Molecular Property Prediction Based on a Multichannel Substructure Graph
Molecular property prediction is important to drug design. With the development of artificial intelligence, deep learning methods are effective for extracting molecular features. In this paper, we propose a multichannel substructure-graph gated recurrent unit (GRU) architecture, which is a novel GRU...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8964313/ |
| _version_ | 1818646443484774400 |
|---|---|
| author | Shuang Wang Zhen Li Shugang Zhang Mingjian Jiang Xiaofeng Wang Zhiqiang Wei |
| author_facet | Shuang Wang Zhen Li Shugang Zhang Mingjian Jiang Xiaofeng Wang Zhiqiang Wei |
| author_sort | Shuang Wang |
| collection | DOAJ |
| description | Molecular property prediction is important to drug design. With the development of artificial intelligence, deep learning methods are effective for extracting molecular features. In this paper, we propose a multichannel substructure-graph gated recurrent unit (GRU) architecture, which is a novel GRU-based neural network with attention mechanisms applied to molecular substructures to learn and predict properties. In the architecture, molecular features are extracted at the node level and molecule level for capturing fine-grained and coarse-grained information. In addition, three bidirectional GRUs are adopted to extract the features on three channels to generate the molecular representations. Different attention weights are assigned to the entities in the molecule to evaluate their contributions. Experiments are implemented to compare our model with benchmark models in molecular property prediction for both regression and classification tasks, and the results show that our model has strong robustness and generalizability. |
| first_indexed | 2024-12-17T00:46:43Z |
| format | Article |
| id | doaj.art-e43c5c1588c0481e8da858ce8e290028 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-17T00:46:43Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-e43c5c1588c0481e8da858ce8e2900282022-12-21T22:09:54ZengIEEEIEEE Access2169-35362020-01-018186011861410.1109/ACCESS.2020.29685358964313Molecular Property Prediction Based on a Multichannel Substructure GraphShuang Wang0https://orcid.org/0000-0002-6779-9537Zhen Li1https://orcid.org/0000-0001-5093-4221Shugang Zhang2https://orcid.org/0000-0002-9774-9709Mingjian Jiang3https://orcid.org/0000-0002-0716-5292Xiaofeng Wang4https://orcid.org/0000-0001-7042-1773Zhiqiang Wei5https://orcid.org/0000-0002-2830-8301Department of Computer Science and Technology, Ocean University of China, Qingdao, ChinaDepartment of Computer Science and Technology, Ocean University of China, Qingdao, ChinaDepartment of Computer Science and Technology, Ocean University of China, Qingdao, ChinaDepartment of Computer Science and Technology, Ocean University of China, Qingdao, ChinaDepartment of Computer Science and Technology, Ocean University of China, Qingdao, ChinaDepartment of Computer Science and Technology, Ocean University of China, Qingdao, ChinaMolecular property prediction is important to drug design. With the development of artificial intelligence, deep learning methods are effective for extracting molecular features. In this paper, we propose a multichannel substructure-graph gated recurrent unit (GRU) architecture, which is a novel GRU-based neural network with attention mechanisms applied to molecular substructures to learn and predict properties. In the architecture, molecular features are extracted at the node level and molecule level for capturing fine-grained and coarse-grained information. In addition, three bidirectional GRUs are adopted to extract the features on three channels to generate the molecular representations. Different attention weights are assigned to the entities in the molecule to evaluate their contributions. Experiments are implemented to compare our model with benchmark models in molecular property prediction for both regression and classification tasks, and the results show that our model has strong robustness and generalizability.https://ieeexplore.ieee.org/document/8964313/Molecular graphmolecular property predictionsubstructure-graph |
| spellingShingle | Shuang Wang Zhen Li Shugang Zhang Mingjian Jiang Xiaofeng Wang Zhiqiang Wei Molecular Property Prediction Based on a Multichannel Substructure Graph IEEE Access Molecular graph molecular property prediction substructure-graph |
| title | Molecular Property Prediction Based on a Multichannel Substructure Graph |
| title_full | Molecular Property Prediction Based on a Multichannel Substructure Graph |
| title_fullStr | Molecular Property Prediction Based on a Multichannel Substructure Graph |
| title_full_unstemmed | Molecular Property Prediction Based on a Multichannel Substructure Graph |
| title_short | Molecular Property Prediction Based on a Multichannel Substructure Graph |
| title_sort | molecular property prediction based on a multichannel substructure graph |
| topic | Molecular graph molecular property prediction substructure-graph |
| url | https://ieeexplore.ieee.org/document/8964313/ |
| work_keys_str_mv | AT shuangwang molecularpropertypredictionbasedonamultichannelsubstructuregraph AT zhenli molecularpropertypredictionbasedonamultichannelsubstructuregraph AT shugangzhang molecularpropertypredictionbasedonamultichannelsubstructuregraph AT mingjianjiang molecularpropertypredictionbasedonamultichannelsubstructuregraph AT xiaofengwang molecularpropertypredictionbasedonamultichannelsubstructuregraph AT zhiqiangwei molecularpropertypredictionbasedonamultichannelsubstructuregraph |