SNL’s Tucker Carlson is so good, this has been my mental reaction these last weeks every time I see some non sense research practice

Also #singlecell RNaseq for sure 🧪🧬💻

Have a relaxing weekend, everyone who went to Cannes! #stats

Steve Quake quotes Brenner at #cshsymposium2026 : “biology is drowning in a sea of data and starving for knowledge”, arguing that never has this been more apt than now.

A survey of 2600 Ph.D. students, postdocs, and other researchers in 65 countries found that top three complaints were that supervisors: 1) used dismissive communication 2) provided little feedback 3) ignored team members’ well-being The main thing they wanted was empathy

When there’s a meeting and the first presenter is full python matplotlib, and the second presenter is full #rstats ggplot viz. I always feel like

🚀 Big updates are here! A new round of #rstats #easystats releases is now live on CRAN, featuring major performance boosts and new features for parameters, bayestestR, performance, modelbased, and see. Update via CRAN or grab the latest dev versions with: `easystats::install_latest()`

Same if you only use animal models but put the human disease term in the title

In a meta-analysis of 210 biomedical AI studies that statistically compared models under cross-validation, 97% used invalid statistical tests. Here's our new preprint doi.org/10.64898/202... led by @tianchu.bsky.social @hetuli.bsky.social @shaoshiz.bsky.social @nichols.bsky.social 1/N

"a paper today is a Frankensteined-together patchwork of supplementary files that run longer than the manuscript, datasets in repositories...code posted to GitHub without documentation. The paper has been replaced....by a mess" www.thetransmitter.org/from-bench-t...

tinyurl.com/reuxynmc Very excited to see this work I was a small part of published! We analysed two large scale colocalisation datasets: OpenTargets data and an analysis of immune-mediated disease GWAS/immune cell eQTLs seeking to understand the "colocalisation gap". Some highlights:

A flaw in Person-Δ may be overstating progress in single-cell perturbation prediction models. Pearson warned about this in the 19th century: reusing the same controls induces spurious correlation. Split the controls, and much of the claimed prediction power fades: www.biorxiv.org/content/10.6...

New #tidyomics blog post! 🎉 #tidySummarizedExperiment greatly simplifies your omics workflows, giving tidyverse capabilities. We have improved its performance 30x on large datasets with tens of thousands of samples! tidyomics.github.io/tidyomicsBlo... Plz share! @mpibiochem.bsky.social

The normative fraud that Errington describes is something I've seen my whole career. A biologist once told me, if p is very small, it can't be a false positive. Teams condition on null being false, see anything that makes p small as truth-seeking. These traditions spread bc they get you published.

Me every week #rstats

Grad Students, PI, Postdocs

🧬 New preprint: "The Human Pleiotropic Map of GWAS Associations and Therapeutic Implications" Why do some genetically supported drug targets succeed in the clinic while others fail? Across 100,526 GWAS, the same evidence flags constrained gene functions rarely safe to modulate.

GWAS meta-analysed summary statistics for 72 long-term conditions from the GEMINI collaborative updated to v1.1 🌟 The input GWAS are the same but we have fixed several minor issues since the original release v1.0 See our GitHub for links (incl. to Zenodo) github.com/GEMINI-multi...

I agree. I can see only a very limited number of scenarios in which it is useful and/or helps catch errors/data manipulation. Sharing detailed analysis pipelines and raw data is so much more important for reproducibility.

📊 Five #ggplot2 functions I wish I'd known about earlier 📊 I've written a short blog (with examples) of some of the lesser-known {ggplot2} functions and arguments that make it easier to create better charts! Link: nrennie.rbind.io/blog/five-gg... #RStats #DataViz

Top priority #Copenhagen outing last week-end: scout a French bakery that uses the PROPER terminology. They sell the CHOCOLATINE not the blasphemous "pain au cholocat". Excellent quality too. share.google/u8M0bK5qpkAl...

Hold habits die hard #rstats

Bell curves.

My all-time favourite method of debugging is going to bed and looking at it again the next morning.

Maybe journals aren’t the root cause of problems in academia just a symptom…

Exam prep for a research and statistics exam in 2026! sachaepskamp.com/PL2132_games...

