I’m excited to announce that Biosphere is coming out of stealth today with $8.8m in Seed funding led by @LowerCarbon & VXI Capital, with support from Founders Fund, GS futures, Caffeinated Capital & B37.

We compared the calibration of various machine learning uncertainty estimation methods for protein engineering. No method excels across all scenarios, and uncertainty-based strategies for optimization often did not outperform methods without uncertainty.

Yo folks, it is time to apply to *hands down the best conference in microbial evolution and ecology*: grc.org/microbial-popu… We begin reviewing apps in 10 days! We strongly encourage everyone to give a poster. DM w/ questions.

A framework for evaluating how well generative models of protein structure match the distribution of natural structures. Possu Huang Lab

Check out Artem's precomputed interactive AF2BIND (small molecule binding site) predictions for the human proteome. 🤩

The Ukrainian School in Evolutionary Biology is online! Talks which are happening now and previous talks can be viewed on YouTube: youtube.com/@biologicnijfa… #biology #evolution #school ESEB

What happens to bacterial genomes when they're cloned into yeast? Do they transcribe, do they get silenced, do they get packaged away from the rest of the DNA? This cool study congrats to the Koszul lab team at Pasteur for this great work science.org/doi/10.1126/sc…

Uh if I’m 100% soft money and bringing in all my own grant money: how come I don’t get to set my own salary…? 🤔

Information storage across a microbial community using universal RNA barcoding go.nature.com/440gL5E

Duckworth goes on Fox and Baier reads her statement calling Hegseth a “f**king liar”

If you are interested in phage satellites, we hope you'll enjoy this. Fun collaboration with the Rocha, Seed, Bikard, and Chen labs! nature.com/articles/s4157…

Structure models generate proteins with higher structure confidence and more plausible energy distributions, but limited diversity and sequence biases. Sequence models generate diverse and novel proteins with lower structural confidence. Neither are good at generating enzymes.

1/ Hagfish slime is a deploy-on-demand hydrogel: tiny glands eject mucin vesicles + coiled protein “threads.” In seawater, vesicles burst, threads unspool, and a viscous mesh balloons to ~10,000x its original volume.

12/ Hagfish slime is a clear example of how protein sequence drives structure and delivers context-dependent mechanics. Understanding these natural principles can unlock new design spaces for creating dynamic, high-performance biomaterials.

Look at this perfect loaf of bread I made with very good slices

The exciting next talk in our Protein Structure Evolution (ProSE) seminar series is tomorrow! Dr. Klara Hlouchova will speak on "Protein Structure Before LUCA" May 13, 3pm BST // 4PM CET Registration link: tinyurl.com/prose-seminar4 Let me know if you have any questions.

Steered generation for protein optimization: On datasets with ~10^2 measurements, steering a discrete diffusion model outperforms RL on a protein language model for generating improved variants. Jason Yang Wenda Chu Frances Arnold Yisong Yue

Have you thought about Pico lately?

CASP is getting cut by NIH... 😢 (Anyone with extra funds wanna help support perhaps the most important competition of the century?) science.org/content/articl…

In your PhD you learn like 3 things and the rest of your career can be just pointing out one of those same 3 things again and again.