mastouille.fr est l'un des nombreux serveurs Mastodon indépendants que vous pouvez utiliser pour participer au fédiverse.
Mastouille est une instance Mastodon durable, ouverte, et hébergée en France.

Administré par :

Statistiques du serveur :

582
comptes actifs

#ReproducibleResearch

1 message1 participant0 message aujourd’hui

New server, new #introduction:

👋 Hi! I'm Eric! I'm a scientific programmer & educator who writes mostly :rstats: code. I currently work part time at #UniversityOfArizona where I help researchers by making R packages, #Shiny apps, automated data workflows, and training them on #ReproducibleResearch practices. I also mentor data scientists for @Posit Academy. I am #OpenToWork as a contractor if any of those skills sound useful to you.

My background is in plant chemical ecology and population ecology and for my PhD I studied #tea and did field work in China. I love tea (although I've been drinking coffee more and more lately) and practice #GongFuCha when I can. I love #foraging and tasting new things.

I currently live in #Tucson but the #BayArea will always be my home.

Suite du fil

Reproducible Crisis in Science?

Yes, we have a technological crisis!

A example?

Many results from biology rely on flow cytometry techniques. And Becton, Dickinson and Company (BD Biosciences) provides such instruments. They decide themselves based on financial criteria, e.g., maintenance or avaibility. Their instruments are fully proprietary. And we – scientific – have no control (power) on these instruments; cash cow end-user!

#OpenScience and #ReproducibleResearch ask that such technology must be fully a Common Good – copyright and patent belonging to people, us, all.

You want to contribute in Science? Don’t work on yet another variant of variant already known result;

Instead be radical: Work on technology! Less personal reward in the short term; more collective reward in the long term.

When would we have open and free flow cytometry instruments?

Yesterday @rougier and I gave an informal #ReproducibleResearch workshop at Inria with ~20 scientists, half of whom working in HPC, the rest of them working on user interfaces, computer graphics, number theory, and robotics—very insightful to see how each community approaches these issues.

@rougier had excellent examples to get the discussion started. I followed up on computational reproducibility with #HPC examples and a touch of #Guix and #SoftwareHeritage:
gitlab.inria.fr/lcourtes/ateli

GitLabLudovic Courtès / Atelier recherche reproductible · GitLabGitlab at Inria

Keeping track of your analyses just got easier!

Galaxy now integrates with RSpace, allowing you to send datasets and histories directly to your electronic lab notebook. No more jumping between tools or losing track of results.

Learn more: galaxyproject.org/news/2025-06

galaxyproject.orgRSpace is now integrated with GalaxyAnalyze data with Galaxy, keep it under control on RSpace

Blog post: What Guix could offer in computational medical environments?

French national agency for secure drug and medicine (ANSM) requires for a medical device to have unambiguous identifications:

1. reference of the product
2. reference of the maker
3. serial number

Well, through my lenses applied to software, it reads:

1. #SoftwareHeritage identifier (#SWHID)
2. #Guix

and 3. is redundant. 😁

Well, a quick summary of a 30min talk I gave past week.

Thanks my previous colleague Sam from APHP to give me the opportunity to brainstorm on this topic. 🤩

simon.tournier.info/posts/2025

simon.tournier.infoWhat Guix could offer in computational medical environments?

Did you know? During the #SnakemakeHackathon2025 we had a staggering 194 work items!

It took a while, but now we are gathering contribution reports and present them online as a ResearchEquals (fediscience.org/@ResearchEqual) collection:

researchequals.com/collections

The first 10 are online and I will post some highlights in the coming weeks.

fediscience.orgFediScience.org

Asking for news from a colleague now in Univ. of Tours – long time, no see.

Nicolas pointed me a recent publication where… guess what? … they archived the source code in #SoftwareHeritage.

pubmed.ncbi.nlm.nih.gov/394876

Section « Open Research, data availability statement » reads:

« Anonymized raw data, the source code of the analyses, instructions to reproduce the computing environment, unformatted analysis reports and figures are available online (gitlab.com/nivall/car-lacates) and long term preserved at Software Heritage (archive.softwareheritage.org/s). »

And guess what else? The computational environment is managed using #Guix.

It makes my day! 😎

I’m happy to see #Guix and #SoftwarHeritage mentioned in the European Journal Of Clinical Investigation.

