Mann Lab

Discovery proteomics → clinical diagnostics? ADAPT-MS makes it possible by adapting to each sample's protein coverage on-the-fly. No imputation, no fixed panels, multiple diagnostic questions from one measurement. Out now in @natcomms.nature.com
www.nature.com/articles/s41...

Congratulations Dr. Vu!

In remembrance of Peer Bork  | EMBL EMBL and its community are deeply saddened by the death of Peer Bork, the organisation’s Interim Director General.

Very sad to hear about Peer Bork's untimely passing. I have known him since my time at EMBL in the 1990s - a brilliant mind and wonderful collaborator. His contributions to bioinformatics were transformative. The community has lost a giant.
www.embl.org/news/embl-an...

a red background with the words happy holidays and a christmas tree on it ALT: a red background with the words happy holidays and a christmas tree on it

Happy Holiday Season!

Celebrating an incredible 2025 @mannlab.bsky.social!
PhD graduations, prestigious awards including Matthias Mann's
@nationalacademies.org election, launch of innovative methods and bioinformatics tools, and celebrating 20 years
@mpibiochem.bsky.social
mannlabs.github.io/research-hig...

Congrats to all the well-deserving winners of this year's #JSPAward and a shout-out to Thierry Nordmann @tnordmann.bsky.social for his outstanding research on #TEN published in @nature.com
www.nature.com/articles/s41...

We can know more than we can tell, but now Al can capture it. Our multimodal agent analyzes lab videos to generate protocols & catch common lab errors. Preserving tacit knowledge in proteomics. Now in @molsystbiol.org @patiskowronek.bsky.social #Google
link.springer.com/10.1038/s443...

We congratulate @georgwa.bsky.social from the @mannlab.bsky.social on his sucessful #PhD defense!

Honored to be included in this reflection on 20 excellent years of MolSystBiol! From our first phosphotyrosine interactome in 2005 to proteome-wide networks - scale has grown exponentially. MS proteomics + AI is transforming systems-level understanding. Exciting times!
embopress.org/doi/pdf/10.1...

Elegant chemistry and cryo-EM explain the selective Lys8 ubiquitylation. A clear example of how metabolite sensing and the ubiquitin system intersect.

Great to see Brenda Schulman’s team post their new bioRxiv paper. MS-based proteomics helped uncover a beautifully simple mechanism: cysteine levels tune metabolism by flipping the inverse stability of LRRC58 and its substrate CDO1.
www.biorxiv.org/content/bior...

Fantastic time at #HUPO2025 in Toronto! Congratulations to our talented Early Career researchers! Proud winners: Oeller Marc Oeller🥇 1st Place Poster, Kathrin Korff 🥈 Runner-up Poster, and @carolineweiss.bsky.social 🎤 3-Minute Thesis award! The future of #proteomics research is bright!

If you are at HUPO 2025, catch up with our team @mannlab.bsky.social to hear about our exciting work in single-cell proteomics, immunopeptidomics, spatial proteomics, and the latest advances in mass spectrometry technology!

The diagnostic potential of proteomics and machine learning in Lyme neuroborreliosis - Nature Communications Researchers incorporate mass spectrometry-based proteomics and machine learning to assess the potential of a less invasive diagnostic approach for Lyme neuroborreliosis, a common nervous system infect...

In Nature Communications: MS-based proteomics + machine learning to diagnose Lyme neuroborreliosis with 92% accuracy (CSF) and 80% (blood) - paving the way for earlier, less invasive testing.
www.nature.com/articles/s41...

Awesome talk by @erictopol.bsky.social at the Bavarian Academy of Sciences and Humanities (BAdW). A positive and hopeful take on #AI in medicine. Thanks for highlighting #DVP, too 😊

AlphaDIA is open-source and free for academic and commercial users. We are currently preparing version 2.0 - stay tuned!
#AlphaDIA #Bioinformatics #DeepLearning #MassSpectrometry

AlphaDIA enables DIA transfer learning for feature-free proteomics - Nature Biotechnology An open-source platform for data-independent acquisition proteomics adapts predicted libraries to experimental settings.

