1
|
Cui X, Chen X, Li Z, Gao Z, Chen S, Jiang R. Discrete latent embedding of single-cell chromatin accessibility sequencing data for uncovering cell heterogeneity. NATURE COMPUTATIONAL SCIENCE 2024; 4:346-359. [PMID: 38730185 DOI: 10.1038/s43588-024-00625-4] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 09/25/2023] [Accepted: 04/05/2024] [Indexed: 05/12/2024]
Abstract
Single-cell epigenomic data has been growing continuously at an unprecedented pace, but their characteristics such as high dimensionality and sparsity pose substantial challenges to downstream analysis. Although deep learning models-especially variational autoencoders-have been widely used to capture low-dimensional feature embeddings, the prevalent Gaussian assumption somewhat disagrees with real data, and these models tend to struggle to incorporate reference information from abundant cell atlases. Here we propose CASTLE, a deep generative model based on the vector-quantized variational autoencoder framework to extract discrete latent embeddings that interpretably characterize single-cell chromatin accessibility sequencing data. We validate the performance and robustness of CASTLE for accurate cell-type identification and reasonable visualization compared with state-of-the-art methods. We demonstrate the advantages of CASTLE for effective incorporation of existing massive reference datasets in a weakly supervised or supervised manner. We further demonstrate CASTLE's capacity for intuitively distilling cell-type-specific feature spectra that unveil cell heterogeneity and biological implications quantitatively.
Collapse
Affiliation(s)
- Xuejian Cui
- Ministry of Education Key Laboratory of Bioinformatics, Bioinformatics Division at the Beijing National Research Center for Information Science and Technology, Center for Synthetic and Systems Biology, Department of Automation, Tsinghua University, Beijing, China
| | - Xiaoyang Chen
- Ministry of Education Key Laboratory of Bioinformatics, Bioinformatics Division at the Beijing National Research Center for Information Science and Technology, Center for Synthetic and Systems Biology, Department of Automation, Tsinghua University, Beijing, China
| | - Zhen Li
- Ministry of Education Key Laboratory of Bioinformatics, Bioinformatics Division at the Beijing National Research Center for Information Science and Technology, Center for Synthetic and Systems Biology, Department of Automation, Tsinghua University, Beijing, China
| | - Zijing Gao
- Ministry of Education Key Laboratory of Bioinformatics, Bioinformatics Division at the Beijing National Research Center for Information Science and Technology, Center for Synthetic and Systems Biology, Department of Automation, Tsinghua University, Beijing, China
| | - Shengquan Chen
- School of Mathematical Sciences and LPMC, Nankai University, Tianjin, China.
| | - Rui Jiang
- Ministry of Education Key Laboratory of Bioinformatics, Bioinformatics Division at the Beijing National Research Center for Information Science and Technology, Center for Synthetic and Systems Biology, Department of Automation, Tsinghua University, Beijing, China.
| |
Collapse
|
2
|
Cao Y, Zhao X, Tang S, Jiang Q, Li S, Li S, Chen S. scButterfly: a versatile single-cell cross-modality translation method via dual-aligned variational autoencoders. Nat Commun 2024; 15:2973. [PMID: 38582890 PMCID: PMC10998864 DOI: 10.1038/s41467-024-47418-x] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/23/2023] [Accepted: 03/28/2024] [Indexed: 04/08/2024] Open
Abstract
Recent advancements for simultaneously profiling multi-omics modalities within individual cells have enabled the interrogation of cellular heterogeneity and molecular hierarchy. However, technical limitations lead to highly noisy multi-modal data and substantial costs. Although computational methods have been proposed to translate single-cell data across modalities, broad applications of the methods still remain impeded by formidable challenges. Here, we propose scButterfly, a versatile single-cell cross-modality translation method based on dual-aligned variational autoencoders and data augmentation schemes. With comprehensive experiments on multiple datasets, we provide compelling evidence of scButterfly's superiority over baseline methods in preserving cellular heterogeneity while translating datasets of various contexts and in revealing cell type-specific biological insights. Besides, we demonstrate the extensive applications of scButterfly for integrative multi-omics analysis of single-modality data, data enhancement of poor-quality single-cell multi-omics, and automatic cell type annotation of scATAC-seq data. Moreover, scButterfly can be generalized to unpaired data training, perturbation-response analysis, and consecutive translation.
