. Correcting this and some other issues solved all problems. I also want to use the opportunity to give you (and all others working with imcsegmentation) a big thanks for helping us all around the world working with the handling of IMC data! I have just a question at the end: what is the difference between the new segmentationpipeline and the new one?

[nilseling]

Hi @StiThor, I'm glad that it worked! And thanks for your kind words! The ImcSegmentationPipeline v2 (old) and ImcSegmentationPipeline v3 (new) are not so much different. The changes can be seen in the CHANGELOG. The main difference is that we deprecated the imctools package and offer an internal imcsegpipe package for data handling. The maintenance changed (all of this was build by @votti - to biggest thanks to him!) and changing the architecture slightly helps us to maintain it. We also extended the documentation and provide example data. I hope that clarifies the change. Feel free to close the issue and open a new one if you run into problems again.

[StiThor]

I could also see that there was a "hot pixel fixing" step in the new script?

[nilseling]

In the old version this was done as part of the first CellProfiler pipeline. The problem was that CellProfiler did not write correct multi-channel images so we had to strip it out of the pipeline and add it to the pre-processing script.

[StiThor]

Okei thanks :) Im now trying to get the Steinbock working using Docker since this workflow enables Deepcells. But this of course is not without problems. Im using windows and have been able to start docker properly by installing Ubuntu and wsl2. I can`t seem to understand how to proceed. I follow whats written here: https://bodenmillergroup.github.io/steinbock/v0.6.2/install/docker/

[StiThor]

I run the command "docker run ghcr.io/bodenmillergroup/steinbock:0.6.2" in Ubuntu which installs what I assume is the package for running Steinbock. But after Steinbock is installed non of the following commands works. How do I proceed? Best, Stian

[nilseling]

Hi @StiThor,

I'm closing this issue now as it is not related to the ImcSegmentationPipeline anymore. Please open a new issue at https://github.com/BodenmillerGroup/steinbock/issues. Also please make sure that you are using the newest steinbock v0.14.2 and follow the documentation at https://bodenmillergroup.github.io/steinbock/latest/.