A Deep Learning Framework Integrating Graph Theory and Dynamical Systems for Predictive Modeling of Drug-Target Interactions
International Journal of Drug Delivery Technology · 04.06.2026 · 16 (45s)
DOI: 10.25258/ijddt.16.45s.70 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Drug Delivery Technology, online erschienen am 04.06.2026. Verfasst von Durai Selvaraj und 9 weiteren Autor:innen.
Abstract (Originalsprache)
Accurate prediction of drug-target interactions (DTIs) is critical for drug repurposing and discovery. Traditional computational methods often overlook the complex, time-evolving nature of molecular binding. This paper proposes a novel deep learning framework, Graph-Dynamical Net (GDyn-Net), which integrates graph theory for structural representation and dynamical systems theory for modeling temporal binding processes. We model drugs and targets as bipartite heterogeneous graphs with node features derived from molecular fingerprints and protein sequences. A Graph Neural Network (GNN) extracts spatial embeddings, while a Neural Ordinary Differential Equation (Neural ODE) module captures the continuous-time dynamics of interaction probabilities. Experiments on three benchmark datasets (Davis, KIBA, BindingDB) show that GDyn-Net achieves an AUC of 0.972 and AUPR of 0.965, outperforming state-of-the-art models. Our framework provides interpretable latent trajectories of drug-target binding, offering new insights into the affinity maturation process.
Zitieren
APA
Selvaraj, D., Krishnapriya, R. K., P, S., C, M. T., Karunakaran, S., Muthu, V., Praveen, N., Vengatesh, T., Gandhiraja, R., & Anbuselvan, B. (2026). A Deep Learning Framework Integrating Graph Theory and Dynamical Systems for Predictive Modeling of Drug-Target Interactions. International Journal of Drug Delivery Technology, 16(45s). https://doi.org/10.25258/ijddt.16.45s.70
Vancouver
Selvaraj D, Krishnapriya RK, P S, C MT, Karunakaran S, Muthu V, et al. A Deep Learning Framework Integrating Graph Theory and Dynamical Systems for Predictive Modeling of Drug-Target Interactions. International Journal of Drug Delivery Technology. 2026;16(45s). doi:10.25258/ijddt.16.45s.70
BibTeX
@article{Selvaraj2026,
author = {Selvaraj, Durai and Krishnapriya, R. Karunia and P, Soumya and C, Mujeeburahman T. and Karunakaran, S. and Muthu, V. and Praveen, Neti and Vengatesh, T. and Gandhiraja, R. and Anbuselvan, B.},
title = {A Deep Learning Framework Integrating Graph Theory and Dynamical Systems for Predictive Modeling of Drug-Target Interactions},
journal = {International Journal of Drug Delivery Technology},
year = {2026},
volume = {16},
number = {45s},
doi = {10.25258/ijddt.16.45s.70}
}
EUVIMED