Collapse
Affiliation(s)
- Yichuan Cao
- School of Mathematical Sciences and LPMC, Nankai University, Tianjin, 300071, China
| | - Xiamiao Zhao
- School of Mathematical Sciences and LPMC, Nankai University, Tianjin, 300071, China
| | - Songming Tang
- School of Mathematical Sciences and LPMC, Nankai University, Tianjin, 300071, China
| | - Qun Jiang
- MOE Key Laboratory of Bioinformatics and Bioinformatics Division of BNRIST, Department of Automation, Tsinghua University, 100084, Beijing, China
| | - Sijie Li
- School of Mathematical Sciences and LPMC, Nankai University, Tianjin, 300071, China
| | - Siyu Li
- School of Statistics and Data Science, Nankai University, Tianjin, 300071, China
| | - Shengquan Chen
- School of Mathematical Sciences and LPMC, Nankai University, Tianjin, 300071, China.
| |
Collapse
|
3
|
Li S, Li Y, Sun Y, Li Y, Chen X, Tang S, Chen S. EpiCarousel: memory- and time-efficient identification of metacells for atlas-level single-cell chromatin accessibility data. Bioinformatics 2024; 40:btae191. [PMID: 38588573 PMCID: PMC11037479 DOI: 10.1093/bioinformatics/btae191] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/28/2023] [Revised: 03/02/2024] [Accepted: 04/05/2024] [Indexed: 04/10/2024] Open
Abstract
SUMMARY Recent technical advancements in single-cell chromatin accessibility sequencing (scCAS) have brought new insights to the characterization of epigenetic heterogeneity. As single-cell genomics experiments scale up to hundreds of thousands of cells, the demand for computational resources for downstream analysis grows intractably large and exceeds the capabilities of most researchers. Here, we propose EpiCarousel, a tailored Python package based on lazy loading, parallel processing, and community detection for memory- and time-efficient identification of metacells, i.e. the emergence of homogenous cells, in large-scale scCAS data. Through comprehensive experiments on five datasets of various protocols, sample sizes, dimensions, number of cell types, and degrees of cell-type imbalance, EpiCarousel outperformed baseline methods in systematic evaluation of memory usage, computational time, and multiple downstream analyses including cell type identification. Moreover, EpiCarousel executes preprocessing and downstream cell clustering on the atlas-level dataset with 707 043 cells and 1 154 611 peaks within 2 h consuming <75 GB of RAM and provides superior performance for characterizing cell heterogeneity than state-of-the-art methods. AVAILABILITY AND IMPLEMENTATION The EpiCarousel software is well-documented and freely available at https://github.com/biox-nku/epicarousel. It can be seamlessly interoperated with extensive scCAS analysis toolkits.
Collapse
Affiliation(s)
- Sijie Li
- School of Mathematical Sciences and LPMC, Nankai University, Tianjin 300071, China
| | - Yuxi Li
- School of Mathematical Sciences and LPMC, Nankai University, Tianjin 300071, China
| | - Yu Sun
- Institute of Health Service and Transfusion Medicine, Beijing 100850, China
| | - Yaru Li
- Institute of Health Service and Transfusion Medicine, Beijing 100850, China
| | - Xiaoyang Chen
- MOE Key Laboratory of Bioinformatics and Bioinformatics Division of BNRIST, Department of Automation, Tsinghua University, Beijing 100084, China
| | - Songming Tang
- School of Mathematical Sciences and LPMC, Nankai University, Tianjin 300071, China
| | - Shengquan Chen
- School of Mathematical Sciences and LPMC, Nankai University, Tianjin 300071, China
| |
Collapse
|
4
|
Tang S, Cui X, Wang R, Li S, Li S, Huang X, Chen S. scCASE: accurate and interpretable enhancement for single-cell chromatin accessibility sequencing data. Nat Commun 2024; 15:1629. [PMID: 38388573 PMCID: PMC10884038 DOI: 10.1038/s41467-024-46045-w] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/21/2023] [Accepted: 02/12/2024] [Indexed: 02/24/2024] Open
Abstract
Single-cell chromatin accessibility sequencing (scCAS) has emerged as a valuable tool for interrogating and elucidating epigenomic heterogeneity and gene regulation. However, scCAS data inherently suffers from limitations such as high sparsity and dimensionality, which pose significant challenges for downstream analyses. Although several methods are proposed to enhance scCAS data, there are still challenges and limitations that hinder the effectiveness of these methods. Here, we propose scCASE, a scCAS data enhancement method based on non-negative matrix factorization which incorporates an iteratively updating cell-to-cell similarity matrix. Through comprehensive experiments on multiple datasets, we demonstrate the advantages of scCASE over existing methods for scCAS data enhancement. The interpretable cell type-specific peaks identified by scCASE can provide valuable biological insights into cell subpopulations. Moreover, to leverage the large compendia of available omics data as a reference, we further expand scCASE to scCASER, which enables the incorporation of external reference data to improve enhancement performance.
