Lightnews — Scholar-powered news

Tristan Bepler @tbepler.bsky.social · Aug 26

Our preprint on sequence-to-property learning and zero-shot fitness prediction with PoET-2 is live: arxiv.org/abs/2508.04724

PoET-2 is also open sourced on github: github.com/OpenProteinA...

Thanks to the @openprotein.bsky.social team!

Understanding protein function with a multimodal retrieval-augmented foundation model

Protein language models (PLMs) learn probability distributions over natural protein sequences. By learning from hundreds of millions of natural protein sequences, protein understanding and design capa...

arxiv.org

3

Reposted by Tristan Bepler

openprotein.bsky.social @openprotein.bsky.social · Jul 25

Boltz-1 & Boltz-2 now live via GUI & APIs! Predict protein, protein–RNA/DNA/ligand structures with confidence scores & binding affinity metrics for virtual screening. Compare finetuned models in the new overview page to find your best performer fast.
www.openprotein.ai/early-access...

3 6

Reposted by Tristan Bepler

openprotein.bsky.social @openprotein.bsky.social · Jun 25

Product update: Indel Analysis lets you score insertions/deletions across your sequence using PoET-2. You can now also compare multiple 3D structures in Mol* to evaluate design alternatives.
Sign up now: www.openprotein.ai/early-access...

1 1

Tristan Bepler @tbepler.bsky.social · Jun 20

How would you use a tool like this? Do you design or screen indels in your work? 4/4

Tristan Bepler @tbepler.bsky.social · Jun 20

Indels are still a major challenge for variant effect prediction and protein design. PoET-2 has significantly improved the state-of-the-art for functional and clinical indel variant effect prediction. 3/4

PoET-2 is state-of-the-art for zero-shot variant effect prediction of both DMS indels and clinical indels in ProteinGym.

1

Tristan Bepler @tbepler.bsky.social · Jun 20

It supports screening deletions, insertion sites, and replacement sites. Explore viable shortened proteins, or insert new structural or functional sequences like localization signals or structural tags. 2/4

1

Tristan Bepler @tbepler.bsky.social · Jun 20

Why does no one in AI protein engineering work on indels?

We’re solving this at OpenProtein.AI. Check out our upcoming indel design tool! 🤩 1/4

@openprotein.bsky.social

1 1 3

Reposted by Tristan Bepler

Pascal Notin @pascalnotin.bsky.social · May 8

Have we hit a "scaling wall" for protein language models? 🤔 Our latest ProteinGym v1.3 release suggests that for zero-shot fitness prediction, simply making pLMs bigger isn't better beyond 1-4B parameters. The winning strategy? Combining MSAs & structure in multimodal models!

1 7 25

Tristan Bepler @tbepler.bsky.social · May 14

Great to see this comparison with genome language models. The hype around these models seems to have strongly outstripped where they actually are in comparison with protein models.

1

Reposted by Tristan Bepler

openprotein.bsky.social @openprotein.bsky.social · May 13

Product update: PoET-2 now supports structure inputs for enhanced prediction and design via Python APIs. Check out our new inverse folding tutorial to see it in action.
🔗 docs.openprotein.ai/walkthroughs...

Sign up for OpenProtein.AI: www.openprotein.ai/early-access...

Inverse Folding with PoET-2 for Generation of Novel Luciferases — OpenProtein-Docs documentation

docs.openprotein.ai

1 1

Tristan Bepler @tbepler.bsky.social · May 12

Huge thanks to the @openprotein.bsky.social team! We've got more exciting PoET-2 updates to come 🚀

Tristan Bepler @tbepler.bsky.social · May 12

Learn more about PoET-2 in our whitepaper: www.openprotein.ai/a-multimodal...

A multimodal foundation model for controllable protein generation and representation learning

PoET-2 is a next generation protein language model that transforms our ability to engineer proteins by learning from nature’s design principles. Through its unique multimodal architecture and training...

www.openprotein.ai

1

Tristan Bepler @tbepler.bsky.social · May 12

Sign up for OpenProtein.AI (free for academic use): www.openprotein.ai/early-access...

and install the python client to get started: github.com/OpenProteinA...

