can't seem to find a place to store startup environmental variables [with containers]

[nandan75]

I want to use my newly assembled bacterial genomes with some reference genomes to generate a phylogeny tree. I found your tool very interesting.

I am using a singularity container to run GToTree in my login on a HPC server

I was trying to run the test singularity exec $base_path/singularity_containers/gtotree_1.8.4--h9ee0642_0.sif gtt-test.sh

when I was asked to
The 'GToTree_HMM_dir' variable is not set :( Use gtt-data-locations to check and configure.

When trying to set the paths (I can access these paths),
singularity exec $base_path/singularity_containers/gtotree_1.8.4--h9ee0642_0.sif gtt-data-locations set

I get the error
We can't seem to find a place to store startup environmental variables
:( '
This is likely due to permission restrictions in the conda environment
location, and in your home location.

Can someone assist?

thanks :-)

[AstrobioMike]

Hi there, @nandan75 :)

Thanks for your interest in trying GToTree, but I'm terribly sorry to say I haven't had the opportunity to configure it to work in a container like with docker or singularity :/

I have things built and tested as a conda environment to be the primary way of using it. And I don't have a setup where I can troubleshoot and iron things out in your situation.

It kills me to say it, but I don't think I can help. Is there any chance you can use a conda/mamba installation instead of running it with singularity? Or are you restricted due to the HPC you have access to?

I can also say for most applications it will also run fine on a standard laptop with a conda installation

[nandan75]

Hi @AstrobioMike

Thanks for your prompt response.

I am trying to integrate the tool along with a few others in a nextflow pipeline.
So ideally I do not want the users to perform any installations for any of the tools which they use in the pipeline - thats where the singularity use!!

Hmm I understand your position.

But again appreciate your response.

Of course any hints in the future are welcome.

Best luck.

Nandan

[AstrobioMike]

I totally understand! I have used it with nextflow with conda, but not with singularity :/

Any chance you could email me the gtotree_1.8.4--h9ee0642_0.sif you've already made, so maybe i could try poking at it in a container if i get the chance in the future? You could send it to MikeLee@bmsis.org

[nandan75]

ok. I have not used conda with nextflow- I am assuming it is equally easy doing so? I will give it a try if you think it is.

Also I got the singularity image for your tool from https://quay.io/repository/biocontainers/gtotree

Cheers mate.

Nandan

[AstrobioMike]

Oh that's right, I forgot it was on bioconda, ha

It is similarly easy using nextflow with conda, just environment yaml files instead of sif files or pointing to containers via links. But I'm not sure if they can be used together (like singularity for some tasks and conda for others). It might be all one or the other, which might complicate what you've already set up

Sorry for the GToTree deficiency here. Hopefully one day I'll be able to address it

Thanks for writing in about it, Nandan :)

[nandan75]

Thanks . You have been a great help. Appreciate it. I will check the conda way and see if works with my other singularity containers. Theoretically it should, as I am using an independent tool per nextflow process. But I am no nextflow expert. So I will try, Thanks, Nandan

…

On Fri, 1 Mar 2024, 12:57 pm Mike Lee, ***@***.***> wrote: Oh that's right, I forgot it was on bioconda, ha It is similarly easy using nextflow with conda, just environment yaml files instead of sif files or pointing to containers via links. But I'm not sure if they can be used together (like singularity for some tasks and conda for others). It might be all one or the other, which might complicate what you've already set up Sorry for the GToTree deficiency here. Hopefully one day I'll be able to address it Thanks for writing in about it, Nandan :) — Reply to this email directly, view it on GitHub <#86 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AC4A5CLAKHBEE4RUYZZT53LYV7N7DAVCNFSM6AAAAABEA4MDG6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNZSGMYTENBXGI> . You are receiving this because you were mentioned.Message ID: ***@***.***>