Sign in

CoSy.Bio

@cosybio-uhh.bsky.social
27 followers 1 following 37 posts

Computational Systems Biomedicine research group at the University Hamburg

PostsRepliesMedia
CoSy.Bio @cosybio-uhh.bsky.social · 15/06/2026
We are happy to have hosted the PROBE WP3 meeting in Hamburg last Friday. Six months in, we shared updates and discussed end-to-end integration of the federated database network and foundation models. Thanks to all partners for a great exchange! #PROBE #AI #OAI
000
CoSy.Bio @cosybio-uhh.bsky.social · 01/06/2026
Last week, we came together for the Steering Committee meeting of the @daibetes.bsky.social project. It was exciting to how far this project has come at its halfway point! #AI #Diabetes
010
CoSy.Bio @cosybio-uhh.bsky.social · 11/05/2026
From data harmonization to federated #AI training 🚀 On 5–6 May, we met at Sapienza University of Rome for the @daibetes.bsky.social / @microb-ai-ome.bsky.social #Hackathon. Thanks to Simon Süwer and Julian Klemm for leading the session, and to everyone who brought ideas and energy to the event!
021
CoSy.Bio @cosybio-uhh.bsky.social · 13/04/2026
#AlternativeSplicing is key to understanding disease, but high-depth data is limited. We tested #DeepLearning tools to bridge the gap and show their added value in analysis pipelines, e.g. for initial hypothesis development. 🔗https://academic.oup.com/bib/article/27/1/bbaf705/8419938
010
CoSy.Bio @cosybio-uhh.bsky.social · 16/03/2026
Our new #Optimization algorithm, the Farthest better or Nearest worse Optimizer (FNO), introduces a “jump and explore” mechanism to skip low-quality regions and focus on high-quality search spaces. #MachineLearning Check it out: link.springer.com/article/10.1...
link.springer.com
Farthest better or nearest worse optimizer: a novel metaheuristic algorithm - Artificial Intelligence Review
A novel metaheuristic optimization algorithm, namely the Farthest better or Nearest worse Optimizer (FNO) algorithm, is proposed in this paper. The idea behind the FNO algorithm is derived from the qu...
011
CoSy.Bio @cosybio-uhh.bsky.social · 09/03/2026
Anti-CD20 therapy may do more than target B cells. Single-cell profiling in patients with #MultipleSclerosis uncovers an IL1Bhigh monocyte population with central nervous system infiltrating potential that’s low in untreated patients but expands after treatment.
https://onlinelibrary.wiley.com/doi/full/10.1002/ana.78004
000
CoSy.Bio @cosybio-uhh.bsky.social · 23/12/2025
While reaching for cookies, we stumbled on a treat of a different kind: DIGGER 2.0 for exploring the functional role of protein interactions in #AlternativeSplicing! ☃️🍪 Now supporting the mouse interactome and the functional enrichment tool NEASE #CoSyAdventcalender 👉 exbio.wzw.tum.de/digger/
exbio.wzw.tum.de
Home | DIGGER
000
CoSy.Bio @cosybio-uhh.bsky.social · 18/12/2025
🐉 Have you ever met a DRaCOoN? This season’s new species doesn’t breathe fire, it reveals pathway-level changes in gene co-expression across conditions. A Python tool + web app with built-in benchmarking for robust #Transcriptomics analysis. #CoSyAdventcalender
110
CoSy.Bio @cosybio-uhh.bsky.social · 17/12/2025
☕ Grab a cup of hot chocolate and enjoy our next holiday reading! Our study shows that exercise therapy and education (GLA:D®) lead to real, measurable improvements for people with #KneeOsteoarthritis beyond normal fluctuations over time. #CoSyAdventcalender 🔗 www.mdpi.com/2077-0383/14...
mdpi.com
010
CoSy.Bio @cosybio-uhh.bsky.social · 16/12/2025
🎄✨ Today’s #CoSyAdventCalender highlight: FedscGen! A federated, privacy-preserving way to correct batch effects in distributed #scRNAseq datasets. 🔗 genomebiology.biomedcentral.com/articles/10....
genomebiology.biomedcentral.com
FedscGen: privacy-preserving federated batch effect correction of single-cell RNA sequencing data - Genome Biology
Single-cell RNA-seq data from clinical samples often suffer from batch effects, but data sharing is limited due to genomic privacy concerns. We present FedscGen, a privacy-preserving communication-eff...
