specify snakelike version during installation?

[ceesu]

Instructions for installation currently start with

mamba create -c conda-forge -c bioconda --name snakemake snakemake snakedeploy

However, it seems like with the latest version of snakemake (8.2.3), running snakemake leads to the error

NotImplementedError: Remote providers have been replaced by Snakemake storage plugins. Please use the corresponding storage plugin instead (snakemake-storage-plugin-*).

[Funkoverload]

Same issue here. When I downgrade to snakemake 7.32.4, I get another errror:

File "/extra/sra/.snakemake/conda/6ced698129ee201eb5ebbc98e1ca2dab_/lib/python3.12/sqlite3/init.py", line 70, in getattr
raise AttributeError(f"module {name!r} has no attribute {name!r}")
AttributeError: module 'sqlite3' has no attribute 'OptimizedUnicode'

[dlaehnemann]

Instructions for installation currently start with

mamba create -c conda-forge -c bioconda --name snakemake snakemake snakedeploy

However, it seems like with the latest version of snakemake (8.2.3), running snakemake leads to the error

NotImplementedError: Remote providers have been replaced by Snakemake storage plugins. Please use the corresponding storage plugin instead (snakemake-storage-plugin-*).

How exactly are you invoking the workflow? This sounds like it might be an issue with using a cluster system. For infos on moving from snakemake 7 to snakemake 8, have a look here (especially at the info on storage providers):
https://snakemake.readthedocs.io/en/latest/getting_started/migration.html#migrating-to-snakemake-8

[dlaehnemann]

Same issue here. When I downgrade to snakemake 7.32.4, I get another errror:

File "/extra/sra/.snakemake/conda/6ced698129ee201eb5ebbc98e1ca2dab_/lib/python3.12/sqlite3/init.py", line 70, in getattr raise AttributeError(f"module {name!r} has no attribute {name!r}") AttributeError: module 'sqlite3' has no attribute 'OptimizedUnicode'

The 7.32.4 issue is probably an incompatibility with python3.12, see here for the deprecation notice of the above attribute:
python/cpython#92547 (comment)

I remember that some python3.12 incompatibilities should be fixed in snakemake 8, so this might go away once you get the workflow running with snakemake 8. Otherwise, a quick fix can be to downgrade the version of python used manually, to get sqlite3 to still have the respective attribute:

mamba activate snakemake
mamba install python=3.10

[ceesu]

Thanks so much @dlaehnemann. I got it to work by deleting everything and using the old version of snakemake:

mamba create -c conda-forge -c bioconda -n name snakemake=7.32.4

I also found that using any version of snakemake older than 7.3.4 also causes an f string error, seems to be related to snakemake/snakemake #2480

@Funkoverload I also that error with version 2.0.0 of this pipeline, but seems that it was fixed most recent version by #69 . maybe you can try installing with

snakedeploy deploy-workflow https://github.com/snakemake-workflows/rna-seq-star-deseq2 . 

Instead of what's given in the instructions, i.e. snakedeploy deploy-workflow https://github.com/snakemake-workflows/rna-seq-star-deseq2 . --tag v2.0.0

[dlaehnemann]

Ah, yes. I have already forgotten about #69 and the similar error there. I have now created a new release, so that the following should also pull in the latest changes (and provide them via a stable tag version):

snakedeploy deploy-workflow https://github.com/snakemake-workflows/rna-seq-star-deseq2 . --tag v2.1.0

It should take a day for the command update in the snakemake workflow catalog, but with the release already done, the command I provide here should already work.

I'll close this issue for now. Feel free to reopen if any of this re-occurs or open new issues for new problems.

[dlaehnemann]

This workflow should now also work with snakemake >=v8, starting from tag="v2.1.1". The underlying issue was fixed in #76 .