John Bogovic

Graph neural networks recover interpretable circuit models from neural activity
saalfeldlab.github.io/flyvis-gnn/
@HHMIJanelia
@AllenInstitute
to be presented at @CosyneMeeting

Calling all OME Community members, contributors, and lurkers! Please take a minute to answer the OME 2026 Community Survey (docs.google.com/forms/d/e/1F...) to help us prepare for discussing the future of OME governance during the OME 2026 Community Meeting at the end of April.

We opened 4 PhD positions at @humantechnopole.bsky.social together with Polimi (see 👇).

My lab offers a PhD project on "Computational Morphogenesis": use state-of-the-art machine learning methods together with realistic simulations to build digital twins of developing tissue.

Apply by March 26.

Lack of reviewers threatens robustness of neuroscience literature Simple math suggests that small groups of scientists can significantly bias peer review.

Broad peer review is crucial for a healthy scientific literature, but neuroscientists turn down review requests too often. Simple math suggests that small groups of scientists can significantly bias the literature, writes @jvoigts.bsky.social.

#neuroskyence

www.thetransmitter.org/publishing/l...

learningfromscratch march 16th, workshop day 1 @ cosyne 2026

Excited to be co-organising a #cosyne2026 workshop with Alison Comrie on 'algorithms for learning from scratch'! With a great line-up of speakers, we'll be tackling the question of what processes enable naive biological & artificial agents to adapt to new situations. Info here: tinyurl.com/4u8enf7k

Yep, it lags a little. But still quite impressive considering that each "pixel" is a 150 MPixel multires image. Overall the total number of 'adressable' pixels in this 2D image are 225x225x150 Mpix = 7.6 Tera pixels. #ImageJ #Fiji #BigDataViewer

Are you working at the interface of #neuroscience and ML/AI? Share your research at Janelia in Nov! 🧠 🤖 janelia.news/NAI26

Graph neural networks uncover structure and functions underlying the activity of simulated neural assemblies Graph neural networks trained to predict observable dynamics can be used to decompose the temporal activity of complex heterogeneous systems into simple, interpretable representations. Here we apply t...

This is work that we presented at last year's #Cosyne workshop on #GNN s sites.google.com/bu.edu/gnnwo.... Better late than never. You can reproduce everything with the associated notebooks. I think it's a good start to learn how to use GNNs to infer something about NNs. arxiv.org/abs/2602.13325

I am joining @joss-openjournals.bsky.social as an editor! I wrote a little bit about how much I hate academic publishing and why this felt like the right thing to dedicate some volunteer time to. pages.erickratamero.com/personal-web...

beautiful protist ExM images here from @dudinlab.bsky.social and @gautamdey.bsky.social

Research Software Engineer - Spatial-omics The Oliver Stegle team and the Bioimage Analysis Support Team at EMBL Heidelberg are looking for a research software engineer to contribute to the development and maintenance of highly-used open-sourc...

We are hiring a Research Software Engineer to enhance and maintain the SpatialData framework.

We will work together with colleagues at EMBL, DKFZ, GSK, and across the scverse community.

🔗 Deadline: 15/02/2026
embl.wd103.myworkdayjobs.com/en-US/EMBL/d...

Introduction to the Cell Painting Assay - AskErinDearBeth Episode 23 YouTube video by Ask Erin, Dear Beth

@bethcimini.bsky.social and I talk about the Cell Painting Assay a LOT, especially on our podcast Ask Erin/Dear Beth, so we figured it was about time to talk about what it actually is! youtu.be/9-d6DoHtgUY
h/t @drannecarpenter.bsky.social
#science #biology #cellpainting #microscopy #imageanalysis

The OME meeting has followed me from New England to Germany! consistently my favorite scientific meeting of the year, with the nicest community anywhere. Join us there!

This course looks great. Amazing organisms + friendly and highly expert instructors, several of whom I have had the privilege of getting to know during my time so far here in Heidelberg. Strong recommend!

Scientific Workflows Developer The European Molecular Biology Laboratory (EMBL) is Europe’s life sciences laboratory – an intergovernmental organisation with more than 110 independent research groups and service teams covering the ...

