LIDER: cell embedding based deep neural network classifier for supervised cell type identification
Background Automatic cell type identification has been an urgent task for the rapid development of single-cell RNA-seq techniques. Generally, the current approach for cell type identification is to generate cell clusters by unsupervised clustering and later assign labels to each cell cluster with ma...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
PeerJ Inc.
2023-08-01
|
| Series: | PeerJ |
| Subjects: | |
| Online Access: | https://peerj.com/articles/15862.pdf |
| _version_ | 1797421232991240192 |
|---|---|
| author | Yachen Tang Xuefeng Li Mingguang Shi |
| author_facet | Yachen Tang Xuefeng Li Mingguang Shi |
| author_sort | Yachen Tang |
| collection | DOAJ |
| description | Background Automatic cell type identification has been an urgent task for the rapid development of single-cell RNA-seq techniques. Generally, the current approach for cell type identification is to generate cell clusters by unsupervised clustering and later assign labels to each cell cluster with manual annotation. Methods Here, we introduce LIDER (celL embeddIng based Deep nEural netwoRk classifier), a deep supervised learning method that combines cell embedding and deep neural network classifier for automatic cell type identification. Based on a stacked denoising autoencoder with a tailored and reconstructed loss function, LIDER identifies cell embedding and predicts cell types with a deep neural network classifier. LIDER was developed upon a stacked denoising autoencoder to learn encoder-decoder structures for identifying cell embedding. Results LIDER accurately identifies cell types by using stacked denoising autoencoder. Benchmarking against state-of-the-art methods across eight types of single-cell data, LIDER achieves comparable or even superior enhancement performance. Moreover, LIDER suggests comparable robust to batch effects. Our results show a potential in deep supervised learning for automatic cell type identification of single-cell RNA-seq data. The LIDER codes are available at https://github.com/ShiMGLab/LIDER. |
| first_indexed | 2024-03-09T07:13:57Z |
| format | Article |
| id | doaj.art-1687af24dd644d59a991a5fef4e441a4 |
| institution | Directory Open Access Journal |
| issn | 2167-8359 |
| language | English |
| last_indexed | 2024-03-09T07:13:57Z |
| publishDate | 2023-08-01 |
| publisher | PeerJ Inc. |
| record_format | Article |
| series | PeerJ |
| spelling | doaj.art-1687af24dd644d59a991a5fef4e441a42023-12-03T08:49:57ZengPeerJ Inc.PeerJ2167-83592023-08-0111e1586210.7717/peerj.15862LIDER: cell embedding based deep neural network classifier for supervised cell type identificationYachen Tang0Xuefeng Li1Mingguang Shi2Hefei University of Technology, Hefei, ChinaHefei University of Technology, Hefei, ChinaHefei University of Technology, Hefei, ChinaBackground Automatic cell type identification has been an urgent task for the rapid development of single-cell RNA-seq techniques. Generally, the current approach for cell type identification is to generate cell clusters by unsupervised clustering and later assign labels to each cell cluster with manual annotation. Methods Here, we introduce LIDER (celL embeddIng based Deep nEural netwoRk classifier), a deep supervised learning method that combines cell embedding and deep neural network classifier for automatic cell type identification. Based on a stacked denoising autoencoder with a tailored and reconstructed loss function, LIDER identifies cell embedding and predicts cell types with a deep neural network classifier. LIDER was developed upon a stacked denoising autoencoder to learn encoder-decoder structures for identifying cell embedding. Results LIDER accurately identifies cell types by using stacked denoising autoencoder. Benchmarking against state-of-the-art methods across eight types of single-cell data, LIDER achieves comparable or even superior enhancement performance. Moreover, LIDER suggests comparable robust to batch effects. Our results show a potential in deep supervised learning for automatic cell type identification of single-cell RNA-seq data. The LIDER codes are available at https://github.com/ShiMGLab/LIDER.https://peerj.com/articles/15862.pdfCell embeddingStacked denoising autoencodersDeep neural network classifierCell type identification |
| spellingShingle | Yachen Tang Xuefeng Li Mingguang Shi LIDER: cell embedding based deep neural network classifier for supervised cell type identification PeerJ Cell embedding Stacked denoising autoencoders Deep neural network classifier Cell type identification |
| title | LIDER: cell embedding based deep neural network classifier for supervised cell type identification |
| title_full | LIDER: cell embedding based deep neural network classifier for supervised cell type identification |
| title_fullStr | LIDER: cell embedding based deep neural network classifier for supervised cell type identification |
| title_full_unstemmed | LIDER: cell embedding based deep neural network classifier for supervised cell type identification |
| title_short | LIDER: cell embedding based deep neural network classifier for supervised cell type identification |
| title_sort | lider cell embedding based deep neural network classifier for supervised cell type identification |
| topic | Cell embedding Stacked denoising autoencoders Deep neural network classifier Cell type identification |
| url | https://peerj.com/articles/15862.pdf |
| work_keys_str_mv | AT yachentang lidercellembeddingbaseddeepneuralnetworkclassifierforsupervisedcelltypeidentification AT xuefengli lidercellembeddingbaseddeepneuralnetworkclassifierforsupervisedcelltypeidentification AT mingguangshi lidercellembeddingbaseddeepneuralnetworkclassifierforsupervisedcelltypeidentification |