That the AI vibe-coding crowd can't even figure out how to run stuff in the background is too on the nose.

These things are driving around all over Austin, TX. Such a hare-brained design.

There are numerous AI tells here but the most obvious one is the bus is on the wrong side of the road.

The dog looks a bit confused.

The illustration is obviously bad, but did you notice the mustache?

If you and eight of your friends are in a room with Jeff Bezos, the average net wealth in the room is going to be ~$20 billion dollars.

As I've said before: It's mind-boggling that the biggest software company in the world is so bad at software. Microsoft software was low quality in the 1990s, and now 30 years later the software is still low quality.

I am happy to share a review I recently wrote on the design of peptide binders. It gives an overview of experimentally validated tools and discusses the challenges of why peptide design is more difficult than the design of classical protein binders. www.chimia.ch/chimia/artic...

Congrats to @jbloomlab.bsky.social for getting elected to the NAS!

Congrats to @jrossibarra.bsky.social for getting elected to the NAS!

The pogo sticks are bad for the lawn?

The last 60% of the movie is the people navigating the bureaucracy trying to get the university to accept the artwork without clearly documented provenance. The final scene is when the dean declares the the university can now finally accept the donation and it will be accessioned soon.

Interesting physics fact: You can indeed fly supersonic without generating much of a boom (which has been the reason that you couldn't operate Concorde over land). However, you have to fly fairly slowly. LA to NYC is more like 4 hours rather than 2.5. See next post for details.

While it's correct that F, T, K awards are the mechanisms by which the NIH supports trainees *directly*, the majority of trainees who are supported by the NIH are supported by awards to PIs (R01, R35, etc.). Any reduction in those awards affects trainees as well.

It definitely doesn't work in my regino right now.

I was going to make a joke about how unstable BlueSky is recently and whether the developers have considered turning it off and back on, but realistically nobody is going to see my post since BlueSky doesn't really work right now, so what's the point?

It's very much a "pick your dystopian tech overloads" kind of situation but one of the benefits of ordering from Amazon is this won't happen.

And the secret to getting a lot done without compromising quality is not caring about quality unless when and until it matters. 1/2

Yes I do. Free email with 1GB of storage. All in a smooth web interface you could use from anywhere.

I've been using random seeds for years but I have no idea how they work. Seeds somehow(?) make the same random numbers? So I figured it out! New post includes an interactive PRNG generator, lava lamps, lottery fraud, @random.org, Chaucer, and Minecraft #rstats www.andrewheiss.com/blog/2026/04...

Just because you don't see them doesn't mean they don't happen. Check out what's going on in the backyard of that large mansion at the edge of town, in particular at night.

Preprint #1: Intrinsic dataset features drive mutational effect prediction by protein language models www.biorxiv.org/content/10.6... @clauswilke.com

In the second category is handling castor bean products, such as castor oil.

That's exactly how I experienced Mastodon. I hated it.

Copy paste into the address bar of your browser for a random generative artwork. No AI involved.

NIH is killing continuous submissions. It was the one perk you had as a standing member on a review panel. One fewer reason to be sad my term is ending this summer. grants.nih.gov/grants/guide...

Ten years ago, I predicted that Google Scholar would never fix their preprint bug. Here we are, a decade later, and the bug is still there, and now it's causing Google Scholar to claim that a preprint of mine is not publicly accessible. 🤦 blog.genesmindsmachines.com/p/the-google...

Ten years ago, I predicted that Google Scholar would never fix their preprint bug. Here we are, a decade later, and the bug is still there, and now it's causing Google Scholar to claim that a preprint of mine is not publicly accessible. 🤦 blog.genesmindsmachines.com/p/the-google...

I no longer get invited to social gatherings because I keep making people watch Angine de Poitrine's KEXP live show and giving the ▲ greeting. #microtonalfacemelter

This is the best evidence we have to date that aliens do indeed exist. www.youtube.com/watch?v=Uo4U...

In grad school I was reading about reproducibility and loved the ideas but obviously didn’t really get it because I spent an embarrassing number of weeks running my entire data pipeline everyday so I could pick up work at the end. It took 10-15 minutes every day.

Fascinating 🧵. I know incredibly smart people who care about the craft and have WILDLY different views of reproducibility (RP) 1. RP means everything must be recreated from scratch; persist nothing 2. RP means saving every intermediate to trace lineage and “drop in” at any point 1/

We recently had a conversation here about reproducible data analysis pipelines. I put my thoughts into a longer-form article. Whatever you do, it's critical you save and archive your final processed data right before plotting. blog.genesmindsmachines.com/p/creating-r...

The proteina-complexa paper is the 2nd this month suggesting that high K+E content, a staple of ProteinMPNN designs, is predictive of poor expression (the bits in bio benchmark paper showed a few weeks ago). In their case, they measure poor sequence recovery in phage display

I feel strongly this is a terrible idea. I battle it with my students all the time. Examples: - Can you quickly make this minor change to this figure? That'll take 30 min. to rerun all the preprocessing. - Can you send me the raw data to this figure that contains 5 points? That'll be 10 TB of data.

The Google Scholar preprint bug strikes again: I now think this is a version of it. GS can't find the PMC version of my article, even though it clearly exists, and hence harasses me about failing accessibility mandates. 1/

Excellent point. If you haven't personally cured cancer, don't opine on whether AI can do it.

Google Scholar claims an article we posted on bioRxiv is not publicly available. Has anybody encountered this? Is there something to be done? @richardsever.bsky.social

Why is this a news story? Cargill uses computer vision to make their butchering process 0.5% more efficient. Next up: Chemical engineer at DuPont discovers that a slight adjustment in buffer chemistry results in 0.5% less waste during nylon production.

The reverse is also true though. Americans want to do "Europe" in a week and it's absolutely too big for that.

Protein language models are bad at mutational effect prediction blog.genesmindsmachines.com/p/protein-la...

I have a joke about grant funding, but it didn't make the payline.

I always feel uncomfortable when I'm in one of these gondolas. They're attached to the cable with a strong spring. It seems a bit sketchy. (Though in fairness, gondola crashes are extremely rare.) www.bbc.com/news/article...

Wow, looks great! Looking forward to nicely formatted equations in ggplot2 axis labels.

NVidia announces new chip for space data centers. The chip works great in space. How you get it there, how you power it, how you cool it, those are your problems. But once you've figured them out, the NVidia chip will work just fine.

Some innovative ideas for sequence generation with protein language models.

Yup, this is the SpringerNature policy. If you're NIH funded you _must_ pay for Gold OA. Think twice before sending your paper there.

In biological data finding meaningful ways to do training-test splits without leaking information can be tricky as readouts can be highly correlated, causing over-optimism. Here a nice example based on protein language models 👇

Good news, protein engineering researchers, mutation effect predictions remain an unsolved problem.