EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Structure-Based Drug Design Framework using Graph Neural Networks and Molecular Dynamics Simulation

Haruto Takahashi, Ren Ito

American Journal of Artificial Intelligence and Neural Networks · 28.02.2026 · 7 (1) · 1-12

DOI: 10.71465/ajainn3543 ↗

Kurz gefasst

Fachartikel in der Zeitschrift American Journal of Artificial Intelligence and Neural Networks, online erschienen am 28.02.2026. Verfasst von Haruto Takahashi und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The process of discovering novel pharmaceutical compounds is traditionally a capital-intensive and time-consuming endeavor, often plagued by high attrition rates in late-stage clinical trials. Structure-based drug design has emerged as a pivotal methodology to mitigate these risks by leveraging the three dimensional information of target proteins. However, conventional docking algorithms frequently struggle with the accurate prediction of binding affinities due to their reliance on static structural representations and simplified scoring functions. This paper proposes a novel framework that synergizes the representational power of Graph Neural Networks with the temporal resolution of Molecular Dynamics simulations. By treating molecular complexes as topological graphs, the proposed deep learning architecture captures intricate spatial dependencies and chemical interactions that are often overlooked by linear descriptors. Furthermore, the integration of short-trajectory simulations allows for the incorporation of entropic contributions and conformational flexibility into the predictive model. The experimental results demonstrate that this hybrid approach significantly outperforms traditional docking baselines and standalone static deep learning models in predicting binding free energies. The framework offers a robust solution for virtual screening campaigns, effectively balancing computational efficiency with physical accuracy.

Zitieren

APA

Takahashi, H., & Ito, R. (2026). A Structure-Based Drug Design Framework using Graph Neural Networks and Molecular Dynamics Simulation. American Journal of Artificial Intelligence and Neural Networks, 7(1), 1-12. https://doi.org/10.71465/ajainn3543

Vancouver

Takahashi H, Ito R. A Structure-Based Drug Design Framework using Graph Neural Networks and Molecular Dynamics Simulation. American Journal of Artificial Intelligence and Neural Networks. 2026;7(1):1-12. doi:10.71465/ajainn3543

BibTeX
@article{Takahashi2026,
  author = {Takahashi, Haruto and Ito, Ren},
  title = {A Structure-Based Drug Design Framework using Graph Neural Networks and Molecular Dynamics Simulation},
  journal = {American Journal of Artificial Intelligence and Neural Networks},
  year = {2026},
  volume = {7},
  number = {1},
  pages = {1--12},
  doi = {10.71465/ajainn3543}
}
Kostenlos starten →