Big thanks to all co-authors, collaborators, and early testers! @danielji.bsky.social @zhuowenzhao.bsky.social @kyleharrington.com @rahelwoldeyes.bsky.social

Done with annotation? Use copick-torch for PyTorch-compatible data loading to train your model!

Docs, tutorials, and code at 👇
🔗 copick.github.io/copick/
🐧 github.com/copick

A figure showing an example prompt and the generated script.

copick includes an MCP-server and batch curation tools: describe your curation pipeline in plain language, let an AI agent compose a curation pipeline for you. No coding required.

A diagram showing the full copick workflow, from obtaining data from the cryoET portal to deposition of new or curated annotations.

copick works across storage backends: local filesystems, HPC, S3/cloud, or directly from the cryoET Data Portal. Plugins for @chimerax.ucsf.edu and @napari.org handle all the details, so multiple users can annotate the same data simultaneously. No overwriting, no headaches, easy import/export.

Excited to share our paper about copick, a dataset API and toolkit for collaborative annotation and analysis of #cryoET data! Whether you're picking particles or curating segmentations, copick reduces friction and brings #OME-Zarr to cryoET without breaking pipelines.
🧵👇

doi.org/10.1002/pro.70578

#UMAP in VR for #3DThursday! In this zebrafish embryogenesis dataset, RNA-seq + ATAC-seq across 94k+ cells reveal changing regulatory programs over time.

🥼 Yang-Joon Kim, Loic Royer @loicaroyer.bsky.social , Matthias Haury
📝 www.biorxiv.org/content/10.1...
🏛️ Biohub @biohub.org
🎬 #syGlass v2.6.0

easymode

New Title Alert: easymode- a collection of general pretrained networks for cellular cryoET.

Learn more here: buff.ly/mLbcUbZ

#SBGrid #SBGridSoftware

Streaming OME-Zarr data directly into syGlass for #3DThursday🌍🔬

With #syGlass v2.6.0, datasets can stream from disk, S3, or HTTPS with no massive downloads required.

Huge thanks to the community building open imaging standards:
@openmicroscopy.org

#OMEZarr #Microscopy #OpenScience #Bioimaging

A screenshot of a dataset page on the cryoET data portal with the metadata panel opened and the View All Info button highlighted.

A paper on this is to follow, a variation was described here: www.nature.com/articles/s41...

An abridged version of the method can also be found directly on the cryoET portal. Click "View All Info" on the page of one of the datasets contained in this deposition.

Four representative slices of cryo-tomograms showing annotated organelles.

New on the CryoET Data Portal: ~27,000 tomograms of affinity-captured lysosomes from HEK293T cells across 4 physiological states. Includes raw data, AreTomo3 reconstructions & Membrain-Seg annotations. Openly available for reuse!

cryoetdataportal.czscience.com/depositions/...

This one was quite the journey! The paper describing the #ChlamyDataset is finally out and on the cover of Mol Cell!

This beautiful rendering made by co-author @jessheebner.bsky.social and Holly Peterson shows an instance of mitochondrial fission found in the dataset 😍

[Maybe long thread ahead]

Tired of knowing what you are looking for in your tomograms?
Why not try bluesky-less Frosina's new self-supervised algorithm for semantic segmentation and particle picking for #teamtomo (includes a new denoiser without the need for odd/even tomos)

Code is on GitHub if you want to try it out

Works with local OME-Zarr files, or as a bonus: stream sub-tilt stacks directly from the cryoET Data Portal: cryoetdataportal.czscience.com

Massive thanks to @danielji.bsky.social who did the heavy lifting implementing this over the summer! 🙌

GitHub - czimaginginstitute/zarr-particle-tools: Particle extraction and reconstruction from OME-Zarr tilt series. Particle extraction and reconstruction from OME-Zarr tilt series. - czimaginginstitute/zarr-particle-tools

Hi #teamtomo! Want to use OME-Zarr for your tilt series but can't integrate it with your existing sub-tomogram averaging pipelines?

Try zarr-particle-tools - RELION-style extraction & reconstruction built for OME-Zarr-based workflows!

📦 pip install zarr-particle-tools
🔗 github.com/czimagingins...

Stunning cover to celebrate the incredible work from @loicaroyer.bsky.social's team.

Learn more about Ultrack, a powerful tool for fast, accurate and scalable cell tracking: biohub.org/life-science... 🧪

Biohub launches initiative combining frontier AI & frontier biology The large-scale scientific effort combines frontier AI with frontier biology to accelerate discovery and solve disease.

