Harsh Dutta

selfie of a sleepy biologist with his research poster on an insect lipid flippase

If you're around for the Just-B poster session today at #bps2026 Exhibition Halls, do drop by B-56 if you're interested in lipid transporters!

Incredibly excited to share our findings on the fruit fly lipid flippase! Please join the #bps2026 poster session at 1.45 PM today and the Membrane Transporters Platform on Tuesday! Big thanks to the organisers for their recognition and @sandipbasak.bsky.social for his generous guidance.

An impressive number of graduate applications this year used the phrase “curiosity-driven approach” 🤔

Asymmetric gating of a homopentameric ion channel GLIC revealed by cryo-EM | PNAS Pentameric ligand-gated ion channels (pLGICs) are vital neurotransmitter receptors that are key therapeutic targets for neurological disorders. Alt...

We are excited to share our latest work on GLIC published in @pnas.org, where we developed a #cryoEM data processing strategy that allowed us to isolate and reconstruct multiple asymmetric states, revealing detailed insights into the channel’s activation mechanism. tinyurl.com/mwtb4rpj

It is like magic. No one really wants to know how a trick is done. they want to believe it is real. With LLM they are more interested in talking about what it MIGHT do, or what they think it can do. Then what it actually does.

thanks for sharing this

Atkinson Hyperlegible Font - Braille Institute Read easier with Atkinson Hyperlegible Font, crafted for low-vision readers. Download for free and enjoy clear letters and numbers on your computer!

periodic reminder of the existence of Atkinson Hyperlegible, a free font available from the Braille Institute designed to improve readability for people with low vision

I use it in talks because it's pretty and also because, as an audience member, I am perpetually squinting at people's slides

the horrors persist, but so DOI (a doi is a persistent identifier for academic papers, e.g. 10.1000/182)

#academicsky #librarysky #libsky #skybrarians #medlibs do u guys like persistent identifier memes?

OpenAI doesn't report a single number for its climate impacts - not anywhere, in any way, shape or form.

It sounds kind of obvious but the fact that a company with such incredible electricity hunger isn't disclosing basic information about its energy consumption and associated emissions is v bad..

Years after the idea and preliminary experiments, Claire Coupland's amazing structure of the endogenous vacuolar- or vesicular-ATPase in its NATIVE vesicle is now in bioRxiv "High resolution cryo-EM of V-ATPase in native synaptic vesicles"
www.biorxiv.org/content/10.1...

im on every app. im in your city, in your neighborhood. the sales associate who handed you a large when you asked for a medium? that was me. i am now behind you.

yes, thanks for the discussion!

this is also correctly predicted

yep correct, the small protein is the green box in bottom right. This is a large membrane protein. The interacting region of the first protein can be verified by homology modelling. All the high confidence regions predicted by AF3 can be verified by homology, and inter domain packing is also good.

PAE plot from AF3

PAE looks like this. The iPTM score is also 0.74, so these helices are really unreliable, and they take up a considerable amount of amino acids in the total sequence.

ah I see, thanks

thanks, yeah I figured because manual inspection of the sequence and IDR prediction server also spit out similar disordered behaviour for these regions. Most of these are just poly-Ns, some need to be checked more carefully.

ohh that low confidence in the middle of the curl..yikes.

ohh do you have a visual example of 'curled' models?

A 500 kDa protein model generated by AlphaFold 3

AF3 model of the protein with the binding partner

AF3 predictions for a 500 kDa protein without (left) and with (right) a known 40 kDa binding partner. Apparently disordered regions collapsing into low confidence cytosolic helices. Is AF hallucinating these helices?

thanks!

🙌🙌

haha it is art as long as it provokes a human emotion, pain in this case

yep, it was apparently a fabrication issue because we didn't notice anything strange while handling and preparing this grid, it was only clear at the screening stage

bad cryoEM grid imaged at low mag

bad cryoEM grid imaged at high mag

pain, just literal pain

FEBS Press Clinical multidrug resistance is caused by a group of integral membrane proteins that transport hydrophobic drugs and lipids across the cell membrane. One class of these permeases, known as multidrug...

ABC transporters contributing to drug resistance in bacteria: doi.org/10.1016/S001...

trying to fix the student numbers right here 😬 hopefully my friends pop in too

Been thinking about creating a collection of good protein structure figures, as inspiration for my own work.

#1 www.nature.com/articles/s41...