I’m excited to share our significantly-updated preprint on de novo antibody design, where we now demonstrate the structurally accurate design of scFvs (in addition to VHHs) with RFdiffusion!

Finally, for the past year or so, some of us have been working to build the next generation of these tools at Xaira Therapeutics (@Xaira_Thera), with the aim of fully realising the promise of therapeutics design. If you’re interested in being a part of it, do get in touch!.

In particular, thank you to the other lead authors, @naterbennett0, @RobertRagotte, @AndrewJBorst, @dejsee, Connor Weidle, @wanderingriti and Yutong Yu! And thanks as always to David Baker for the leadership and for creating such a vibrant and exciting scientific environment!.

Altogether, we believe this work is a major step towards routine therapeutic antibody design. This was an amazing collaboration with so many passionate and bright scientists at @UWproteindesign and beyond. Thanks to all the authors!.

The second is a challenging peptide-MHC complex with clinical relevance. We demonstrate that RFdiffusion can design scFvs that bind. What’s more, mutating a single amino acid on the target peptide totally abolished binding, highlighting the specificity possible with RFdiffusion.

The first has no known VHH or antibody binder to it in the training set. Two structures (one high resolution) demonstrate structural accuracy of the designs.

For many clinical applications, two-Fv antibodies are preferred (as these are the human antibody format). We therefore combined our structure-based design method and novel experimental techniques to design scFv binders to two targets.

A year ago, we posted a preprint demonstrating that a fine-tuned RFdiffusion network could design de novo VHHs (nanobodies). We’ve extended that work, and demonstrate the structurally accurate design of a VHH to an epitope without such a binder in the PDB (i.e., the training set)

We are also releasing inference code for the whole pipeline, free and fully open source!.

This is work performed at @UWproteindesign in collaboration with a plethora of amazing scientists!.

All the big things happening to all the best Davids!.Congratulations David 😍.

Congratulations to all - incredibly deserved! It was a total and constant honour working with David. His passion for protein design and its applications, as well as his uncanny attention to detail continue to inspire me. And nothing we do today would be possible without AlphaFold.

Very exciting to see this work out! It still feels quite amazing that you can take three functional bits of protein and bring them all together with novel scaffolds, all while retaining their native function. Congratulations to @karla_mcastro for leading this super cool project!.

A great day for science!.Congratulations to the teams at @GoogleDeepMind @IsomorphicLabs!!.

RT @brianltrippe: I am pleased to share that I have accepted a position @Stanford to start as an assistant professor in the department of s….

We're extremely grateful and excited to have the backing to try and realise this vision! I'm very excited to work with our growing team of amazing people in the UK and US. Read more about the project here:.

I'm excited to announce that I'm part of the team at Xaira Therapeutics!. This project has been some time in the making. I'm convinced that generative modelling, and ML for biology more generally, will play a pivotal role in the next generation of therapeutics. 1/2.

RT @MathieuEmile: @CalSleeper really love travelling on your sleeper trains! Why not allow for booking 1 berth in cabins whilst keeping the….

Congrats to my amazing PhD supervisor Manu!!.

RT @json_yim: Our review on diffusion models for protein structures and docking is out @HannesStaerk @GabriCorso Bowen Jing @BarzilayRegina….