Prediction of enhancer-promoter interactions via natural language processing
Abstract Background Precise identification of three-dimensional genome organization, especially enhancer-promoter interactions (EPIs), is important to deciphering gene regulation, cell differentiation and disease mechanisms. Currently, it is a challenging task to distinguish true interactions from o...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2018-05-01
|
| Series: | BMC Genomics |
| Subjects: | |
| Online Access: | http://link.springer.com/article/10.1186/s12864-018-4459-6 |
| _version_ | 1818109850658275328 |
|---|---|
| author | Wanwen Zeng Mengmeng Wu Rui Jiang |
| author_facet | Wanwen Zeng Mengmeng Wu Rui Jiang |
| author_sort | Wanwen Zeng |
| collection | DOAJ |
| description | Abstract Background Precise identification of three-dimensional genome organization, especially enhancer-promoter interactions (EPIs), is important to deciphering gene regulation, cell differentiation and disease mechanisms. Currently, it is a challenging task to distinguish true interactions from other nearby non-interacting ones since the power of traditional experimental methods is limited due to low resolution or low throughput. Results We propose a novel computational framework EP2vec to assay three-dimensional genomic interactions. We first extract sequence embedding features, defined as fixed-length vector representations learned from variable-length sequences using an unsupervised deep learning method in natural language processing. Then, we train a classifier to predict EPIs using the learned representations in supervised way. Experimental results demonstrate that EP2vec obtains F1 scores ranging from 0.841~ 0.933 on different datasets, which outperforms existing methods. We prove the robustness of sequence embedding features by carrying out sensitivity analysis. Besides, we identify motifs that represent cell line-specific information through analysis of the learned sequence embedding features by adopting attention mechanism. Last, we show that even superior performance with F1 scores 0.889~ 0.940 can be achieved by combining sequence embedding features and experimental features. Conclusions EP2vec sheds light on feature extraction for DNA sequences of arbitrary lengths and provides a powerful approach for EPIs identification. |
| first_indexed | 2024-12-11T02:37:48Z |
| format | Article |
| id | doaj.art-278e9654f5ee492fa2a5f34854a5904f |
| institution | Directory Open Access Journal |
| issn | 1471-2164 |
| language | English |
| last_indexed | 2024-12-11T02:37:48Z |
| publishDate | 2018-05-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Genomics |
| spelling | doaj.art-278e9654f5ee492fa2a5f34854a5904f2022-12-22T01:23:41ZengBMCBMC Genomics1471-21642018-05-0119S2132210.1186/s12864-018-4459-6Prediction of enhancer-promoter interactions via natural language processingWanwen Zeng0Mengmeng Wu1Rui Jiang2MOE Key Laboratory of Bioinformatics; Bioinformatics Division and Center for Synthetic & Systems BiologyMOE Key Laboratory of Bioinformatics; Bioinformatics Division and Center for Synthetic & Systems BiologyMOE Key Laboratory of Bioinformatics; Bioinformatics Division and Center for Synthetic & Systems BiologyAbstract Background Precise identification of three-dimensional genome organization, especially enhancer-promoter interactions (EPIs), is important to deciphering gene regulation, cell differentiation and disease mechanisms. Currently, it is a challenging task to distinguish true interactions from other nearby non-interacting ones since the power of traditional experimental methods is limited due to low resolution or low throughput. Results We propose a novel computational framework EP2vec to assay three-dimensional genomic interactions. We first extract sequence embedding features, defined as fixed-length vector representations learned from variable-length sequences using an unsupervised deep learning method in natural language processing. Then, we train a classifier to predict EPIs using the learned representations in supervised way. Experimental results demonstrate that EP2vec obtains F1 scores ranging from 0.841~ 0.933 on different datasets, which outperforms existing methods. We prove the robustness of sequence embedding features by carrying out sensitivity analysis. Besides, we identify motifs that represent cell line-specific information through analysis of the learned sequence embedding features by adopting attention mechanism. Last, we show that even superior performance with F1 scores 0.889~ 0.940 can be achieved by combining sequence embedding features and experimental features. Conclusions EP2vec sheds light on feature extraction for DNA sequences of arbitrary lengths and provides a powerful approach for EPIs identification.http://link.springer.com/article/10.1186/s12864-018-4459-6Enhancer-promoter interactionsThree-dimensinal interactionsNatural language processingUnsupervised learning |
| spellingShingle | Wanwen Zeng Mengmeng Wu Rui Jiang Prediction of enhancer-promoter interactions via natural language processing BMC Genomics Enhancer-promoter interactions Three-dimensinal interactions Natural language processing Unsupervised learning |
| title | Prediction of enhancer-promoter interactions via natural language processing |
| title_full | Prediction of enhancer-promoter interactions via natural language processing |
| title_fullStr | Prediction of enhancer-promoter interactions via natural language processing |
| title_full_unstemmed | Prediction of enhancer-promoter interactions via natural language processing |
| title_short | Prediction of enhancer-promoter interactions via natural language processing |
| title_sort | prediction of enhancer promoter interactions via natural language processing |
| topic | Enhancer-promoter interactions Three-dimensinal interactions Natural language processing Unsupervised learning |
| url | http://link.springer.com/article/10.1186/s12864-018-4459-6 |
| work_keys_str_mv | AT wanwenzeng predictionofenhancerpromoterinteractionsvianaturallanguageprocessing AT mengmengwu predictionofenhancerpromoterinteractionsvianaturallanguageprocessing AT ruijiang predictionofenhancerpromoterinteractionsvianaturallanguageprocessing |