Multi-scaled self-attention for drug–target interaction prediction based on multi-granularity representation
Abstract Background Drug–target interaction (DTI) prediction plays a crucial role in drug discovery. Although the advanced deep learning has shown promising results in predicting DTIs, it still needs improvements in two aspects: (1) encoding method, in which the existing encoding method, character e...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
BMC
2022-08-01
|
Series: | BMC Bioinformatics |
Subjects: | |
Online Access: | https://doi.org/10.1186/s12859-022-04857-x |
_version_ | 1798038090429759488 |
---|---|
author | Yuni Zeng Xiangru Chen Dezhong Peng Lijun Zhang Haixiao Huang |
author_facet | Yuni Zeng Xiangru Chen Dezhong Peng Lijun Zhang Haixiao Huang |
author_sort | Yuni Zeng |
collection | DOAJ |
description | Abstract Background Drug–target interaction (DTI) prediction plays a crucial role in drug discovery. Although the advanced deep learning has shown promising results in predicting DTIs, it still needs improvements in two aspects: (1) encoding method, in which the existing encoding method, character encoding, overlooks chemical textual information of atoms with multiple characters and chemical functional groups; as well as (2) the architecture of deep model, which should focus on multiple chemical patterns in drug and target representations. Results In this paper, we propose a multi-granularity multi-scaled self-attention (SAN) model by alleviating the above problems. Specifically, in process of encoding, we investigate a segmentation method for drug and protein sequences and then label the segmented groups as the multi-granularity representations. Moreover, in order to enhance the various local patterns in these multi-granularity representations, a multi-scaled SAN is built and exploited to generate deep representations of drugs and targets. Finally, our proposed model predicts DTIs based on the fusion of these deep representations. Our proposed model is evaluated on two benchmark datasets, KIBA and Davis. The experimental results reveal that our proposed model yields better prediction accuracy than strong baseline models. Conclusion Our proposed multi-granularity encoding method and multi-scaled SAN model improve DTI prediction by encoding the chemical textual information of drugs and targets and extracting their various local patterns, respectively. |
first_indexed | 2024-04-11T21:35:28Z |
format | Article |
id | doaj.art-14aebf18090a445f9d562c066f3f85c0 |
institution | Directory Open Access Journal |
issn | 1471-2105 |
language | English |
last_indexed | 2024-04-11T21:35:28Z |
publishDate | 2022-08-01 |
publisher | BMC |
record_format | Article |
series | BMC Bioinformatics |
spelling | doaj.art-14aebf18090a445f9d562c066f3f85c02022-12-22T04:01:45ZengBMCBMC Bioinformatics1471-21052022-08-0123111510.1186/s12859-022-04857-xMulti-scaled self-attention for drug–target interaction prediction based on multi-granularity representationYuni Zeng0Xiangru Chen1Dezhong Peng2Lijun Zhang3Haixiao Huang4School of Information Science and Technology, Zhejiang Sci-Tech UniversityCollege of Computer Science, Sichuan UniversityCollege of Computer Science, Sichuan UniversitySichuan Zhiqian Technology Co., LtdSichuan Provincial Commission of Politics and LawAbstract Background Drug–target interaction (DTI) prediction plays a crucial role in drug discovery. Although the advanced deep learning has shown promising results in predicting DTIs, it still needs improvements in two aspects: (1) encoding method, in which the existing encoding method, character encoding, overlooks chemical textual information of atoms with multiple characters and chemical functional groups; as well as (2) the architecture of deep model, which should focus on multiple chemical patterns in drug and target representations. Results In this paper, we propose a multi-granularity multi-scaled self-attention (SAN) model by alleviating the above problems. Specifically, in process of encoding, we investigate a segmentation method for drug and protein sequences and then label the segmented groups as the multi-granularity representations. Moreover, in order to enhance the various local patterns in these multi-granularity representations, a multi-scaled SAN is built and exploited to generate deep representations of drugs and targets. Finally, our proposed model predicts DTIs based on the fusion of these deep representations. Our proposed model is evaluated on two benchmark datasets, KIBA and Davis. The experimental results reveal that our proposed model yields better prediction accuracy than strong baseline models. Conclusion Our proposed multi-granularity encoding method and multi-scaled SAN model improve DTI prediction by encoding the chemical textual information of drugs and targets and extracting their various local patterns, respectively.https://doi.org/10.1186/s12859-022-04857-xDrug–target interactionDeep learningSelf-attention networksRepresentations learning |
spellingShingle | Yuni Zeng Xiangru Chen Dezhong Peng Lijun Zhang Haixiao Huang Multi-scaled self-attention for drug–target interaction prediction based on multi-granularity representation BMC Bioinformatics Drug–target interaction Deep learning Self-attention networks Representations learning |
title | Multi-scaled self-attention for drug–target interaction prediction based on multi-granularity representation |
title_full | Multi-scaled self-attention for drug–target interaction prediction based on multi-granularity representation |
title_fullStr | Multi-scaled self-attention for drug–target interaction prediction based on multi-granularity representation |
title_full_unstemmed | Multi-scaled self-attention for drug–target interaction prediction based on multi-granularity representation |
title_short | Multi-scaled self-attention for drug–target interaction prediction based on multi-granularity representation |
title_sort | multi scaled self attention for drug target interaction prediction based on multi granularity representation |
topic | Drug–target interaction Deep learning Self-attention networks Representations learning |
url | https://doi.org/10.1186/s12859-022-04857-x |
work_keys_str_mv | AT yunizeng multiscaledselfattentionfordrugtargetinteractionpredictionbasedonmultigranularityrepresentation AT xiangruchen multiscaledselfattentionfordrugtargetinteractionpredictionbasedonmultigranularityrepresentation AT dezhongpeng multiscaledselfattentionfordrugtargetinteractionpredictionbasedonmultigranularityrepresentation AT lijunzhang multiscaledselfattentionfordrugtargetinteractionpredictionbasedonmultigranularityrepresentation AT haixiaohuang multiscaledselfattentionfordrugtargetinteractionpredictionbasedonmultigranularityrepresentation |