Patrick Bryant

The future of drug design is in AI. RareFoldGPCR: Agonist Design Beyond Natural Amino Acids.
Paper: www.biorxiv.org/content/10.1...
Code: github.com/patrickbryan...

RareFold: Structure prediction and design of proteins with noncanonical amino acids Protein structure prediction and design have traditionally been limited to the 20 canonicalamino acids. Expanding this space to include noncanonical amino acids (NCAAs) offers newopportunities for pro...

Update: RareFold 🧬
Our AI framework for protein design with 29 noncanonical AAs now shows designed binders (linear + cyclic) are non-immunogenic in patient-derived assays — paving the way for safe next-gen peptide therapeutics.
👉https://www.biorxiv.org/content/10.1101/2025.05.19.654846v2

GitHub - patrickbryant1/EvoBind: In silico directed evolution of peptide binders with AlphaFold In silico directed evolution of peptide binders with AlphaFold - patrickbryant1/EvoBind

This is the first study from my lab - providing the first PoC for both linear and cyclic untargeted binder design! EvoBind is available here: github.com/patrickbryan...

Our study where we develop EvoBind2: Design of linear and cyclic peptide binders from protein sequence information is now published! www.nature.com/articles/s42...

Cool! Congrats @proteinator.bsky.social 🎉

We have much more coming in this space where we can identify target interfaces and inhibit the interactions - all using protein structure prediction!

Now published: our study on human-pathogen protein-protein interactions! We identify 30 interactions with an expected TM-score ≥0.9, tripling the structural coverage in these networks. One novel interaction was validated with mass spectrometry. journals.plos.org/ploscompbiol...

Our latest work is out: we designed dual GLP1R/GCGR agonists—cyclic peptides that activate both metabolic receptors, entirely from sequence alone.
This has never been done before. www.biorxiv.org/content/10.1...

You can also design in Colab now: colab.research.google.com/github/patri...

The WT binder is 1.8 uM which means that we create as good binders but with new modes of binding for a target where these NCAA interactions are completely unknown 😎

Thanks! We will release a lot of new tech this year - stay tuned! We are only in the beginning of protein design I think

There is also an audio summary generated by Science Cast here: sciencecast.org/casts/4kj8lr...

Just like we have used EvoBind to e.g. create functional HIV inhibitors in a single shot (biorxiv.org/content/10.1...) we can now do this with an expanded vocabulary to have more chemical possibilities and avoid e.g. immune recognition and degradation

RareFold supports 49 different AAs.
The 20 regular, and 29 rare ones: MSE, TPO, MLY, CME, PTR, SEP,SAH, CSO, PCA, KCX, CAS, CSD, MLZ, OCS, ALY, CSS, CSX, HIC, HYP, YCM, YOF, M3L, PFF, CGU,FTR, LLP, CAF, CMH, MHO.
You can simply specify which you want to use and design!

Happy to release our breakthrough AI-model: RareFold, which predicts and designs proteins with noncanonical AAs. With EvoBindRare, we designed linear & cyclic peptide binders with high affinity & novel binding modes, wet lab validated.

📄 biorxiv.org/content/10.1...
💻 github.com/patrickbryan...