Integrative analysis of single-cell genomics data by coupled nonnegative matrix factorizations

Duren, Zhana; Chen, Xi; Zamanighomi, Mahdi; Zeng, Wanwen; Satpathy, Ansuman T.; Chang, Howard Y.; Wang, Yong; Wong, Wing Hung

doi:10.1073/pnas.1805681115

Cited by 159 publications

(110 citation statements)

References 28 publications

Supporting

Mentioning

110

Contrasting

Order By: Relevance

“…We then clustered mESCs and 2i cells using coupled Non-negative Matrix Factorization (NMF) 19 , based on the correlative information of transcriptome and chromatin accessibility of each individual cell. Notably, when clustering was based on either the differentially expressed genes or the accessible regions identified by coupled NMF, two distinct clusters were always observed ( Supplementary Fig.…”

Section: Resultsmentioning

confidence: 99%

Parallel Bimodal Single-cell Sequencing of Transcriptome and Chromatin Accessibility

Xing

Farran

Yao

et al. 2019

Preprint

View full text Add to dashboard Cite

Accessibility, Co-accessibility. RUNNING TITLE: Multi-omic single-cell sequencing of gene expression and open chromatin SUMMARY We developed ASTAR-Seq (Assay for Single-cell Transcriptome and Accessibility Regions) integrated with automated microfluidic chips, which allows for parallel sequencing of transcriptome and chromatin accessibility within the same single-cell. Using ASTAR-Seq, we profiled 192 mESCs cultured in serum+LIF and 2i medium, 424 human cell lines including BJ, K562, JK1, and Jurkat, and 480 primary cells undergoing erythroblast differentiation. Integrative analysis using Coupled NMF identified the distinct subpopulations and uncovered sets of regulatory regions and the respective target genes determining their distinctions. Analysis of epigenetic regulomes further unravelled the key transcription factors responsible for the heterogeneity observed.

show abstract

Section: Resultsmentioning

confidence: 99%

Parallel Bimodal Single-cell Sequencing of Transcriptome and Chromatin Accessibility

Xing

Farran

Yao

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…First, a more well-crafted module, such as modules considering dropout evens, can be introduced to better characterize the scRNA-seq data, and thus further improves the performance of our method. Second, our method can be extended to incorporate other types of functional genomics data such as chromatin accessibility [23,24]. Finally, drawing on the idea of VPAC, we can integrate the feature selection module with other modules to endow the method with the ability to balance the feature selection and prediction steps, and thus extract features that are more conducive to the cell type classification [25].…”

Section: Enclasc Enables Cross-species Classificationmentioning

confidence: 99%

EnClaSC: A novel ensemble approach for accurate and robust cell-type classification of single-cell transcriptomes

Chen

Jiang

2019

Preprint

View full text Add to dashboard Cite

BackgroundIn recent years, the rapid development of single-cell RNA-sequencing (scRNA-seq) techniques enables the quantitative characterization of cell types at a single-cell resolution. With the explosive growth of the number of cells profiled in individual scRNA-seq experiments, there is a demand for novel computational methods for classifying newly-generated scRNA-seq data onto annotated labels. Although several methods have recently been proposed for the cell-type classification of single-cell transcriptomic data, such limitations as inadequate accuracy, inferior robustness, and low stability greatly limit their wide applications. ResultsWe propose a novel ensemble approach, named EnClaSC, for accurate and robust celltype classification of single-cell transcriptomic data. Through comprehensive validation experiments, we demonstrate that EnClaSC can not only be applied to the self-projection within a specific dataset and the cell-type classification across different datasets, but also scale up well to various data dimensionality and different data sparsity.We further illustrate the ability of EnClaSC to effectively make cross-species classification, which may shed light on the studies in correlation of different species.EnClaSC is freely available at https://github.com/xy-chen16/EnClaSC. ConclusionsEnClaSC enables highly accurate and robust cell-type classification of single-cell transcriptomic data via an ensemble learning method. We expect to see wide applications of our method to not only transcriptome studies, but also the classification of more general data.

show abstract

“…Non-negative matrix factorization (NMF) techniques have emerged as powerful tools to identify the cellular and molecular features that are associated with distinct biological processes from single cell data (Cleary et al, 2017;Zhu et al, 2017;Clark et al, 2019;Welch et al, 2019;Kotliar et al, 2019;Duren et al, 2018). Bayesian factorization approaches can mitigate local optima and leverage prior distributions to encode biological structure in the features Stein-O'Brien et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

CoGAPS 3: Bayesian non-negative matrix factorization for single-cell analysis with asynchronous updates and sparse data structures

Sherman

Gao

Fertig

2019

Preprint

View full text Add to dashboard Cite

Motivation: Bayesian factorization methods, including Coordinated Gene Activity in Pattern Sets (CoGAPS), are emerging as powerful analysis tools for single cell data. However, these methods have greater computational costs than their gradient-based counterparts. These costs are often prohibitive for analysis of large single-cell datasets. Many such methods can be run in parallel which enables this limitation to be overcome by running on more powerful hardware. However, the constraints imposed by the prior distributions in CoGAPS limit the applicability of parallelization methods to enhance computational efficiency for single-cell analysis. Results: We upgraded CoGAPS in Version 3 to overcome the computational limitations of Bayesian matrix factorization for single cell data analysis. This software includes a new parallelization framework that is designed around the sequential updating steps of the algorithm to enhance computational efficiency. These algorithmic advances were coupled with new software architecture and sparse data structures to reduce the memory overhead for single-cell data. Altogether, these updates to CoGAPS enhance the efficiency of the algorithm so that it can analyze 1000 times more cells, enabling factorization of large single-cell data sets. Availability: CoGAPS is available as a Bioconductor package and the source code is provided at github.com/FertigLab/CoGAPS. All efficiency updates to enable single-cell analysis available as of version 3.2.

show abstract

Integrative analysis of single-cell genomics data by coupled nonnegative matrix factorizations

Cited by 159 publications

References 28 publications

Parallel Bimodal Single-cell Sequencing of Transcriptome and Chromatin Accessibility

Parallel Bimodal Single-cell Sequencing of Transcriptome and Chromatin Accessibility

EnClaSC: A novel ensemble approach for accurate and robust cell-type classification of single-cell transcriptomes

CoGAPS 3: Bayesian non-negative matrix factorization for single-cell analysis with asynchronous updates and sparse data structures

Contact Info

Product

Resources

About