Hi, I am new to Nextflow and I would highly appreciate your help. I am running a custom rna-seq pipeline, made by ex-empoyee, on HPC. While the first tool of the pipleine executes correctly (fatsp), the pipeline breaks while trying to run the second tool, with the error message:

Lmod has detected the following error: The following module(s) are unknown:

"star/2.7.2"

Please check the spelling or version number. Also try "module spider ..."

It is also possible your cache file is out-of-date; it may help to try:

$ module --ignore_cache load "star/2.7.2"

Also make sure that all modulefiles written in TCL start with the string

#%Module

When I check the nf script, star version is hardcoded:

process Star_Mapping_rna {

module "star/2.7.2:samtools"

....

}

and when I check module avail on the cluster the different version of star is present. My first guess was that this was causing the code to break, but when I checked if versions matched for the first tool in the script (fastp), the versions are not the same and yet that part is executed without problems (fastp/0.22.0 in nf script vs. fastp/1.0.1/gcc on the cluster). Do you have any idea what is going on? Thank you!

You can change the used module via a configuration file:

process {
    withName: Star_Mapping_rna {
        module = "<name of module>"
    }
}

If there is no

star/2.7.2

module,

module

will compain. Why this does not happen with fastp, I do not know, my guess would be that the correct version is available somehow, can you simply do

module load fastp/0.22.0

and see if it works?

@Niklas Schandry you are right,

module load fastp/0.22.0

loads fastp! @Nicolas Vannieuwkerke the pipeline works this way, thank you guys so much!

Glad to hear that! Keep in mind that using

module

really hinders the reproducibility of your pipeline so I would suggest migrating to containers (Docker, Singularity, Podman...) if you ever plan to update the pipeline

Thank you for the advice!