New paper "Variant selection to maximize variance explained in cis-Mendelian randomization" led by @AngZhou3 and published at @HGGAdvances: pubmed.ncbi.nlm.nih.gov/41548043/, with contribution from @vkarhune.bsky.social, Haodong, Janne, Ash, and Eric. Brief thread:

New article @bmj.com reviews procedures for eligibility and treatment assignment for #targettrial emulation. "Starting right: aligning eligibility and treatment assignment at time zero when emulating a target trial" Authors include @miguelhernan.org. Read the article: www.bmj.com/content/392/...

Two new preprints on the efx of GLP-1s just dropped. They use different samples, data, & designs. But they come to the same conclusion on one point: GLP-1s do not generate meaningful short- to medium-run reductions in non-GLP-1 medical spending, despite clear health benefits.

Have you seen George Davey Smith telling him the smoking findings were completely implausible at a seminar in Edinburgh 1994? youtu.be/K9pyS7EGCV8?...

How to use LLMs in research work - a really nice introduction to coding with LLMs for research by Kohei Kawaguchi. Code and docs available in a github repo #linkoftheday note.com/mixingale/n/...

I've had this on my reading list for awhile, but I finally got around to it and it is excellent It is a great introduction to bounds for missing data (that I wish I read much earlier when I was first learning about them!!) www.sciencedirect.com/science/arti...

Paul Goldsmith-Pinkham, Peter Hull, Michal Koles\'ar: Leniency Designs: An Operator's Manual https://arxiv.org/abs/2511.03572 https://arxiv.org/pdf/2511.03572 https://arxiv.org/html/2511.03572

We wrote an article explaining why you shouldn't put several variables into a regression model and report which are statistically significant - even as exploratory research. bmjmedicine.bmj.com/content/4/1/.... How did we do?

Dismantling the Institutional Review Board (IRB), that oversees all human subjects research, is bad. Really bad.

Though it’s traumatic for me to bring this up, I was targeted by a group allied with Turning Point USA in 2018, with the help of Tucker Carlson on Fox. My home address circulated online and I got death threats. 1/

I wrote something about statistics under authoritarianism

Recently I learnt that EMBL-EBI has a free online course about AlphaFold — how it works, strengths and limitations, how to use it — and it's very good! There are also little quizzes and interactives. www.ebi.ac.uk/training/onl...

Jana Christofer and I wrote a piece on #papermills for Laborjournal. We discuss challenges of paper mills and AI for scientific publishing. Science needs immediate solutions to correct it and prevent it from massively produced fraud. @image-integrity.bsky.social www.laborjournal.de/rubric/essay...

🚨ICYMI 🫀🧬 from @mariosgeorgakis.bsky.social & co 📄Single-cell transcriptome-wide Mendelian randomization and colocalization analyses uncover cell-specific mechanisms in atherosclerotic cardiovascular disease

I probably won't have time to make a video out of this. But the materials are available online, and should be pretty self-explanatory. Hope this is useful to some people. privefl.github.io/statgen-cour...

Due both to federal research cuts and lack of funds in state govs., a number of universities are facing hard financial times. Indiana University is handling this in part by canceling or suspending 116 academic programs, tinyurl.com/28x2cysu. 🧵 1/

🎉(1/4) Excited to share that our latest paper is now published! We investigated genetic factors linked to antidepressant side effects in 13,000 individuals from the @estbiobank.bsky.social , leveraging data from questionnaires and clinical notes using NLP. www.nature.com/articles/s41...

about that: www.sciencedirect.com/science/arti...

Thanks @timpmorris.bsky.social et al - a wonderful course at UCL on Simulation - highly recommend. And huge congratulations to my colleague Ryan for completing this complex methodological review - will be so important for #DIAMOND going forward. 🎉👏🏼

Harvard today, your institution tomorrow. It's all part of the Project 2025 plan to destroy high education in America. All leading US universities depend on federal funding and tuition dollars from international students. They severely curtailed the former. Now they're eliminating the latter.

🎫Registration open for 1-day hybrid event: “Target trial emulation and other frameworks: The role and potential of observational data for evaluating effects of interventions”. 📆26th June. tinyurl.com/dash-tte. Hosted by LSHTM Centre for Data & Statistical Science for Health. @lshtm-dash.bsky.social

Watching Scott Zimmerman and Erin Ferguson lead a workshop on reproducible research and multiverse analysis. It's their practice for a workshop at #EpiResearch #SER 2025. epiresearch.org/annual-meeti... Drawing on Erin's paper: link.springer.com/article/10.1... So cool.

📣📣 Thrilled to see this cool work finally out in the wild! "Genome-wide analyses of variance in blood cell phenotypes provide new insights into complex trait biology & prediction" www.nature.com/articles/s41... Loads of cool findings including MR for alcohol usage -> increased variance in BC traits

I have been in biomedical science for over a decade & I independent of that have followed @joshtpm.bsky.social and @talkingpointsmemo.com for political commentary. Those worlds now converge as Josh is one of the key political reporters with relentless focus on what is happening at NIH/NSF.

