Ciencia en alianza: @GSK y ANII consolidan una cooperación que genera impacto🤝
A través del Fondo Sectorial ANII - GSK, esta alianza público-privada ha permitido que equipos locales lideren investigaciones de nivel mundial en enfermedades complejas, consolidando nuestro ecosistema científico como un referente de generación de conocimiento
Leé la nota de Forbes Uruguay 📰👇
Hogares unipersonales: 15% en 1985 → 29% en 2023 (gente mayor, divorcios, etc.) y seguirán creciendo.
Ojo con efectos no intencionados de restringir oferta:
+ presión sobre pisos familiares, alquileres, etc
Hay que CONSTRUIR + vivienda y facilitar acceso.
Welcome to the Lab of the Future! 🧬🤖
Excited to share LUMI-lab, out today in @CellCellPress — a self-driving platform that pairs an AI foundation model with a robotic lab to autonomously discover ionizable lipids (LNPs) for mRNA delivery.
The core problem: Designing lipid nanoparticles (LNPs) is hard. The chemical space of ionizable lipids is vast, experimental cycles are slow, and — critically — historical LNP datasets are far too small to train a predictive model from scratch. Most AI approaches in this space hit a wall immediately: not enough data to learn from.
Our solution: lab-in-the-loop foundation model learning. Instead of training on LNP data alone, LUMI starts as a transformer-based foundation model pretrained across broad chemical space, building rich molecular representations before it ever sees a single LNP experiment. Then it enters a closed loop with a robotic synthesis platform: predict → synthesize → assay → update. Each round of real wet-lab experiments fine-tunes the model, which then proposes smarter candidates for the next round. The lab isn't just validating AI predictions — it's actively teaching the model, continuously.
What happened when we let it run: LUMI-lab autonomously synthesized and screened 1,700+ ionizable lipids in human bronchial epithelial cells. The top candidate — LUMI-6 — features a brominated lipid tail, a structural motif that had been largely overlooked in LNP design. LUMI found it without being told where to look. When formulated into LNPs and delivered intratracheally to mice, LUMI-6 achieved 20.3% gene editing efficiency in lung epithelial cells — a compelling result for one of the hardest-to-reach therapeutic targets, directly relevant to diseases like cystic fibrosis and alpha-1 antitrypsin deficiency.
Why this matters beyond LNPs: This is a proof of concept for a broader thesis — that foundation model pretraining + active learning + robotic experimentation can overcome the data scarcity bottleneck that plagues AI-driven discovery in biology. You don't need a massive domain-specific dataset to start. You need a model that can generalize, a lab that can generate the right data, and a loop that connects them.
Huge congratulations to first authors Yue Xu, @HAOTIANCUI1, and Kuan Pang, and to the entire @BowenLi_Lab team. Grateful to our collaborators at @UHN and @UofTPharmacy, and to Princess Margaret Cancer Centre Research @PMResearch_UHN.
📄 Paper: https://t.co/5tfD3dkSr7
Today we announced that the U.S. Food and Drug Administration’s (FDA) Center for Biologics Evaluation and Research (CBER) has notified Moderna that it will not initiate a review of the biologics license application (BLA) for our investigational influenza vaccine, mRNA-1010, and has issued a Refusal-to-File letter.
Read more: https://t.co/BYkEdHyha0
Today we share a technical report demonstrating how our drug design engine achieves a step-change in accuracy for predicting biomolecular structures, more than doubling the performance of AlphaFold 3 on key benchmarks and unlocking rational drug design even for examples it has never seen before.
Head to the comments to read our blog.
Not that anyone asked, but the country with the World's lowest life expectancy (Chad) has a higher life expectancy than the country with the highest life expectancy 120 yrs ago (Sweden)
Think Science (and especially Medicine) has somehow failed you? Think again
#ScienceMatters
In the 1970s and 1980s the doomer movement targeted gene editing with the same kind of unhinged paranoid rhetoric and apocalyptic end-of-society screaming.
They delayed life saving medicines and prevented advances in saving lives by decades.
We are only now starting to get to the point where we can really use genetic engineering to cure diseases.
Below you can see trials that edit the gene for high cholesterol. This was save many many lives and dramatically bring down the costs of health care on individuals and society if we had more medicine like this that targeted complete fixes instead of a life long pill subscription.
