Adlung Lab

Adlung Lab

@adlunglab.com

Data-driven & knowledge-based. Rooted in Hamburg, Germany. Systems biology / Immunology. Fat + Gut <3 Account handled by π @LorenzAdlung.com www.adlunglab.com

993 followers359 following877 posts

Very smart approach, I'll definitely try it out (and encouraged my group and others to do so, too!) Congrats, Florian!

Florian Ingelfinger@florianingelfinger.bsky.social

1/3 Deep generative modeling coming to cytometry! 🔬 CytoVI is now in Nature Methods: integrate flow cytometry, mass cytometry & CITE-seq, predict missing markers, and discover disease-associated cell states. www.nature.com/articles/s41...

I was very humbled to meet Ada several times during my life in Israel and also afterwards at #LINO23. She was a true idol in terms of dedication and thoroughness for me as a non-academic lower working-class man. May her memory be a blessing. www.nature.com/artic... 🧪

nature.comAda Yonath obituary: Crystallographer who overcame ‘impossible’ challenge of mapping the ribosomeNature - Nobel laureate persisted despite widespread derision and revolutionized structural biology by protecting fragile crystals.

I really wonder why people use these AI tools journals offer to summarise the main findings of a research paper instead of just reading the abstract?

Nice reference of our work: "For instance, Kilian et al. used longitudinal flow cytometry and scRNA-seq data from a murine DSS colitis model to parameterize ordinary differential equation (ODE)-based models of intestinal cell-population dynamics, ...

Cool Science Advances paper on the effects of regular exercise on immune cells in the human peripheral blood: antigen-presentation is enhanced. While (auto)inflammation is tamed down, there's increased epigenetic preparedness of cytotoxic T and NK cells: www.science.org/doi/... 🧪

This (abstract and editor's summary) reads like a super cool paper combining ideas of quorum sensing in macrophages with metabolic control. Can't access it in full on the train but looking forward to reading it: www.science.org/doi/10.1126/... 🧪 Congrats @nicolasavilalab.bsky.social

science.orgTissue mitochondrial activity dictates the macrophage pool sizeTissue mitochondrial activity controls macrophage abundance by regulating CSF1 availability, thereby enabling efficient waste removal.

I don't like that they're opening up the competition to industry while restricting "openness" of the code and the science... Still a great challenge. We also host a team at @uni-hamburg.de

Arc Institute@arcinstitute.org

We share our rationale and evaluation approach for the competition in a commentary published today in Cell. Our goal is for this event to build a new community of machine learning researchers, computational biologists, and experimentalists. www.cell.com/cell/fulltex...

That's a very smart idea to experimentally address one of the most concerning bottlenecks in the field! Thanks for sharing, @labwaggoner.bsky.social. I'd have missed it otherwise.

Waggoner Lab@labwaggoner.bsky.social

@biorxivpreprint.bsky.social Functional Pairing of TCR-peptide-MHC Interactomes by Single-cell Clonal Expansion www.biorxiv.org/content/10.6... @stanforduniversity.bsky.social

Intriguing @nature.com paper from @yaleschoolofmed.bsky.social on commensal-derived Acetylcholine, which educates mucosal immunity. Among many other cool things, they monocolonize germ-free mice with more than 100 individual bacterial strains... www.nature.com/articles/s41... 🧪

nature.comCommensal-derived acetylcholine enhances mucosal immune education - NatureA diet–microbiome–host axis strengthens mucosal immune defences and reinforces host–microbiota mutualism.

Also multi-level convergence: Many strains produce Acetylcholine (ACh), from different dietary choline sources. Mice colonized with ACh-producing strains exhibited enhanced intestinal immunoglobulin A production, altered microbiota composition and increased resistance to enteric infection.

