New from Arc Institute: the first functional AI-generated genomes. This is a conceptual breakthrough, but it may also unlock new strategies for combating antibiotic resistance.

Evo-designed genomes are here!! HUGE congrats to Samuel King for fearlessly bringing this project to life. Check out the thread to learn more! ⬇️

Scientists have created the first ever viruses designed by AI, and they’re capable of hunting down and killing strains of E. coli go.nature.com/4mppvIi

We may have been making a lot of progress in genome design, but we are not done with protein design just yet lol

Efficient generation of epitope-targeted de novo antibodies with Germinal biorxiv.org/content/10.110… #biorxiv_synbio

In another preprint from the @brianhie Lab and Xiaojing Gao, they introduce Germinal, a generative AI system for de novo antibody design. Germinal produces functional nanobodies in just dozens of tests, making custom antibody design more accessible than ever before.

Today, we report Germinal, a method for efficient de novo antibody design, with Santiago Mille and Xiaojing Gao. Germinal achieves success rates of 4-22% across diverse epitopes. We make the work fully open, without doing lame things like posting a preprint without methods. 🧵

Having often dealt with binder-limited projects, we sought a more accessible source for nanobodies than yeast display or llama. Here we introduce Germinal, computationally designing antibody-like binders with such a hit rate that only tens need to be screened for each target.

Super excited to share what we’ve been working on in collaboration with Xiaojing Gao over the past few months on de novo antibody design. Check out this great thread by our team lead Santiago Mille highlighting the technical aspects of the pipeline!

Entering my PhD, de novo antibody design was a grand challenge I thought would not be solved without huge increases in affinity data and Ab-Ag structure. Only 2 years later, we provide the first open-source recipe to get antibody binders, almost magically, out of a computer (1/3)

Ensembles of models are unlocking new successes in de novo antibody design, congratulations to the team! The day these experiments first worked, the lab roared like a stadium.

When structure-based models alone are too rigid and sequence-based models alone are not specific enough, put them together for a phenomenal antibody design tool!

Still boggles my mind that we can leave frameworks as-is and only edit CDRs and still reliably get nM binders. Complete opposite of what I would have expected given how hard it is to graft CDRs from one FW to another (such as humanizing mouse mAbs). First RFantibody and now this

Great work by Brian Hie & Xiaojing Gao labs. Looks like a big leap over SOTA with very clever bootstrapping of design & optimizing using multiple models with complementary strengths & weaknesses. Super expt. validation platform as well.

Excited to share that my PhD thesis work is out in Science Magazine today. We demonstrate robust rearrangement of the human genome using bridge recombinases, performing programmable insertions, excisions, and inversions at megabase-scale.

Combine multimer structure prediction and an antibody language model to design de novo antibodies with nanomolar binding affinity. Santiago Mille John Wang Talal Widatalla Claudia Driscoll Haoyu Dai Xiaowei Zhang Xiaojing Gao Brian Hie

Over the past week, Arc Institute published three new discoveries that I’m very proud of. • The world's first functional AI-generated genomes. Using Evo 2 (the largest biology ML model ever trained, which Arc released in partnership with NVIDIA in February), Arc scientists

Excited to share this work from my PhD! See our structure below of a computationally designed IL-21 mimic with anti-tumor efficacy in mice. A collaboration between the David Baker, Stephanie & Mike Dougan, Warren Leonard, and Chris Garcia labs.

it's aliiveeeeeee thank you Arc Institute Santiago Mille et. al.