(Non-identifiability is the equivalent of looking at an equation 'A x B = 20' and claiming you know exactly what value A and B are.)

Theorem of the Day (November 9, 2025) : Netto’s Conjecture (Dixon’s Theorem) Source : Theorem of the Day / Robin Whitty pdf : buff.ly/eWQ4C4N notes : buff.ly/ElCux2H #mathematics #maths #math #theorem

“changes in statistical significance are often not themselves statistically significant. … even large changes in significance levels can correspond to small, nonsignificant changes in the underlying quantities.” www.tandfonline.com/doi/abs/10.1... #Statistics

In a field where evidence suggests that a high proportion of papers have serious flaws, a database of "potentially important" papers which appear *not* to have serious flaws might be useful, as well as identifying those that do.

In 1982, I was in the Jackson Laboratory Research Training Program. As part of the program, we got to attend the Short Course lectures. At the end of one lecture by a young researcher, this old guy very viciously and aggressively lit into him. Turned out that the nasty old guy was Jim Watson.

A fantastic piece. Fascinating that Watson was another victim of The Bell Curve and became entrenched by his undeserved arrogance and hubris

Okay, here are some first reflections on Watson. Watson's life is a tragedy, really of Shakespearean proportions. He did not, as most bios will tell you, do one great thing when he was young and then collect laurels for it for the next 60 years. His career arc was unlike any in science.

Rarely in the entire history of science, has QC been a topic of such passion, importance and impact. If only the French and Americans had adopted such rigor when they messed up the design of that multi billion $ telescope because one was using the metric system, the other the imperial system.

New paper on everyone’s favourite topic, QC! We show why you should do genotype-level QC on your WGS data www.biorxiv.org/content/10.1... Very real quotes about this paper - “The most exciting, mind-blowing paper of the year!” “On a par with Fisher 1918” “I read it every night. Just so beautiful”

Bluetorial-Jim Watson I met Jim Watson a few times but did not know him well. However, I was greatly influenced by his book “The Double Helix”. He was a complicated human being with some very, very bad features, but some good contributions. What follows is my personal perspective. 1/41

Probably the papers I share the most often are @f2harrell.bsky.social 1996 on how to effectively develop multivariate models and @benvancalster.bsky.social 2019 paper on Calibration. Both extremely well written and informative. What are you most shared stats paper? #statistics #statssky #academicsky

✨Hello folks!✨ We're back in your feed with an amazing paper spotlight by @kirahoeffler.bsky.social & colleagues📰🔦 "Optimizing genetic ancestry adjustment in DNA methylation studies: a comparative analysis of approaches" 🔗http://bit.ly/49syndi Read our overview below! 🧵1/6

📈 #TrendingWithImpact: A new #research paper was #published on September 10, titled “Longitudinal associations of epigenetic aging with cognitive aging in Hispanic/Latino adults from the Hispanic Community Health Study/Study of Latinos.” 🔗 doi.org/10.18632/agi... #aging #quote #epigenetics

Insights on downstream immunity from blood methylome signatures in children exposed to maternal type 1 diabetes #T1D #Genetics #Epigenetics www.nature.com/articles/s42...

New pre-print out on peer-review milling! Together with @deevybee.bsky.social and M. Angeles Oviedo Garcia. Enjoy! doi.org/10.1101/2025...

If people are using the correct statistics, casual inference and study design, they will quickly realize they don't have the funds to design the study, the time to learn the statistics, or the right data to answer the actual question they are interested in. Then they will never publish or get tenure

Excited to launch an openRxiv partnership with the scientist-run AI review service qed (@qedscience.bsky.social), the brainchild of @odedrechavi.bsky.social 1/n openrxiv.org/enabling-rev...

My prediction is that LLM peer review will slow down science. It will do this for precisely the same reasons that contemporary peer review does and some extra ones. Start by reading @hansonmark.bsky.social thread below, then read on. 🧵

Excited to share our latest work on the factors that determine what genes we find (and don't find!) in GWAS and burden tests. We describe a critical concept that we call *specificity*. Led by Jeff Spence and Hakhamanesh Mostafavi:

Now a nice series of papers using genetics to "uncouple" excess adiposity from its adverse metabolic effects. Here for metabolic traits www.nature.com/articles/s42..., here including non metabolic traits elifesciences.org/articles/72452 & here using partitioned PRS www.nature.com/articles/s41...

Great news! We've figured out something for 'trending towards significance' to mean. It's when the significance falls away before you quite finish your now-add-another-control-and-retest process. "Guys, we _nearly_ got from a bivariate to a defensible analysis with significance intact. Next time!"

Did you know (robust) loess fitting can fail if the data is already smooth? I made a notebook that shows the flawed fit (red) along with a possible improvement (blue), using Cleveland's original demonstration curve. #stats observablehq.com/@xangregg/lo...

How do GWAS and rare variant burden tests rank gene signals? In new work @nature.com with @hakha.bsky.social, @jkpritch.bsky.social, and our wonderful coauthors we find that the key factors are what we call Specificity, Length, and Luck! 🧬🧪🧵 www.nature.com/articles/s41...

