The last blog post I wrote was about Life Science Support on HPC clusters. Honestly? It was more of a rant. Not a good blog post.

So, someone suggested I delete it, which I did. It took me a long time to recover. Now, I have re-written this blog post. I think it is better. I weighted every phrase. It still has some rant flavour.

Here it is: https://blogs.fediscience.org/rupture-de-catenaire/2025/12/15/empowering-life-scientists-what-hpc-support-should-look-like/

The upcoming articles will have more of an example character. But I still will not be able to update on a regular basis.

#Bioinformatics #HPC #ReproducibleComputing #Snakemake #Nextflow

@Dutch_Reproducibility_Network

In fact, I am a #Snakemake co-maintainer and teacher. I was not aware of WorkflowHub - and that was an omission on my part. We actually support and favour this kind of registration: https://snakemake.readthedocs.io/en/stable/snakefiles/deployment.html#uploading-workflows-to-workflowhub

In my original post, I also neglected to mention the integration of WorkflowHub with #RO-Crate and in turn, the integration of RO-Crates with nanopubs. I am actively working on a better support for #nanopub and RO-Crates with @fbartusch. The question, how I teach that stands: The #HPC world (at least my bubble) is not really supportive for #ReproducibleComputing . All #OpenScience shenanigans are frowned upon. And PIs in my vicinity are still on this level: https://phdcomics.com/comics/archive.php?comicid=1531 - so, how do we educate the educators?

Just a minute ago, I accidentally discovered Workflowhub (https://about.workflowhub.eu/) when browsing through a few Nanopubs. I was not aware of this repository.

A first thought: "Yet another repository to register #FAIR materials? Seriously?" Then again, I am all in favour of #reproducibleComputing and want to see more of it.

On the other hand: transparency in our work is central to the #OpenScience idea. NOT cluttering a file space so that we can be transparent in our work is paramount.

Lately, we have seen #nanopub, #ROcrates, and apparently WorkflowHub, too. (This is by no means a comprehensive list.) I wonder: Are we getting too many solutions to keep up-to-date? Are we risking frightening young researchers with the high standards we are setting? We are by no means where we want to be in terms of #transparency and #reproducibility in research, or are we?

Note, that I understand how competing systems and tool sets arise. Still, my nonrhetorical questions (1 & 2) stand.

Why? Why are so many admins tinkering with their SLURM setup?

I say tinkering because there are understandable amendments and breaking changes.

A breaking change is when you require developments made on "your" cluster really, really hard to port onto another. When admins do that, they were tinkering with the SLURM source code.

😱😱😱😱😱😱😱😱😱😱😱😱😱😱

#HPC #academicchatter #reproducibleComputing

RE: https://fediscience.org/@snakemake/115611862667755622

Now, this is huge!

Thanks to a contribution from Cade Mirchandani (Santa Cruz, CA), whom I met at this year's #SnakemakeHackathon users can now supply a partition profile. So, instead of wrangling #SLURM partition information into a workflow profile (indicated with --workflow-profile), we can now have a global file to contain this information.

I added a time conversion function, such that the SLURM time format is obeyed, too.

There are several other development needs, before we continue in this direction (e.g. parsing SLURM partition information directly). But a task to be done is summing this up for non-users, e.g. administrators, is due too.

In any case, I think this merits a new major version.

#Snakemake #HPC #ReproducibleComputing

This took a while. After the new version of the Snakemake paper (a rolling paper on F1000) came out, the DOI now is "working" 🥳 :

https://doi.org/10.12688/f1000research.29032.3

From my point of view, it particularly describes the working with various #HPC batch systems. And: Development did not cease. If you want to follow our announcement bot for updates: @snakemake

#Snakemake #ReproducibleComputing #DataAnalysis #OpenScience #WorkflowManagement

Just submitted a talk for FOSDEM (been invited). They asked to attach an icon-image for the talk. So I drew one. The compute racks are difficult to identify as such, but this is as far as my aquarelle skills go.

#Snakemake #ReproducibleComputing #HPC