Pre-training graph neural networks for link prediction in biomedical networks
Motivation: Graphs or networks are widely utilized to model the interactions between different entities (e.g. proteins, drugs, etc.) for biomedical applications. Predicting potential interactions/links in biomedical networks is important for understanding the pathological mechanisms of various compl...
| Main Authors: | , , , , , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/162781 |
| _version_ | 1811689539371008000 |
|---|---|
| author | Long, Yahui Wu, Min Liu, Yong Fang, Yuan Kwoh, Chee Keong Chen, Jinmiao Luo, Jiawei Li, Xiaoli |
| author2 | School of Computer Science and Engineering |
| author_facet | School of Computer Science and Engineering Long, Yahui Wu, Min Liu, Yong Fang, Yuan Kwoh, Chee Keong Chen, Jinmiao Luo, Jiawei Li, Xiaoli |
| author_sort | Long, Yahui |
| collection | NTU |
| description | Motivation: Graphs or networks are widely utilized to model the interactions between different entities (e.g. proteins, drugs, etc.) for biomedical applications. Predicting potential interactions/links in biomedical networks is important for understanding the pathological mechanisms of various complex human diseases, as well as screening compound targets for drug discovery. Graph neural networks (GNNs) have been utilized for link prediction in various biomedical networks, which rely on the node features extracted from different data sources, e.g. sequence, structure and network data. However, it is challenging to effectively integrate these data sources and automatically extract features for different link prediction tasks. Results: In this article, we propose a novel Pre-Training Graph Neural Networks-based framework named PT-GNN to integrate different data sources for link prediction in biomedical networks. First, we design expressive deep learning methods [e.g. convolutional neural network and graph convolutional network (GCN)] to learn features for individual nodes from sequence and structure data. Second, we further propose a GCN-based encoder to effectively refine the node features by modelling the dependencies among nodes in the network. Third, the node features are pre-trained based on graph reconstruction tasks. The pre-trained features can be used for model initialization in downstream tasks. Extensive experiments have been conducted on two critical link prediction tasks, i.e. synthetic lethality (SL) prediction and drug–target interaction (DTI) prediction. Experimental results demonstrate PT-GNN outperforms the state-of-the-art methods for SL prediction and DTI prediction. In addition, the pre-trained features benefit improving the performance and reduce the training time of existing models. Availability and implementation: Python codes and dataset are available at: https://github.com/longyahui/PT-GNN. |
| first_indexed | 2024-10-01T05:49:42Z |
| format | Journal Article |
| id | ntu-10356/162781 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T05:49:42Z |
| publishDate | 2022 |
| record_format | dspace |
| spelling | ntu-10356/1627812022-11-09T00:25:11Z Pre-training graph neural networks for link prediction in biomedical networks Long, Yahui Wu, Min Liu, Yong Fang, Yuan Kwoh, Chee Keong Chen, Jinmiao Luo, Jiawei Li, Xiaoli School of Computer Science and Engineering Joint NTU-UBC Research Centre of Excellence in Active Living for the Elderly (LILY) Engineering::Computer science and engineering Neural Networks Biomedical Networks Motivation: Graphs or networks are widely utilized to model the interactions between different entities (e.g. proteins, drugs, etc.) for biomedical applications. Predicting potential interactions/links in biomedical networks is important for understanding the pathological mechanisms of various complex human diseases, as well as screening compound targets for drug discovery. Graph neural networks (GNNs) have been utilized for link prediction in various biomedical networks, which rely on the node features extracted from different data sources, e.g. sequence, structure and network data. However, it is challenging to effectively integrate these data sources and automatically extract features for different link prediction tasks. Results: In this article, we propose a novel Pre-Training Graph Neural Networks-based framework named PT-GNN to integrate different data sources for link prediction in biomedical networks. First, we design expressive deep learning methods [e.g. convolutional neural network and graph convolutional network (GCN)] to learn features for individual nodes from sequence and structure data. Second, we further propose a GCN-based encoder to effectively refine the node features by modelling the dependencies among nodes in the network. Third, the node features are pre-trained based on graph reconstruction tasks. The pre-trained features can be used for model initialization in downstream tasks. Extensive experiments have been conducted on two critical link prediction tasks, i.e. synthetic lethality (SL) prediction and drug–target interaction (DTI) prediction. Experimental results demonstrate PT-GNN outperforms the state-of-the-art methods for SL prediction and DTI prediction. In addition, the pre-trained features benefit improving the performance and reduce the training time of existing models. Availability and implementation: Python codes and dataset are available at: https://github.com/longyahui/PT-GNN. This work was supported by the National Natural Science Foundation of China [61873089]; and the Key Program of National Natural Science Foundation of China [62032007]. 2022-11-09T00:25:11Z 2022-11-09T00:25:11Z 2022 Journal Article Long, Y., Wu, M., Liu, Y., Fang, Y., Kwoh, C. K., Chen, J., Luo, J. & Li, X. (2022). Pre-training graph neural networks for link prediction in biomedical networks. Bioinformatics, 38(8), 2254-2262. https://dx.doi.org/10.1093/bioinformatics/btac100 1367-4803 https://hdl.handle.net/10356/162781 10.1093/bioinformatics/btac100 35171981 2-s2.0-85128705246 8 38 2254 2262 en Bioinformatics © The Author(s) 2022. Published by Oxford University Press. All rights reserved. |
| spellingShingle | Engineering::Computer science and engineering Neural Networks Biomedical Networks Long, Yahui Wu, Min Liu, Yong Fang, Yuan Kwoh, Chee Keong Chen, Jinmiao Luo, Jiawei Li, Xiaoli Pre-training graph neural networks for link prediction in biomedical networks |
| title | Pre-training graph neural networks for link prediction in biomedical networks |
| title_full | Pre-training graph neural networks for link prediction in biomedical networks |
| title_fullStr | Pre-training graph neural networks for link prediction in biomedical networks |
| title_full_unstemmed | Pre-training graph neural networks for link prediction in biomedical networks |
| title_short | Pre-training graph neural networks for link prediction in biomedical networks |
| title_sort | pre training graph neural networks for link prediction in biomedical networks |
| topic | Engineering::Computer science and engineering Neural Networks Biomedical Networks |
| url | https://hdl.handle.net/10356/162781 |
| work_keys_str_mv | AT longyahui pretraininggraphneuralnetworksforlinkpredictioninbiomedicalnetworks AT wumin pretraininggraphneuralnetworksforlinkpredictioninbiomedicalnetworks AT liuyong pretraininggraphneuralnetworksforlinkpredictioninbiomedicalnetworks AT fangyuan pretraininggraphneuralnetworksforlinkpredictioninbiomedicalnetworks AT kwohcheekeong pretraininggraphneuralnetworksforlinkpredictioninbiomedicalnetworks AT chenjinmiao pretraininggraphneuralnetworksforlinkpredictioninbiomedicalnetworks AT luojiawei pretraininggraphneuralnetworksforlinkpredictioninbiomedicalnetworks AT lixiaoli pretraininggraphneuralnetworksforlinkpredictioninbiomedicalnetworks |