Biohub

Biohub

@biohub.org

Combining frontier AI & frontier biology to help scientists cure or prevent disease

3.1K followers 110 following90 posts

ZFIN’s imaging data is immensely valuable to #zebrafish 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/gene2fish

The work was led by Biohub’s @verajanssen.bsky.social and @loicaroyer.bsky.social, in collaboration with the software engineering team. Read the preprint here: www.biorxiv.org/content/10.6...

biorxiv.orggene2fish: a multi-gene spatial expression browser for in situ hybridization data across zebrafish developmentSingle-cell RNA-sequencing (scRNA-seq) atlases of zebrafish development have catalogued marker genes for hundreds of cell types, but transcriptomic data alone cannot reveal their expression patterns i...

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 labels to segment 3D cellular structures, cutting annotation time by 10x. chanzuckerberg.github.io/saber/

Read the paper in Cell Patterns, led by @kimanius.bsky.social, and learn how it addresses a major bottleneck in biomedical research: www.cell.com/patterns/ful...

cell.comLarge-scale CryoET segmentation from minimal expert annotations with foundation video modelsAnnotating 3D microscopy data is a major bottleneck in biomedical research. Schwartz et al. introduce SABER, a workflow that turns sparse expert notes on 2D slices into full 3D semantic segmentations....

Shawn Zheng leads Biohub’s work to streamline cryo-electron tomography, which images cells in 3D at near-atomic resolution. His team’s AreTomoLive automates cryoET reconstruction and denoising in real time and cuts processing from months to days. Learn more: sbgrid.org/software/tal...

Biohub reposted

Constantin Ahlmann-Eltze

@const-ae.bsky.social

I’m super excited that Isomorphic labs is part of the virtual biology initiative that will invest a total of 1.8$ billion (!!) to generate the data needed to train the next generation of machine leaving models for biology 🔬🧬💻

Biohub@biohub.org

Last week, Biohub, the U.S. Department of Energy, the National Institutes of Health, and new funding partners announced a major expansion of an international effort to generate and make accessible the data enabling predictive AI models of biology. bit.ly/3UhjoxD

bit.lyOpen data for predictive AI models of biology: $1.8 billion committedBiohub, DOE, NIH, and partners will generate open, standardized data to train AI models that predict how cells respond to interventions.

Last week, Biohub, the U.S. Department of Energy, the National Institutes of Health, and new funding partners announced a major expansion of an international effort to generate and make accessible the data enabling predictive AI models of biology. bit.ly/3UhjoxD

bit.lyOpen data for predictive AI models of biology: $1.8 billion committedBiohub, DOE, NIH, and partners will generate open, standardized data to train AI models that predict how cells respond to interventions.

View full thread

DOE will invest more than $500 million over five years in lab measurement, modeling and computation. NIH will coordinate the contribution of relevant datasets, repositories, and knowledge bases developed through more than $500 million in prior federal investment.

In addition, Google DeepMind, Isomorphic Labs, and Meta are collectively investing $300 million in the Virtual Biology Initiative to create the technologies and multi-modal datasets needed to build predictive models of life.

Last week, Biohub, the U.S. Department of Energy, the National Institutes of Health, and new funding partners announced a major expansion of an international effort to generate and make accessible the data enabling predictive AI models of biology. bit.ly/3UhjoxD

bit.lyOpen data for predictive AI models of biology: $1.8 billion committedBiohub, DOE, NIH, and partners will generate open, standardized data to train AI models that predict how cells respond to interventions.

Together, the organizations are investing $1.8 billion 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.

Biohub reposted

@uermel.bsky.social

New on the #cryoET Data Portal: three datasets benchmarking the crossed laser phase plate (xLPP) ⚡ Data by Reza Paraan and the @biohub.org team with collaborators at UC Berkeley. Full credits at the end of the 🧵 cryoetdataportal.czscience.com/depositions/10363

Biohub Investigator @jameszou.bsky.social built a “virtual biotech”—37K AI agents that analyzed 57K clinical trials and found clues to what makes drugs succeed. The New York Times traces its roots to a collaboration sparked by our scientist John Pak: bit.ly/4hLrpnm

Biohub reposted

Loïc A. Royer 💻🔬🧪

@loicaroyer.bsky.social

What if any dataset, any size, any number of dimensions, opened in a browser tab from a link? 🔬🧪💻 Luxar is out today: write it in Python, share it as a link, explore it in any browser. Open source. 🧵 @biohub.org Preprint: doi.org/10.5281/zen... Code: github.com/royerlab/luxar

Three webinars on the science behind ESMC interpretability, ESMFold2, and binder design are now on YouTube. Hear from the researchers who built the models, and view live demos and tutorials that you can apply to your own research. Watch the playlist: https://bit.ly/4h1IObl

bit.lyBiohub Webinars: ESM Protein AI Models, Structure Prediction & Design - YouTubeExplore the ESM family of protein AI models with Biohub → https://biohub.ai/ This playlist collects Biohub's technical webinars on the ESM family of open pro...

