Nathan Spano retweeted
New @RowanSci miniblog! A while ago, I ran some MD simulations of the daraxonrasib tri-complex system and ended up finding some very interesting hydration dynamics in the binding pockets.
Check it out: rowansci.com/blog/daraxonras…
Nathan Spano retweeted
having a touch of the 'tism had a good run, but that era is rapidly coming to an end
Nathan Spano retweeted
Suppression of Aging in Mice by the Hormone Klotho | Science science.org/doi/10.1126/scie…
Nathan Spano retweeted
JUST IN: 🇺🇸 Elon Musk announces he will give 1 of his SpaceX $SPCX shares to "every kid in America" next year.
"This will be the first of many."
Nathan Spano retweeted
Mice engineered to overexpress klotho lived 20–30% longer with no signs of illness doi.org/10.1126/science.1112…
Nathan Spano retweeted
Un investigador de Anthropic acaba de decirle a una sala llena de doctorandos que su plan de carrera ya no existe.
La charla se titula "El fin del Antropoceno" y los datos son brutales:
→ El cómputo para entrenar modelos se multiplicó por 10 millones en la última década
→ La duración de las tareas que una IA completa sola se duplica cada pocos meses
→ Desde hace 7 meses los modelos resuelven problemas abiertos de Erdős que los matemáticos no habían logrado cerrar
→ En San Francisco, tener "timelines largos" significa pensar que la AGI llega en 10 años
Su proyección: el cómputo de 2030 será 10.000 veces mayor que el de 2025.
Todavía estamos al principio de la exponencial.
Sus 3 consejos para los estudiantes:
1. Usa la IA cada día. Si estás sobre la exponencial, nunca te quedas atrás de ella.
2. Elige proyectos que mejoren cuando el cómputo escala. El resto quedará resuelto antes de que los termines.
3. Deja de planificar tu carrera como si fuera estable. La opción segura ya no compra seguridad, ya no existe.
Y una frase que resume todo:
"Escribir es pensar. Si externalizas tu escritura, externalizas tu pensamiento."
La charla completa aquí ⬇️
Nathan Spano retweeted
three glaucoma patients got one injection of OSK genes into the eye, the yamanaka factor stuff, first human trial data ever. at 8 weeks two of the three improved on the visual field test
no serious side effects. the safety board just cleared the next dose, 3x higher
Nathan Spano retweeted
Former MIT scientist wins $6M defamation verdict against colleague who accused him of sexual misconduct trib.al/q1cVlXE
Nathan Spano retweeted
Replying to @stats_feed
Nathan Spano retweeted
Ex professore del Mit lasciò la cattedra perché accusato di molestie da una collega: il tribunale gli dà ragione. Risarcimento record trib.al/pHcoBMh
Nathan Spano retweeted
Nathan Spano retweeted
AI neurons run on simple math. Real neurons are far weirder.
A new open dataset fitted 1,000+ PyTorch-ready neuron models to real mouse and human cells. Does brain-like diversity make AI better? It's still an open question, but now it's testable.
bit.ly/3WfGAwR

ALT Illustration showing a connection between neuroscience and machine learning, featuring diagrams of neurons and circuits. Text reads: "A new bridge between neuroscience and machine learning. Artificial neurons run on simple math, but real brain cells are far richer. Can brain-like diversity make AI better?
Nathan Spano retweeted
Brain rhythms behind memory weaken in Alzheimer’s, but why? A new study from @GladstoneInst and the Allen Institute shows cells fire in waves, like passing a baton. In Alzheimer’s the handoff breaks. Boosting Nav1.1 restored the beat and improved memory.
Nathan Spano retweeted
CLAUDE MYTHOS JUST MOVED UP THE TIMELINE FOR QUANTUM COMPUTING
Anthropic published it this morning. They gave Claude eight standard quantum subroutines, a checker that verifies the answer is right, and told it to find cheaper circuits.
The big one is elliptic-curve point addition. That’s the step a quantum computer would run to break the keys behind Bitcoin and most of the internet. Claude found a design with 17x fewer gates than the best one humans had published, using 12% more qubits, it ran for 53 hours with almost no human input.
Second one, a decoder inside Google’s own optimization algorithm. Claude cut the gate count 12x against Google’s construction and got the scratch qubits from 665 down to 19.
Every estimate of when quantum computers become useful, or dangerous, assumes humans design the circuits by hand over years. If an AI does it in a weekend, every one of those timelines gets shorter. Anthropic’s own words: this could bring useful quantum computing “within reach sooner.”
The people on the paper were Harvard’s Mikhail Lukin and Susanne Yelin. Lukin co-founded QuEra.
$INFQ $IONQ $QNT $QBTS
GOOGLE IS ONE STEP CLOSER TO A FAULT TOLERANT QUANTUM COMPUTER
Every useful quantum computer needs a supply of something called magic states. They’re the ingredient that lets a machine run real algorithms instead of simple ones, they can’t be reused, and making them is so expensive that some estimates say 95% of a future quantum computer’s capacity would go to manufacturing them. That’s been the quiet bottleneck behind every roadmap in the industry.
Today Google published a paper in Science showing a new way to make them on its Willow chip. Instead of building a magic state from scratch at huge cost, you start with a rough one, grow it inside a small protected patch of the chip, measure it over and over, throw away every attempt that shows an error, and keep the survivors. They call it cultivation.
The numbers were a 40-fold cut in errors versus the raw state, and the highest-fidelity magic state ever produced on hardware, 99.99%. They also showed the grown state can be transplanted from one type of error correcting code into another, which is the plumbing a real machine needs.
Google’s own caveat is that it has to work at much larger scale before it counts for practical computing. Fair, but the bottleneck everyone assumed would be the expensive part just got a lot cheaper, on a chip that exists.
$INFQ $IONQ $QNT $GOOGL

































