Pairwise Heuristic Sequence Alignment Algorithm Based on Deep Reinforcement Learning
<italic>Goal:</italic> Various methods have been developed to analyze the association between organisms and their genomic sequences. Among them, sequence alignment is the most frequently used method for comparative analysis of biological genomes. We intend to propose a novel pairwise seq...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Open Journal of Engineering in Medicine and Biology |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9340257/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849422682107412480 |
|---|---|
| author | Yong-Joon Song Dong Jin Ji Hyein Seo Gyu-Bum Han Dong-Ho Cho |
| author_facet | Yong-Joon Song Dong Jin Ji Hyein Seo Gyu-Bum Han Dong-Ho Cho |
| author_sort | Yong-Joon Song |
| collection | DOAJ |
| description | <italic>Goal:</italic> Various methods have been developed to analyze the association between organisms and their genomic sequences. Among them, sequence alignment is the most frequently used method for comparative analysis of biological genomes. We intend to propose a novel pairwise sequence alignment method using deep reinforcement learning to break out the old pairwise alignment algorithms. <italic>Methods:</italic> We defined the environment and agent to enable reinforcement learning in the sequence alignment system. This novel method, named DQNalign, can immediately determine the next direction by observing the subsequences within the moving window. <italic>Results:</italic> DQNalign shows superiority in the dissimilar sequence pairs that have low identity values. And theoretically, we confirm that DQNalign has a low dimension for the sequence length in view of the complexity. <italic>Conclusions:</italic> This research shows the application method of deep reinforcement learning to the sequence alignment system and how deep reinforcement learning can improve the conventional sequence alignment method. |
| format | Article |
| id | doaj-art-5d1a7f2f387943dbafa35a048b3decd4 |
| institution | Kabale University |
| issn | 2644-1276 |
| language | English |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Journal of Engineering in Medicine and Biology |
| spelling | doaj-art-5d1a7f2f387943dbafa35a048b3decd42025-08-20T03:30:57ZengIEEEIEEE Open Journal of Engineering in Medicine and Biology2644-12762021-01-012364310.1109/OJEMB.2021.30554249340257Pairwise Heuristic Sequence Alignment Algorithm Based on Deep Reinforcement LearningYong-Joon Song0https://orcid.org/0000-0001-5751-3171Dong Jin Ji1https://orcid.org/0000-0002-2093-6469Hyein Seo2https://orcid.org/0000-0002-5722-7957Gyu-Bum Han3Dong-Ho Cho4https://orcid.org/0000-0001-9849-4392School of Electrical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, South KoreaSchool of Electrical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, South KoreaSchool of Electrical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, South KoreaSchool of Electrical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, South KoreaSchool of Electrical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, South Korea<italic>Goal:</italic> Various methods have been developed to analyze the association between organisms and their genomic sequences. Among them, sequence alignment is the most frequently used method for comparative analysis of biological genomes. We intend to propose a novel pairwise sequence alignment method using deep reinforcement learning to break out the old pairwise alignment algorithms. <italic>Methods:</italic> We defined the environment and agent to enable reinforcement learning in the sequence alignment system. This novel method, named DQNalign, can immediately determine the next direction by observing the subsequences within the moving window. <italic>Results:</italic> DQNalign shows superiority in the dissimilar sequence pairs that have low identity values. And theoretically, we confirm that DQNalign has a low dimension for the sequence length in view of the complexity. <italic>Conclusions:</italic> This research shows the application method of deep reinforcement learning to the sequence alignment system and how deep reinforcement learning can improve the conventional sequence alignment method.https://ieeexplore.ieee.org/document/9340257/Deep reinforcement learningglobal alignmentpairwise alignmentsequence alignmentsequence comparison |
| spellingShingle | Yong-Joon Song Dong Jin Ji Hyein Seo Gyu-Bum Han Dong-Ho Cho Pairwise Heuristic Sequence Alignment Algorithm Based on Deep Reinforcement Learning IEEE Open Journal of Engineering in Medicine and Biology Deep reinforcement learning global alignment pairwise alignment sequence alignment sequence comparison |
| title | Pairwise Heuristic Sequence Alignment Algorithm Based on Deep Reinforcement Learning |
| title_full | Pairwise Heuristic Sequence Alignment Algorithm Based on Deep Reinforcement Learning |
| title_fullStr | Pairwise Heuristic Sequence Alignment Algorithm Based on Deep Reinforcement Learning |
| title_full_unstemmed | Pairwise Heuristic Sequence Alignment Algorithm Based on Deep Reinforcement Learning |
| title_short | Pairwise Heuristic Sequence Alignment Algorithm Based on Deep Reinforcement Learning |
| title_sort | pairwise heuristic sequence alignment algorithm based on deep reinforcement learning |
| topic | Deep reinforcement learning global alignment pairwise alignment sequence alignment sequence comparison |
| url | https://ieeexplore.ieee.org/document/9340257/ |
| work_keys_str_mv | AT yongjoonsong pairwiseheuristicsequencealignmentalgorithmbasedondeepreinforcementlearning AT dongjinji pairwiseheuristicsequencealignmentalgorithmbasedondeepreinforcementlearning AT hyeinseo pairwiseheuristicsequencealignmentalgorithmbasedondeepreinforcementlearning AT gyubumhan pairwiseheuristicsequencealignmentalgorithmbasedondeepreinforcementlearning AT donghocho pairwiseheuristicsequencealignmentalgorithmbasedondeepreinforcementlearning |