EUVIMEDEUVIMED entdecken
Open Access

Multiscale graph equivariant diffusion model for 3D molecule design

Lu Chen iD, Yan Li, Yanjie Ma iD, Lin Gao, Liang Yu iD

Science Advances · 18.04.2025 · 11 (16)

DOI: 10.1126/sciadv.adv0778 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Science Advances, erschienen am 18.04.2025. Verfasst von Lu Chen und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science and Technology, Xidian University, Xi'an 710071, Shaanxi, China. Die Arbeit wurde bisher 17-mal zitiert (Crossref) und nennt 70 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Three-dimensional molecular generation is critical in drug design. However, current methods often rely on point clouds or oversimplified interaction models, limiting their ability to accurately represent molecular structures. To address these challenges, this paper proposes the multiscale graph equivariant diffusion model for 3D molecule design (MD3MD). MD3MD partitions molecular conformations into multiscale graphs, assigning different weights to capture atomic interactions across scales. This framework guides the diffusion process, enabling high-quality 3D molecular generation. Experimental results demonstrate that MD3MD excels in both unconditional and conditional generation tasks, producing diverse, stable, and innovative molecules that meet specified conditions. Visualization highlights MD3MD’s ability to learn domain-specific patterns and generate molecules distinct from existing datasets while maintaining distributional consistency. By effectively exploring chemical space, MD3MD surpasses previous methods in generating innovative and chemically diverse molecules, offering a notable advancement in the field of molecular design.

Autor:innen und Institutionen

  • School of Computer Science and Technology, Xidian University, Xi'an 710071, Shaanxi, China.
  • School of Management, Xi'an Polytechnic University, Xi'an 710000, Shaanxi, China.

ORCID-Profile: Lu Chen, Yanjie Ma, Liang Yu

Zitieren

APA

Chen, L., Li, Y., Ma, Y., Gao, L., & Yu, L. (2025). Multiscale graph equivariant diffusion model for 3D molecule design. Science Advances, 11(16). https://doi.org/10.1126/sciadv.adv0778

Vancouver

Chen L, Li Y, Ma Y, Gao L, Yu L. Multiscale graph equivariant diffusion model for 3D molecule design. Science Advances. 2025;11(16). doi:10.1126/sciadv.adv0778

BibTeX
@article{Chen2025,
  author = {Chen, Lu and Li, Yan and Ma, Yanjie and Gao, Lin and Yu, Liang},
  title = {Multiscale graph equivariant diffusion model for 3D molecule design},
  journal = {Science Advances},
  year = {2025},
  volume = {11},
  number = {16},
  doi = {10.1126/sciadv.adv0778}
}
Kostenlos starten →