000
CoSy.Bio @cosybio-uhh.bsky.social · 15/12/2025
Are you ready for the next #CoSyAdventcalender reading? Say hello to FedProt, a privacy-preserving tool for multi-center differential protein analysis. #Proteomics #FederatedLearning 🔗 www.nature.com/articles/s43...
nature.com
Privacy-preserving multicenter differential protein abundance analysis with FedProt - Nature Computational Science
In this Resource, the authors present FedProt, a tool that enables privacy-preserving, federated differential protein abundance analysis across multiple institutions. Its results match the results of ...
000
CoSy.Bio @cosybio-uhh.bsky.social · 11/12/2025
You better watch out, you better not bias … We tested 17 normalization + 2 batch correction methods in proteomics and built PRONE to reveal how much they change results. Normalization matters and so does picking the right one. #CoSyAdventcalender academic.oup.com/bib/article/...
academic.oup.com
Systematic evaluation of normalization approaches in tandem mass tag and label-free protein quantification data using PRONE
Abstract. Despite the significant progress in accuracy and reliability in mass spectrometry technology, as well as the development of strategies based on i
000
CoSy.Bio @cosybio-uhh.bsky.social · 10/12/2025
❄️ Not every snowflake becomes a snowman and not every model belongs in the clinic. We mapped 16 scenarios to judge when a prognostic model is “good enough to build or deploy, using a structured benefit-risk approach. #PrognosticModels #HealthcareAI #CoSyAdventcalender 👉 jmir.org/2025/1/e69341
jmir.org
To Implement or Not to Implement? A Commentary on the Pitfalls of Judging the Value and Risks of Personalized Prognostic Statistical Models
Prognostic models in medicine have garnered significant attention, with established guidelines governing their development. However, there remains a lack of clarity regarding the appropriate circumsta...
000
CoSy.Bio @cosybio-uhh.bsky.social · 09/12/2025
Oh the viruses may be frightful, but the DIPs are so delightful …This year we explored defective interfering particles (DIPs), viral mutants that block viral replication and could become novel #antivirals. #CoSyAdventcalender Learn more: academic.oup.com/nargab/artic... @mathbiotanja.bsky.social
academic.oup.com
Meta-analysis of genomic characteristics for antiviral influenza defective interfering particle prioritization
Abstract. Defective interfering particles (DIPs) are viral deletion mutants that hamper virus replication and are, thus, potent novel antiviral agents. To
011
CoSy.Bio @cosybio-uhh.bsky.social · 08/12/2025
🎁 This holiday season, we unwrap one of our highlights: @daibetes.bsky.social ! By combining privacy-perserving #AI & virtual twins, the project is working to transform #Type2DiabetesCare #CoSyAdventcalender
100
CoSy.Bio @cosybio-uhh.bsky.social · 01/12/2025
🎄 The first light on our holiday research tree: StrucTFactor! This deep learning method predicts #TranscriptionFactors using 3D structure, shining 17% brighter than existing methods. ✨ #CoSyAdventcalender 🎁 Learn more: academic.oup.com/bioinformati...
academic.oup.com
Transcription factor prediction using protein 3D secondary structures
AbstractMotivation. Transcription factors (TFs) are DNA-binding proteins that regulate gene expression. Traditional methods predict a protein as a TF if th
000
CoSy.Bio @cosybio-uhh.bsky.social · 01/12/2025
The holiday season is here! 🎄 Cookies, chocolate … and research! This advent calendar serves up our research highlights from 2025, a sweet countdown of computational biomedicine highlights. 🍪📄 → Follow us for daily advent calendar posts! #CoSyAdventcalender #Bioinformatics
000
CoSy.Bio @cosybio-uhh.bsky.social · 06/10/2025
We present FedscGen, a federated, #PrivacyPreserving method (built on scGen) batch-effect correction tool to help analyze distributed #scRNAseq datasets. genomebiology.biomedcentral.com/articles/10....
010
CoSy.Bio @cosybio-uhh.bsky.social · 29/09/2025
✨ Congratulations to Faren, the winner of the Young Investigator Award #RExPO25! This is a truly well-deserved recognition of her outstanding research and impact. 👏 @repo4eu.bsky.social
010
CoSy.Bio @cosybio-uhh.bsky.social · 22/09/2025
#Acne and other #SkinDisorders are often linked to the sebaceous gland (SG). We provide a detailed molecular map of the human SG, revealing dynamic differentiation of sebocytes and identifying new #GeneMarkers, providing a valuable reference for future research. www.jidonline.org/article/S002...
000
CoSy.Bio @cosybio-uhh.bsky.social · 15/09/2025
🚀🦠 Exciting discussions at the @microb-ai-ome.bsky.social Steering Committee Meeting in Vienna! Great to see progress at the intersection of #MicrobiomeResearch and #AI!
