Looks like electron cryo-microscopy/tomography has finally lost the battle

Somewhat related: I was looking for a JS library to read MRC files a while ago, but couldn't find one (sorry if I missed it). So you can now “npm install @warpem/mrc-parser” github.com/warpem/mrc-p...

I had Claude make this as a mini-game for Relay's disconnect screen (like Chrome's dinosaur), but then pivoted to stand-alone. In this game, you have to find Euler angles matching the target projection, i.e. cosplay RELION. It can be pretty hard. Works on mobile too! warpem.github.io/euler-hunt/

Once you install peonping.com, don't forget to mute the volume if you decide to claude something during a meeting

A cats lock key that lets you quickly disable the keyboard when a cat approaches it would be a lot more useful than the antiquated caps lock

What is the current resolution record for in situ nucleosomes? The best unofficial result I remember seeing was around 7 Å, but I feel like I've heard a better figure mentioned somewhere already

California is famous for its sunsets, but sunny early mornings are the real magic here

Do we have “I am a robot” CAPTCHAs for Moltbook yet?

Super happy to finally see native vertical tabs in Chrome, but why why why is the + button all the way at the bottom? #uxsky

If someone's age is a prime number, are they in their prime?

This exciting new method has a bright future!

Great CV advice in this thread! I'll also add that for computational scientists, a GitHub/GitLab link is a must. Sure, there are people whose work I know well enough to not care about their repos, but don't bet on that.

2026 will bring great tilt series alignments for the ribosome-deficient among us 🎆

Here is why friends don't let friends calculate FSC on denoised half-maps. I took EMD-17961, denoised the maps using Warp's Noise2Map, and calculated the FSC between various things. With noise gone from the half-maps, their consistency is no longer related to SSNR. Denoised map vs. model still works

Don't get me wrong: the original IsoNet was the coolest thing to happen in cryoET in at least a decade. But after the initial “wow” moment it's time to make this topic more sciency

I'm not superstitious but I never set my TV's volume to 13

I'm used to seeing driverless Waymos at this point, but seeing a Waymo with no driver *and* no passengers roll down a hill still gave me a moment of “OMG THE PARKING BRAKE FAILED” panic

Cool, definitely a visual improvement! But do I understand correctly the FSC was calculated between 2 *denoised* half-maps? You can't quantify resolution like that, only consistency.

😂

OMG they will go so well with my Bliss ugly Christmas sweater

It took me several weeks to notice how perfect this arrangement was

The things Apple will do to procrastinate improving Siri

Ugh, I'm starting to get spam texts with em dashes

Happy to announce "Technical Innovations Using #SerialEM: A Community Meeting", a new format of hybrid meeting David Mastronarde and I are organizing in April 2026. For more info, a preliminary list of speakers, pre-registration etc. visit www.nexperion.net/semcm2026

The North Bay is so packed with Porsches that the non-Porsches (Norsches?) are the real exotics there

Important to remember the theoretical size limit is based on assumptions and a model that may or may not be complete. It isn't some indisputable universal truth.

A fun introduction to problems arising from having to do Cartesian things on a sphere: www.bowerbyte.com/posts/blocky...

You're absolutely right!

K-Pop Demon Hunters OST playing on the radio? Cool!

I just found out the latest version of MacOS finally implements the motion cue dots iOS has had for a few years now, and it's doing wonders for me when I need to use my laptop on bus rides

I don't know what cursed token sequence puts Claude into stupid mode, but it's impossible to recover from it and you have to start a new conversation

First flight lesson survived!

Whoever gets the resolution of raw tomograms to <10 Å within the sampled parts of Fourier space would probably deserve a Nobel. Or they'd get one anyway for the crazy new physics they discovered to achieve that.

I had a dream last night where some online store had the 5090 FE in stock at MSRP. It was a happy dream 🥹

He's currently beta-testing “faucet with a sensor that doesn't register your hands until you bring them in the least fortunate position, then sends water full-blast horizontally at them to be deflected to your crotch area to make it look like you peed yourself” in select locations

Y'all marketing people need to stop handing out reusable water bottles at every event

Came for the memes, stayed for the science

I've got bad news for #teamtomo: earlier today, our resident headcrab attacked @lorenzlamm.bsky.social, turning his Mem🧠 into 🧟‍♂️🧠

When you wake up at 4am because your spouse used the allyourblanketarebelongtous cheat code, and your brain puts on a slideshow to make sure you won't fall asleep again, but you need to make slides for a talk tomorrow so it's a literal slideshow

Been thinking of trying WarpTools but worried about having to start processing from scratch? @rasmusjensen.bsky.social wrote this handy guide for moving projects from desktop Warp to the latest version which runs on linux warpem.github.io/warp/referen...

Damn, this is brilliant. But I feel like the authors didn't answer the most important question: can you make tiny soap bubbles with this??

If Anthropic turns a profit this quarter, it will be mostly due to me burning tokens in Claude Code

Don't miss this workshop at a top-secret location in central London!

I must consider the possibility that I like Melbourne even more than Heidelberg

It's amazing how your brain instantly adjusts to everything being upside-down in Australia. But you can't fool a camera!

Hey Australians (and everyone familiar with the place)! I'll have a few days in Melbourne after #Tomo10 next week. Can you recommend anything fun to do/explore while I'm there?

I'm sorry, ChatGPT, but Claude and I are just happier together

De novo tilt series alignment accuracy is still the #1 problem, even with thinner samples. Can't recognize particles if they're mangled by misalignment.

Looks like way less to me 😕

People who set their car to make a loud beep upon locking: WHY???