5/ Ezgi Sen, a great PhD student in the Rippe and Stegle labs, led the downstream analysis: Atera clearly resolved 15 main cell clusters on one CRC section. Non-malignant epithelium splits into absorptive and secretory lineages with LGR5+ stem cells at the crypt bases.
Two weeks ago, MAIGE group leader @brianfclarke.bsky.social and Katharina, together with Dominik (Steinmetz lab) joined #Perturb2026, exploring new technologies and datasets. Highlight: Katharina’s first short talk on large-scale perturbation screens in cardioids and great discussions at our poster!
Join us for a postdoc in AI in human genetics at @embl.org in close collaboration with Adrian Cortes (GSK) and @brianfclarke.bsky.social - exploring the effects of rare variants using population-scale cohorts and single-cell readouts.
🔗 Apply now: embl.wd103.myworkdayjobs.com/en-US/EMBL/d...
Do you want to do your #postdoc in #cancerresearch with the most prestigious fellowship in Europe? Apply to DKFZ #MSCA MasterClasses by May 18. 26 Projects in cancer research including #cellandmolecularbiology #genomics #bioinformatics #immunology. @horizoneu.bsky.social
👉Apply here: t1p.de/nx6tu
Welcome @bbbranjan.bsky.social! His career across single-cell genomics, 🐁 biology and software dev has been about bridging wet-lab with reliable analysis workflows. At @embl.org he will lead a collab with @vrmlnroel.bsky.social on modelling GxE interactions to dissect risk factors for human disease.
And the CAMDA Health Privacy Challenge has entered its second edition, ready for your submissions.
🚀 Get involved and find more details: benchmarks.elsa-ai.eu?ch=8&com=int...
Thanks to CAMDA for providing the platform, and to the organizing team and all participants who helped make this effort possible.
📄 Our preprint is now available on bioRxiv - explore it here: doi.org/10.64898/202...
We’re excited to announce the 2nd edition of the Health Privacy Challenge. 🍅🫐
What started last year as a CAMDA community challenge has grown into a benchmarking study of generative models for biologically useful and privacy-preserving synthetic omics, focusing on trade-offs across evaluation axes.👇
We are happy to have contributed to this collaborative effort combining experimental scale with computational integration. Many thanks to the whole team - especially Dewi Moonen, @anniquec.bsky.social, Lars Steinmetz, Daniel Schreivogel, Stefan Schrod, @oliverstegle.bsky.social and many others.
This work provides a genome-scale view of how noncoding variants contribute to immune disease and shows how dispersed risk loci converge on shared transcriptional programs.
📜 Preprint: www.biorxiv.org/content/10.6...
By systematically testing immune disease risk loci in the relevant cell type, we for the first time mapped the entire causal chain from variant → CRE → gene → regulatory network.
How do immune disease-relevant variants rewire gene regulation in CD4+ T cells?
In a collaboration led by Daniel Schraivogel and Lars Steinmetz at @embl.org, we combined two large-scale CRISPRi screens (4.1M cells) to map the downstream cascades of thousands of SNPs.
And huge thanks to everyone who helped to make this possible. 🤝
@elsa-ai.eu @embl.org @ellisheidelberg.bsky.social
It was encouraging to see conversations emerge across disciplines, with a shared interest in advancing trustworthy AI for healthcare and continuing to build this community together 🔥
Thanks to everyone who joined, spoke, presented posters, and brought such great energy!
Discussions started from the human dimension of trust between end users and researchers, and moved toward translating uncertainty and explainability into real accountability, and how to stress-test models and develop meaningful benchmarks and metrics that strengthen trust in practice.
We wrapped up our in-person @elsa-ai.eu TrustworthyAI4Health 2026 Workshop at @embl.org on Monday. The workshop brought together experts and a community of enthusiastic researchers to explore pathways for building trust in AI for healthcare.
A fuller write-up is coming soon!
📢 Keynote reveal: We’re excited to announce Prof. Dr. Oliver Stegle as a keynote speaker at HMC Conference 2026 – Metadata in Action!
📢 3rd Keynote reveal!
We’re excited to welcome Prof. Dr. Oliver Stegle as keynote speaker at #HMCConference2026
Director & Co-Spokesperson of #GHGA, @oliverstegle.bsky.social advances secure, #FAIR infrastructures for data-driven biomedical research. 🧬
steglelab.bsky.social ghga.bsky.social
Working on trans-eQTL mapping with population-scale scRNA-seq? Meet LIVI, our latest framework that enables efficient trans-eQTL mapping at single-cell resolution across cohorts of hundreds to thousands of donors.
Preprint: doi.org/10.64898/202...
Summary in the 🧵 below by @danaivagiaki.bsky.social
We are hiring a Research Software Engineer at @embl.org Heidelberg to enhance and maintain open-source infrastructure for spatial omics.
🔗 Deadline: 15/02/2026
embl.wd103.myworkdayjobs.com/en-US/EMBL/d...
Contact us by email if you meet the eligibility criteria and are interested in applying based on our project outline. Your email should include your project proposal, a CV and two reference letters. We look forward to hearing from you.
Project Outline Nr. 58: drive.google.com/file/d/1YZbs...
Important information:
- A minimum of one first-author publication is required
- Applicants must develop their own research proposal based on our project outline
- Provide your CV, two reference letters and your project proposal
- Expected start date: between August 1, 2026, and January 1, 2027
SCP enables the quantification of thousands of proteins per individual cell, thereby revealing cell states, regulatory mechanisms, and functional heterogeneity. Join us in developing a framework to analyze comprehensive SCP datasets (up to 6000 quantified proteins per cell!).
Together with Jeroen Krijgsveld, we are looking for a postdoc to help us unlock single-cell proteomics (SCP) through the establishment of a scalable computational framework.
For more information, see the @hlsalliance.bsky.social Postdoc Program:
www.health-life-sciences.de/opportunitie...
We're organizing a one-day, in-person workshop on trustworthy healthcare AI, with leading experts discussing key dimensions such as robustness, fairness, privacy, explainability, and clinical validation.
Register now and submit your poster: elsa-ai.eu/elsa-worksho...
📅 March 9, Heidelberg DE
Many thanks to all authors: @dbdimitrov.bsky.social, Stefan Schrod, @mrohbeck.bsky.social, and @oliverstegle.bsky.social.
@embl.org | @dkfz.bsky.social | @uniheidelberg.bsky.social | @sangerinstitute.bsky.social
We further describe how these methods achieve the 3 key aims of causal modelling:
1️⃣ Understand perturbation responses
2️⃣ Extrapolate to unseen conditions
3️⃣ Guide future experiments
We map how these methods commonly utilise partial views of causal signatures (perturbations ⚡, temporal ⏳, spatial 📍& multi-omics 🧬) and rely on a shared core modelling concepts (from Disentanglement to Mechanistic Discovery).
Explore the tools covered in our review in an online repository: interp-extrap-perturb.readthedocs.io 💻
This database is continuously growing, and we invite everyone to submit methods and help us keep the repository up-to-date.
We review and connect >150 methods to help you choose the most suitable method for a given biological task and dataset.
Our perspective further proposes a unifying ontology to structure and organise these methods across tasks, assumptions, and modelling concepts.
