Preprint: Characterization of the dual roles of senescent-like T cells that arise during healthy and unhealthy ageing www.biorxiv.org/content/10.1... @drsianh.bsky.social

NEW: The doctor behind breakthrough Parkinson’s research was among the scientists purged from the National Institutes of Health, the US’s leading medical research agency. www.wired.com/story/doctor...

Decoding sequence determinants of gene expression in diverse cellular and disease states [updated] Predicts gene expression from DNA via single-cell data; reveals regulation & variant effects.

Enhancing CAR-T cell activity prediction via fine-tuning protein language models with generated CAR sequences [new] Fine-tuning ProtLM predicts CAR-T activity. Augmented CAR seqs enable task-specific adaptation.

The CDC has literally been gutted today. There’s no need for “April Fools” when reality is a nightmare for every human on this planet. Thousand of scientists fired. Entire departments erased. 1/

Happy to share a blog post I wrote on our new computational approach LogiCAR Designer, which identifies logic-gated antigen circuits for precise, next-generation CAR therapies. 🎯🧬 mlandbio.substack.com/p/from-singl...

Can we engineer smarter CAR-T cells that target cancer with precise logic? 🎯🧬 So excited to share the heart of my PhD work: 🌟 LogiCAR designer, a framework that identifies logic-gated multi-antigen circuits for next-generation cell therapies 🧩🧵 www.biorxiv.org/content/10.1...

Single-cell-guided identification of logic-gated antigen combinations for designing effective and safe CAR therapy [new] Finds logic-gated antigen combos in single-cell data for safer CAR-T, exceeding shared methods.

Big news: we are setting up a new non-profit organization to run bioRxiv and medRxiv. It's called openRxiv [no it's not a new preprint server; it's dedicated organization to oversee the servers] openrxiv.org 1/n

We are experiencing an assault on science unparalleled by anything I’ve seen in my life. It’s not one issue or another anymore, the entire institution is under attack by the most powerful individuals in the country. This Friday, where will you be? standupforscience2025.org

Bizarre tumor immunology accident. Surgeon injures hand while resecting malignant histiocytoma and finds an inflamed mass at the site of prior injury. Tumor cells transplanted, patient & surgeon HLA mismatched causing intense yet insufficient host-vs-graft response: www.nejm.org/doi/full/10....

One main pain point: interdisciplinary work is often judged by the highest standards in *both* disciplines. This creates an almost impossible burden, effectively extinguishing creative work. Sure, some standards are there for a reason, but more often than not, they are just stock critiques.

This paper is the result of an incredibly fun collaboration with Sandro Santagata and Peter Sorger. Led by an amazing postdoc @tanjinakader.bsky.social and JiaRen Lin, we provide a pre-cancer atlas of Fallopian tube precursors to High-Grade Serous Ovarian Cancer. The data is also on cbioportal!

Wouldn't have guessed that a PCSK9 blocker (used for LDL cholesterol lowering) might someday be indicated to prevent breast cancer in individuals with a common genomic variant www.cell.com/cell/fulltex...

Could one envision a synthetic receptor technology that is fully programmable, able to detect diverse extracellular antigens – both soluble and cell-attached – and convert that recognition into a wide range of intracellular responses, from gene expression and real-time fluorescence to modulation..

One of the alluring and theoretical strategies for extending healthspan and longevity is to rejuvenate the thymus gland, promote an intact immune system. Now there's a way to do that in aged mice. www.science.org/doi/10.1126/...

WaPo Edito: Be thankful for the applications of AI in medicine. More accurate detection of cancers (breast, prostate, skin, brain), faster diagnosis of strokes, sepsis, heart attacks, faster MRIs, full-body in 40 minutes. Much more to come over the next years. www.washingtonpost.com/opinions/202...

Medically adapted foundation models (think Med-*) turn out to be more hot air than hot stuff. Correcting for fatal flaws in evaluation, the current crop are no better on balance than generic foundation models, even on the very tasks for which benefits are claimed. arxiv.org/abs/2411.04118

Happy to share our new paper, which was just published in Nature Communications. In-context learning enables general-purpose vision language models to perform well on medical image analysis tasks. We don't need to train a neural network at all - just provide a handful of examples at inference time.

Incredible experience at the NCI HTAN Data Jamboree! Grateful to work with a brilliant team to develop HTANalyzer, a multi-agent LLM that helps researchers query and analyze HTAN spatial transcriptomics data through natural language queries. Check out our project github: github.com/NCI-HTAN-Jam...

—Please give me an A.I. that automatically detects, interprets all types of medical. images (scans, retina, endoscopy, skin) that I can interact with, can integrate data from other sources, saves a lot of time, and improves accuracy —Sure, here it is www.nature.com/articles/s41...

A genome-scale map of genetic interaction profiles building upon cell morphology. Great work by Florian Heigwer and collaborators. #Genomics #Imaging “A global genetic interaction network by single-cell imaging and machine learning” www.sciencedirect.com/science/arti...