Malte Petersen

@biodatacore@genomic.social
190 Followers
171 Following
669 Posts
IT infrastructure project lead, former high-performance bioinformatician and #genomics data scientist
Githubhttps://github.com/mptrsen
ORCiDhttps://orcid.org/my-orcid?orcid=0000-0001-7601-9873

@HeidiSeibold has a fine new "6 Steps Towards Reproducible Research". But don't wait to the end for Step 6, "Publish your research outputs". This one can take time, much more than fits comfortably between

"Your paper has been accepted, please check the proofs carefully by Friday"
and
"Your #researchdata is now available for public download"

https://doi.org/10.5281/zenodo.12744715

6 Steps Towards Reproducible Research

A short book with 6 steps that get you closer to making your work reproducible.  PDF and epub versions are available    

Zenodo
#Genomics #SnakeMake
Always nice to see "X of X steps (100%) done" after fighting with the cluster and docker and snakemake to scale an analysis. Short-term growing pains for long-term scalability.

@HeidiSeibold
brand/reputation.
As much as I like the Dagstuhl idea, it is ONE lighthouse. Most folks in science will never be able to go there. So it is very much like a Ivy-league secluded place, or Davos.

We need to make Open Science accessible, by everyone. That means we need many places, that can facilitate quality research, and that even mingles with early career Scientists (possibly even Undergrads?)

I do not think centralized structures are the answer, hence build a federated model

The idea of having a dedicated venue for #OpenScience, akin to Dagstuhl and Oberwolfach, would be amazing, wouldn't it?

We could use this place to create systemic change, allow exhausted researchers to reboot, run hackathons, summer schools, seminars, workshops, or celebrate our achievements in improving the quality of research... 💭

Read more about my daydreaming in this week's newsletter post!

https://buff.ly/3QRaxxH

Dreaming about a perfect space to work on Open Science 💭

new sweatshirt, “rage against the machine learning”

Turns out that #Microsoft a big customer of #Azure #HPC. This is a great example of that; Surface laptops are engineered in Azure.

The fact that the public can also use some of that HPC capability (Genoa-X, InfiniBand, etc) is incidental in some ways.

https://azure.microsoft.com/en-us/blog/azure-high-performance-computing-leads-to-developing-amazing-products-at-microsoft-surface/

Azure high-performance computing leads to developing amazing products at Microsoft Surface | Microsoft Azure Blog

Microsoft Azure HPC plays a vital role in enabling product level simulation models for the Microsoft Surface team. Learn more.

Microsoft Azure Blog

Want to set up your research or code projects in a #FAIR and #reproducible way?

There are very useful templates available. I collected a couple of them in this post.

https://buff.ly/4agJzah

Setting up a FAIR and reproducible project

(comic) Automate all the things!

Work Chronicles

Attending #deRSE24 in #wurzburg tomorrow. Conference on research software engineering in Germany. Going to present our #Snakemake on #HPC teaching material together with @rupdecat

No idea how to return with the rail #strike on Thursday...

Ten simple rules for collaborating with wet lab researchers for computational researchers https://arxiv.org/abs/2402.18348 #qbioOT
Ten simple rules for collaborating with wet lab researchers for computational researchers

Computational biologists are frequently engaged in collaborative data analysis with wet lab researchers. These interdisciplinary projects, as necessary as they are to the scientific endeavour, can be surprisingly challenging due to cultural differences in operations and values. In these Ten Simple Rules guide we aim to help dry lab researchers identify sources of friction; and provide actionable tools to facilitate respectful, open, transparent and rewarding collaborations.

arXiv.org