IMO, the real cost of low success rates is not the time wasted on reviews but the time researchers collectively waste developing grant proposals that will not be funded. 1/2

LLM‑Assisted Software Engineers Can Only Be Mad (And Will Block Smartasses) An impromptu essay on the impossible equilibrium of AI‑augmented coding (1)

👀 Absolute banger paper. Leveraging cattle/pig GWAS/eQTL to help explain the missing regulatory gap in humans. Consistent with previous works, selection makes identifying relevant eQTLs in humans challenging. www.biorxiv.org/content/10.6...

How I feel logging into a virtual machine, then ssh-ing into a cloud server, then connecting to an Rstudio server to access my data.

“Everything that scientists say hinges on the specifics of what they actually did. If you can’t (or don't) evaluate the details, you can’t evaluate the science. I'm grateful to the authors for providing access to those details.“

I have a SNP joke, but I don’t think you’d get the reference.

they can 100% amplify good judgment. You can put it to work on refactoring, testing, docs, debugging, product consistency, etc. But that requires _you_ to focus on it. And _that_ is hard.

🥇 The €100K Institutional Award goes to the Brazilian Reproducibility Initiative - a nationwide effort to evaluate lab biology research and the largest coordinated replication effort in the field worldwide, catalyzing lasting improvements across Brazil's research ecosystem. Congratulations!

"You've shown why we tell participants not to share their data online" ← So UK Biobank blame cohort members for their own re-identification? From just two quasi identifiers! This is victim-blaming in place of data governance. Yet again, serious questions need asking about UK Biobank governance.

One way we made it fast was by mapping with simpleaf (github.com/COMBINE-lab/...) 🥳 This gives a ~50x speed-up over CellRanger, and natively produces spliced vs unspliced reads. Percent spliced reads is IMO *the* most important QC metric for single nuclei data (link.springer.com/article/10.1...).

"Preprint servers are a time machine, they move everyone forward 12 months and speed up the exchange of ideas" ht @pedrobeltrao.bsky.social www.evocellnet.com/2021/06/a-no...

At this point I think my peer review has gotten to the point where I couldn't be replaced by an LLM, but I could be replaced with copy-and-paste 'The authors should add their statistical methods to the manuscript and carefully identify where and how they aggregate data'

I get contacted multiple times a week by users; it's really exciting to hear about all the cool science🌟 If {ukbrapR} supports your research, please cite it so I can track its impact (this really helps me!) Pro-tip: Use the {grateful} package to automatically include {ukbrapR} in your bibliography

I die a little every time I hear "oh I don’t use projects I just setwd()". Now that I voiced my concern to colleagues, I have to give a presentation on project-oriented workflows, I think it’s also a good time to introduce git. What’s your favorite teaching materials to get started? #rstats

New paper showing that much of the apparent success of protein language models in predicting mutational effects is a mirage: These models mostly memorize sites. 1/ www.biorxiv.org/content/10.6...

quarto-revealjs-editable has a new release with 5.0.0!! It is now much more stable, should now work everywhere, plus: - Rotation support - Undo/Redo - Copy qmd to Clipboard - keyboard controls github.com/EmilHvitfeld... #quarto

Issue 35 of #rdmweekly is out! Including: ➡️ Building Realistic Fake Datasets with Pointblank @richmeister.bsky.social ➡️ Rethinking TXT Files @kbriney.bsky.social ➡️ Economic Benefits of OS @plos.org ➡️ What Is Data Governance? @thegovlab.org and more! Link: rdmweekly.substack.com/p/rdm-weekly...

📢 New preprint out! tl;dr: Publish your code, add clear README's! 138 participants assessed data & code archiving practices across 1861 papers published at 7 @britishecologicalsociety.org journals, identified gaps & offered recommendations for improvement 🔗 doi.org/10.32942/X26...

"Dashboards" are very frustrating for exactly this reason. Needlessly fancy dashboards look impressive to people who don't use them.

OpenAI announces ChatGPT is now at the level of real scientific reviewers, “It just tells you to do single-cell RNAseq in response to every query” says Sam Altman

S4 methods developers when you type @ to access some slots. #rstats