Happy to share our new (open access) article in the Canadian Journal of Economics on people’s willingness to donate DNA for science and its relationship with economic incentives and preferences. A brief thread on what we learned (1/4): doi.org/10.1111/caje...

Big news for health data research and population diversity: The Mexico City Prospective Study data are now available securely to approved researchers worldwide. www.ndph.ox.ac.uk/news/latin-a... Cohort description: www.ctsu.ox.ac.uk/research/pro...

Cool story! Fun fact: The original SARS-CoV-2 strain was uncovered using metagenomics in Jan 2020 (www.nature.com/articles/s41...) which was pivotal for vaccine development efforts. Metagenomics has arguably saved many millions of lives already.

Very excited to give a virtual talk as part of the @eshg.bsky.social webinar series tomorrow at 4pm (CEST). If you want to listen in, register here: wma.eventsair.com/eshg-webinar... I will talk about all the exciting things we can learn from GWAS summary statistics...

How many NIH supported "life-changing" studies have been or will be terminated? This one on the Diabetes Prevention Program Outcomes Study, by @dhruvkhullar.bsky.social @newyorker.com www.newyorker.com/news/the-led...

Here is a link to 60 Minutes NIH segment... www.cbsnews.com/news/scienti...

From @thelancet.bsky.social www.thelancet.com/journals/lan...

New open-access paper in Statistics in Medicine: Inverse Probability of Treatment Weighting Using the Propensity Score With Competing Risks in Survival Analysis. onlinelibrary.wiley.com/doi/10.1002/...

Important column by the brilliant Ninez Ponce at our sister center at UCLA on federal changes and disappearances to key public data. www.milbank.org/quarterly/op...

Why oral health should be prioritized as a key marker of social progress: journals.sagepub.com/doi/10.1177/...

NIH cancels its first and largest study centered on women | Science | AAAS www.science.org/content/arti...

A quantitative approach to evidence triangulation: development of a framework to address rigour and relevance https://www.medrxiv.org/content/10.1101/2024.09.20.24314046v1

New work by Andy Dahl and Michal Sadowski on using GxE to study genetics of drug response now out in Cell Genomics www.cell.com/cell-genomic...

🎙️ Check out our latest episode with @noahgreifer.bsky.social! @epiellie.bsky.social & @lucystats.bsky.social chat with Noah about cool weighting methods, R packages, and more! casualinfer.libsyn.com/site/propens...

🚨 Join Our Team! 🚨 Are you interested in scientific research & population-based #epidemiology? We’re looking for individuals to be part of our team! Find out more about the #job 👉 uni.ms/v8r9x #INFRALINK #hiring @uni-muenster.de @volkswagenstiftung.de @gmds-ev.bsky.social #DGEpi_eV

🚨 New CRAN Task View: Compositional Data Analysis #rstats By Karel Hron, Javier Palarea-Albaladejo, Matthias Templ, Alessandra Menafoglio URL: cran.r-project.org/view=Composi...

*NEW PAPER* PROBAST+AI: an updated quality, risk of bias & applicability assessment tool for prediction models using regression or AI methods PROBAST+AI consists of two distinct parts: - model development (quality assessment tool) - model evaluation (risk of bias tool) www.bmj.com/content/388/...

It’s finally out! People, I’ve been hearing about this paper for so many years, but I am grateful it is out. Andrew Baker, Brantly Callaway, Scott Cunningham, Andrew Goodman-Bacon, Pedro H. C. Sant’Anna www.linkedin.com/posts/andrew... arxiv.org/abs/2503.13323

📢Visit our website to find resources, FAQs, demos, a manual for the software www.gesis.org/gesis-appkit 📧Register for our newsletter to stay informed: lists.gesis.org/mailman/list... 🤝 And please contact us since we also want to be a central hub for mobile researchers and their needs and visions!

Get ready for #DGEpi2025! Abstract submissions are now open for this year's annual #DGEpi conference, which will take place here in Münster! Submission is open until April 14th. Check out our conference website: 2025.dgepi.de @uni-muenster.de

Good summary on microplastics and our health risks nymag.com/intelligence...

Back by popular demand! We just released another call for proposals using RCTs and quasi-experiments. 3-page LOIs are due March 14. Details here: bit.ly/4ambjv3 (1/2)

🚨NEW PAPER! Together with Wuno Akingbuwa (@wonuola.bsky.social) we developed a way to estimate non-linear genetic correlations. It didnt sit well with us that known non-linear relations (e.g. BMI and MDD with a famous U shaped ) are poorly captured in statistical genetics. 🧵

I would also add UKBB + MVP + FinnGen meta-analysis for a subset of harmonized traits: mvp-ukbb.finngen.fi