Twitter Beğeni Hilesi

Let’s do more stuff that works 💪 awesome results.

"Recent generative protein design models have failed to improve protein activity because they learn natural evolution which is not focused on maximizing protein function."

Is there an agent app that can take a paper and make a full deck out of the figures? If not, can someone build it? Would be super helpful for presentations and course slides Feels like it shouldn’t be too hard

The secret to my success? I make all of my papers look as impressive as possible without conveying any useful information whatsoever. They fly right into Nature and Cell and nobody can scoop us because there's no actual information in the paper.

I thought this was one of the most provocative images from the EVOLVEpro paper by Omar Abudayyeh and Jonathan Gootenberg. Supported by experimental data, they assert that a protein's fitness and its activity are lowly, or even negatively, correlated. How could this be? Shouldn't nature

See Simon Barnett and our team at DIMENSION's thinking on protein fitness vs. activity and how frontier work from Jonathan Gootenberg and Omar Abudayyeh is pushing machine learning theory towards few-shot experimental practice for protein design.

🎉 Today, Arc welcomes our newest Core Investigator, Dr. Felix Horns! Dr. Horns joins the Arc community as a Core Investigator and Stanford University as Assistant Professor of Genetics. The Horns lab works at the interface of synthetic biology and genomics to discover the fundamental

Leading minds in aging research attended the LBF and Foresight Longevity Workshop. Join us at Vision Weekend USA & continue the Longevity Biotech conversation. Get your tickets here: foresight.org/vw2024us/ Max Novendstern Omar Abudayyeh Alexander Fedintsev Nikola Markov

Thanks to all who joined for our fourth Boston longevity biotech mixer! Deep gratitude to our speakers Ulrike Trendelenburg from Novartis DARe and Omar Abudayyeh + Jonathan Gootenberg from Harvard Medical School; our co-sponsors Hevolution Foundation; and our gracious hosts at Boston Consulting Group. Can’t wait for next time!

Why is it that patents last only 20 years, but copyrights can stretch over 150 years? 🤔 Is this really about protecting creativity, or is it more about big corporations? Seems like the Mickey Mouse lobby might have something to do with it... Crazy politics.stackexchange.com/questions/1810…

zero shot binder design is not drug discovery zero shot binder design constitutes hit discovery at best, and is more academic exercise than translational practice the present day value of machine learning for drug discovery is in improving cost/time/quality of design cycles

September 3rd — Kaiyi Jiang , EVOLVEpro September 17th — Jeff Ruffolo , ProseLM October 1st — Amy Lu , CHEAP October 15th — Kapil Devkota , Ray-gun October 29th — Andre Cornman , The OMG dataset & gLM With more announced soon✨

Announcing our first speaker of the fall!! Join us this Tuesday, 9/03, to hear Kaiyi Jiang present EVOLVEpro: rapid protein evolution by few-shot learning with a PLM read the paper: biorxiv.org/content/10.110… as always, sign up on our website to get zoom links via email or DM us!