Accurate structural modelling of protein-glycan recognition with GlycoWeave

Abstract

Protein-glycan recognition underlies diverse biological processes, yet accurate prediction of protein-glycan complex structures remains challenging owing to the chemical diversity, conformational flexibility and sparse experimental structural coverage of glycans. Here we present GlycoWeave, a glycan-specific diffusion framework that integrates chemically constrained conformational sampling, multiscale molecular representations and multi-source learning. By operating in internal-coordinate space, GlycoWeave preserves glycan connectivity and stereochemistry while efficiently exploring bound conformations. A heterogeneous equivariant network couples atom-level interaction geometry with monosaccharide-level structural context, whereas multi-source, multi-head training integrates experimental complexes with structures refined using classical and quantum mechanics-informed potentials. On a benchmark of 698 protein-glycan complexes, GlycoWeave achieves a 71.9% Top-25 high-quality success rate, compared with 19.2% for AlphaFold 3, while requiring only 0.34 s per sampled pose. GlycoWeave further preserves glycan structural validity, recovers native protein-glycan interactions and reproduces linkage-specific conformational preferences, maintaining high accuracy across diverse glycan architectures and protein environments. These results establish GlycoWeave as an accurate, efficient and chemically faithful framework for modelling protein-glycan recognition. GlycoWeave is freely available at https://cadd.zju.edu.cn/glycoweave.

FunFOLD5: combining localised templates and structural homology for accurate predictions of biologically relevant protein–ligand interactions
Recep Adiyaman, Liam James McGuffin
Bioinformatics10 Oct 2026Peer-reviewedCC BY 4.0Full textMolecular biology & gene regulationCrossrefPMC
Structural modelling and functional analysis support lipid binding by the dimeric E. coli DedA protein YqjA
T. J. Paige, J. Chen, A. Le Roy +9
bioRxiv20 Sep 2026PreprintCC BYFull textMolecular biology & gene regulationbioRxiv
Molecular architecture of surface-active ionic liquids governs protein structural integrity and drug recognition at engineered carbon quantum dot nano–bio interfaces
Iaimanda Nonglamin, Sugam Kumar, Sivaprasad Mitra
Frontiers in Molecular Biosciences9 Nov 2026Peer-reviewedCC BY 4.0Molecular biology & gene regulationFrontiers
Accurate quantification of protein concentrations and state populations by 2D 1H-N Q-BEST NMR
Yacine Ouali, Beate Bersch, Bernhard Brutscher
Journal of Biomolecular NMR14 Oct 2026Peer-reviewedCC BY 4.0Full textBiophysics Structural biologyHALarXiv
Structural basis of selective phospholipid recognition and local membrane deformation by CTL1
H. Jiang, A. Martinez-Jimenez, K. A. Hossain +10
bioRxiv2 Oct 2026PreprintAll rights reservedFull textBiophysicsbioRxiv

Cite

BibTeX RIS

BibTeX

@misc{zhao2026accurate,
  title = {{Accurate structural modelling of protein-glycan recognition with GlycoWeave}},
  author = {Huifeng Zhao and Jintu Zhang and Yifei Xiao and Youqiao Qian and Silong Zhai and Zichang Jin and Li Wang and Yuchao Wang and Hang Dong and Yifan Zhao and Zhenxing Wu and Hongyan Du and Yuansheng Huang and Bowei Zhao and Xiaorui Wang and Shukai Gu and Jike Wang and Zhongjun Li and Zhongtang Li and Tingjun Hou and Yu Kang},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.37410},
  eprint = {2610.37410},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.37410},
  note = {CC BY-NC-SA 4.0},
}

RIS

TY  - UNPB
TI  - Accurate structural modelling of protein-glycan recognition with GlycoWeave
AU  - Huifeng Zhao
AU  - Jintu Zhang
AU  - Yifei Xiao
AU  - Youqiao Qian
AU  - Silong Zhai
AU  - Zichang Jin
AU  - Li Wang
AU  - Yuchao Wang
AU  - Hang Dong
AU  - Yifan Zhao
AU  - Zhenxing Wu
AU  - Hongyan Du
AU  - Yuansheng Huang
AU  - Bowei Zhao
AU  - Xiaorui Wang
AU  - Shukai Gu
AU  - Jike Wang
AU  - Zhongjun Li
AU  - Zhongtang Li
AU  - Tingjun Hou
AU  - Yu Kang
PY  - 2026
DA  - 2026/10/27
PB  - arXiv
DO  - 10.48550/arxiv.2610.37410
UR  - https://arxiv.org/abs/2610.37410
AB  - Protein-glycan recognition underlies diverse biological processes, yet accurate prediction of protein-glycan complex structures remains challenging owing to the chemical diversity, conformational flexibility and sparse experimental structural coverage of glycans. Here we present GlycoWeave, a glycan-specific diffusion framework that integrates chemically constrained conformational sampling, multiscale molecular representations and multi-source learning. By operating in internal-coordinate space, GlycoWeave preserves glycan connectivity and stereochemistry while efficiently exploring bound conformations. A heterogeneous equivariant network couples atom-level interaction geometry with monosaccharide-level structural context, whereas multi-source, multi-head training integrates experimental complexes with structures refined using classical and quantum mechanics-informed potentials. On a benchmark of 698 protein-glycan complexes, GlycoWeave achieves a 71.9% Top-25 high-quality success rate, compared with 19.2% for AlphaFold 3, while requiring only 0.34 s per sampled pose. GlycoWeave further preserves glycan structural validity, recovers native protein-glycan interactions and reproduces linkage-specific conformational preferences, maintaining high accuracy across diverse glycan architectures and protein environments. These results establish GlycoWeave as an accurate, efficient and chemically faithful framework for modelling protein-glycan recognition. GlycoWeave is freely available at https://cadd.zju.edu.cn/glycoweave.
ER  -

Abstract

Related

Cite