PI at @HelmholtzMunich, fellow at @TU_Muenchen, pushing the boundaries of computational methods for human genetics & health AI.
Munich, Germany
Joined May 2015
- Tweets22
- Following340
- Followers285
- Likes193
Pinned Tweet
Now out in @GenomeBiology!
HistoGWAS combines histology foundation models, statistical genetics and generative AI to characterize variant effects on tissue morphology, enabling genetic analysis at the tissue scale.
Led by @skc_2017
@HelmholtzMunich @PioneerCampus
New preprint: RisQ, a multimodal AI framework for learning shared representations of human health across diseases and time. It provides a new framework to study the shared organization of disease susceptibility.
🔗doi.org/10.64898/2026.07.07.…
Huge congratulations to co-first authors Paul Hager, Benedikt Roth, Niklas Bühler, and Diyuan Lu! Great collaborative effort across the labs of @fabian_theis, @DanielRueckert, @ja_schnabel, and ours!
📢 Deadline extension: RECOMB-Genetics 2026
The RECOMB-Genetics 2026 workshop will take place in Thessaloniki, Greece 🇬🇷 on May 25, 2026, as part of the RECOMB satellite meetings, just before the main RECOMB conference (May 26–29).
📅 Submission deadline: April 2 (AoE)
👉 tinyurl.com/recomb-genetics-…
@RECOMBconf
New preprint: REECAP, a framework using imaging foundation models for disease genetics analysis. We benchmark it against alternative image-derived representations and link loci to retinal morphology. Led by @shilova_lubov, collab with @BjoernEskofier.
🔗 doi.org/10.1101/2025.11.19.2…
Huge thanks to all co-authors and collaborators for their contributions. Daniel Sens, Ayshan Aliyeva, Shubham Chaudhary, Qiaohan Xu, Emmanuelle Salin, Johannes Schiefelbein, Ben Asani, Oana Veronica Amarie, Elida Schneltzer, @asegre, @ja_schnabel, @caina89
🚀 In @genomeresearch!
BayesRVAT integrates pathogenicity predictions, e.g. from AI sequence models, to model gene disruption-response in rare variant association studies. Collab with @caina89 & @Tkaraletsos.
@HelmholtzMunich @PioneerCampus @czbiohub
🔗doi.org/10.1101/gr.280689.12…
Paolo Casale retweeted
🚨 RECOMB-Genetics Deadline 🚨
The RECOMB-Genetics 2025 workshop is in Seoul 🌏 on April 25, 2025, just before the main RECOMB conference (April 26–29).
📅 Submission deadline: Feb 14, 2025 (AoE)
👉recomb.org/recomb2025/recomb…
Submit your work and join us for this exciting event!
🚨 UPCOMING DEADLINES 🚨
⭐️Highlights: January 27, 2025
📋Posters: January 27, 2025
RECOMB-Seq: January 24 (abstracts)
RECOMB-Microbiome: January 31
RECOMB-P&E: February 5
RECOMB-CG: February 7
RECOMB-Genetics: February 14
RECOMB-CCB: February 14
recomb.org/recomb2025/
Paolo Casale retweeted
AI is completely insane! (From the lab of this years winner of the Nobel prize)
nature.com/articles/s41586-0…
🚨 RECOMB-Genetics Deadline 🚨
The RECOMB-Genetics 2025 workshop is in Seoul 🌏 on April 25, 2025, just before the main RECOMB conference (April 26–29).
📅 Submission deadline: Feb 14, 2025 (AoE)
👉recomb.org/recomb2025/recomb…
Submit your work and join us for this exciting event!
🚨 UPCOMING DEADLINES 🚨
⭐️Highlights: January 27, 2025
📋Posters: January 27, 2025
RECOMB-Seq: January 24 (abstracts)
RECOMB-Microbiome: January 31
RECOMB-P&E: February 5
RECOMB-CG: February 7
RECOMB-Genetics: February 14
RECOMB-CCB: February 14
recomb.org/recomb2025/
Paolo Casale retweeted
After 4 fantastic years of research, with @caina89 & @heinig_matthias, I’m thrilled to share the first tissue specific map of apparent mtDNA heteroplasmy in humans and their relationships with donor age and mtDNA gene expression across 49 human tissues!👇
biorxiv.org/content/10.1101/…
Thrilled to announce that my group at @HelmholtzMunich and @PioneerCampus will contribute to the transformative scientific endeavor of advancing virtual cell models with @Tkaraletsos and @cziscience. Truly honored to collaborate with such a pioneering and brilliant team! 🧬🔬
Excited to share we are partnering with new AI residents @StephanMandt, @FrancescoLocat8, @fpcasale and their labs at @UCIrvine, @ISTAustria, and @HelmholtzMunich in support of our work at @cziscience to build virtual cell models that can accelerate scientific discovery and progress.
Paolo Casale retweeted
We present MultiOrg: an #organoid dataset tailored for #object_detection tasks with #uncertainty quantification. Over 400 high-resolution 2D #microscopy images and multi-expert annotations of more than 60,000 organoids.
Paper: arxiv.org/abs/2410.14612
Data: kaggle.com/datasets/christin…
Paolo Casale retweeted
Looking to enhance rare variant association testing? Catch my platform talk on FuncRVP - a Bayesian rare variant association framework leveraging functional gene embeddings.
Nov 7, 11:30-11:45, Room 501. #ASHG24
Paolo Casale retweeted
I’m thrilled to introduce MemBrain V2(github.com/teamtomo/membrain…), our latest research tool developed by @LorenzLamm and co-supervised by @bengeliscious, right here at the beautiful Princeton campus. It’s exciting to see how widely it’s being adopted by the community!
Tingying Peng (@p_tingying) of @HelmholtzMunich presented MemBrain v2, “an end-to-end tool for analysis of membranes in cryo-electron tomography” #FEMPL3
ALT Tingying Peng speaking at podium
Paolo Casale retweeted
PhD students—Do this and 90% of your paper problems will disappear.
The secret to getting your papers accepted isn't what you think.
It's about...
10 rules to structure your papers like you mean business.
I wish I knew these when I started my PhD.
Here they are:
ALT Table 1 from Mensh, B., & Kording, K. (2017). Ten simple rules for structuring papers. PLOS Computational Biology, 13(9), e1005619.
As a physicist-turned-AI scientist, I’m thrilled to see this! Physics principles are at the core of AI breakthroughs, and AI models are dynamic physical systems!!! 🌍🔬 #AI #Physics #NeuralNetworks #NobelPrize2024
BREAKING NEWS
The Royal Swedish Academy of Sciences has decided to award the 2024 #NobelPrize in Physics to John J. Hopfield and Geoffrey E. Hinton “for foundational discoveries and inventions that enable machine learning with artificial neural networks.”