"This work identifies EP2 signaling in tissue-resident macrophages as a central regulator of organ-wide aging through its control of senescent neutrophil clearance, reframing aging as a failure of active cellular clearance rather than passive degeneration." www.science.org/doi/10.1126/... 🧪

science.orgRestored clearance of senescent neutrophils by tissue-resident macrophages limits organ agingAging disrupts tissue homeostasis across organ systems. Here, we identify tissue-resident macrophages (TRMs) as central coordinators of age-related organ decline through impaired clearance of senescen...

Great review on the discovery of antimicrobial peptides and other novel antibiotics. Thanks for sharing. #WeeklyRead 🧪

Cell Host & Microbe@cp-cellhostmicrobe.bsky.social

Mining the code of life for new antibiotics Review charts classical screening to digital discovery, where genomes, proteomes & libraries become programmable resources. Virtual screening, deep learning & generative AI are reshaping the antibiotic pipeline www.cell.com/cell-host-mi...

Very happy to see this out - made a humble contribution analyzing some sequencing data to see our hypothesis confirmed in vitro and in a human cohort. 🧪

Bile acid retention in efferocytic #macrophages shapes their inflammatory status during #cholangitis. New study from Amirah Al Jawazneh, Imke Liebold, Stephanie Leyk, Lidia Bosurgi (Universitätsklinikum Hamburg Eppendorf) et al.: rupress.org/jem/article/... @pjsaez.bsky.social #inflammation

Uncertainty-quantification in neural ODEs: www.nature.com/artic... 🧪

nature.comA novel approach to quantify out-of-distribution uncertainty in Neural and Universal Differential Equations | npj Systems Biology and ApplicationsDynamical systems play a central role across the quantitative sciences, offering a powerful mathematical framework to describe, analyze, and predict the evolution of complex processes over time. In systems biology, dynamical systems provide a foundation for modeling and predicting the intricate behaviors of biological systems. Recent advances in data-driven approaches, such as Neural Ordinary Differential Equations (NODEs) and Universal Differential Equations (UDEs), have enabled the development of models that are either fully or partially data-driven. Integrating data-driven components into dynamical systems amplifies the challenge of generalization beyond training data, highlighting the need for robust methods to quantify uncertainty in out-of-distribution (OOD) scenarios—i.e., conditions not encountered during training. In this work, we investigate the reliability of uncertainty quantification (UQ) based on ensembles of models in the reconstruction of dynamical systems. We show that standard ensembles (i.e., models trained independently with different random initializations) risk producing overconfident predictions in previously unseen scenarios, as the models in the ensemble tend to exhibit similar behaviors. To address this issue, we propose a novel ensemble construction method for NODEs and UDEs that fosters diversity in the reconstructed vector field across models within specific regions of the state space, while maintaining explicit control over the fit on the training set. We first evaluate our method on numerical test cases derived from three models commonly used as benchmarks for data-driven reconstruction of dynamical systems: the Lotka–Volterra model, the damped oscillator, and the Lorenz system. We then apply the method to a biologically motivated model of cell apoptosis, considering more realistic conditions such as partial observability of the system outputs and noise in the training dataset. Overall, our results show that the proposed method improves the reliability of UQ in previously unseen scenarios compared with standard ensembles, especially where the latter exhibit overconfidence.

Adlung Lab reposted

🚀 What an amazing #SBMC2026! From inspiring welcome addresses, to the MTZ Award ceremony, scientific sessions, panel discussion, poster sessions, and Poster Awards — thank you to everyone who made these three days so special. Wonderful science, great discussions, and even better networking! 📸✨

Was really cool to give a "triplet talk" with my clinical partners last week at the Danish Diabetes and Endocrine Academy

😦 I gave this to DeepL: Our core international scientific collaborations span Israel, Poland, Sweden, the UK, Switzerland, Austria, France and the US. What came out: Our core international scientific collaborations span Israel, Poland, Sweden, the UK, Switzerland, Austria and France.

In a meeting yesterday, someone said: "Quick-and-dirty approaches are usually just dirty" – and I thought this was great sentiment.

Adlung Lab reposted