chronicling the phenomenology of information superfluidity \ cto @transcriptabio — hyperscaled transcriptomics
San Francisco, CA, 🌎
Joined January 2010
- Tweets613
- Following1.3K
- Followers227
- Likes4.4K
Clayton Mellina retweeted
Drag-and-drop makes the digital graspable, but building it well is fiddly enough that many don’t bother. Declarative Dragging (with Josh Horowitz & Jeff Heer, #uist2026 best paper award), lets you make an element draggable just by listing the states dragging it can reach.
Clayton Mellina retweeted
Today we announced results from our research collaboration with @SynfiniInc, which produced a validated-quality lead candidate targeting MSH3, a gene increasingly understood to drive the progression of Huntington’s disease, in just a few weeks.
news.transcriptabio.com/p/tr…
Clayton Mellina retweeted
When you go skiing in the winter, and see those big machines spitting out snow, know that they are spitting out purified proteins made from dead bacteria.
I didn’t realize this until recently! Basically, they take Pseudomonas syringae cells — not engineered; a natural microbe found on plants — and grow them up in huge bioreactors. The cells make ice-nucleating proteins, which sit on their outer membranes and help ice crystals form. The cells are concentrated, freeze-dried, and sterilized until only these proteins remain.
The reason these proteins are useful is because they help water molecules form ice more quickly. Water can remain liquid even below 0°C, and only freezes when enough molecules form into a crystal. The proteins have, on their surfaces, repeating snippets of amino acids that help water molecules organize into that shape.
So these snow machines just blast super cold water with these proteins, and the two hit the air, mix together, and immediately freeze into snow.
Building a custom model of my (new) house with claude/codex. Fused Matterport data from the listing with floorplans and tape measurements.
First use case is to build the model itself, so feature #1 is 3d anchors I can reference in agent chat.
You need to be able to point at things.
Clayton Mellina retweeted
After years of working in electronics, this feels like a full circle moment. Many late nights later we are releasing Atlas Fields Studios, free for everyone
fields-studio.arenaphysica.c…
The Fields Studio App lets you explore electromagnetic behavior around PCB designs
#buildinpublic – at New York, NY
Clayton Mellina retweeted
We are thrilled to announce the inaugural Stanford Virtual Embryo Challenge at NeurIPS 2026 @StanfordAILab @NeurIPSConf , in collaboration with Laude Institute @LaudeInstitute , UCSD, and Harvard!
Join us in building AI models that predict how life takes shape, across space, scale, time, and perturbation: virtualembryo.ai/challenge.
We welcome participants from academia and industry, independent researchers, and even AI agent scientists!!!
Our vision is inspired by and complementary to the virtual-cell efforts championed by @arcinstitute . But life is more than a collection of individual cells.
Why Virtual Embryos? See below 🧵👇
Clayton Mellina retweeted
hot take but i think AI is literally on the verge of solving california’s megafire problem
> california already has 1,000+ AI-monitored wildfire cameras that have spotted 900+ fires before the first 911 call
> now the first 3 satellites in Google’s FireSat network just reached orbit to catch fires those cameras miss
> FireSat is designed to spot fires just 5 x 5 meters, roughly 1/400th the size current wildfire satellites can detect
> once complete, the network will scan the planet every 20 minutes
> together, cameras below and satellites above are creating a 24/7 early-warning system for finding tiny fires before they can grow into megafires
> but the missing piece is getting water there just as fast
> in one california test, an autonomous Black Hawk found staged fires, flew to water and filled a 324-gallon bucket
> it then calculated its own drop and extinguished it with no human in the loop
> NASA is also building the traffic-control system needed to coordinate aircraft like this through darkness and heavy smoke
> california already stations 70+ firefighting aircraft around the state, positioned to reach even remote fires within 20 minutes
> the goal is to automate that entire chain
> so AI finds the fire, launches the aircraft and fully extinguishes it within 10 minutes without waiting for a human
> each system is designed to do that across an area larger than SF + oakland + san jose combined
> scale that across california and dangerous fires get extinguished before they can become megafires
yes... obviously california will always have natural fires
and scaling this statewide might take a decade
but i'm increasingly convinced the $100B-entire-community-destroying fires will soon be a thing of the past 🤞
Clayton Mellina retweeted
Okay so I did something.
Drawn annotations that reflow with text as it re-shapes based on device, text size, wrapping, margins, etc.
Still work to do- but damn. Content-retaining notes across devices, orientations, screen sizes, fonts, or text size.
