Generating protein hydrogels with customizable stress relaxation behavior via deep learning-driven entanglement design.
6 affiliations
- 1Department of Chemical and Biological Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong.
- 2Department of Data Science, City University of Hong Kong, Kowloon, Hong Kong.
- 3Beijing National Laboratory for Molecular Sciences, Key Laboratory of Polymer Chemistry & Physics of Ministry of Education, Center for Soft Matter Science and Engineering, College of Chemistry and Molecular Engineering, Peking University, Beijing, P. R. China.
- 4School of AI for Science (AI4S), Shenzhen Graduate School, Peking University, Shenzhen, P. R. China.
- 5Department of Chemical and Biological Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong. [email protected].
- 6Department of Chemical and Biological Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong. [email protected].
Abstract
Protein hydrogels are promising artificial extracellular matrices (ECMs) for 3D stem cell and organoid culture due to their favorable stress relaxation behavior (a decrease in stress in response to strain). Inter-chain entangled motifs, in which different protein chains are interlaced, represent a powerful strategy to synthesize such hydrogels. However, designing these motifs with tailored properties such as binding energy remains a major challenge due to the difficulty of simultaneously controlling these properties while ensuring entanglement. Here, we introduce TangleDiff, a deep learning framework for the de novo design of homodimeric entangled proteins with programmable features. TangleDiff generates diverse foldable entangled sequences with an in-silico success rate exceeding 70%, markedly outperforming current models (~1%). By conditioning TangleDiff on inter-chain binding energy, we generate novel protein dimers whose binding energies closely match the specified ranges, with approximately 70% of successful designs conforming to expected values. We experimentally validate TangleDiff by designing nine homodimers targeting various binding energies; seven successfully form hydrogels, with stress relaxation dynamics correlated with specified binding energies. This work establishes a general strategy for entangled protein design, opening avenues for entanglement-based biomaterial innovation.
Related
- Modeling and Characterization of Stress Relaxation Behavior of EVA–CNT Nanocomposites Under Uniaxial Loading
- Silicon‑Noded Cross-Ligation Platform Enabled by Customizable and Moisture-Stable Bis(polyfluoroalkoxy)silanes
- Flame-Exposed Aluminium Foil as a Low-Cost, Customizable TEM Grid for Nanomaterial Imaging
- Design-guided synthesis of customizable carbonized electrodes for maximizing charge storage in carbon-based supercapacitors
- Efficient Customizable Synthesis Planning with Graph Grammars, with an application to Elastase B inhibitor candidates
BibTeX
@article{deng2026generating,
title = {{Generating protein hydrogels with customizable stress relaxation behavior via deep learning-driven entanglement design.}},
author = {Puqing Deng and Yutong Wu and Hong Kiu Francis Fok and Linyan Li and Wen-Bin Zhang and Fei Sun and Hanyu Gao},
journal = {Nature communications},
year = {2026},
month = sep,
doi = {10.1038/s41467-026-77607-9},
url = {https://europepmc.org/article/MED/42907456},
note = {CC BY},
}
RIS
TY - JOUR
TI - Generating protein hydrogels with customizable stress relaxation behavior via deep learning-driven entanglement design.
AU - Puqing Deng
AU - Yutong Wu
AU - Hong Kiu Francis Fok
AU - Linyan Li
AU - Wen-Bin Zhang
AU - Fei Sun
AU - Hanyu Gao
PY - 2026
DA - 2026/09/21
JO - Nature communications
DO - 10.1038/s41467-026-77607-9
UR - https://europepmc.org/article/MED/42907456
AB - Protein hydrogels are promising artificial extracellular matrices (ECMs) for 3D stem cell and organoid culture due to their favorable stress relaxation behavior (a decrease in stress in response to strain). Inter-chain entangled motifs, in which different protein chains are interlaced, represent a powerful strategy to synthesize such hydrogels. However, designing these motifs with tailored properties such as binding energy remains a major challenge due to the difficulty of simultaneously controlling these properties while ensuring entanglement. Here, we introduce TangleDiff, a deep learning framework for the de novo design of homodimeric entangled proteins with programmable features. TangleDiff generates diverse foldable entangled sequences with an in-silico success rate exceeding 70%, markedly outperforming current models (~1%). By conditioning TangleDiff on inter-chain binding energy, we generate novel protein dimers whose binding energies closely match the specified ranges, with approximately 70% of successful designs conforming to expected values. We experimentally validate TangleDiff by designing nine homodimers targeting various binding energies; seven successfully form hydrogels, with stress relaxation dynamics correlated with specified binding energies. This work establishes a general strategy for entangled protein design, opening avenues for entanglement-based biomaterial innovation.
SN - 2041-1723
ER -