Do you teach #rstats? Do your students complain about how lame and old-fashioned dplyr is? Don't worry: I have the solution for you: github.com/hadley/genzp.... genzplyr is dplyr, but bussin fr fr no cap.

I greatly appreciate Ochoa group's long-lasting effort to define the quantity first and trying to estimate it. They've been stressing that PCs are unnecessary in LMMs are in fact harmful in certain cases. I have some thoughts. (1/n) academic.oup.com/genetics/adv...

they ask what's the minimum to pass? as if i keep a secret rubric of disappointment i say do your best work they hear write more words

Let’s talk about rare variant association tests - we have a new method that helps you discover more with the same data! It’s called *FlexRV* to signal our approach: “flexibly modeling rare variant pathogenicity improves gene discovery for complex traits”. www.biorxiv.org/content/10.1...

🚨 NEWS! The GWAS Catalog Diagram is back up! 🚨 Smoother, faster loading for filtered views! 🗺️ Explore now 👉 www.ebi.ac.uk/gwas/diagram Tell us what you think! Are there any features you would like? Help us to make it useful for the #gwas community

Asymmetries of the brain and body - are they linked genetically? In a new study we sequenced the genomes of people with their visceral organs reversed on the left-right axis, in a rare congenital condition called situs inversus totalis: www.biorxiv.org/content/10.1... Short thread below ⬇️ 1/5

New: GWAS of serurm antibody levels. Interesting findings include genetically correlated traits with hard-to-find shared causal variants, and apparently genetically uncorrelated traits sharing causal variants that operate in inconsistent directions www.medrxiv.org/content/10.1...

... 6. Validate findings in truly independent datasets 7. Account for biological and technical variation 8. Distinguish between correlation and causation 9. Make your analysis reproducible and transparent 10. Collaborate early with biostatisticians

1. Design your study with statistics in mind 2. Understand your data structure and sources of noise 3. Explore the data before statistical testing 4. Perform correction for multiple-hypothesis testing 5. Emphasize effect sizes and confidence intervals over P values ...

I am perplexed by this major new study of NHS England data among children on multiple accounts a) It writes "risk is higher 12 months after infx than 4wks after vax" without showing that this comparison is meaningful (ie risk 12m after infx is linked to infx, not vax) 1/

There's been a significant fall in obesity rates in the US, linked to weight-loss drugs, according to a recent @gallup.com survey. After peaking at nearly 40% in 2022, the adult obesity rate has fallen to 37%. This means 7.6 million fewer people living with obesity. news.gallup.com/poll/696599/...

How to fit a model without estimating any parameters? Use a formula with y ~ 0 + offset(...) This works with any formula interface, and with #tidymodels this can be passed to add_model(). #rstats

Theorem of the Day (November 5, 2025) : The Merton College Theorem Source : Theorem of the Day / Robin Whitty pdf : buff.ly/nFZCPhg notes : buff.ly/gt8qXnE #mathematics #maths #math #theorem

"ah the ubiquitous but also ominous awk duplicate remover awk '!a[$0]++' this sweet baby is the love child of awk's power and terseness. the pinnacle of awk one liners. short but powerful and arcane all at once. " unix.stackexchange.com/questions/15... #Awk #Unix

Glad to share our full paper on why deployed clinical prediction models should not discard predicted risks in favour of thresholds with @rexwp.bsky.social and @aidybarnett.bsky.social now up on @jclinepi.bsky.social. A brief explainer (1/4): #statsky #rstats authors.elsevier.com/a/1m1D83BcJQ...

a bit off topic, but I think you might be interested: [2509.06917] Paper2Agent: Reimagining Research Papers As Interactive and Reliable AI Agents share.google/EXK2F7c7EZda...

Elevated mutation near crossovers inhibits the evolution of recombination https://www.biorxiv.org/content/10.1101/2025.11.01.685904v1

Wrote up a little intervention post/explanation for my class about why using LLMs for trying to learn programming (as first time learners!) is bad and detrimental datavizf25.classes.andrewheiss.com/news/2025-11...

"Gentlemen, that is surely true, it is absolutely paradoxical, we can't understand it, and we don't know what it means, but we have proved it, and therefore we know it must be the truth."– Benjamin Peirce (1809–1880) #quote #mathematics #maths #math

For 61 years the #BBCWorldService has been broadcasting the latest in science via its weekly Science in Action programme. That dies in the next half hour, with this final edition, reflecting on the fall in trust in expertise driven by malign interests over recent years.

#mathsky #statistics

"The mathematician's patterns, like the painter's or the poet's must be beautiful; the ideas, like the colours or the words must fit together in a harmonious way. Beauty is the first test [...]" – Godfrey Harold Hardy (1877–1947) #quote #mathematics #beauty #maths #math

I was marginally non-significant once.

Cell Decoder: decoding cell identity with multi-scale explainable deep learning. #SingleCellOmics #CellCharacterization #GenomeBiology 🧪🧬 🖥️ genomebiology.biomedcentral.com/articles/10....

Submission: I did a thing. Reviewer: Too bad it’s not the thing I imagined while half-reading your abstract. Revision: What a coincidence, it is now.