Today @nature.com, it's #AlphaGenome, to decipher and determine functionality of the regulatory (very challenging) variants in our genome.
Another big step of AI for advancing life science
nature.com/articles/s41...
Our lab at Yale @yalemedicine.bsky.social seeks #postdocs with in vivo expertise to pioneer research at the intersection of tissue remodeling & aging. Work with us to uncover immunological & vascular drivers of ovarian aging, applying single-cell, spatial omics, and ML! jobrxiv.org/job/yale-uni...
🔭 Next steps are to explore how to integrate this model into clinical workflows and evaluate its impact prospectively. We’re also interested in expanding to other ICU populations and incorporating additional data streams like imaging and clinician notes!
In failed extubation cases, the model would’ve advised against extubation 35% of the time, showing promise as a second opinion tool 💡. Top predictors from SHAP and ablation testing include plateau pressure, heart rate, PaCO2, aligning with clinical intuition.
Our best model, an LSTM 🤖 , predicted next-day extubation with AUROC 0.87 in both our internal and external validation hospitals. It often flagged patients as ready for extubation days before it actually happened, suggesting potential to wean earlier.
We trained models on 37 clinical features (vitals, labs, meds, vent settings)🫁 collected from midnight–8 AM ⏰ , so predictions are ready for morning rounds. We carefully annotated data, reviewing hundreds of charts.
We built ML models to predict ICU patients’ readiness for extubation, a decision that’s critical and time-sensitive. Too early = failure. Too late = complications.
🚨 Our paper is out in Scientific Reports! Amazing collaboration with Alec Peltekian, @catgaohow.bsky.social, and Ankit Agrawal.
www.nature.com/articles/s41...
🔭 Next steps are to explore how to integrate this model into clinical workflows and evaluate its impact prospectively. We’re also interested in expanding to other ICU populations and incorporating additional data streams like imaging and clinical notes!
In failed extubation cases, the model would’ve advised against it 35% of the time, showing promise as a second opinion tool 💡. Top predictors from SHAP and ablation testing include plateau pressure, heart rate, PaCO2, aligning with clinical intuition.
Our best model 🤖 , an LSTM, predicted next-day extubation with AUROC 0.87 in both our internal and external validation hospitals. It often flagged patients as ready for extubation days before it actually happened, suggesting potential to wean earlier.
We trained models on 37 clinical features (vitals, labs, meds, vent settings)🫁 collected from midnight–8 AM ⏰ , so predictions are ready for morning rounds. We carefully annotated data, reviewing hundreds of charts.
We built ML models to predict ICU patients’ readiness for extubation, a decision that’s critical and time-sensitive. Too early = failure. Too late = complications.
Today "a milestone in the evolution of personalized therapies for rare & ultra-rare inborn errors of metabolism"
—the 1st human to undergo custom genome editing
—from decades of NIH funded research
www.nejm.org/doi/full/10....
@nejm.org
www.nejm.org/doi/full/10....
www.nytimes.com/2025/05/15/h...
What if LLMs could “read” & “write” biology? 🤔
Introducing C2S‑Scale—a Yale and Google collab: we scaled LLMs (up to 27B!) to analyze & generate single‑cell data 🧬 ➡️ 📝
🔗 Blog: research.google/blog/teachin...
🔗 Preprint: biorxiv.org/content/10.1...
Activism works. Case in point: #LongCovid
@nature.com
www.nature.com/articles/d41...
From 2010 to 2016 (latest data I have ), NIH research contributed to EVERY drug approved by the FDA
"Despite being preventable and highly curable, TB continues to have devastating health, social, and economic impacts globally." - Cenyun Guan, MPH '25 (@yaleemd.bsky.social)
#WorldTBDay2025
99% of new medicines developed by the pharmaceutical industry depend on NIH research jamanetwork.com/journals/jam...
Anne Carpenter writes about the science cuts: “this is like suddenly announcing that you will pay for doctors and nurses but not the hospital building they work in” for the LaPorte Herald Dispatch in IN 🌽
🧪🧬🔬🏠
@drannecarpenter.bsky.social
www.lpheralddispatch.com/opinion/gues...
@thejohnnyyu.bsky.social, @therealnima.bsky.social, and I, are excited to tell you about Tahoe-100M! The largest publicly available single-cell dataset that measures the effect of 1200 genes on 50 cell line models. The Vevo team has outdone itself. #Tahoe100M www.biorxiv.org/content/10.1...
A very innovative approach to cancer: engineering fat cells to convert to beige fat to starve tumors, outcompeting 5 types of cancer in experimental models
www.nature.com/articles/s41... @naturebiotech.bsky.social y.social @nadavahituv.bsky.social tuv.bsky.social
Post the amazing science things you have done with federal funding.