Big BIG congrats to @mbadiagraset.bsky.social. She led this work beautifully - thoughtful, rigorous, and relentless 💪.
There’s more coming from her! Stay tuned.

Our latest paper is out! We built a large-scale mutational map of amylin to understand how sequence drives #amyloid formation, a process linked to type 2 #diabetes🩸📈📉. It’s a step towards connecting molecular mechanisms with human genetic variation in a systematic way.

We hope this is a small step forward towards understanding how aggregation can be tuned to fulfil different functions rather than disease!

By coupling this with functional assays, we discover that necroptotic signaling requires a “sweet spot” 🍭 of amyloid formation. Variants that disrupt this balance are rare in the human population, consistent with evolutionary sequence optimization.

We find that both proteins rely on an evolutionarily conserved aliphatic tetrad to nucleate amyloids, but the prioritization of aliphatic interfaces differs: RIPK3 relies primarily on one interface, whereas RIPK1 engages two.

@marianomartin.bsky.social mutagenised mouse and human RIPK1 & RIPK3, the proteins that assemble into amyloid structures to build the necrosome and signal inflammatory cell death 💀.

This time we went “functional.” 🙂
After years studying pathological amyloids, we followed RIPK1/3 into the other side of aggregation - where amyloids help cells decide when to die.
🔗🔗 🔗 www.biorxiv.org/content/10.6...

We hope this is a small step forward towards understanding how aggregation can be tuned to fulfil different functions rather than disease!

By coupling this with functional assays, we discover that necroptotic signaling requires a “sweet spot” 🍭 of amyloid formation. Variants that disrupt this balance are rare in the human population, consistent with evolutionary sequence optimization.

We find that both proteins rely on an evolutionarily conserved aliphatic tetrad to nucleate amyloids, but the prioritization of aliphatic interfaces differs: RIPK1 relies primarily on one interface, whereas RIPK3 engages two.

@marianomartin.bsky.social mutagenised mouse and human RIPK1 & RIPK3, the proteins that assemble into amyloid structures to build the necrosome and signal inflammatory cell death 💀.

Characterizing protein sequence determinants of nuclear condensates by high-throughput pooled imaging with CondenSeq Nature Methods - CondenSeq is an imaging-based, high-throughput platform for characterizing condensate formation within the nuclear environment, uncovering the protein sequence features that...

In the nucleus, many intrinsically disordered proteins (IDPs) form condensates. What IDP sequence features drive this behavior? We developed CondenSeq, a high-throughput approach to measure nuclear condensate formation, and applied it to ~14,000 IDPs to find out!

rdcu.be/eq975

Our lab @ibecbarcelona.eu is hiring both 👩‍💻 and 🧪 profiles (or mixed!) to mechanistically probe disordered and aggregation prone proteins at scale using combinatorial strategies. Get in touch!

This one is close to my ❤️ as we learnt SO MUCH in the process. We hope this approach will be employed to address the transition states 🎇 of many other reactions - amyloid and not! Well done @immunobananna.bsky.social, Mireia and @ajfaure.bsky.social!

New cocktail 🍹 in town - in collaboration with @benlehner.bsky.social : Massive combinatorial mutagenesis + amyloid kinetics selection + thermodynamic modelling showing us the portrait 🎨 of an amyloid transition state. Read about it in this great thread:

Confocal microscopy photo

A black hole in the early universe? Sometimes I like to imagine that I operate a telescope, not a microscope 😁. The hole in the middle of the image is actually a blood vessel. The stars are astrocytes🟡 in the human brain, interacting with beta-amyloid🔵. The brain is from a donor with #Alzheimer.

Candle-lit commuting sponsored by @rodalies.info 🕯️🕯️🕯️

Full slide deck here from our talk yesterday on the gender divide and relationship recession, featuring many more countries and plenty of economic data too: www.dropbox.com/scl/fi/1lp1d...

Medical Research Funding — Unbreaking How the administration is breaking the government, and what that means for all of us.

I urge everyone to read Unbreaking’s new page on everything the govt is doing to destroy & diminish medical research funding.

It’s the best thing I’ve read on this topic: @lizneeley.bsky.social & co have such done an incredible job.

unbreaking.org/issues/medic...

And what a great talk to end #VariantEffect25 ! Thanks to the amazing crowd that travelled to Barcelona and made MSS25 such a fantastic meeting. Lots to think about now! 🧦🗺️☀️

Fantastic talk #VariantEffect25 by @pollyfordyce.bsky.social on how much we can understand about sequence-function relationships once we are able to factor in kinetic 🏎️rates. Microfluidics 💧💧for the win!

@pollyfordyce.bsky.social at #VariantEffect25

Good views, food and even better company at #VariantEffect25

The first poster session just started! #VariantEffect25

Welcome back from the lunch break. Kicking off the afternoon session at #VariantEffect25 we welcome ​​Patrick Hsu @pdhsu.bsky.social all the way from @arcinstitute.org – Bridging biology and AI for genome design 💥🧬🤖 #SynBio #AI Let’s gooo! 🚀

It is a deeply inspiring story, reminding us why we do what we do.

Welcome back from the coffee break and to the last session of Workshop 1 of #VariantEffect25. @pascalnotin.bsky.social is giving us an overview of protein fitness prediction models! #AI #ProteinML

Next up at #VariantEffect25 Workshop 1 we have Noelia Ferruz! From Rosetta to Flow Matching: The Evolution of Protein Design -How Should You Begin Today?

Let’s get this party started! Kicking off workshop 1 of #VariantEffect25 we have Albert Escobedo — From Fitness to Free Energy: Thermodynamic Modeling of Protein DMS Data for Accurate Genetic Prediction with MoCHI 🍡🧬

Our "Atlas of Variant Effects 2030 Roadmap" is live: zenodo.org/records/1542...

1/n