New study from Praneet Prakash, Vona, and Raymond E. Goldstein (DAMTP, Cambridge University) reveals how tiny swimmers like rotifers and algae have their own unique swimming styles — some pull water in while others push it away — leading to distinct surface interactions! @ go.aps.org/3BXMNDs

Paper with Praneet Prakash, Yasa Baig and Francois Peaudecerf on the algae-bacteria microcosm is just out: pnas.org/doi/10.1073/pn…

Protein function often depends on protein dynamics. To design proteins that function like natural ones, how do we predict their dynamics? Hannah Wayment-Steele and I are thrilled to share the first big, experimental datasets on protein dynamics and our new model: Dyna-1! 🧵

DeTikZify is out on Hugging Face Synthesizing Graphics Programs for Scientific Figures and Sketches with TikZ

BASED ✌️ turns 1! One year since its launch at NeurIPS 2023 — and it's helped shape the new wave of efficient LMs. ⚡️ Fastest linear attention kernels 🧠 405B models trained on 16 GPUs 💥 Inspired Mamba-v2, RWKVs, MiniMax Checkout our retrospective below!

1/ What's the worst enemy of a plasmid? Another plasmid! Our new study demonstrates synthetic plasmids engineered as "Trojan horses" to eliminate plasmids encoding antibiotic resistance. Duke Center for Quantitative Biodesign Duke Biomedical Engineering science.org/doi/10.1126/sc…

Research is for everyone. Grateful to WBEZ to talk about why research is important and why stopping hurts people and our leadership as the #1 country for innovation. wbez.org/reset-with-sas…

I'm at #ICLR2025 this week to present our work on🔬high-precision algorithm learning🔬with Transformers! Stop by our poster session Thursday afternoon! 🔗arxiv.org/abs/2503.12295 With Jess Grogan, Owen Dugan , Ashish Rao, Simran Arora , Atri Rudra, and hazyresearch!

new Principal InvestiGATOR alert! Thrilled to be committed to join the University of Florida, Physics as an assistant professor this coming Fall.

Research cuts impact everyone. Northwestern CSB trying to get the word out: syntheticbiology.northwestern.edu/csb-leaders-vo…

So the new pope is a mathematician. Does he believe every set of reals is measurable? The Solovay model requires the existence of an inaccessible cardinal.

RLXF is on bioRxiv! We introduce a PPO-based workflow to align the logits of any protein language model away from evolutionary plausibility and towards biochemical function. Not only does this seem to be more robust than other preference optimization methods... (1/3)

Reinforcement learning with experimental feedback (RLXF) shifts protein language models so that they generate sequences with improved properties. Nathaniel Blalock sri Romero lab

Extremely awesome work from some really brilliant folks at the the Gilpin Group! Transformers naturally learn numerical matrix patterns when embedding chaotic time series--wild!

Super cool stuff on context compression! Maybe like game boy cartridges though, maybe blowing on my GPU cartridge will also help it perform better...

Just one more reminder that the NMR ensembles deposited in the PDB != dynamics of a protein... The ensemble represents *uncertainty* due to lack of experimental data when solving a structure from NMR. Using ensembles to benchmark ML/AI models of dynamics isn't a good idea.

Was extremely fun to work on this paper with Jerry Liu and finally fulfilling our 7 year plan from year one of undergrad to write a paper together! One of many I hope!

🚀 Introducing PhysiX: One of the first large-scale foundation models for physics simulations! PhysiX is a 4.5B parameter model that unifies a wide range of physical systems, from fluid dynamics to reaction-diffusion, outperforming specialized, state-of-the-art models.

I'm excited to share my work on the Virtual Lab, a collaboration between LLM-powered AI agents and human researchers to identify nanobody binders to SARS-CoV-2, published in nature. nature.com/articles/s4158… Our code is publicly available at github.com/zou-group/virt…

“What’s past is prologue” — excited about chromatin recording by synthetically engineering recruitment of adenine methyltransferases in living cells. Will enable one to correlate past states with subsequent fate decisions. New work from the virtuosic Yodai Takei See thread.