Doomer delusions have real body count and always have.
But people never learn.
Like Sauron, the Doomers get defeated in one generation, but come back in a new form with a new paranoid target to dump their fears and anxieties into.
The only thing we have to fear is fear itself.
Gracias de corazón a todos los que participaron en la colecta organizada por el @IPMontevideo. Con mucha alegría les cuento que ya reunimos los fondos para continuar mi tratamiento en el exterior. Gracias una vez más por tanto apoyo!!!
Un fármaco uruguayo contra la obesidad, que "engaña" al tejido adiposo. El hallazgo, concretado por una start-up alojada en el Institut Pasteur , es único en su clase, reveló el Director Ejec del IP , Carlos Batthyany . Son casi 4 minutos muy recomendables. La entrevista completa en https://t.co/FEAgLSuMFn
Nuestra democracia tiene un debe pendiente, de larga data, que es evaluar políticas sociales para evidenciar dónde realmente se logran impactos significativos. Tenemos un sin número de programas pero poco se sabe de sus impactos. Sin evaluar se invierte inercialmente.
holy shit
MIT researchers just turned skin cells directly into neurons without stem cell intermediate, 100-fold efficiency boost, and they actually worked when transplanted into mouse brains
1/
I taught an LLM to optimize proteins. It proposed a better carbon capture enzyme.
Introducing Pro-1, an 8b param reasoning model trained using GRPO towards a physics based reward function for protein stability.
It takes in a protein sequence + text description + previous experimental results, reasons over the information given in natural language, and proposes modifications to improve the stability of the given sequence.
🧵(1/n)
Top 5 advances in medicine in February (🧵)
1. The first large trial to show that oxygen can improve outcomes in stroke
Patients in this trial had acute ischaemic stroke (caused by a large vessel blockage in the front part of the brain)
They received either 100% oxygen at 10 litres/minute (via a non-rebreather mask), or a sham treatment (roughly equivalent to breathing oxygen from air), for 4 hours
All patients were treated with thrombectomy (mechanical clot removal), and that's the key: it's thought that oxygen 'freezes' the penumbra (the salvageable brain tissue) before blood flow is restored by thrombectomy
That also explains why this hasn't been found before: previous trials were done before thrombectomy was widely used
This trial was done in only one country, in a select group of patients, so we'll need broader trials: these are underway
https://t.co/Y0gr6RAV5B
Top 5 advances in medicine this week (🧵)
1. Gene therapy provided sight to children born blind
The 4 children had a rare form of blindness, caused by mutations in the retina gene, AIPL1
A functional copy of this gene was introduced, in one eye per patient, using the other eye as a control
All of the children started to see (though imperfectly): they started to draw and write, moved around on their own, and played with others
This is not the first time gene therapy has been used to treat blindness (there is even an approved therapy for inherited RPE65 mutations)
This approach is now being rolled out more widely to children affected by AIPL1 deficiency
Study: https://t.co/rYBkqPUPeo
Glad to set foot in Montevideo, 30 years after my predecessor Jacques Delors.
Throughout this time, Uruguay has always been a trusted, like-minded friend.
Today it is also a key hub for trade and investment in Mercosur.
Thank you @LuisLacallePou for bringing us together.
10. A method to get antisense oligonucleotides (ASOs) across the blood-brain barrier, to deplete pathological proteins
These oligonucleotide transport vehicles (OTVs) had better brain distribution, and target knockdown, than ASOs delivered into the CSF
https://t.co/i9Wr2z9ky3
Reminder
We are over 3 years and almost 13 BILLION doses of the SARS-CoV-2 vaccines (including mRNA vaccines) given worldwide.
There have been NO mass deaths.
NO mass side effects.
NO research proving issues.
NO widespread panic.
NO mass birth defects.
NO mass sterilizations.
NO sweeping new government mandates.
NO permanent lockdowns.
NO permanent vaccine passports.
Anyone who's claiming that they were "proved right" about vaccine conspiracy nonsense are speaking from a place of profound paranoia and ignorance.
The mRNA tech has proven itself a valid concept, anyone who says otherwise is living in a delusional land of podcasts and YT conspiracies.