120
CoSy.Bio @cosybio-uhh.bsky.social · 11/09/2025
Studying #BoneTissue proteins can be likened to finding a needle in a haystack of collagen. Our new protocol eliminates collagen interference without the use of enzymes, yielding 68% more proteins identified. #proteomics pubs.acs.org/doi/10.1021/...
000
CoSy.Bio @cosybio-uhh.bsky.social · 08/09/2025
We introduce DRaCOoN 🐉: a Python package and web tool designed for pathway-level #DifferentialCoexpression analysis. It features a unique benchmarking framework and demonstrates robust performance across various #Transcriptomics datasets. bmcbioinformatics.biomedcentral.com/articles/10....
000
CoSy.Bio @cosybio-uhh.bsky.social · 21/08/2025
Presenting DIGGER 2.0: we extended our #AlternativeSplicing software DIGGER with new functionalities, including support of the mouse interactome and NEASE for #EnrichmentAnalysis! Check it out here: exbio.wzw.tum.de/digger/
000
CoSy.Bio @cosybio-uhh.bsky.social · 07/08/2025
When is "good enough" actually good enough for #PrognosticModels? Our commentary tackles this question with a structured analysis of 16 scenarios + practical approaches for development decisions for healthcare-related prognostic models. www.jmir.org/2025/1/e69341
000
CoSy.Bio @cosybio-uhh.bsky.social · 28/07/2025
Defective interfering particles (DIPs) as potential novel #Antivirals! Our meta-analysis of 20 #Influenza datasets provides insights for selecting candidates for antiviral treatment design. academic.oup.com/nargab/artic...
000
CoSy.Bio @cosybio-uhh.bsky.social · 21/07/2025
We compared various models predicting changes in #KneePain of #Osteoarthritis patients due to participation in supervised education and exercise and propose a concise 6-variable model for personalized prediction. rehab.jmir.org/2025/1/e60162
000
CoSy.Bio @cosybio-uhh.bsky.social · 30/06/2025
Introducing DREAMER, a groundbreaking tool that identifies mechanisms of #AdverseDrugReactions (ADRs) by linking drug and disease phenotypes to protein interaction networks. Here, DREAMER revealed #MolecularMechanisms behind 67 ADRs. www.sciencedirect.com/science/arti...
000
CoSy.Bio @cosybio-uhh.bsky.social · 10/06/2025
We wrapped up a productive @daibetes.bsky.social Steering Committee Meeting last week in Boston! Great discussions on how privacy-preserving #AI can advance #DiabetesResearch and #PersonalizedPatientCare.
010
CoSy.Bio @cosybio-uhh.bsky.social · 02/06/2025
Soluble guanylate cyclase (sGC) protects cells from #OxidativeStress. Our research shows ta cGMP-dependent reduction of #ReactiveOxygenSpecies (ROS) upon soluble guanylate cyclase activation via a downregulation of NOX4/5 via NF-κB. Our findings have significant implications for #StrokeRecovery!
100
CoSy.Bio @cosybio-uhh.bsky.social · 12/05/2025
Watch the new video on dAIbetes! This @horizoneu.bsky.social project leverages #AI and #FederatedLearning to transform type 2 diabetes treatment. Discover how virtual twins and secure data sharing are enabling personalized care. #DigitalHealth #DiabetesResearch daibetes.eu/news/details...
daibetes.eu
dAIbetes video - dAIbetes
dAIbetes as a whole, just have a look at the ⇒ video
000
CoSy.Bio @cosybio-uhh.bsky.social · 10/03/2025
Say goodbye to endless PDF scrolling and hello to faster discoveries! 🌟💊 Check out how DrugRepoChatter is revolutionizing #DrugRepurposing research!
020
CoSy.Bio @cosybio-uhh.bsky.social · 10/02/2025
Meet BioChatter! CoSy.Bio contributed to this open source framework for interfacing with #LLMs in biomedical research while ensuring user safety and privacy. Explore BioChatter at www.nature.com/articles/s41... @saezlab.bsky.social
nature.com
A platform for the biomedical application of large language models - Nature Biotechnology
Nature Biotechnology - A platform for the biomedical application of large language models
010
CoSy.Bio @cosybio-uhh.bsky.social · 03/02/2025
Predicting drug responses is crucial for #PersonalizedMedicine. Our study reveals that #TranscriptionFactor activity-based methods outperform others in this task . #MachineLearning www.nature.com/articles/s41...
020