Biohub reposted

Loïc A. Royer 💻🔬🧪

@loicaroyer.bsky.social

AI has become such a powerful tool, enabling entire areas of science and biology. I got to talk about this as part of a new PBS special on how AI and biotech are transforming medicine. Overall, the documentary is quite interesting! It's worth a watch! www.pbs.org/video/data-c...

pbs.orgData Cure: Stories of Healing and Hope | Episode 1A race to stop superbugs before medicine enters an era where antibiotics no longer work.

Biohub scientist Dr. Angela Oliveira Pisco worked with a team of genomics and software experts to more than double the number of cells in the Tabula Sapiens dataset, and increase the number of cell types represented, to 700.

Cell - a Cell Press journal@cp-cell.bsky.social

Now online! Tabula Sapiens 2.0: A comprehensive transcriptomic atlas of human cell types

dlvr.itTabula Sapiens 2.0: A comprehensive transcriptomic atlas of human cell typesAn expanded atlas of human cells maps the expression of nearly all human transcription factors across 701 cell-type/tissue combinations and uncovers the heterogeneous nature of senescence across tissues and cell types.

🔬New in @cellpress.bsky.social from Biohub Investigator Bo Huang: AI-designed proteins built from scratch (dubbed "Rhobins") work in living cells and now match or even surpass fluorescent tags refined over decades of natural protein engineering. Read the paper: www.cell.com/cell/fulltex...

Organoids can mimic human organs, but we study them mostly with static, endpoint assays. A new perspective argues for instrumented tissues, which are engineered human tissue woven with electrical, chemical, mechanical, and optical sensor arrays, to move faster from disease mechanism to therapy.

Read the piece, co-authored by Csaba Forro, Suji Choi, and others from Biohub, in @cp-cellbiomat.bsky.social: doi.org/10.1016/j.ce...

doi.orgRedirecting

This is the largest dataset contributed to the Billion Cells Project to date. It is an important step toward the scale and coordination needed to build the Virtual Biology Initiative.

Gladstone Institutes@gladstoneinst.bsky.social

An unprecedented map of 22 million immune cells provides the functional rulebook for virtual biology and next-generation immunotherapies. @ucsanfrancisco.bsky.social @marsonlab.bsky.social @stanford.edu @biohub.org @cp-cell.bsky.social @10xgenomics.bsky.social

gladstone.orgScientists Decode Gene Circuits of Human Immunity, Fueling New Era of Disease ResearchAn unprecedented map of 22 million immune cells provides the functional rulebook for virtual biology and next-generation immunotherapies.

Tumors aren't one disease—they're patchworks of coexisting cancer cell states. Two new Nature Genetics papers show these are AI-mappable (ARACNe+VIPER) and nearly identical across patients. We found drug combos hitting multiple states at once, doubling survival in a rare brain tumor. bit.ly/45TNIkb

bit.lyWhy one drug isn't enough: AI reveals cancer's secret weaponTumors are patchworks of distinct malignant cell states, each with its own survival strategy. With AI, Biohub scientists are discovering drug combinations that defeat them all at once.

Proteins are the machinery of life. Scientists have cataloged billions of protein sequences—but their biology is still mostly unknown. ESM Atlas is a new way in. https://bit.ly/4dJcF6G

Biohub reposted

@tattabio.bsky.social

SeqHub protein searches now show which regions of a predicted structure are tied to specific biological functions or evolutionary patterns, powered by @biohub.org's SAE features. 🧵

🎉 Congrats to Scott Fraser, Biohub’s President of Dynamic Imaging, on receiving the Microscopy Society of America’s Distinguished Scientist Award - Biological Sciences, in recognition of his transformative contributions in bioimaging and molecular analysis!

Biohub reposted

@jshendure.bsky.social

Replying to @jshendure.bsky.social

Preprint: biorxiv.org/content/10.6... This is the product of many years of effort, and particularly amazing summer teamwork by Qi Yu, Haedong Kim, @seidels.bsky.social @cxqiu.bsky.social and many others from Shendure Lab & SeaHub (a @alleninstitute.org @biohub.org @uwmedicine.bsky.social colab) 2/n

biorxiv.orgIn vivo reconstruction of the cell lineage history of a developing mouse with DNA Typewriter, from zygote to late organogenesisThe complete cell lineage of C. elegans, mapped over four decades ago, was tractable because the animal is small, transparent, and lineage invariant. Most animals are none of these, having orders of m...

