Hierarchical discrete representations for coarse-to-fine protein conformation generation
Bioinformatics · 2026
DOI: 10.1093/bioinformatics/btag611 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, 2026. Verfasst von SeokJun On und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Artificial Intelligence, Hanyang University, Seoul 04763, Republic of Korea. Hauptthemen laut MEDLINE: Protein Conformation, Proteins, Computational Biology. Die Arbeit nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Institute for Information & communications Technology Planning & Evaluation, Korean government.
Abstract (Originalsprache)
Motivation Protein conformation generation remains a fundamental challenge in structural biology and machine learning. Proteins are highly dynamic macromolecules, and their functions are governed not only by static 3D structures but also by their conformational flexibility. Therefore, generating a diverse ensemble of physically plausible structures is crucial for modeling conformational heterogeneity, characterizing intrinsically disordered proteins (IDPs), and supporting structure-based drug discovery. However, existing generative models often operate in continuous coordinate space. This approach makes it difficult to impose structural priors or handle complex, long-range dependencies, frequently leading to a significant trade-off where models must sacrifice structural validity to ensure conformational diversity. Results To overcome these limitations, we propose a novel approach that learns hierarchical discrete representations of protein structures using vector quantization. Inspired by the success of VQ-VAE models in vision, we discretize residue-level structural contexts into learnable codebooks at multiple levels of granularity. Our framework first constructs a coarse-grained scaffold that captures global topology and secondary structures, and subsequently conditions on this scaffold to progressively refine the fine-grained local geometry. This coarse-to-fine generation mechanism enables both efficient sampling and highly accurate reconstruction. Through extensive experiments, we demonstrate that our method outperforms state-of-the-art models such as ESMDiff across challenging benchmark datasets, including BPTI MD trajectories and conformational-changing pairs (apo/holo and fold-switch). Our model mitigates the trade-off between conformational diversity and structural validity, generating realistic ensembles while providing interpretable and scalable representations of protein geometry. Ultimately, this work offers a powerful new paradigm for flexible biomolecular modeling. Availability and implementation The code is available at https://github.com/vanha9/hierarchical_conformation_generation. Archived at https://zenodo.org/records/21817083.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Department of Artificial Intelligence, Hanyang University, Seoul 04763, Republic of Korea.
- Department of Computer Science, Hanyang University, Seoul 04763, Republic of Korea.
ORCID-Profile: SeokJun On, Yujin Jeong, Eun-Sol Kim
Förderung
- Institute for Information & communications Technology Planning & Evaluation
- Korean government
Zitieren
APA
On, S., Jeong, Y., Kim, K. H., Kang, K., & Kim, E. S. (2026). Hierarchical discrete representations for coarse-to-fine protein conformation generation. Bioinformatics. https://doi.org/10.1093/bioinformatics/btag611
Vancouver
On S, Jeong Y, Kim KH, Kang K, Kim ES. Hierarchical discrete representations for coarse-to-fine protein conformation generation. Bioinformatics. 2026. doi:10.1093/bioinformatics/btag611
BibTeX
@article{On2026,
author = {On, SeokJun and Jeong, Yujin and Kim, Kang-Hyeon and Kang, Kyungheon and Kim, Eun-Sol},
title = {Hierarchical discrete representations for coarse-to-fine protein conformation generation},
journal = {Bioinformatics},
year = {2026},
doi = {10.1093/bioinformatics/btag611}
}
EUVIMED