RFK Jr. rejects cornerstone of health science: Germ theory. arstechnica.com/health/2025/...

Whooping cough cases are rising, and doctors are bracing for another tough year. There have been 8,485 cases reported so far in 2025, according to preliminary data from the CDC. That’s twice as many cases as this time last year. apnews.com/article/whoo...

Final point here: this post is from an anonymous NIH employee, and it feels critical to raise these voices precisely because the Trump administration has silenced credible NIH officials and their leadership is not representing them. donmoynihan.substack.com/p/the-nih-bu...

What can you do? Make noise. The ask is “Leave NIH funding at last year’s (FY24) level. No DOGE cuts. No recission to codify cuts. Hands off NIH.” If they are going to vote to cut cancer research, lets make that clear. donmoynihan.substack.com/p/the-nih-bu...

New at Can We Still Govern: An anonymous NIH employee maps out the GOP budget gameplan to permanently gut our most important science agency. Impoundment, delay and red tape will create artificial "savings" that become the new benchmark for NIH budgets. donmoynihan.substack.com/p/the-nih-bu...

What if the reference human genome had been paywalled? Yesterday I heard genomics pioneer Bob Waterston tell of a Perkin-Elmer CEO who envisioned a pay-for-access genome that would have smothered the incredible recent progress in biotechnology. www.thisgenomiclife.org/p/what-if-th...

The notice and COMMENT PERIOD OPENED TODAY AND CLOSES MAY 23. Anyone who would like to register an objection or comment should go to this page and click on the green “public comment” button at the top: www.federalregister.gov/documents/20... 6/n

Again, cancelling NSF & other federal competitive grants in higher education is massive contract breaking which undermines the most basic tenets of the rule of law & property rights in a democratic society.

What if LLMs could “read” & “write” biology? 🤔 Introducing C2S‑Scale—a Yale and Google collab: we scaled LLMs (up to 27B!) to analyze & generate single‑cell data 🧬 ➡️ 📝 🔗 Blog: research.google/blog/teachin... 🔗 Preprint: biorxiv.org/content/10.1...

ai.stanford.edu/postdoctoral... Postdoc candidates with strong expertise in AI for bio should apply for this new Stanford SAIL fellowship. We encourage applicants who would like to work with multiple SAIL faculty (we are part of SAIL as well). Come join us

A disturbing, but alas, not surprising, report from the Washington Post. [Gift link] wapo.st/4jtCPdw

🚨I could not be more excited to share our new preprint on saturation genome editing of the small nuclear RNA (snRNA) RNU4-2: www.medrxiv.org/content/10.1... A super fun collaboration with incredible duo @gregfindlay.bsky.social @joachimdejonghe.bsky.social from @crick.ac.uk 🧬🖥️🩺 🧵1/12

I am looking forward to spending the next year learning about 1st millennium Arabic and Islamic literature:

Drosophila biologists still going strong with the best gene names in the business: pubmed.ncbi.nlm.nih.gov/40215271/

Here's an interesting thought-instead of speculating about super-intelligent AGI, let's think about the implications of artificial intelligence as *normal* tech. New essay from the authors of AI Snake Oil: knightcolumbia.org/content/ai-a...

A very smart take on what "smarter than human AI" really means, from the inimitable @glichfield.bsky.social "Stop trying to compare AI to humans, or to anything from the movies, and instead just keep asking: What does it actually do?" www.bloomberg.com/opinion/feat...

I tell my kids they'll need to reinvent themselves during their careers. In grad school I hadn't written code since high school and worked in a lab with only two computers. A few of us taught ourselves Perl because it seemed useful and now 40% of my work is computational. And now it's AI:

Imagine being this family 100 years ago–4 of 8 siblings died of spinal muscular atrophy as babies. Within the past decade, we learned how to cure this disease. One of the treatments, ASOs, is being brought to bear on cancer.

Some reading for the weekend: Bad AI will be worse than p-hacking, same genes but different mutation patterns in autism and cancer, open datasets for drug development, and a couple more. www.thisgenomiclife.org/p/this-weeks...

Yes, more books like this:

OK, so this year won't be a total loss: www.nytimes.com/2025/04/09/b...

in case you're wondering, "what's the harm in claiming an extinct species has been brought back from the dead" when it most certainly has not, our interior secretary is already using it to justify taking animals off the endangered species list

Times seem rough but I'm glad we live in an era when we develop drugs by *designing* them rather hoping we'll get lucky and find a natural compound to cure a fatal disease. The era of drug design is still young & antisense oligos are an exciting part of it. www.thisgenomiclife.org/p/treating-d...

We released our preprint on the CREsted package. CREsted allows for complete modeling of cell type-specific enhancer codes from scATAC-seq data. We demonstrate CREsted’s robust functionality in various species and tissues, and in vivo validate our findings: www.biorxiv.org/content/10.1...

The confirmation process for a quack doctor with no administrative experience to run Medicare and Medicaid made barely a ripple because it’s not even among the administration’s top 100 idiocies.

