Conrad young @_RadBio_

What if a binder could not only bind a target, but selectively recognize one conformational state over another? Today, we’re excited to share AlloGen, our experimentally-validated framework for conformation-selective binder design! 🐙 📜: arxiv.org/abs/2606.05474 🤗: huggingface.co/ChatterjeeLab/… 🧵👇🏾

Building autonomous agents for scientific discovery? 🧬🤖 @GoogleDeepMind Science Skills is now available on GitHub. We've open-sourced this specialized toolkit to accelerate your agentic workflows with scientific grounding and higher token efficiency. Download now ↓ github.com/google-deepmin…

Scientists from our university have used a neural network to create all-atom models of proteins, as well as the dynamic movements that govern their function. Their comprehensive yet simplified approach resolves a major bottleneck in biology. actu.epfl.ch/news/ai-genera…

New papers alert! Last week was a big week for new papers in target discovery. One research paper in @SciReports Nature Scientific Reports on multi-modal target identification and benchmarking of target ID systems. And another Mega Review paper in @NatRevDrugDisc Nature Reviews Drug Discovery (NRDD) on target discovery and assessment using generative AI with a consortium of experts from many institutions specializing in target discovery. At the end of the day, target discovery must translate into disease-modifying and ideally curative treatments confirmed by late-stage clinical trials. There are over 60 target discovery philosophies that worked in the past to some extent and AI can help choose the right approach for a specific objective and provide more confidence when working with absolutely novel targets. Check out the links in the comments.

keyboard_arrow_left Previous keyboard_arrow_right Next

We quietly launched the Symphony repo on Github last month, and it's already accumulated 15.5k stars! Excited to share this post that dives into it even deeper: a library that lets you use Codex to orchestrate work normally done by teams of engineers. openai.com/index/open-sou…

AI can now design antibodies that bind with atomic precision, but not ones that cells can produce. Our preprint closes this gap, delivering a structural principle, an AI-guided rescue pipeline, and adalimumab variants with 20-100x in vivo potency. biorxiv.org/content/10.648…

FlashAffinity: Bridging the Accuracy-Speed Gap in Protein-Ligand Binding Affinity Prediction biorxiv.org/content/10.648… #biorxiv_molbio

You can now give your agent deep knowledge of millions of papers in one line with #paperclip!📎 >8 million papers natively indexed for agents. Much more thorough + often 10x faster than standard deep research. Just add the paperclip mcp (instruction below).

The MedGemma 1.5 technical report is out 👇 arxiv.org/pdf/2604.05081…

We released Claude Opus 4.6 just two months ago. Today we're sharing some info on our new model, Claude Mythos Preview.

keyboard_arrow_left Previous keyboard_arrow_right Next

In collaboration with the Institute for Protein Design, University of Washington led by Prof. David Baker, we are launching a new research fellowship to advance open-source AI for protein structure prediction and design! More 👇

タンパク質・リガンド設計ツール30種以上を統合したPythonフレームワークBioPipelinesが公開されました RFdiffusion・ProteinMPNN・Boltz2等を数行コードで連携することが可能で、JupyterからSlurmまで同一スクリプトで動作するため、運用がしやすくなりそうです ソースはリプ🔽

Another amazing post from Nick Boyd and Sam Guns at Escalante Bio, finetuning and RLing BoltzGen on a small dataset, but showing extremely strong results. This process, borrowed from LLMs, combines the best of hallucination and generative into one! blog.escalante.bio/teaching-gener…

Terminal Conjugation Enables Nanopore Sequencing of Peptides | Journal of the American Chemical Society pubs.acs.org/doi/10.1021/ja…

If you want to work at the absolute frontier of molecular foundation models: I am hiring for the structure team, get in touch!

Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞) @teortaxesTex

4 months ago

if you forgot, the CEO and founder of Isomorphic is Demis Hassabis. So ByteDance has just caught up (and slightly exceeded) AlphaFold 3, and the original team is… significantly ahead.

8 28 662 107K 251

The Iso team has cooked something incredible: our new technical report unveils the latest results from our drug design engine, the IsoDDE, progressing far beyond AlphaFold 3. This breaks new ground compared to AF and other similar methods by a significant degree across all key benchmarks. 1/7

We worked with @Ginkgo to connect GPT-5 to an autonomous lab, so it could propose experiments, run them at scale, learn from the results, and decide what to try next. That closed loop brought protein production cost down by 40%.

When medicine laureate Tu Youyou was 16, she contracted tuberculosis and had to take a two-year break from school. “This experience led me to make a decision to choose medical research for my advanced education and career – if I could learn and have (medical) skills, I could not only keep myself healthy but also cure many other patients,” she said. #WomenInScience #NobelPrize

Interesting CELLULAR AGING BREAKTHROUGH Scientists may have found a molecular blueprint to revive old cells. A new UCSF study identified four key transcription factors, E2F3, EZH2, STAT3, and ZFX, that can push aged cells back into a youthful state. By reactivating this gene control network, old human cells regained higher energy levels, increased cell division, and younger gene expression patterns. In aged mice, activating one of these factors reduced liver fat and scarring and improved glucose metabolism, making old tissue function more like young tissue. This discovery suggests aging isn’t just damage, it’s a loss of coordinated gene regulation that may be partially reversible. Instead of replacing cells or turning them into stem cells, this approach hints at resetting cells from within, opening a new path toward therapies focused on extending healthspan, not just lifespan.