MSEDDI: Multi-Scale Embedding for Predicting Drug—Drug Interaction Events
A norm in modern medicine is to prescribe polypharmacy to treat disease. The core concern with the co-administration of drugs is that it may produce adverse drug—drug interaction (DDI), which can cause unexpected bodily injury. Therefore, it is essential to identify potential DDI. Most existing meth...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | International Journal of Molecular Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1422-0067/24/5/4500 |
| _version_ | 1797615273098870784 |
|---|---|
| author | Liyi Yu Zhaochun Xu Meiling Cheng Weizhong Lin Wangren Qiu Xuan Xiao |
| author_facet | Liyi Yu Zhaochun Xu Meiling Cheng Weizhong Lin Wangren Qiu Xuan Xiao |
| author_sort | Liyi Yu |
| collection | DOAJ |
| description | A norm in modern medicine is to prescribe polypharmacy to treat disease. The core concern with the co-administration of drugs is that it may produce adverse drug—drug interaction (DDI), which can cause unexpected bodily injury. Therefore, it is essential to identify potential DDI. Most existing methods in silico only judge whether two drugs interact, ignoring the importance of interaction events to study the mechanism implied in combination drugs. In this work, we propose a deep learning framework named MSEDDI that comprehensively considers multi-scale embedding representations of the drug for predicting drug—drug interaction events. In MSEDDI, we design three-channel networks to process biomedical network-based knowledge graph embedding, SMILES sequence-based notation embedding, and molecular graph-based chemical structure embedding, respectively. Finally, we fuse three heterogeneous features from channel outputs through a self-attention mechanism and feed them to the linear layer predictor. In the experimental section, we evaluate the performance of all methods on two different prediction tasks on two datasets. The results show that MSEDDI outperforms other state-of-the-art baselines. Moreover, we also reveal the stable performance of our model in a broader sample set via case studies. |
| first_indexed | 2024-03-11T07:23:00Z |
| format | Article |
| id | doaj.art-d0f1bcefb809402a8a7b53540204bc5a |
| institution | Directory Open Access Journal |
| issn | 1661-6596 1422-0067 |
| language | English |
| last_indexed | 2024-03-11T07:23:00Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | International Journal of Molecular Sciences |
| spelling | doaj.art-d0f1bcefb809402a8a7b53540204bc5a2023-11-17T07:49:19ZengMDPI AGInternational Journal of Molecular Sciences1661-65961422-00672023-02-01245450010.3390/ijms24054500MSEDDI: Multi-Scale Embedding for Predicting Drug—Drug Interaction EventsLiyi Yu0Zhaochun Xu1Meiling Cheng2Weizhong Lin3Wangren Qiu4Xuan Xiao5Department of Computer, School of Information Engineering, Jingdezhen Ceramic University, Jingdezhen 333403, ChinaDepartment of Computer, School of Information Engineering, Jingdezhen Ceramic University, Jingdezhen 333403, ChinaDepartment of Computer, School of Information Engineering, Jingdezhen Ceramic University, Jingdezhen 333403, ChinaDepartment of Computer, School of Information Engineering, Jingdezhen Ceramic University, Jingdezhen 333403, ChinaDepartment of Computer, School of Information Engineering, Jingdezhen Ceramic University, Jingdezhen 333403, ChinaDepartment of Computer, School of Information Engineering, Jingdezhen Ceramic University, Jingdezhen 333403, ChinaA norm in modern medicine is to prescribe polypharmacy to treat disease. The core concern with the co-administration of drugs is that it may produce adverse drug—drug interaction (DDI), which can cause unexpected bodily injury. Therefore, it is essential to identify potential DDI. Most existing methods in silico only judge whether two drugs interact, ignoring the importance of interaction events to study the mechanism implied in combination drugs. In this work, we propose a deep learning framework named MSEDDI that comprehensively considers multi-scale embedding representations of the drug for predicting drug—drug interaction events. In MSEDDI, we design three-channel networks to process biomedical network-based knowledge graph embedding, SMILES sequence-based notation embedding, and molecular graph-based chemical structure embedding, respectively. Finally, we fuse three heterogeneous features from channel outputs through a self-attention mechanism and feed them to the linear layer predictor. In the experimental section, we evaluate the performance of all methods on two different prediction tasks on two datasets. The results show that MSEDDI outperforms other state-of-the-art baselines. Moreover, we also reveal the stable performance of our model in a broader sample set via case studies.https://www.mdpi.com/1422-0067/24/5/4500drug—drug interactionknowledge graphgraph neural networkself-attention mechanism |
| spellingShingle | Liyi Yu Zhaochun Xu Meiling Cheng Weizhong Lin Wangren Qiu Xuan Xiao MSEDDI: Multi-Scale Embedding for Predicting Drug—Drug Interaction Events International Journal of Molecular Sciences drug—drug interaction knowledge graph graph neural network self-attention mechanism |
| title | MSEDDI: Multi-Scale Embedding for Predicting Drug—Drug Interaction Events |
| title_full | MSEDDI: Multi-Scale Embedding for Predicting Drug—Drug Interaction Events |
| title_fullStr | MSEDDI: Multi-Scale Embedding for Predicting Drug—Drug Interaction Events |
| title_full_unstemmed | MSEDDI: Multi-Scale Embedding for Predicting Drug—Drug Interaction Events |
| title_short | MSEDDI: Multi-Scale Embedding for Predicting Drug—Drug Interaction Events |
| title_sort | mseddi multi scale embedding for predicting drug drug interaction events |
| topic | drug—drug interaction knowledge graph graph neural network self-attention mechanism |
| url | https://www.mdpi.com/1422-0067/24/5/4500 |
| work_keys_str_mv | AT liyiyu mseddimultiscaleembeddingforpredictingdrugdruginteractionevents AT zhaochunxu mseddimultiscaleembeddingforpredictingdrugdruginteractionevents AT meilingcheng mseddimultiscaleembeddingforpredictingdrugdruginteractionevents AT weizhonglin mseddimultiscaleembeddingforpredictingdrugdruginteractionevents AT wangrenqiu mseddimultiscaleembeddingforpredictingdrugdruginteractionevents AT xuanxiao mseddimultiscaleembeddingforpredictingdrugdruginteractionevents |