Big news! The Data Management team/Data Science Centre at EMBL (i.e. my team) is hiring for a position based in Heidelberg (but working across all sites!). We are looking for a scientific workflows developer that will focus on multimodal pipelines. embl.wd103.myworkdayjobs.com/en-US/EMBL/j...

Deep Learning for Microscopy Image Analysis Topics The following will be covered extensively during lectures, exercises, and project work: Image denoising and restoration (fully supervised and self-supervised) Image translation (e.g.,

🚨 Alarm!!! 🚨
AI/ML course for microscopy image analysis!!! 🧐
In 2026 at Janelia (@hhmijanelia.bsky.social), no tuition, housing and meals provided! Isn’t that borderline unbelievable?!?

20 students, ~14 TAs and lecturers
🗓️ June 4-18 2026

✍️ Jan 15 2026 ✍️

🔁 pls!!

www.janelia.org/you-janelia/...

On the left, the image shows a schematic of a fly head, ring neurons and EPG neurons together with some calcium imaging frames. On the right is a photo of a fly on a ball in virtual reality and another schematic of a VR system.

📢 Join us, the Haberkern lab, @uni-wuerzburg.de for a postdoc studying neural circuit mechanisms of navigation. You’ll spearheading neurophysiology experiments on our brand new 2P!

⏳ Apply by 28th February 2026

Details: www.haberkernlab.de/docs/ENPostd...

#neuroscience #academicjobs #postdoc

In addition to the Comments linked to below, this news article ("Technology Feature") ties together perspectives from many additional workers in the field, including yours truly: www.nature.com/articles/s41...

Toward better 2D - 3D registration in BigWarp BigWarp has so far focused on 2D-2D or 3D-3D registration, and not on 2D-3D registration. After a conversation with @Christian_Tischer , we thought some simple UI changes could make the support for 2...

Do you have 2D-to-3D image registration challenges? Let me know about them here
forum.image.sc/t/toward-bet...

🤘🤖🤘

Hackathon participants group picture

We had an amazing OME-NGFF hackathon last week, Wednesday to Friday, following the inspiring OME-NGFF symposium. What a fantastic week! It’s hard to describe how motivating it is to be surrounded by so many talented, curious & friendly people who care deeply about open, FAIR bioimaging data.
1/14 🧵

We wrapped up the symposium part of our 2025 International OME-NGFF workshop on Tuesday. What a phenomenal two days! I'm humbled by the incredible positive energy, talent, and enthusiasm that filled this event.

Hackathon group picture in the lecture room

Hackathon group picture in the restaurant (where not everyone fit on the picture unfortunately)

Josh Moore giving a talk at the hackathon

Matt McCormick asking a question

This week, we're hosting the #OME_NGFF_Hackathon 2024 at the BioVisionCenter at the University of Zurich. We're bringing together 40 motivated researchers and specialist for #OMEZarr workflows, #OMEZarr Python tools & #OMEZarr Java tools. (1/3)

This is what hacking *actually* looks like. #OME_NGFF_Hackathon

🌭?

An Introduction to OME-Zarr for Big Bioimaging Data

Today I'm announcing a new digital textbook 📖🖥️, "An Introduction to OME-Zarr for Big Bioimaging Data".

ome-zarr-book.readthedocs.io

Do you want to learn more about 25+ bioimage analysis tools, 100% for free? Sign up to attend Halfway to I2K today! tinyurl.com/Halfway-to-I...
Wishing you'd submitted a tool or workshop but missed the deadline? GREAT NEWS, we just extended one week! Submit by Friday - airtable.com/app2zpB8d82r...

Registration for Halfway to I2K is now OPEN! And workshop submission is still too (for a bit)! Join 20+ presenters and counting, and hundreds of attendees in learning about image analysis, whether you're a complete beginner or a total pro.

Over at `ome-zarr-models` we're working on implementing support for the new coordinate systems too - here's an example of a transformation graph from multiple tiled images that can be transformed into a common "world" coordinate system. 🟥 ➡️ 🌎

The #OME-Zarr RFC-5 proposal allows you to transform multiple images into the same coordinateSystem. Trying this out with napari at github.com/will-moore/n.... Cool that you can offset T-dimension as well as XYZ!