Daniel E. Weeks
LightNews LightNews
banner
statgendan.bsky.social
Daniel E. Weeks
@statgendan.bsky.social
650 followers 1.2K following 110 posts
Statistical geneticist. Professor of Human Genetics and Biostatistics at the University of Pittsburgh. Assiduously meticulous.
Posts Replies Media Videos
Reposted by Daniel E. Weeks
hurkasabine.bsky.social
Tables in publications are often the worst. Even worse is when cells are merged and data is formatted with lines, fonts, and next to each other, as this is only for visualizing data, not for data use. I hate them. If you're using XLSX files as a supplement, give them a useful format!
February 15, 2026 at 9:41 AM
Reposted by Daniel E. Weeks
itaiyanai.bsky.social
The 3 hardest things to learn as a scientist:
1. Trust the data.. especially when it’s not what you expected,
2. Trust the data.. allowing it to change your direction,
3. Trust the data.. but not too much: test with new data at every turn.
itaiyanai.bsky.social Itai Yanai @itaiyanai.bsky.social · May 31
90% of doing science is being open to new ideas.
February 19, 2026 at 1:14 AM
Reposted by Daniel E. Weeks
blekhman.bsky.social
Yes, they can hallucinate papers that don't exist, discuss results that seem to be imaginary, and can be confusing and inconsistent. But talking to tenured professors may still be helpful
January 14, 2025 at 10:30 PM
Reposted by Daniel E. Weeks
vpascat-biostat.bsky.social
🚨 Proud to share that our paper on the genetic links between type 2 diabetes (T2D) and hypertension is now published in @natcomms.nature.com!
We dissect the genetic relationship between T2D and hypertension and show it is not driven by a single shared pathway. 🧬
(1/N)
February 18, 2026 at 6:51 PM
Reposted by Daniel E. Weeks
f2harrell.bsky.social
#Statistics thought of the day: If you think you can find new disease subtypes by empirically clustering patients, think again: www.fharrell.com/post/cluster... #StatsSky
The Burden of Demonstrating Statistical Validity of Clusters – Statistical Thinking
Patient clustering, often described as the finding of new phenotypes, is being used with increasing frequency in the medical literature. Most of the applications of clustering of observations are not ...
www.fharrell.com
February 18, 2026 at 12:40 PM
Reposted by Daniel E. Weeks
f2harrell.bsky.social
Wow this is a breakthrough in validation of predictive models that fully accounts for point estimates used in predictions being only point estimates. Great work @richarddriley.bsky.social and colleagues! #Statistics #StatsSky
richarddriley.bsky.social Richard Riley (R²) @richarddriley.bsky.social · 4d
⭐ NEW PAPER ⭐ on a Bayesian approach to sample size calculations for external validation of risk prediction models - account for uncertainty in the assumed true performance of the model - plus calculate assurance probabilities & the value of information
onlinelibrary.wiley.com/doi/10.1002/...
February 16, 2026 at 12:45 PM
statgendan.bsky.social
“on average, only about 60% of the information is retained after dichotomization.”
February 16, 2026 at 11:32 PM
Reposted by Daniel E. Weeks
paturley.bsky.social
Over the course of 2025, UKB researchers were forced to move all of their data analysis to the cloud rather than working with local copies. What was the effect of this move on scientific output? Here is a simple first pass at answering that question.
February 16, 2026 at 10:27 PM
Reposted by Daniel E. Weeks
drmowinckels.io
New neuromapr #rstats package is available from GitHub! Highly experimental, early adopters and bug identifiers are super welcome to report issues!

github.com/lcbc-uio/neu...

#neuroscience #rstats #neuroimaging
February 15, 2026 at 11:00 AM
Reposted by Daniel E. Weeks
shaicarmi.bsky.social
Very interesting perspective article by @caina89.bsky.social et al on genetic studies in psychiatry.

The authors argue that studies increasingly rely on "shallow" phenotyping (self-reported), leading to biases in estimation of genetic relationships between disorders.

www.nature.com/articles/s41...
The predicament of heritable confounders - Nature Genetics
This Perspective argues that diagnoses derived from self-reports, electronic health records and self-administered questionnaires introduce heritable bias that confounds the interpretation of data from...
www.nature.com
February 15, 2026 at 11:49 AM
statgendan.bsky.social
“How do we convince people that learning, itself, is the goal, and to put in real effort when there may be easy ways to game the evaluation system (i.e. grades)?“

