Daniel E. Weeks

The Atlantic is active and becoming more so! Welcome to the heart of hurricane season!

(Loop from Google Weather Lab: Cyclones)

grading final projects and I've never ever seen so much brazen LLM output. **Most stuff is still great!** only ≈20ish% is pure LLM (text like "Lifenpsctarece.years" in completely made up images!)

This is the most disheartening time I've had grading, like, ever. What are we even doing anymore.

Hypermutable hotspot enables the rapid evolution of self/non-self recognition genes in Dictyostelium Cells require highly polymorphic receptors to perform accurate self/non-self recognition. In the amoeba Dicytostelium discoideum, polymorphic TgrB1 & TgrC1 proteins are used to bind sister cells and e...

I'm excited to announce our new biorxiv preprint, wherein we investigate the evolution of the weirdest genetic locus I've ever seen! Behold the tgr genes of the social amoeba, which mediate self/non-self discrimination during facultative multicellularity 🐅 🧵 1/
www.biorxiv.org/content/10.1...

Fun fact: G*Power was created to support this kind of analysis, which is called a 'compromise power analysis'. With large samples, it is often very sensible to lower the alpha level, as @maxmaier.bsky.social and I explain: journals.sagepub.com/doi/10.1177/...

Our new AJHG paper is online. We develop a new TWAS tool that uses local-ancestry information to improve power of gene mapping in admixed samples. Lead author Taylor Head (now at MD Anderson) did a phenomenal job spearheading this work!

Looks very nice to me!

That’s it I’m firing R because my results weren’t significant

Polygenic Hazard Score for Predicting Age-associated Risk of Alzheimer’s Disease in European Populations: Development and Validation Objectives Polygenic hazard score (PHS) models can be used to predict the age-associated risk for complex diseases, including Alzheimer’s disease (AD). In this study, we present an improved PHS model ...

In this new EADB preprint led by Ole Andreassen, we present an improved polygenic hazard score model for AD that demonstrates enhanced predictive accuracy for age of onset in European populations compared to alternative models.

www.medrxiv.org/content/10.1...

Comprehensive presentation of the "Theorem of the Day", starting with a statement of this theorem. Fermat’s Last Theorem : If x, y, z and n are integers satisfying x^n + y^n = z^n, then either n ≤ 2 or xyz = 0.

Theorem of the Day (August 2, 2025) : Fermat’s Last Theorem
Source : Theorem of the Day / Robin Whitty
pdf : buff.ly/L3PnOb5
notes : buff.ly/KxJUhw1

#mathematics #maths #math #theorem

You link #phenotype 🦜 to #genotype 🧬 with #comparative #genomics 💻?

This #review is for you 📜: authors.elsevier.com/sd/article/S...

We review new #methods, remaining #challenges and #future directions and highlight recent key studies.

Thanks @hillermich.bsky.social!

Please share! 🙂

After 10 years of work, a complete telomere-to-telomere gap-free genome for C. elegans finally exists: it has 106 Mb rather than the textbook 100.3 Mb, and up to 366 additional genes.
genome.cshlp.org/content/35/8...

Really excited to share the first paper from my PhD - it’s all about assumptions in modelling and the history of early population genetics… 🧵

doi.org/10.1111/ahg....

65 Genomes Expand Our Picture Of Human Genetics Researchers closely examined the genomes of 65 individuals to paint a more complex, and more complete, picture of human genetic diversity.

Complete genomes alert! @glennislogsdon.bsky.social, @christinebeck.bsky.social, and I were on @scifri.bsky.social today talking about "Complex genetic variation in nearly complete human genomes"
📄 www.nature.com/articles/s41...
📻 www.sciencefriday.com/segments/65-...

Inferring chromosome segregation error stage and crossover in trisomic disorders with application to Down syndrome - Nature Communications The authors report a computational framework for determining the meiotic/mitotic origin of nondisjunction (NDJ) in trisomies without parental data. Applying this to Down syndrome, they uncover links b...

New paper alert! Our team published in @natcomms.nature.com on MeiHMM - a hidden Markov model that identifies meiotic nondisjunction error types in chromosomal abnormalities, like Down Syndrome, without parental data! #Meiosis #Genomics #ComputationalBiology #StJudeResearch

doi.org/10.1038/s414...

