Combining frontier AI & frontier biology to help scientists cure or prevent disease
Joined June 2017
- Tweets6.1K
- Following616
- Followers32.3K
- Likes4.9K
Biohub scientists built a model that recognizes sepsis from blood-cell images without being told what to look for.
Remo-ID, trained only on sample-level labels, got 43 of 47 samples right in an early study.
Led by Paul Lebel and team: biohub.org/blog/ai-sepsis-de…
biohub retweeted
The next frontier for AI reasoning could be understanding how human cells behave.
@biohub's @lorigoler explains why new biological datasets could help AI make scientific discoveries:
"The idea is to sort of create this almost an oracle that's sort of like a generalizable reasoning model for the cell and for biology."
400 million people live with a rare disease. Fewer than 5% have an approved therapy. Applications are open for Cycle 4 of Rare As One, funding patient-led organizations working on rare respiratory disorders, rare immune dysregulation disorders, and sarcomas. Apply by 12/15: bit.ly/RAO-cycle-4
New on the cryoET Data Portal: three cryo-electron tomography datasets benchmarking the dual laser phase plate (xLPP). The microscope captured PP7 virus-like particles in an E. coli cell, and results confirm the unprecedented resolution enabled by the xLPP technology.
The portal puts the data in researchers’ hands to explore, evaluate, and build on. Check it out: cryoetdataportal.czscience.c…
After reconstructing a 3D volume of a sample from a series of images, Biohub scientists use an image processing technique called subtomogram averaging to determine structures of macromolecules (such as proteins) in situ. With the xLPP, structures approach atomic resolution.
Learn more about the laser phase plate:
biohub.org/news/laser-phase-…
🧵 The #zebrafish imaging data from @ZFINmod is immensely valuable to researchers, but users are limited to browsing it one gene at a time.
gene2fish, a new Biohub tool, lets you compare dozens of genes side by side, saving hours of time. Try it on Zebrahub: zebrahub.sf.czbiohub.org/gen…
The work was led by Biohub’s Vera Janssen and @loicaroyer, in collaboration with the software engineering team. 🐟
Read the preprint here: biorxiv.org/content/10.64898…
biohub retweeted
Some astute observers have noted that our @biohub API for ESMFold2 doesn't seem as accurate as some other API services. This is because we predict only in single sequence mode.
Now a max accuracy option exists for in the API to be able to quickly fold your favorite protein!
biohub retweeted
Single-cell atlases give us marker genes. Where are they expressed in the embryo? 🐟🧬🔬
gene2fish shows 53,595 in situ images of 8,263 zebrafish genes side by side, in one gene × stage grid. Free, in the browser. Led by Vera Janssen at @biohub 🧵
🧵 CryoET reveals molecular organization inside cells, but annotation limits analysis at scale. SABER (Segment Anything Based Electron Recognition) pairs a foundation video model with sparse expert labels to segment 3D cellular structures, thus reducing annotation time by 10x and addressing a major bottleneck in biomedical research.
Access the tool:
chanzuckerberg.github.io/sab…
Read the paper in @Patterns_CP:
cell.com/patterns/fulltext/S…
The Virtual Biology Initiative is a $1.8B commitment to generate the right data to make accurate predictive models of biology, unlocking a greater understanding of disease and opening new paths for cures.
@alexrives joined @labenz on The Cognitive Revolution podcast to discuss this coordinated international effort, and more.
Tune in: cognitiverevolution.ai/a-wor…
biohub retweeted
Biohub & partners just committed $1.8B to AI-ready biological data 🧬
The goal: a "world model of life" – give it a cell's state + inputs, get back a distribution of what happens next.
@alexrives, Head of Science at @biohub thinks that's attainable in ~5 years 👇
Shawn Zheng leads a computational group working to streamline cryo-electron tomography (cryoET), a cutting-edge technology that can image cellular structures in 3D at near-atomic resolution.
His team developed AreTomoLive, which automates cryoET reconstruction and denoising in real time and cuts processing from months to days.
“It feels like we’re on the eve of another revolution,” he says in this @SBGrid profile.
Read more:
sbgrid.org/software/tale/pro…
biohub retweeted
“Together, the organizations are investing $1.8B in funding, data, computation, and new measurement technology, the largest coordinated commitment to generating AI-ready biological data to date. The result will be an open resource for the research community that provides the foundation for greater understanding and ultimately treatment of human diseases.” 👏🏼👏🏼👏🏼
As drug design becomes better / faster / cheaper, knowing which targets to pursue will become even more important. Hope this resource helps our industry increase our collective probability of clinical success by picking the best disease targets
Thank you @biohub @NIH @ENERGY @GoogleDeepMind @IsomorphicLabs @Meta and all the contributing investigators and labs
🧵 Biohub, @ENERGY, @NIH, and new funding partners today announced a major expansion of an international effort to generate and make accessible the data enabling predictive AI models of biology. Together, the organizations are investing $1.8 billion in funding, data, computation and new measurement technology - largest coordinated commitment to map the biology of the cell to power development of digital models of biology. bit.ly/3UhjoxD
biohub retweeted
Why AlphaFold Didn't Solve Protein Folding: The Bitter Lesson of Biology, Virtual Cells, & Curing All Disease latent.space/p/biohub-deepmi…
@GoogleDeepMind’s @pushmeet and @biohub’s @salcandido explain why AlphaFold's breakthrough hasn't solved protein dynamics, why scaling AI compute and data alone isn't enough to understand biology, how protein language models contain hidden knowledge about structure, function, and evolution, why future frontier models may understand other AI systems better than humans, how better biological data and virtual cells could unlock the next wave of breakthroughs, and what it will take to achieve 10–100x faster drug discovery on the path to curing all disease.
Foundational models are evolving, and next-gen AI could unlock new ways of doing scientific discovery.
At AIxBio, @pushmeet and @salcandido discussed data and scaling, dynamics, function and design, interpretability, and the path to the clinic, moderated by Brandon Anderson of @latentspacepod.
Predicting a protein’s structure is only part of understanding it. The next frontier is connecting structure to dynamics and function: how proteins change shape, respond to their environment, and interact within biological systems.