#AI #Learning #Education #LLMs
robp.bsky.social Rob Patro @robp.bsky.social · 5d
I’ve written a post about my recent experiences (successes) with AI coding models; the experiences that caused me to re-evaluate my initial judgements, the surprise I had at what can be accomplished, & some fears I have about these tools. Discussion welcome! combine-lab.github.io/blog/2026/02...
COMBINE-lab - The skeptic’s guide to generative AI assisted coding
An easy-to-use, flexible website template for labs, with automatic citations, GitHub tag imports, pre-built components, and more.
combine-lab.github.io
February 15, 2026 at 7:41 AM
Reposted by Daniel E. Weeks
toddgureckis.bsky.social
i'm kinda excited about markup languages for cooking: cooklang.org
Cooklang – Recipe Markup Language
Cooklang is a markup language for recipes.
cooklang.org
February 15, 2026 at 4:35 AM
Reposted by Daniel E. Weeks
mollyprz.bsky.social
The program for the 2026 New York Population Genetics meeting, hosted by at the @simonsfoundation.org on March 9th 2026, is now up: events.simonsfoundation.org/event/7c91dd....
NY Population Genetics meeting
events.simonsfoundation.org
February 12, 2026 at 7:26 PM
Reposted by Daniel E. Weeks
cedricboeckx.bsky.social
Very nice work led @drghawkes.bsky.social,
showing convergence between rare and common genetic
associations, using whole-genome sequencing data to evaluate the
contribution of rare non-coding variants for commonly studied anthropometric traits 🧪🧬
www.nature.com/articles/s41...
Whole-genome sequencing analysis of anthropometric traits in 672,976 individuals reveals convergence between rare and common genetic associations - Nature Communications
Most GWAS have focused on common variants or rare protein coding variants. Here, the authors interrogate the contribution of rare non-coding variants for anthropometric traits, identifying new genes a...
www.nature.com
February 14, 2026 at 8:18 AM
Reposted by Daniel E. Weeks
profsimonfisher.bsky.social
Please tell friends & colleagues about our unique course “Genetics & Neurobiology of Language” July 27-Aug 3 2026. Expert tutors, interactive talks, panel discussions, all in a beautiful setting. Scholarships available: meetings.cshl.edu/courses.aspx...
@cshlnews.bsky.social @cshlbanbury.bsky.social
Group photograph of faculty and participants of the very first Cold Spring Harbor summer course on Genetics and Neurobiology of Language in 2014, taken as the sun was going down at the Banbury Campus, Lloyd Harbor.
February 13, 2026 at 5:01 PM
statgendan.bsky.social
Juan Celedon shared this favorite quote during his seminar today:

"Never trust anyone's data, particularly your own"

#DataScience #QualityControl #Science
February 13, 2026 at 7:40 PM
Reposted by Daniel E. Weeks
cognitivecamz.bsky.social
Our multi-ancestry GWAS of EXTernalizing conditions (ADHD, substance use...) in ~4M people reveals neurodevelopmental risk, drug-repurposing targets, and yields one of the strongest psychiatric polygenic indices yet! 🧬🎉 doi.org/10.64898/202...
Genomic insights into substance use and disinhibitory disorders
Externalizing spectrum disorders- spanning attention-deficit/hyperactivity disorder, conduct disorder, substance use disorders, and other disorders characterized by disinhibition - frequently co-occur...
doi.org
February 11, 2026 at 10:33 AM
Reposted by Daniel E. Weeks
datavisfriendly.bsky.social
#rstats RIP, John Fox
@yihui.org just published this lovely tribute to John Fox and his work

