Drug-target interaction prediction based on graph convolutional autoencoder with dynamic weighting residual GCN

Abstract Background The exploration of drug-target interactions (DTIs) is a critical step in drug discovery and drug repurposing. Recently, network-based methods have emerged as a prominent research area for predicting DTIs. These methods excel by extracting both topological and feature information...

Full description

Saved in:
Bibliographic Details
Main Authors: Ming Zeng, Min Wang, Fuqiang Xie, Zhiwei Ji
Format: Article
Language:English
Published: BMC 2025-07-01
Series:BMC Bioinformatics
Subjects:
Online Access:https://doi.org/10.1186/s12859-025-06198-x
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849341936133996544
author Ming Zeng
Min Wang
Fuqiang Xie
Zhiwei Ji
author_facet Ming Zeng
Min Wang
Fuqiang Xie
Zhiwei Ji
author_sort Ming Zeng
collection DOAJ
description Abstract Background The exploration of drug-target interactions (DTIs) is a critical step in drug discovery and drug repurposing. Recently, network-based methods have emerged as a prominent research area for predicting DTIs. These methods excel by extracting both topological and feature information from DTIs networks, thereby achieving superior DTIs prediction performance. However, the majority of existing GCN-based methods utilize shallow graph neural networks, which are incapable of extracting higher-level semantic information. Additionally, the current training of models lacks an effective guiding mechanism, leading to the insufficient improvement of network’s representation capabilities. Results In this paper, we propose a graph convolutional autoencoder model, named DDGAE, for DTIs prediction. We develop a DWR-GCN module, which incorporates dynamic weighting graph convolution with residual connection, to improve the representation capability for DTI heterogeneous networks. Further, to improve the learning efficiency of the model, we devise a dual self-supervised joint training mechanism. Specifically, this mechanism integrates DWR-GCN and a graph convolutional autoencoder into a cohesive system, enhancing both the learning performance and stability of DDGAE. Conclusion Experimental results show that DDGAE significantly outperforms several SOTA models in DTIs prediction, achieving optimal performance and the reliability of our method is verified by case study.
format Article
id doaj-art-c9b441325e6d409dbc981ebfca8af0c6
institution Kabale University
issn 1471-2105
language English
publishDate 2025-07-01
publisher BMC
record_format Article
series BMC Bioinformatics
spelling doaj-art-c9b441325e6d409dbc981ebfca8af0c62025-08-20T03:43:31ZengBMCBMC Bioinformatics1471-21052025-07-0126112910.1186/s12859-025-06198-xDrug-target interaction prediction based on graph convolutional autoencoder with dynamic weighting residual GCNMing Zeng0Min Wang1Fuqiang Xie2Zhiwei Ji3School of Mathematics and Computer Science, Gannan Normal UniversitySchool of Intelligent Manufacturing and Future Energy, Gannan Noraml UniversitySchool of Mathematics and Computer Science, Gannan Normal UniversityCollege of Artificial Intelligence, Nanjing Agricultural UniversityAbstract Background The exploration of drug-target interactions (DTIs) is a critical step in drug discovery and drug repurposing. Recently, network-based methods have emerged as a prominent research area for predicting DTIs. These methods excel by extracting both topological and feature information from DTIs networks, thereby achieving superior DTIs prediction performance. However, the majority of existing GCN-based methods utilize shallow graph neural networks, which are incapable of extracting higher-level semantic information. Additionally, the current training of models lacks an effective guiding mechanism, leading to the insufficient improvement of network’s representation capabilities. Results In this paper, we propose a graph convolutional autoencoder model, named DDGAE, for DTIs prediction. We develop a DWR-GCN module, which incorporates dynamic weighting graph convolution with residual connection, to improve the representation capability for DTI heterogeneous networks. Further, to improve the learning efficiency of the model, we devise a dual self-supervised joint training mechanism. Specifically, this mechanism integrates DWR-GCN and a graph convolutional autoencoder into a cohesive system, enhancing both the learning performance and stability of DDGAE. Conclusion Experimental results show that DDGAE significantly outperforms several SOTA models in DTIs prediction, achieving optimal performance and the reliability of our method is verified by case study.https://doi.org/10.1186/s12859-025-06198-xDrug-target interactionDynamic weighting convolutional residual connectionDual self-supervised joint training mechanismGraph convolutional autoencoderGenerative adversarial network
spellingShingle Ming Zeng
Min Wang
Fuqiang Xie
Zhiwei Ji
Drug-target interaction prediction based on graph convolutional autoencoder with dynamic weighting residual GCN
BMC Bioinformatics
Drug-target interaction
Dynamic weighting convolutional residual connection
Dual self-supervised joint training mechanism
Graph convolutional autoencoder
Generative adversarial network
title Drug-target interaction prediction based on graph convolutional autoencoder with dynamic weighting residual GCN
title_full Drug-target interaction prediction based on graph convolutional autoencoder with dynamic weighting residual GCN
title_fullStr Drug-target interaction prediction based on graph convolutional autoencoder with dynamic weighting residual GCN
title_full_unstemmed Drug-target interaction prediction based on graph convolutional autoencoder with dynamic weighting residual GCN
title_short Drug-target interaction prediction based on graph convolutional autoencoder with dynamic weighting residual GCN
title_sort drug target interaction prediction based on graph convolutional autoencoder with dynamic weighting residual gcn
topic Drug-target interaction
Dynamic weighting convolutional residual connection
Dual self-supervised joint training mechanism
Graph convolutional autoencoder
Generative adversarial network
url https://doi.org/10.1186/s12859-025-06198-x
work_keys_str_mv AT mingzeng drugtargetinteractionpredictionbasedongraphconvolutionalautoencoderwithdynamicweightingresidualgcn
AT minwang drugtargetinteractionpredictionbasedongraphconvolutionalautoencoderwithdynamicweightingresidualgcn
AT fuqiangxie drugtargetinteractionpredictionbasedongraphconvolutionalautoencoderwithdynamicweightingresidualgcn
AT zhiweiji drugtargetinteractionpredictionbasedongraphconvolutionalautoencoderwithdynamicweightingresidualgcn