How hard was it to get the laser phase plate to work? Like “trying to hit a bullet with a bullet” hard, says Biohub scientist David Agard. Check out the latest episode of the Uncovered podcast from @cenmag.bsky.social.

C&EN (Chemical & Engineering News)@cenmag.bsky.social

Learn about a device that boosts the contrast in electron microscopy using the most intense laser beam yet made, and the story of its path from an idea to an invention that could open new lines of inquiry in structural biology and biochemistry. #chemsky 🧪

cen.acs.orgUncovered: Record-setting laser boosts cryo-electron microscopyPhase-contrast device could allow in-cell proteomics

Can science give you superpowers? Don’t miss Biohub Investigators @profhernandez.bsky.social and KC Huang at San Diego Comic-Con tonight! #SDCC

Christopher J Hernandez@profhernandez.bsky.social

I’m excited about my upcoming talk at San Diego Comic Con in a panel entitled “How science may give you superpowers” with KC Huang and Tom Galloway. #SDCC #Makeitcannon #Originstory www.comic-con.org/cc/things-to...

comic-con.orgProgramming Schedule - Comic-ConCheck out the programming schedule for Comic-Con in San Diego!

🚨 New preprint: We reprocessed 77k zebrafish RNA-seq datasets from across the global research community and found hidden viruses in 10% of them—including a fish virus whose proteins are 91% identical to human Influenza B—then built a generalizable pipeline. bit.ly/4vxFF72

Introducing RNAquarium 🐟: a discovery engine built to scale across any species with public data, extracting multi-scale and multi-organism biology from sequences that already exist and underscoring the value of bulk RNA-seq data for modeling 🧬 bit.ly/4wqLRPD

bit.lyRNAquarium PortalExplore, visualize, and interact with RNAquarium project data.

7.5M people in the U.S. live with psoriasis. Many still struggle despite treatment. Our team combined AI + CRISPR to find overlooked biological pathways — then tested an existing asthma drug. It worked. This is frontier AI and frontier biology working together. 🧬 Learn more: bit.ly/44QSiyW

bit.lyA Genome-Wide CRISPR Screen Reveals New Psoriasis TargetsBiohub researchers combined a genome-wide CRISPR screen with AI to uncover an unexpected psoriasis drug target: the receptor for oxytocin.

Excited to be at #ICML 2026! We’ll share three papers on virtual cell modeling, single-cell data generation, and immune recognition. → Virtual Cells Need Context, Not Just Scale icml.cc/virtual/2026... → scLDM icml.cc/virtual/2026... → DecoderTCR icml.cc/virtual/2026...

🎉Congrats to the Remoscope team, 1st place winner of the Medical Device category of the Nebius AI Discovery Awards! Developed at Biohub, Remoscope is an AI-powered diagnostic platform that can rapidly image and analyze blood and other biological fluids to diagnose disease.

Learn more about Remoscope in this 2024 article: biohub.org/blog/ai-powe...

biohub.orgAI powers a better way to diagnose malaria - BiohubCZ Biohub San Francisco’s Remoscope shows promise in clinical study.

The deadline for this funding opportunity is fast approaching: applications are due July 16!

Biohub@biohub.org

Better vitrification = better cryo-ET. We've opened an RFA for two-year grants to advance vitrification techniques for biological samples—welcoming applications from researchers in heat transfer, cryogenics, and materials science. Apply here: bit.ly/4uip6eJ

How do you track thousands of cells in complex systems, in 3D & over time? We are launching a Kaggle challenge to push the frontier of AI-powered cell tracking, building on advances like Ultrack & the DaXi lightsheet microscope from the @loicaroyer.bsky.social lab. bit.ly/4bfdzWE

Biohub reposted

Loïc A. Royer 💻🔬🧪

@loicaroyer.bsky.social

1/ 🧬🔬 How does one cell become an entire animal? We can now film it in 3D 🐟, but turning those movies into accurate cell tracks is brutally hard. Enter our new #Kaggle challenge: $60K prize pool, to help crack 3D+time cell tracking at scale! 🧵👇 www.kaggle.com/competitions... @biohub.org

What happens when physicists turn their tools on living systems? That's the question at the heart of the Physics of Life Symposium—and this year marks its fifth edition. Submit an abstract or register to attend by Aug. 14: bit.ly/4xJIPH7

The 2nd Chicago BioEngineering Conference (CBEC) is back — Sept. 17–18 in downtown Chicago! 🧬 Omics. Bioelectronics. AI × bio. A stellar speaker lineup. Two days of science that matters. Space is limited → register now: bit.ly/4vqlXLj

Cryo-ET gives unparalleled views of cellular structures, but targeted sample preparation by cryo-FIB milling has been a constraint. No longer! A new workflow yields close to 100% targeting success for centrioles & other small structures. Read the preprint ⤵️ www.biorxiv.org/content/10.6...

