Researchers unveil TangleDiff to design entangled protein hydrogels
A team of authors introduced TangleDiff, a deep‑learning framework that creates homodimeric entangled proteins with programmable features. In silico, TangleDiff generates diverse foldable sequences with a success rate exceeding 70%, far above the roughly 1% reported for current models. When conditioned on inter‑chain binding energy, about 70% of the successful designs fall within the targeted…
Key points
- TangleDiff achieves in‑silico success exceeding 70%, versus ~1% for existing models.
- Conditioning on binding energy yields approximately 70% of designs matching target ranges.
- Experimental validation of nine homodimers produced seven hydrogels with expected stress‑relaxation behavior.
The researchers experimentally tested nine designed homodimers covering various binding energies. Seven of these homodimers formed hydrogels, and the observed stress‑relaxation dynamics aligned with the specified binding‑energy values. The work demonstrates a general strategy for entangled protein design, potentially expanding biomaterial innovation for 3D stem‑cell and organoid culture.
Model page: TangleDiff →
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us.
More in Research
All →- OpenAI's GPT-6 Astra and Anthropic's Claude Fable fail safety test in RoboHarm benchmark · 2 src
- TatBLiMP benchmarks Tatar linguistic minimal pairs · 1 src
- Recursive language models generalize out of domain, study shows · 1 src
- Reviser proposes cursor-based text generation · 1 src
- SAGE system raises grant review agreement to kappa 0.58, beating baseline · 1 src
Comments
via GitHub Discussions