Mann Lab
@mannlab.bsky.social
The Mann Lab is a pioneer in mass spectrometry-based proteomics. Posts represent personal views from lab members and Matthias Mann
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 #Bioinformatics #DeepLearning #MassSpectrometry
October 22, 2025 at 1:36 PM
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 #Bioinformatics #DeepLearning #MassSpectrometry
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
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
September 24, 2025 at 1:39 PM
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
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
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
September 24, 2025 at 1:34 PM
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
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
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...
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...
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...
doi.org
September 24, 2025 at 12:22 PM
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...
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
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
September 24, 2025 at 12:22 PM
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
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
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...
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...
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...
www.biorxiv.org
September 24, 2025 at 12:22 PM
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...
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
We also ship a benchmark dataset of Golgi morphologies and use it to compare image featurization tools: #ConvNeXt, #SubCell, #CellProfiler
September 24, 2025 at 12:22 PM
11/
We also ship a benchmark dataset of Golgi morphologies and use it to compare image featurization tools: #ConvNeXt, #SubCell, #CellProfiler
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
✨ 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
September 24, 2025 at 12:22 PM
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
✨ 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
✨ 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
September 24, 2025 at 12:22 PM
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
✨ 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
✨ 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
September 24, 2025 at 12:22 PM
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
✨ 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
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
September 24, 2025 at 12:22 PM
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
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.
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.
September 24, 2025 at 12:22 PM
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.
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 🖥️
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 🖥️
September 24, 2025 at 12:22 PM
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 🖥️
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 🖥️
4/
Problem: microscopy images are messy, fragmented, and hard to use for ML
Solution: scPortrait standardizes them into a new .h5sc single-cell image format turning 🔬microscopy images into a reusable resource for integrative cell modeling
Problem: microscopy images are messy, fragmented, and hard to use for ML
Solution: scPortrait standardizes them into a new .h5sc single-cell image format turning 🔬microscopy images into a reusable resource for integrative cell modeling
September 24, 2025 at 12:22 PM
4/
Problem: microscopy images are messy, fragmented, and hard to use for ML
Solution: scPortrait standardizes them into a new .h5sc single-cell image format turning 🔬microscopy images into a reusable resource for integrative cell modeling
Problem: microscopy images are messy, fragmented, and hard to use for ML
Solution: scPortrait standardizes them into a new .h5sc single-cell image format turning 🔬microscopy images into a reusable resource for integrative cell modeling
3/
Microscopy images are:
📈 easy to acquire across scales (organism → subcellular)
🖥️ information-rich (cellular architecture, tissue structure, perturbation responses)
= 🚀 ideal fuel for foundation models of cell behavior
Microscopy images are:
📈 easy to acquire across scales (organism → subcellular)
🖥️ information-rich (cellular architecture, tissue structure, perturbation responses)
= 🚀 ideal fuel for foundation models of cell behavior
September 24, 2025 at 12:22 PM
3/
Microscopy images are:
📈 easy to acquire across scales (organism → subcellular)
🖥️ information-rich (cellular architecture, tissue structure, perturbation responses)
= 🚀 ideal fuel for foundation models of cell behavior
Microscopy images are:
📈 easy to acquire across scales (organism → subcellular)
🖥️ information-rich (cellular architecture, tissue structure, perturbation responses)
= 🚀 ideal fuel for foundation models of cell behavior
2/
AI has had major breakthroughs (#alphafold #chatgpt) & computational models can now detect patterns in complex datasets without external guidance 🧠🖥️
🧬 biological datasets often contain entangled information making them complex to interpret →🧠🖥️ + 🧬 = unlock new biology
AI has had major breakthroughs (#alphafold #chatgpt) & computational models can now detect patterns in complex datasets without external guidance 🧠🖥️
🧬 biological datasets often contain entangled information making them complex to interpret →🧠🖥️ + 🧬 = unlock new biology
September 24, 2025 at 12:22 PM
2/
AI has had major breakthroughs (#alphafold #chatgpt) & computational models can now detect patterns in complex datasets without external guidance 🧠🖥️
🧬 biological datasets often contain entangled information making them complex to interpret →🧠🖥️ + 🧬 = unlock new biology
AI has had major breakthroughs (#alphafold #chatgpt) & computational models can now detect patterns in complex datasets without external guidance 🧠🖥️
🧬 biological datasets often contain entangled information making them complex to interpret →🧠🖥️ + 🧬 = unlock new biology
1/
Tomorrow’s large scale cross-modality models will further unlock biology, but they need standardized inputs. With @sophia-maedler.bsky.social & @nik-as.bsky.social, we built #scPortrait, @scverse.bsky.social package to turn microscopy images into single-cell image datasets for multimodal modeling
Tomorrow’s large scale cross-modality models will further unlock biology, but they need standardized inputs. With @sophia-maedler.bsky.social & @nik-as.bsky.social, we built #scPortrait, @scverse.bsky.social package to turn microscopy images into single-cell image datasets for multimodal modeling
September 24, 2025 at 12:22 PM
1/
Tomorrow’s large scale cross-modality models will further unlock biology, but they need standardized inputs. With @sophia-maedler.bsky.social & @nik-as.bsky.social, we built #scPortrait, @scverse.bsky.social package to turn microscopy images into single-cell image datasets for multimodal modeling
Tomorrow’s large scale cross-modality models will further unlock biology, but they need standardized inputs. With @sophia-maedler.bsky.social & @nik-as.bsky.social, we built #scPortrait, @scverse.bsky.social package to turn microscopy images into single-cell image datasets for multimodal modeling
September 10, 2025 at 3:16 PM