Proteomic assays assess thousands of proteins but we lack understanding of their origins New preprint explores determinants of 3,000 circulating protein levels in UK Biobank (N=43,240) 👉genetic, modifiable & technical parameters🧬🥘🧪 👉explained variance 0-80% (median 23%)📊

1/ Excited to see moscot-tools.org published in nature! We scaled Optimal Transport (OT) in single-cell genomics & added multimodality together with spatiotemporal trajectory inference, finding exciting new biology in the pancreas! 🚀 Read at nature.com/articles/s4158…

Kick-off of “Industria del Coneixement” 2024 projects from Generalitat de Catalunya. OneCareAI has been granted with Producte and Innovadors funds! Davide Cirillo BSC-CNS

📢#BSCSeminar: Multi-view and Multi-modal Foundation Models for Drug Discovery 🗣 Michal Rosen-Zvi, IBM Research 📆 30 January ➡ bsc.es/ZCU #WomeninBSC #EquityinBSC Bioinfo4Women SOMM alliance

🟣IDIBAPS WOMEN AND GIRLS IN SCIENCE DAY🟣 📆12/02 - 11h 📍 Esteve Auditorium, CEK 🎙️"Enhancing Suicide Prevention on Social Media through AI" Ana Freire  🎙️"Sex and Gender Bias in Artificial Intelligence for Health" Davide Cirillo  👉 s.mtrbio.com/ophjzwagod

Thrilled to introduce our Cell Reports @cellreports article, showing how the iconic '#Hallmarks of #Cancer' are located in the space of real human primary tumors. A team effort co-led by Eduard Porta Eduard Porta , Matthew H. Bailey, and yours truly. #SpatialBiology #Visium

🟣How to overcome together sex and gender biases in artificial intelligence?🟣 📆12/02 - 11h 📍 Esteve Auditorium, CEK 🎙️"Sex and Gender Bias in AI for Health" Davide Cirillo 🎙️"Enhancing Suicide Prevention on Social Media through AI" Ana Freire  👉 s.mtrbio.com/gnsdkhpnnp

Now out: "A machine learning approach to predicting dynamical observables from network structure" (royalsocietypublishing.org/doi/10.1098/rs…). Here we use ML techniques to predict dynamic outcomes of networked systems using their structural properties. Kudos to Francisco Rodrigues Thomas Peron et al.

In Milan Human Technopole we opened core-funded computational Group Leader positions in one or more of bioimage analysis, spatial ‘omics, AI for genomes and proteins, computational simulations of cells and tissues. Come to lead your research programme! careers.humantechnopole.it/o/group-leader…

AI applications in biomedicine research 🔬🧬💻🤖 Very interesting talk! Excellent presentation by Davide Cirillo at #SCD2025 SCDigestologia BSC-CNS

[New Paper] Can economic complexity help us understand how countries move towards their Sustainable Development Goals (SDGs)? In a new paper, led by Fengmei Ma and Heming Wang (Heming Wang), from CCL, we explore how countries make progress towards SDGs. Instead of looking at

Impressive advancement in Computational Pathology. A new multimodal foundation model by Faisal Mahmood trained on 47,000 paired histology & genomics, which beautifully shows the multi-modal power of images & DNA & RNA Even though patient genomic data is rare, it's so powerful 🧵

New preprint claims that most existing DNA language models perform just as well with random weights, suggesting that pretraining does nothing (Mistral & DNABERT-2 look like exceptions). We need better DNA language models.

What is a differentially expressed gene? biorxiv.org/content/10.110…

Benchmarking gene embeddings from sequence, expression, network, and text models for functional prediction tasks biorxiv.org/content/10.110…

1/9 🚨 New Paper Alert: Cross-Entropy Loss is NOT What You Need! 🚨 We introduce harmonic loss as alternative to the standard CE loss for training neural networks and LLMs! Harmonic loss achieves 🛠️significantly better interpretability, ⚡faster convergence, and ⏳less grokking!

Biomolecular condensates: It was RNA all along!: Molecular Cell cell.com/molecular-cell…

Agentic AI Meets Medicine!!! 🔬 Excited to announce MedRAX: a groundbreaking Medical Reasoning Agent for Chest X-ray interpretation, now on arXiv! Paper:arxiv.org/abs/2502.02673 Code: github.com/bowang-lab/Med… What is MedRAX? MedRAX is the first versatile AI agent that

The shift from AI as tools to AI as agents has tremendous implications for medicine and healthcare. Eric Topol and I share our perspective on new possibilities and how to prepare for this shift in The Lancet thelancet.com/journals/lance…

Glad to be part of the FUTURE-AI Consortium led by Karim Lekadir ! Have a look at our international consensus guideline for trustworthy and deployable artificial intelligence in healthcare, just published at The BMJ here: bmj.com/content/388/bm…