Clayton Mellina retweeted
Fine-tuning sequence-to-expression deep-learning models on personal genome and transcriptome data. A step toward individualized regulatory genomics. 🤖🧬 @GenomeBiology
link.springer.com/article/10…
Clayton Mellina retweeted
🚀 We are introducing PerturbPair (with @TakaKud0) — a platform that combines parallel Perturb-seq and optical pooled screening (OPS/PerturbView) in primary cells to systematically map at massive scale how genetic perturbations reshape cellular states across modalities.
With wonderful collaborators @TakaKud0, @AnaMeireles, @AntRios, @jchuetter, @MinOta, @ORozenblattRosen, @LeviAGarraway, @KGeiger, @avtarsingh, @jkpritch, and Aviv Regev.
Paper link: biorxiv.org/content/10.64898…
Clayton Mellina retweeted
Today we're announcing ESMFold2, an open scientific engine to power prediction, design, and discovery across protein biology.
The new model delivers state of the art performance on protein interactions, especially antibodies, a critical modality for therapeutics.
We have designed and validated miniprotein binders and single chain antibodies across five therapeutic targets that are important in cancer and immunology. We are seeing very high success rates, and affinities at levels consistent with therapeutic activity.
We’re also releasing an atlas of 6.8 billion proteins, and 1.1 billion predicted structures.
ESMFold2 is built on a state of the art language model that has been trained on billions of protein sequences.
A world model of protein biology emerges through language modeling.
We’ve used the techniques of mechanistic interpretability developed to understand large language models to understand the concepts ESM uses to represent proteins.
The model’s representation space has a compositional organization of features across scales, levels of complexity, and abstraction, that reflects and mirrors the understanding of protein biology developed through a century of empirical science.
This understanding emerges without prior knowledge, just from language modeling of protein sequences.
Language models are becoming a powerful substrate to understand and program biology.
The design of protein interactions is one of the most fundamental problems in biophysics, and has critical implications for the discovery of new medicines. A simple gradient based search with the model was able to discover high-affinity protein binders.
I'm excited by the potential this has to accelerate basic science and the understanding of proteins. And especially for the new avenues it opens up for therapeutic design and medicine.
Clayton Mellina retweeted
AI has transformed how we design therapeutics.
But targeted delivery is still an expensive guessing game.
Today @BobbyHollings and I are launching @deliverome with @beckypferdehirt and @radialscience at @AsteraInstitute, to fix that. 🧵
Clayton Mellina retweeted
Ever wanted to talk to a gene regulatory network? Ever wonder what they could do in a different context, embodiment, space?
New #preprint! @YanboZhang3
"Language Game: Talking to Non-Human Systems"
arxiv.org/abs/2605.16321
An early step in our program of developing ways to talk to organs, cells, molecular networks, and far weirder kinds of agents.
Clayton Mellina retweeted
New #preprint, @PigozziFederico:
arxiv.org/abs/2605.06746
"The Causally Emergent Alignment Hypothesis: Causal Emergence Aligns with and Predicts Final Reward in Reinforcement Learning Agents"
"A hallmark of life on Earth is the ability of agents to exert causal power and be drivers of subsequent events. This is key to cognition at all scales. Causal emergence, measuring the degree to which an agent exerts unique predictive power on its future, is one consequence of causal power. Indeed, recent discoveries have shown that biological agents, even minimal ones, increase their causal emergence after learning new memories. However, there is a major knowledge gap regarding how causally emergent artificial agents are. We focused on Reinforcement Learning (RL) of neural-network agents across an array of environmental conditions, encompassing different algorithms, agent architectures, and six environments arranged on a complexity spectrum. For consistency, we computed the causal emergence of their latent-space representations over their lifetimes. We used the recently proposed {\Phi}ID to estimate causal emergence and tested how it related to learning performance. Our results suggested a Causally Emergent Alignment Hypothesis: successful agents exhibited causal emergence that was consistently predictive of final reward early in training and whose representational dynamics aligned with reward improvement in most tasks. This idea suggests that causal emergence may be a previously undisclosed axis of reorganization of neural representations in RL agents, with the potential to establish causal relationships and interventions that will lead to better RL agents. Our work also highlights the alignment between causal emergence and learning as another way biological and artificial creatures compare."
Clayton Mellina retweeted
Insight as a toy model for mystical experiences.
Our new paper shows how Aha!, insight, Eureka, and mystical experiences can be modelled as realizations varying in intensity and size of the network affected. A small Aha! might restructure a local problem representation, whereas a mystical experience destabilizes a much larger self/world model.