EvolutionaryScale will be joining forces with Biohub to create the first large-scale scientific initiative combining frontier AI & frontier biology. Together, we're building world-class compute, AI research & state-of-the-art technologies to cure or prevent disease. bit.ly/4nOUGO4

Glushkova, Böhm, & Beck @maxplanck.de develop a publicly available computational method to measure the thickness of biological membranes in cryo-electron tomograms. Analysis of algae & human cells reveals systematic membrane thickness variations within & across organelles rupress.org/jcb/article/...

Figure 1

Figure 2

Figure 3

Figure 4

Lessons learned from a Kaggle challenge for particle picking in cryo-electron tomography [new]
Kaggle cryo-ET picking: Over-pick tol, aug, & annot key.

🎉 Our paper “MIFA: Metadata, Incentives, Formats and Accessibility guidelines to improve the reuse of AI datasets for bioimage analysis” is out in @natmethods.nature.com!
Community-driven standards to make bioimage data AI-ready & reusable.
👉 www.nature.com/articles/s41... #AI #Bioimaging #FAIRdata

A realistic phantom dataset for benchmarking cryo-ET data annotation - Nature Methods A standardized, realistic phantom dataset consisting of ground-truth annotations for six diverse molecular species is provided as a community resource for cryo-electron-tomography algorithm benchmarki...

Better ML for cryo-ET starts with better benchmarks.

We built a phantom cryo-ET dataset (~500 tomograms) + hosted a Kaggle challenge.

The result: community models beat expert tools.

Read more in the @natmethods.nature.com article that just came out:
🔗 doi.org/10.1038/s415...

GitHub - apeck12/denoiset: An implementation of Noise2Noise for cryoET data An implementation of Noise2Noise for cryoET data. Contribute to apeck12/denoiset development by creating an account on GitHub.

New Title Alert: DenoisET- an implementation of the Noise2Noise algorithm specifically designed for cryoET data denoising.

Learn more here: buff.ly/8i8i87U

#SBGrid #SBGridSoftware #StructualBiology

🦠🧠 MemBrain update! 🧠🦠
We’ve updated our preprint! It now covers the full MemBrain v2 pipeline for end-to-end membrane analysis in #CryoET: segmentation, particle picking, and spatial statistics.
🔗 Preprint: doi.org/10.1101/2024...
🔗 Code: github.com/CellArchLab/...
🧵(1/6) #TeamTomo

Figure 4. MemBrain v2 end-to-end workflow detects periodic phycobilisome organization. A: Raw tomogram slice of EMD-31244. B: Out-of-the-box MemBrain-seg segmentation (light blue). C: A single membrane instance can be visualized in Surforama and manually annotated with GT phycobilisome positions (magenta). D: MemBrain-pick localizes particles (trained with data from C) on all membranes in the tomogram. E: MemBrain-stats computes Ripley’s O statistic using the positions from D with a bin size of 5nm. The distance between peaks (35 nm) was measured to estimate chain unit spacings.

We have updated our #MemBrain v2 preprint with a lot more details about the MemBrain-pick and MemBrain-stats modules, as well as some application examples!

Stay tuned for the upcoming thread by lead author @lorenzlamm.bsky.social! 🧠🧵

#CryoET #TeamTomo
www.biorxiv.org/content/10.1...

Excited to share our preprint on the molecular architecture of heterochromatin in human cells 🧬🔬w/ @jpkreysing.bsky.social, @johannesbetz.bsky.social,
@marinalusic.bsky.social, Turoňová lab, @hummerlab.bsky.social @becklab.bsky.social @mpibp.bsky.social

🔗 Preprint here tinyurl.com/3a74uanv

I'm super happy that our story is now published!
📖 www.science.org/doi/10.1126/...
But what changed compared to the original preprint?

Also, I feel i should post Movie 1 🎥, that inspired the cover. Back when I did the original bluesky thread, movies were not available.

📣 Competition Launch Alert! Locating Bacterial Flagellar Motors 2025 hosted by Brigham Young University.

kaggle.com/competitions/byu-locating-bacterial-flagellar-motors-2025

The kaggle competition may be over, but we remain excited to see solutions from #TeamTomo!

If you're interested in checking whether your algorithm beats the kaggle winners' please don’t hesitate to give it a go and make a submission to the data portal!

A 4-panel image showing tomographic slices and annotations from CZII's object detection challenge.

Now that the competition is closed, we're releasing all 492 runs, including raw data, tomograms and ground truth annotations as a benchmark dataset for the community. Find the datasets here: cryoetdataportal.czscience.com/depositions/...

Results from CZII's #cryoET object detection challenge on kaggle are in!

Participation exceeded our expectations with 931 Teams from 76 countries competing, generating over 27,000 submissions. For an overview check out our post-competition page: cryoetdataportal.czscience.com/competition

Happy to share our manuscript on the in situ visualization of the copia retrotransposon in its final form today published in @cellcellpress.bsky.social www.cell.com/cell/fulltex.... What’s new?