Sign up for early access | OpenProtein.AI

Join the revolution in protein research with early access to our cutting-edge Open Protein AI platform. Sign up now to explore the future of protein analysis and discovery.

www.openprotein.ai

1

Tristan Bepler @tbepler.bsky.social · May 12

Generative protein sequence design, variant effect prediction, and fine-tuning are now fully supported for PoET-2 with structure and sequence prompts in the @openprotein.bsky.social python client and APIs!

Check out our new walkthrough on inverse folding: docs.openprotein.ai/walkthroughs...

Inverse Folding with PoET-2 for Generation of Novel Luciferases — OpenProtein-Docs documentation

docs.openprotein.ai

1 2 3

Reposted by Tristan Bepler

SynBioBeta @synbiobeta.bsky.social · Feb 21

🧬 Protein Revolution: The Tiny Model Making a Massive Impact!
PoET-2 is changing the game in computational protein design, slashing experimental data needs by 30x! 🚀

learn more: www.synbiobeta.com/read/protein...

#ProteinDesign #BiotechInnovation #AIRevolution

Protein Revolution: The Tiny Model Making a Massive Impact - SynBioBeta

www.synbiobeta.com

2 1

Tristan Bepler @tbepler.bsky.social · Feb 11

Huge thanks to our incredible team @openprotein.bsky.social, especially Tim Truong. This is just the beginning of AI systems that truly understand protein biology.

I can’t wait to see what the community can do with these models! 13/13

2

Tristan Bepler @tbepler.bsky.social · Feb 11

Read more in our white paper: www.openprotein.ai/a-multimodal...
12/13

A multimodal foundation model for controllable protein generation and representation learning

PoET-2 is a next generation protein language model that transforms our ability to engineer proteins by learning from nature’s design principles. Through its unique multimodal architecture and training...

www.openprotein.ai

1 1 2

Tristan Bepler @tbepler.bsky.social · Feb 11

This has huge implications for protein engineering - from more efficient directed evolution and multiproperty optimization to de novo protein design. 11/13

1 1

Tristan Bepler @tbepler.bsky.social · Feb 11

Most importantly, PoET-2 gets us closer to understanding the sequence-structure-function relationship - learning from just a handful of examples to predict properties and design new sequences. 10/13

1 1

Tristan Bepler @tbepler.bsky.social · Feb 11

Beyond predictions, PoET-2 introduces a powerful prompt grammar for protein generation. One model for: free sequence generation, inverse folding, motif scaffolding, and more! 9/13

1 1

Tristan Bepler @tbepler.bsky.social · Feb 11

The results show PoET-2 has learned fundamental principles:
* Improves sequence and structure understanding
* Accurate zero-shot function prediction, especially for insertions and deletions
* 30x less data needed for transfer learning
8/13

1 1

Tristan Bepler @tbepler.bsky.social · Feb 11

This lets us break conventional scaling laws. PoET-2 achieves with 182M parameters what would require trillion-parameter models using standard architectures. 7/13

1 1

Tristan Bepler @tbepler.bsky.social · Feb 11

Enabled by our tiered attention structure, PoET-2 processes sequence families with order equivariance while preserving long-range dependencies within and between sequences, enabling processing of large sequence families with optional structural data. 6/13

1 1

Tristan Bepler @tbepler.bsky.social · Feb 11

Rather than regurgitating databases, PoET-2 "meta-learns" evolutionary principles through in-context learning - inferring structural and functional constraints at inference time from small numbers of examples. 5/13

1 1

Tristan Bepler @tbepler.bsky.social · Feb 11

PoET-2 takes a different approach. Instead of massive scale, we developed a multimodal architecture that learns to reason about sequences, structures, and evolutionary relationships simultaneously. 4/13

1 1 3