Excited to share AlphaDIA's publication in Nature Biotechnology!
Our open-source DIA framework brings deep learning directly to raw MS data with feature-free processing, transfer learning for any PTM, and performance matching top tools.
www.nature.com/articles/s41...

How much hands-on lab expertise gets lost? We developed a multimodal AI agent that turns videos into protocols and detects procedural errors. Making science accessible. Great collaboration with #Google.
Preprint: www.biorxiv.org/content/10.1...
@patiskowronek.bsky.social explains👇

16/
If this interests you:
🔁 Retweet the first post:
bsky.app/profile/mann...
⭐️ Give our github repo a star github.com/MannLabs/scP...
❓ tell us what you are going to do with #scPortrait

15/
This work was a fantastic collaboration:
@mannlab.bsky.social
@fabiantheis.bsky.social
@v-hornung.bsky.social
A big shoutout to all of our co-authors: Alessandro Palma,
Altana Namsaraeva, Ali Oğuz Can, Varvara Varlamova, Mahima Arunkumar, @lukasheumos.bsky.social, @georgwa.bsky.social

scPortrait integrates single-cell images into multimodal modeling Machine learning increasingly uncovers rules of biology directly from data, enabled by large, standardized datasets. Microscopy images provide rich information on cellular architecture and are accessi...

14/
Bottom line:
scPortrait turns microscopy images into a first-class modality for machine learning and multimodal foundation models, alongside RNA and proteomics 🔬💻
🔗 Preprint: doi.org/10.1101/2025...
🔗 GitHub: github.com/MannLabs/scP...

13/
Our extensive documentation and tutorials make scPortrait easy to use and accessible.
And as part of @scverse.bsky.social it’s compatible with existing tools like scanpy, squidpy, bento-tools or Moscot 🚀🐍
mannlabs.github.io/scPortrait/i... #OpenSourceTools #Tutorial #CodeDocumentation

SPARCS, a platform for genome-scale CRISPR screening for spatial cellular phenotypes Forward genetic screening associates phenotypes with genotypes by randomly inducing mutations and then identifying those that result in phenotypic changes of interest. Here we present spatially resolv...

12/
scPortrait already scales:
1️⃣ 120M+ single-cell images from image-based genetic screens
2️⃣ applied on patient-derived datasets to perform AI-driven morphology analysis
Refs:
1️⃣ BioRxiv2023 ➡️ www.biorxiv.org/content/10.1...
2️⃣ Nature 2025➡️ www.nature.com/articles/s41...

11/
We also ship a benchmark dataset of Golgi morphologies and use it to compare image featurization tools: #ConvNeXt, #SubCell, #CellProfiler

10/
✨ Embedding images into transcriptome atlases ✨
We use scPortrait to embed single-cell images from a @10xgenomics.bsky.social Xenium ovarian cancer dataset into the #SCimilarity transcriptome atlas (R2 = 0.65), recovering meaningful cell types

9/
✨ Morphology defined cell states ✨
Image embeddings generated with scPortrait resolve intra- vs extratumoral macrophages with distinct morphologies, linked to anti-inflammatory vs fibroblast-like programs

8/
✨ Transcriptomes from images ✨
Using optimal transport + flow matching, scPortrait generates gene expression directly from CODEX images, capturing canonical marker expression like TCL1A in germinal centers in the tonsil
#CODEX #flowmatching #OT

7/
With standardized single-cell image datasets in place, the key question is: what new biology can we unlock?
We highlight three use-cases for scPortrait

6/
The new .h5sc format provides fast random access to single-cell images for ML training.
It follows #FAIR data principles (findable, accessible, interoperable, reusable) and integrates with @scverse.bsky.social tools via AnnData.

5/
The scPortrait pipeline transforms raw input images step by step:
• stitch FOVs
• segment & extract cells
• output standardized .h5sc single-cell image datasets

From messy pixels → inputs ready for training 🖥️