PubMedGlycolytic activity following anti-CD19 CAR-T cell infusion in non-Hodgkin lymphoma - PubMedEnergy metabolism of chimeric antigen receptor-T cells (CAR-T) activation in humans remains unexplored. As a glycolytic activity surrogate, we investigated the dynamics of peripheral blood (PB) lactate in the first weeks post-CAR-T infusion. In 17 patients treated with CD28 harbording anti-CD19 CAR- …

« The integration of #Guix and #SoftwareHeritage paves the way for more transparency and verification of the whole computational environment involved in scientific research.

Scientific production should be robust to external service failures; for example, being able to audit or reuse scientific findings should not depend on the availability of platforms hosting source code.

That’s why backing package managers with Software Heritage appears vitally important. »

Thanks @swheritage for the spotlight! 😀

softwareheritage.org/2025/05/2

Software Heritage · How Software Heritage ensures reliable Guix deployment - Software HeritageLearn how to ensure your research remains reproducible for years to come with Software Heritage and Guix...
Suite du fil

Hey “Choose Europe for Science”!

Do not only promote « public money, public code », make it an obligation!

Make an obligation that projects funded by Europe only rely on software that people all around the world are allowed to modify, distribute and make derivatives.

This would be a real measure which concretely claims: Science is a pillar for Democracy! There is no other Science than #OpenScience.

#ReproducibleResearch #ChooseEuropeForScience #FAIR
3/3

Suite du fil

Yet another example.

This morning, a colleague of mine: Can you install mixcr.com/mixcr/about ?

1. It’s far to be trivial to be able to audit/verify the correspondence between the source code and the distributed binary. Who trust who? github.com/milaboratory/mixcr

2. It’s not based on the scientific principles! Because the LICENSE reads: « The User is not authorized by this License to modify, distribute or make derivatives of the Software. » github.com/milaboratory/mixcr/

About 1. ok that’s a difficult topic. It’s big and deep in our (technical) society.

About 2. hey it’s easy: « public money, public code ». Researchers paid by public money must only use free software (not gratis software!).

#ReproducibleResearch #OpenScience
2/3

mixcr.comAbout - MiXCRMiXCR: ultimate software platform for analysis of Next-Generation Sequencing (NGS) data for immune profiling.

#OpenScience and #ReproducibleResearch, are they really making concrete progress?

Aside a ridiculous small set of people trying to do their best, the other are well-informed and deeply aware but they just don’t care!

Many researchers and communities hide behind « we cannot do otherwise »… 🤔 If *you* – paid by public money with a permanent position in a public institute – cannot do otherwise, who can?!? Damned!

Somehow I’m tired… Back from holidays one week ago and I already feel empty. 🤯

1/3

Reading « Are you still using OCaml 4.08 or earlier? », as a scientific practitioner I ask myself:

Well, 5 years is an eternity in the computing world – OCaml 4.08 is from 2019 and OCaml 4.07 from 2018, IIRC – but it’s peanuts when we speak about being able to audit and verify some scientific papers relying on these versions.

And there is the balance with the carbon footprint.

The question does not seem to retain 4.08 but to be able to rebuild from scratch 4.08, i.e., being able to rebuild all its dependencies. Thus it asks about the #bootstrap of the compiler and how we preserve it.

anil.recoil.org/notes/deprecat

Anil Madhavapeddy · Are you still using OCaml 4.08 or earlier? If so, we need to knowPar Anil Madhavapeddy

« Réfléchissons collectivement à la lourde tâche d'archivage de certains savoirs spécifiques. Je pense aux travaux aussi de #SoftwareHeritage, qui, en France, archive l'ensemble des logiciels créés dans le monde, avec une initiative européenne à construire sur le sujet. Je souhaite, au-delà de la question de notre capacité, à attirer chercheurs, enseignants-chercheurs, que nous ayons aussi une capacité à préserver, consolider, attirer plateformes de collaboration et données au service de l'intérêt général. »

Bravo @swheritage !

(peu importe ce qu’il y a avant et après ce petit passage 😀)

elysee.fr/emmanuel-macron/2025

#ReproducibleResearch #OpenScience

1/3

elysee.fr · Lancement de « Choose Europe for Science » à La Sorbonne.Par Elysee