Collapse
Affiliation(s)
- Songming Tang
- School of Mathematical Sciences and LPMC, Nankai University, Tianjin, 300071, China
| | - Xuejian Cui
- MOE Key Laboratory of Bioinformatics and Bioinformatics Division of BNRIST, Department of Automation, Tsinghua University, 100084, Beijing, China
| | - Rongxiang Wang
- Department of Computer Science, University of Virginia, Charlottesville, VA, 22903, USA
| | - Sijie Li
- School of Mathematical Sciences and LPMC, Nankai University, Tianjin, 300071, China
| | - Siyu Li
- School of Statistics and Data Science, Nankai University, Tianjin, 300071, China
| | - Xin Huang
- Beijing Key Laboratory for Radiobiology, Department of Radiation Biology, Beijing Institute of Radiation Medicine, 100850, Beijing, China
| | - Shengquan Chen
- School of Mathematical Sciences and LPMC, Nankai University, Tianjin, 300071, China.
| |
Collapse
|
5
|
Zhang W, Jiang R, Chen S, Wang Y. scIBD: a self-supervised iterative-optimizing model for boosting the detection of heterotypic doublets in single-cell chromatin accessibility data. Genome Biol 2023; 24:225. [PMID: 37814314 PMCID: PMC10561408 DOI: 10.1186/s13059-023-03072-y] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/30/2023] [Accepted: 09/22/2023] [Indexed: 10/11/2023] Open
Abstract
Application of the widely used droplet-based microfluidic technologies in single-cell sequencing often yields doublets, introducing bias to downstream analyses. Especially, doublet-detection methods for single-cell chromatin accessibility sequencing (scCAS) data have multiple assay-specific challenges. Therefore, we propose scIBD, a self-supervised iterative-optimizing model for boosting heterotypic doublet detection in scCAS data. scIBD introduces an adaptive strategy to simulate high-confident heterotypic doublets and self-supervise for doublet-detection in an iteratively optimizing manner. Comprehensive benchmarking on various simulated and real datasets demonstrates the outperformance and robustness of scIBD. Moreover, the downstream biological analyses suggest the efficacy of doublet-removal by scIBD.
Collapse
Affiliation(s)
- Wenhao Zhang
- Department of Automation, Xiamen University, Xiamen, 361000, Fujian, China
- National Institute for Data Science in Health and Medicine, Xiamen University, Xiamen, 361000, Fujian, China
| | - Rui Jiang
- Ministry of Education Key Laboratory of Bioinformatics, Research Department of Bioinformatics at the Beijing National Research Center for Information Science and Technology, Center for Synthetic and Systems Biology, Department of Automation, Tsinghua University, Beijing, 100084, China
| | - Shengquan Chen
- School of Mathematical Sciences and LPMC, Nankai University, Tianjin, 300071, China.
| | - Ying Wang
- Department of Automation, Xiamen University, Xiamen, 361000, Fujian, China.
- National Institute for Data Science in Health and Medicine, Xiamen University, Xiamen, 361000, Fujian, China.
- Xiamen Key Laboratory of Big Data Intelligent Analysis and Decision, Xiamen, 361005, Fujian, China.
| |
Collapse
|
6
|
Jiang R, Li Z, Jia Y, Li S, Chen S. SINFONIA: Scalable Identification of Spatially Variable Genes for Deciphering Spatial Domains. Cells 2023; 12:cells12040604. [PMID: 36831270 PMCID: PMC9954745 DOI: 10.3390/cells12040604] [Citation(s) in RCA: 4] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/04/2023] [Revised: 01/29/2023] [Accepted: 02/07/2023] [Indexed: 02/17/2023] Open
Abstract
Recent advances in spatial transcriptomics have revolutionized the understanding of tissue organization. The identification of spatially variable genes (SVGs) is an essential step for downstream spatial domain characterization. Although several methods have been proposed for identifying SVGs, inadequate ability to decipher spatial domains, poor efficiency, and insufficient interoperability with existing standard analysis workflows still impede the applications of these methods. Here we propose SINFONIA, a scalable method for identifying spatially variable genes via ensemble strategies. Implemented in Python, SINFONIA can be seamlessly integrated into existing analysis workflows. Using 15 spatial transcriptomic datasets generated with different protocols and with different sizes, dimensions and qualities, we show the advantage of SINFONIA over three baseline methods and two variants via systematic evaluation of spatial clustering, domain resolution, latent representation, spatial visualization, and computational efficiency with 21 quantitative metrics. Additionally, SINFONIA is robust relative to the choice of the number of SVGs. We anticipate SINFONIA will facilitate the analysis of spatial transcriptomics.
Collapse
Affiliation(s)
- Rui Jiang
- MOE Key Laboratory of Bioinformatics and Bioinformatics Division, BNRIST/Department of Automation, Tsinghua University, Beijing 100084, China
| | - Zhen Li
- MOE Key Laboratory of Bioinformatics and Bioinformatics Division, BNRIST/Department of Automation, Tsinghua University, Beijing 100084, China
| | - Yuhang Jia
- School of Statistics and Data Science, Nankai University, Tianjin 300071, China
| | - Siyu Li
- School of Statistics and Data Science, Nankai University, Tianjin 300071, China
| | - Shengquan Chen
- School of Mathematical Sciences and LPMC, Nankai University, Tianjin 300071, China
- Correspondence:
| |
Collapse
|