Happy to share this paper that Siri Birkeland (@siribirkeland.bsky.social) and I wrote together. It's been really nice to put our heads together over such a complex topic!

Reevaluating the role of education on cognitive decline and brain aging in longitudinal cohorts across 33 Western countries Nature Medicine - In a large cross-national study, education was linked to better memory and larger brain volumes but not to slower cognitive or brain decline with age, suggesting that the...

New in @NatureMedicine Education is not linked to slower memory or brain decline in aging. We analyzed 400,000 memory tests and 15,000 MRIs from 33 countries. Associations likely shaped by childhood schooling and development. @LCBC_UiO @LifebrainEU rdcu.be/ex8iC

I am sort of fascinated that the NIH includes an option that supports compensation of peer reviewers. Do any journals actually pay reviewers currently? I've been reviewing for ~30 years and have never heard of such a thing (and have been bitter about it).

@charleston.bsky.social and colleagues introduce the ancestry-specific expected genetic relationship matrix (as-eGRM), an analysis framework that estimates the relatedness within ancestry components between admixed individuals, in the @ajhgnews.bsky.social latest article: bit.ly/3U5p0aI #ASHG

What do people use for references in Google docs/word? Coming from latex where referencing was solid, have been stung a couple of times by Google docs/zotero losing or messing up which paper each reference number points to. This is too painful to go through again. What are the alternatives?

#rstats #quarto Quarto gotchas!

I've enjoyed using Quarto the last few months, especially for my forthcoming book on linear algebra in data science.

But it's been annoying to deal with bugs in the product. Quarto must still be regarded as beta (and maybe always will). I'll list a few issues. 🧵 1/

NAs in #rstats should have had the special missing value distinctions afforded in other packages like Stata and, @thegodpodcast.com forbid, SAS. Tagged NAs are implemented in {haven} but then break with dplyr::bind_rows() or joins because vctrs think labelled double cannot be combined with numeric.

Rapid fluctuations drive complex gene regulation in bacteria Bacteria respond to environmental factors such as antibiotics by inducing regulatory proteins that switch their target genes on or off. Researchers led by Prof. Erik van Nimwegen have discovered that ...

🧪 Gene regulation in bacteria: The team led by @erikvannimwegen.bsky.social discovered that environmental stimuli cause rapid fluctuations in the concentration of regulatory proteins, leading different target genes to respond differently to the same stimulus. www.biozentrum.unibas.ch/news/detail/...

Leena Peltonen School of Human Genetics in full-swing!

@gosiatrynka.bsky.social
@dgmacarthur.bsky.social
@bpasaniuc.bsky.social
@tuuliel.bsky.social
@hilarycmartin.bsky.social
@sashagusevposts.bsky.social
@zkutalik.bsky.social
@mashaals.bsky.social
@alemedinarivera.bsky.social

🧪BREAKING: NIH has laid out 5 options for funding publications costs on grants (yes, FIVE!). This is hugely impactful.

Make your opinions known by September 15, fellow scientists.
🧪
grants.nih.gov/grants/guide...

Thorough discussion and dissection of ‘panmixia’ in population genetics - the idea that organisms in a group being studied reproduce randomly with one another. It’s an assumption at the heart of many analyses in population genetics even though true ‘panmictic’ populations are rare.

Hints of hidden heritability in GWAS - Nature Genetics Although susceptibility loci identified through genome-wide association studies (GWAS) typically explain only a small proportion of the heritability, a classical quantitative genetic analysis now argu...

Does anyone know who first used the term "Manhattan Plot"?

Wikipedia links to a 2010 paper by Greg Gibson. I haven't yet found an earlier use, but Greg's phrasing suggests that he did not invent the term.
doi.org/10.1038/ng07...

New blog article on Statistical Thinking: confidence limits for bootstrap overfitting-corrected predictive performance measures for regression models, useful for strong internal validations including confidence bands for debiased calibration curves: fharrell.com/post/bootcal #statistics #StatsSky

our p-value
was 0.054
so we described the result
as “approaching significance”
and let the reader
decide
how badly
they wanted to believe

Huge congrats to @audreyhendricks.com for her 2025 ASHG Educator Award! So incredibly well-deserved.

With Tom Lehrer's passing, I suppose this is a moment to share the story of the prank he played on the National Security Agency, and how it went undiscovered for nearly 60 years.