yihui.org/en/2026/02/j...
R.I.P., John Fox - Yihui Xie | 谢益辉
Last November, I learned the very sad news from Michael Friendly that John Fox had passed away. That brought my memory back to 2006 when I emailed John for the first time asking for his help on a &hel...
yihui.org
February 12, 2026 at 2:05 AM
Reposted by Daniel E. Weeks
andrew.heiss.phd
I just whipped up this little #QuartoPub site last week that demonstrates how I teach p-values/hyp-testing through simulation both with live OJS and with #rstats, and I think it's super neat! It has examples for diff-in-means, diff-in-props, and regression slopes nullworlds.andrewheiss.com #statsky
Simulated null distribution for data with a sample size of 100, difference in group means of 5, and a p-value of 0.142 Simulated null distribution of a slope of 0.8 and p-value of 0.002 Finally, we have to decide if the p-value meets an evidentiary standard or threshold that would provide us with enough evidence that we aren’t in the null world (or, in more statsy terms, enough evidence to reject the null hypothesis). There are lots of possible thresholds. By convention, most people use a threshold (often shortened to α) of 0.05, or 5%. But that’s not required! You could have a lower standard with an α of 0.1 (10%), or a higher standard with an α of 0.01 (1%). Statistically significant The p-value is < 0.001 and our threshold for α is 0.05 In a world where there is no relationship between x and y, the probability of seeing a slope of at least 0.901 is < 0.1% Since < 0.001 is less than 0.05, we have enough evidence to say that the slope is statistically significant. Evidentiary standards When thinking about p-values and thresholds, I like to imagine myself as a judge or a member of a jury. Many legal systems around the world have formal evidentiary thresholds or standards of proof. If prosecutors provide evidence that meets a threshold (i.e. goes beyond a reasonable doubt, or shows evidence on a balance of probabilities), the judge or jury can rule guilty. If there’s not enough evidence to clear the standard or threshold, the judge or jury has to rule not guilty. With p-values: If the probability of seeing an effect or difference (or δ) in a null world is less than 5% (or whatever the threshold is), we rule it statistically significant and say that the difference does not fit in that world. We’re pretty confident that it’s not zero. If the p-value is larger than the threshold, we do not have enough evidence to claim that δ doesn’t come from a world of where there’s no difference. We don’t know if it’s not zero. Importantly, if the difference is not significant, that does not mean that there is no difference. It just means that we can’t detect one if there is. If a prosecutor doesn’t provide sufficient evidence to clear a standard or threshold, it does not mean that the defendant didn’t do whatever they’re charged with†—it means that the judge or jury can’t detect guilt.
February 11, 2026 at 9:14 PM
Reposted by Daniel E. Weeks
hetanshah.bsky.social
‘Novice workers who rely heavily on AI to complete unfamiliar tasks may compromise their own skill acquisition… We find that AI use impairs conceptual understanding, code reading, and debugging abilities, without delivering significant efficiency gains on average.’
arxiv.org/pdf/2601.20245
Abstract Al assistance produces significant productivity gains across professional domains, particularly for novice workers. Yet how this assistance affects the development of skills required to effectively supervise AI remains unclear. Novice workers who rely heavily on AI to complete unfamiliar tasks may compromise their own skill acquisition in the process. We conduct randomized experiments to study how developers gained mastery of a new asynchronous programming library with and without the assistance of Al. We find that AI use impairs conceptual understanding, code reading, and debugging abilities, without delivering significant efficiency gains on average. Participants who fully delegated coding tasks showed some productivity improvements, but at the cost of learning the library. We identify six distinct AI interaction patterns, three of which involve cognitive engagement and preserve learning outcomes even when participants receive AI assistance. Our findings suggest that Al-enhanced productivity is not a shortcut to competence and AI assistance should be carefully adopted into workflows to preserve skill formation - particularly in safety-critical domains.
February 3, 2026 at 9:00 PM
Reposted by Daniel E. Weeks
jrossibarra.bsky.social
Fun news! @gcbias.bsky.social and I are teaching a 2-week online population genetics workshop this summer to raise money for the Center for Population Biology at UC Davis. We're trying to gauge interest -- please fill this out if you might be interested! And please share broadly!
Davis Summer Population Genomics Program
Want to learn population genetics? Please fill out this form to indicate your potential interest in a 2-week intensive online summer population genetics course taught by Jeffrey Ross-Ibarra and Graham...
docs.google.com
February 9, 2026 at 6:24 PM
Reposted by Daniel E. Weeks
sashagusevposts.bsky.social
Another Claude project: a static site that pulls in GWAS SNP data from ensemble, multiple public biobanks, open targets, gtex, eqtl catalog, and OMIM.

sashagusev.github.io/gwas_lookup/
February 9, 2026 at 6:50 PM
Reposted by Daniel E. Weeks
robertwarmerdam.bsky.social
🧬 New preprint alert! After years of collaborative work across 52 datasets we are presenting eQTLGen phase 2: a genome-wide eQTL meta-analysis covering 43,301 blood samples: www.medrxiv.org/content/10.6... (1/8)
February 6, 2026 at 10:14 AM
Reposted by Daniel E. Weeks
timfrayling.bsky.social
The Exeter team seem to be the Alex Honnold of human genetics. Most sane people on seeing the challenge of meta analysing >1billion variants & 10,000s aggregates across ~100% of the genome would opt for the safety rope of the exome. Not these guys.
hiwwright.bsky.social Harry Wright @hiwwright.bsky.social · 14d
Excited to share my first preprint on federated conditional analysis of rare single variant and aggregate association tests across six genetically-inferred ancestry groups in All of Us and UK Biobank doi.org/10.64898/202...
February 6, 2026 at 3:46 PM
Reposted by Daniel E. Weeks
stephenturner.us
Footnote: "The em dashes reflect our stylistic preference and should not be interpreted as evidence of Al-assisted text generation."
February 7, 2026 at 6:09 PM