Best popular science book I have read in a long, long time. Right up there with Gleick’s Chaos and Judson’s 8th Day of Creation. I find a lot of pop science boring because it reads like a Wikipedia article dressed up with a bunch of anecdotes. Not this.

Great review of state-of-the-art gene therapies that target the nervous system. There has been a tremendous amount of progress over the last decade. www.sciencedirect.com/science/arti...

One of the developers of support vector machines (after her PhD) on an experience familiar to everyone who went to grad school. Committee asks why she didn’t do X, she said it’s not important, but the real reason is she just wanted to graduate:

Questions answered. It comes from this classic perspective: pubmed.ncbi.nlm.nih.gov/28886340/

Genomics people, does anyone know what paper this image comes from? I took a screenshot years ago and haven't been able to find the original reference again, even with image searching.

It’s always important to update your skills throughout your career. I wrote up some recommendations for how to learn about AI, from basic overviews to graduate-level textbooks: www.thisgenomiclife.org?utm_source=n...

I am fascinated by the fact that the payment processing company Stripe founded its own press to sell books about innovation. Just picked up the memoirs of one of the 20th century's most important tech thinkers, Vannevar Bush, who I've long admired. press.stripe.com/pieces-of-th...

If, like me, you're not a physicist but fascinated by the deep relationship between thermodynamics, computation, and information, this new book is worth checking out. I've been looking for something like this for years. press.princeton.edu/books/hardco...

A friend drove two hours today for her breast cancer chemotherapy only to learn it was canceled due to NIH cuts. Clinical trial ended midstream. No treatment, no restart, no plan, no information. Her bro-in-law's chemo was also canceled.

Our latest preprint is up, describing how a ribosomal protein regulates the expression of its own gene by counteracting an intronic splicing enhancer. The result of some very insightful analysis by @gdstormo.bsky.social www.biorxiv.org/content/10.1...

If you’ve ever wanted to do a deep dive into Mesopotamian, Egyptian, Greek, & Roman classics, the Bible, early Christianity, late pagan poetry, early Medieval natural philosophy, Islamic literature, etc, this podcast is the one to guide you.

The entire business model of American R1 universities is not viable without restoration of federal funding, & every individual institution pretending it can solve this problem alone is also not viable. University presidents need to be speaking out, collectively & loudly about this, in DC, right now.

Today would be yet another good day for university presidents to loudly, collectively, & publicly start defending our institutions & industry. The silence is deafening & frankly, incomprehensible. What exactly are you waiting for? How many Columbias, Maines, & Johns Hopkins will it take?

"Cannot be legal" is a key point that every news story reporting these cuts needs to point out. Litigation takes time but it needs to be said loud and clear by those being targeted that it will be litigated. Right? Are Columbia and Johns Hopkins going to just roll over?

Philosophers and theologians have long debated the the good life. Socrates: seek wisdom & virtue. Augustine: live for God's glory. Our @WashU dean of public health's answer: the foundation of a good life is health & health research makes it possible. jamanetwork.com/journals/jam...

ATTENTION: TALENT AVAILABLE It appears that termination of SROs from institute review branches at NIH is underway (or worse). There are also program officers and SROs who have been terminated over the past month. IF YOU HAVE POSITIONS FOR SKILLED SCIENTIST-ADMINISTRATORS, PLEASE POST INFO HERE

Stand up for science: here are five things I stand for: open.substack.com/pub/thisgeno...

Ladies and gentlemen, the President of the United States: www.nytimes.com/2025/03/07/u...

If this were in good faith they would not be trying to push shit like this through before there is an actually confirmed NIH director.

My youngest daughter was born in 2009. Routine childhood vaccinations given to her birth cohort are expected to prevent 42,000 early deaths and 20 million cases of disease. Add that up for each annual cohort and you get a much healthier country: pubmed.ncbi.nlm.nih.gov/24590750/

In our updated TraitGym preprint (w/ @gonzalobenegas.bsky.social & Gökcen Eraslan), we evaluate Evo 2 on regulatory variants associated with human traits. We see marked performance gains with scale on Mendelian traits, although still a bit behind alignment-based methods. doi.org/10.1101/2025... 1/n

Nice take. I've been part of the Sc2.0 project (we did ChrXI) and now we're looking at whether AI can design sections of our syn chromosome given instructions of what proteins need to be encoded there and the relative expression needed for these. Right now models like Evo 2 can't do that, they are..

Actually we know. No. Generative AI currently cannot write genomes. EVO2 still does not dominate coding DNA. It has a very poor representation of distal regulatory DNA. And what long context information it is learning is not clear at all. bsky.app/profile/yun-...

Interesting last week reads on human genetics, multiomics & precision medicine🧵 1⃣ Coding & non-coding genetic determinants of circulating proteome Integrating WGS & proteomics, this resource offers genome-wide associations of rare variants with 2,907 proteins 🔗 nature.com/articles/s41...