Biohub reposted

C&EN (Chemical & Engineering News)

@cenmag.bsky.social

Phase contrast enables structures of smaller proteins, could drive in-cell proteomics: cen.acs.org/analytical-c... #chemsky 🧪

“The laser phase plate has the highest continuous light intensity in the solar system, outside of the inner regions of the Sun,” says physicist Holger Müller. Learn more about this landmark achievement in microscopy in @cenmag.bsky.social: cen.acs.org/analytical-c...

cen.acs.org15 years, flawless mirrors, and the brightest laser in the world: How physicists brought phase contrast to cryo-EMPhase contrast enables structures of smaller proteins, could drive in-cell proteomics

Check out Alex Rives on Latent Space discussing the science behind ESMFold2, ESMC, and ESM Atlas, and what a world model of protein biology actually makes possible. 🎙️ https://bit.ly/4vwqLy3

The laser phase plate is leveling up cryoEM so we can see biology like never before. Learn more: bit.ly/4vK9LVn

Biohub reposted

@skyjase.bsky.social

met a grad student in Holger’s lab and asked him about his research. my jaw was on the floor about halfway through and he just kept piling on to the bonkers scope and challenge. This project is so insanely cool and ambitious.

Biohub@biohub.org

In a series of 3 papers and preprints, we’re thrilled to share with you the working laser phase plate. In collaboration with research led by Holger Müller at UC Berkeley, this is a huge innovation in imaging to make small and faint objects inside cells visible. bit.ly/4vK9LVn

In a series of 3 papers and preprints, we’re thrilled to share with you the working laser phase plate. In collaboration with research led by Holger Müller at UC Berkeley, this is a huge innovation in imaging to make small and faint objects inside cells visible. bit.ly/4vK9LVn

Dubbed a “Swiss army knife” for biological imaging, MOSAIC’s development was led by Biohub Investigator @supgokul.bsky.social of @ucberkeleyofficial.bsky.social . Learn more about the microscope, and the “AI mind” he’s creating to make sense of all the data ⤵️

news.berkeley.eduHi-res microscopes give biologists petabytes of data. Scientists are creating an AI assistant to make sense of it. - Berkeley NewsAt UC Berkeley, high-resolution microscopes are generating images of cells and embryos day and night, collecting massive amounts of data to train an AI model for living biological systems.

Nature Methods@natmethods.nature.com

MOSAIC: a versatile multimodal microscopy that seamlessly transitions between light-sheet, 2-photon, label-free and super-resolution modalities. @supgokul.bsky.social www.nature.com/articles/s41...

nature.comA multimodal adaptive optical microscope for in vivo imaging from molecules to organisms - Nature MethodsMOSAIC is a versatile multimodal microscope that allows seamless transition between various light-sheet, two-photon, label-free and super-resolution modalities. It is demonstrated in various systems r...

Better vitrification = better cryo-ET. We've opened an RFA for two-year grants to advance vitrification techniques for biological samples—welcoming applications from researchers in heat transfer, cryogenics, and materials science. Apply here: bit.ly/4uip6eJ

Biohub reposted

@manu-leo.bsky.social

📣 new preprint multimodal atlas. Imaging + scRNA, 57M cells. 🧬🔬 Cells are complex dynamical systems — but most ways we measure them destroy them. We asked: how does live imaging compare to scRNA-seq, the field’s gold std? The answer surprised us 🧵 www.biorxiv.org/content/10.6...

👀🔥 @nature.com on why releasing ESMFold2, ESMC, and ESM Atlas openly changes what's possible for the research community

Nature@nature.com

A newly released AI tool has generated an atlas of more than one billion predicted protein structures and billions more protein sequences. go.nature.com/4wS3Pen

go.nature.comMove over, AlphaFold: open source model predicts shape of 1 billion proteinsThe new open-source atlas, generated by an AI tool called ESMFold2, vastly increases the known protein universe.