Adaptive Multi-Kernel Graph Neural Network for Drug-Drug Interaction Prediction
Interdisciplinary Sciences: Computational Life Sciences · 2025-06 · 17 (2) · 409-423
DOI: 10.1007/s12539-024-00684-1 ↗Kurz gefasst
Fachartikel in der Zeitschrift Interdisciplinary Sciences: Computational Life Sciences, online erschienen am 28.01.2025. Verfasst von Linqian Zhao und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science, Qufu Normal University, Rizhao, 276826, China. Hauptthemen laut MEDLINE: Neural Networks, Computer. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 47 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Combination therapy, which synergistically enhances treatment efficacy and inhibits disease progression through the combined effects of multiple drugs, has emerged as a mainstream approach for treating complex diseases and alleviating symptoms. However, drug-drug interactions (DDIs) can sometimes lead to adverse reactions, potentially endangering lives. Therefore, developing efficient and accurate DDI prediction methods is crucial for elucidating drug mechanisms and preventing side effects. Current prediction methods often focus solely on the presence of interactions between drugs when constructing DDI graphs, neglecting the specific types of DDIs. This oversight can result in a decline in predictive performance. To address this issue, we propose an Adaptive Multi-Kernel Graph Neural Network (AMKGNN) for DDI prediction. AMKGNN differentiates DDIs into increase-type and decrease-type interactions, constructing separate increased DDI and decreased DDI graphs as convolutional kernels. AMKGNN employs a graph kernel learning mechanism that adaptively determines the optimal threshold between high-frequency and low-frequency signals in the network to capture node embeddings. Initially, AMKGNN learns drug embedding representations based on these two graph convolutional kernels and various drug features. These representations are then concatenated and input into a deep neural network to predict potential DDIs. The results show that our model achieved AUC and AUPR values above 90% across three sub-tasks on two datasets, significantly outperforming the other five comparison models. Furthermore, ablation experiments and case studies validate the superiority of AMKGNN.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Computer Science, Qufu Normal University, Rizhao, 276826, China.
- School of Computer Science, Qufu Normal University, Rizhao, 276826, China. shangjunliang110@163.com.
- School of Information and Control Engineering, Qingdao University of Technology, Qingdao, 266520, China.
- School of Health and Life Science, University of Health and Rehabilitation Sciences, Qingdao, 266113, China.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Zhao, L., Shang, J., Meng, X., He, X., Zhang, Y., & Liu, J. X. (2025). Adaptive Multi-Kernel Graph Neural Network for Drug-Drug Interaction Prediction. Interdisciplinary Sciences: Computational Life Sciences, 17(2), 409-423. https://doi.org/10.1007/s12539-024-00684-1
Vancouver
Zhao L, Shang J, Meng X, He X, Zhang Y, Liu JX. Adaptive Multi-Kernel Graph Neural Network for Drug-Drug Interaction Prediction. Interdisciplinary Sciences: Computational Life Sciences. 2025;17(2):409-423. doi:10.1007/s12539-024-00684-1
BibTeX
@article{Zhao2025,
author = {Zhao, Linqian and Shang, Junliang and Meng, Xianghan and He, Xin and Zhang, Yuanyuan and Liu, Jin-Xing},
title = {Adaptive Multi-Kernel Graph Neural Network for Drug-Drug Interaction Prediction},
journal = {Interdisciplinary Sciences: Computational Life Sciences},
year = {2025},
volume = {17},
number = {2},
pages = {409--423},
doi = {10.1007/s12539-024-00684-1}
}
EUVIMED