Had recently the fastest ever turnover in a manuscript we submitted. All 3 reviewer reports were for sure heavily AI-generated. All have multiple factual errors (e. g. the authors don't specify X -> it's in the manuscript), very generic and asked for 'validation' in a super broad manner 🙄

Just saw Ubuntu 25.04 isn't covered 😥
Understandable to only support LTS releases

🙏 this is great! Many years ago I had a bunch of apt-get installable r-cran- packages that I tried to make play nice with the rest from bio conductor, great that there's a dedicated repo with ready-to-use debs now that included bioc

Thank you! ♥️

Aging reshapes local tissue interactions but most organs retain a resilient overall architecture.
Our atlas provides a new framework to study tissue structure, resilience, and aging.
🔬 Learn more: rendeiro.group

Among 100+ biological and clinical factors, age was the main driver of microanatomical changes.
About 27% of domains shift in abundance over adulthood, yet others remain remarkably stable.

We developed H&E-UTAG, an unsupervised method that detects recurrent microanatomical domains directly from H&E slides.
It identified 218 domains across 40 tissues and 29 organs — revealing how cellular organization underlies organ function.

I agree that they are more similar than most make it to be, and often conceptually the same just operating on different units (pixels vs cells).
In our UTAG work e.g a key step is the joint unsup. clustering across images weighting more recurrent struct. across images, less image-specific artifacts

, a leading #publichealth concern. @sciimmunology.bsky.social
www.science.org/doi/10.1126/...