New paper in Protein Science! Meet DreamFold: a World Model that generates protein folding pathways and reveals druggable intermediate states: onlinelibrary.wiley.com/doi/…

New @naturecomms paper by @katerbowie @MarkGerstein @jordan_peccia @H2O_Hannah. We study how disinfection shapes microbes in hospital sink drain biofilms. Biofilms regrew in 4 days, enriched for carbapenem-resistant bacteria and multidrug efflux pump genes nature.com/articles/s41467-0…

In our @NatMachIntell paper, we introduce a framework to analyse interpretability in deep learning by drawing on a formal notion of model semantics from the philosophy of science. We illustrate our framework with examples from biomedicine. Read here: rdcu.be/e9uYh

🔐 New open-access paper in Cell Reports Methods! We show that fully homomorphic encryption enables privacy-preserving polygenic risk scores (PRS), allowing secure computation directly on encrypted genomes with near-zero accuracy loss. 📄 cell.com/cell-reports-method…

🚀 New paper in Bioinformatics! Our #ASTRO work led by @dingyao_zhang introduces "ASTRO: Automated Spatial-Transcriptome whole RNA Output", an automated pipeline optimized for whole-transcriptome spatial analysis, especially in challenging FFPE samples. 🔗 doi.org/10.1093/bioinformati…

Our @NeurIPSConf work led by @_YunyangLI “E2Former: An Efficient and Equivariant Transformer with Linear-Scaling Tensor Products” was selected as a spotlight (with score ranked ~17 / 21k submissions). Poster: Thur Dec 4, Exhibit Hall CDE #5512 Online: openreview.net/forum?id=ls5L…

🧠 At our recent Gerstein Lab roster meeting, we took a detour into… personality science! Turns out we’re INT Central 🧪 📌 70% Introverts 📌 83% Intuitives 📌 57% Thinkers Analysts (INTP, INTJ) dominate, far more than the U.S. baseline. #MBTI #INTP #INTJ

4/4 ⚡ WANet + WALoss ⇒ 18 % faster SCF convergence & 1 000 × energy-error reduction vs. SOTA. One model, many properties—HOMO/LUMO, dipoles, electron densities—all from a single predicted Hamiltonian.

2/4 🧩 We introduce Wavefunction-Alignment Loss + WANet, slashing SCF iterations while keeping ab-initio precision for molecules 3× larger than training data.

1/4 🚀 New #ICLR2025 SPOTLIGHT ALERT Gerstein Lab presents “Enhancing the Scalability & Applicability of Kohn-Sham Hamiltonians”—led by  @_YunyangLI & Z Xia & L Huang & J Zhang & @MarkGerstein. Joint work with @MSFTResearch.