Heterogeneous network drug-target interaction prediction model based on graph wavelet transform and multi-level contrastive learning
Abstract Reliable prediction of drug–target interaction (DTI) is essential for accelerating drug discovery, yet remains hindered by data imbalance, limited interpretability, and neglect of protein dynamics. Here, we present GHCDTI, a heterogeneous graph neural framework designed to overcome these ch...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-08-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-16098-y |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849226267717533696 |
|---|---|
| author | Wenfeng Dai Yanhong Wang Shuai Yan Qingzhi Yu Xiang Cheng |
| author_facet | Wenfeng Dai Yanhong Wang Shuai Yan Qingzhi Yu Xiang Cheng |
| author_sort | Wenfeng Dai |
| collection | DOAJ |
| description | Abstract Reliable prediction of drug–target interaction (DTI) is essential for accelerating drug discovery, yet remains hindered by data imbalance, limited interpretability, and neglect of protein dynamics. Here, we present GHCDTI, a heterogeneous graph neural framework designed to overcome these challenges through three synergistic innovations. First, cross-view contrastive learning with adaptive positive sampling improves generalization under extreme class imbalance (positive/negative ratio<1:100). Second, heterogeneous data fusion integrates molecular graphs, protein structure graphs, and bioactivity data via cross-graph attention, enabling interpretable residue-level insights. Third, multi-scale wavelet feature extraction captures both conserved and dynamic structural features by decomposing protein conformations into frequency components. GHCDTI achieves state-of-the-art performance on benchmark datasets (AUC: 0.966 ± 0.016; AUPR: 0.888 ± 0.018) and processes 1,512 proteins and 708 drugs in under two minutes, highlighting its potential for scalable virtual screening and drug repositioning. These results demonstrate GHCDTI’s ability to effectively identify novel drug–target pairs, providing a practical tool for accelerating drug discovery and improving biomedical knowledge integration. |
| format | Article |
| id | doaj-art-43a84bd13a094f2d84ac2bb8364a1943 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-08-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-43a84bd13a094f2d84ac2bb8364a19432025-08-24T11:28:04ZengNature PortfolioScientific Reports2045-23222025-08-0115111710.1038/s41598-025-16098-yHeterogeneous network drug-target interaction prediction model based on graph wavelet transform and multi-level contrastive learningWenfeng Dai0Yanhong Wang1Shuai Yan2Qingzhi Yu3Xiang Cheng4School of Information Engineering, Jingdezhen Ceramics UniversitySchool of Information Engineering, Jingdezhen Ceramics UniversitySchool of Information Engineering, Jingdezhen Ceramics UniversitySchool of Information Engineering, Jingdezhen Ceramics UniversitySchool of Information Engineering, Jingdezhen Ceramics UniversityAbstract Reliable prediction of drug–target interaction (DTI) is essential for accelerating drug discovery, yet remains hindered by data imbalance, limited interpretability, and neglect of protein dynamics. Here, we present GHCDTI, a heterogeneous graph neural framework designed to overcome these challenges through three synergistic innovations. First, cross-view contrastive learning with adaptive positive sampling improves generalization under extreme class imbalance (positive/negative ratio<1:100). Second, heterogeneous data fusion integrates molecular graphs, protein structure graphs, and bioactivity data via cross-graph attention, enabling interpretable residue-level insights. Third, multi-scale wavelet feature extraction captures both conserved and dynamic structural features by decomposing protein conformations into frequency components. GHCDTI achieves state-of-the-art performance on benchmark datasets (AUC: 0.966 ± 0.016; AUPR: 0.888 ± 0.018) and processes 1,512 proteins and 708 drugs in under two minutes, highlighting its potential for scalable virtual screening and drug repositioning. These results demonstrate GHCDTI’s ability to effectively identify novel drug–target pairs, providing a practical tool for accelerating drug discovery and improving biomedical knowledge integration.https://doi.org/10.1038/s41598-025-16098-yHeterogeneous networksGraph wavelet transformHeterogeneous graph convolutional networkContrastive learningAttention mechanism |
| spellingShingle | Wenfeng Dai Yanhong Wang Shuai Yan Qingzhi Yu Xiang Cheng Heterogeneous network drug-target interaction prediction model based on graph wavelet transform and multi-level contrastive learning Scientific Reports Heterogeneous networks Graph wavelet transform Heterogeneous graph convolutional network Contrastive learning Attention mechanism |
| title | Heterogeneous network drug-target interaction prediction model based on graph wavelet transform and multi-level contrastive learning |
| title_full | Heterogeneous network drug-target interaction prediction model based on graph wavelet transform and multi-level contrastive learning |
| title_fullStr | Heterogeneous network drug-target interaction prediction model based on graph wavelet transform and multi-level contrastive learning |
| title_full_unstemmed | Heterogeneous network drug-target interaction prediction model based on graph wavelet transform and multi-level contrastive learning |
| title_short | Heterogeneous network drug-target interaction prediction model based on graph wavelet transform and multi-level contrastive learning |
| title_sort | heterogeneous network drug target interaction prediction model based on graph wavelet transform and multi level contrastive learning |
| topic | Heterogeneous networks Graph wavelet transform Heterogeneous graph convolutional network Contrastive learning Attention mechanism |
| url | https://doi.org/10.1038/s41598-025-16098-y |
| work_keys_str_mv | AT wenfengdai heterogeneousnetworkdrugtargetinteractionpredictionmodelbasedongraphwavelettransformandmultilevelcontrastivelearning AT yanhongwang heterogeneousnetworkdrugtargetinteractionpredictionmodelbasedongraphwavelettransformandmultilevelcontrastivelearning AT shuaiyan heterogeneousnetworkdrugtargetinteractionpredictionmodelbasedongraphwavelettransformandmultilevelcontrastivelearning AT qingzhiyu heterogeneousnetworkdrugtargetinteractionpredictionmodelbasedongraphwavelettransformandmultilevelcontrastivelearning AT xiangcheng heterogeneousnetworkdrugtargetinteractionpredictionmodelbasedongraphwavelettransformandmultilevelcontrastivelearning |