Deconvolution of single-cell multi-omics layers reveals regulatory heterogeneity

Liu, Longqi; Liu, Chuanyu; Wu, Liang; Quintero, Andrés; Yuan, Yue; Wang, Mingyue; Cheng, Mengnan; Xu, Liqin; Dong, Guoyi; Li, Rui; Liu, Yang; Wei, Xiaoyü; Xu, Jiangshan; Chen, Xiaowei; Lu, Haorong; Chen, Dongsheng; Wang, Quanlei; Zhou, Qing; Lin, Xinxin; Li, Guibo; Liu, Shiping; Wang, Qi; Wang, Hongru; Fink, J. Lynn; Gao, Zhengliang; Liu, Xin; Hou, Yong; Zhu, Shida; Yang, Huanming; Ye, Yunming; Chen, Fang; Herrmann, Carl; Eils, Roland; Shang, Zhouchun; Xu, Xun

doi:10.1101/316208

Cited by 27 publications

(29 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This complexity should be taken into account to enhance our interpretation of genetic discoveries in AD. For example, our data on cell-type specific expression of GWAS genes will prompt expression quantitative trait loci (eQTL, 1 ), splice QTL (sQTL, 53 ) and single-cell ATAC-Seq analyses 70 tailored to specific cell populations, to inform functional specialization of AD gene variants and “regional” ( i.e. , cell-type specific) susceptibility to disease.…”

Section: Discussionmentioning

confidence: 99%

A single cell brain atlas in human Alzheimer’s disease

Grubman

Chew

Ouyang

et al. 2019

Preprint

View full text Add to dashboard Cite

26Alzheimer's disease (AD) is a heterogeneous disease that is largely dependent on the complex 27 cellular microenvironment in the brain. This complexity impedes our understanding of how 28 individual cell types contribute to disease progression and outcome. To characterize the 29 molecular and functional cell diversity in the human AD brain we utilized single nuclei RNA-30 seq in AD and control patient brains in order to map the landscape of cellular heterogeneity in 31 AD. We detail gene expression changes at the level of cells and cell subclusters, highlighting 32 specific cellular contributions to global gene expression patterns between control and 33 Alzheimer's patient brains. We observed distinct cellular regulation of APOE which was 34 repressed in oligodendrocyte progenitor cells (OPCs) and astrocyte AD subclusters, and highly 35 enriched in a microglial AD subcluster. In addition, oligodendrocyte and microglia AD 36 subclusters show discordant expression of APOE. Integration of transcription factor regulatory 37 modules with downstream GWAS gene targets revealed subcluster-specific control of AD cell 38 fate transitions. For example, this analysis uncovered that astrocyte diversity in AD was under 39 the control of transcription factor EB (TFEB), a master regulator of lysosomal function and 40 which initiated a regulatory cascade containing multiple AD GWAS genes. These results 41 establish functional links between specific cellular sub-populations in AD, and provide new 42 insights into the coordinated control of AD GWAS genes and their cell-type specific 43 contribution to disease susceptibility. Finally, we created an interactive reference web resource 44 which will facilitate brain and AD researchers to explore the molecular architecture of subtype 45 and AD-specific cell identity, molecular and functional diversity at the single cell level. 46Highlights 47 • We generated the first human single cell transcriptome in AD patient brains 48 • Our study unveiled 9 clusters of cell-type specific and common gene expression 49 patterns between control and AD brains, including clusters of genes that present 50 properties of different cell types (i.e. astrocytes and oligodendrocytes) 51 • Our analyses also uncovered functionally specialized sub-cellular clusters: 5 52 microglial clusters, 8 astrocyte clusters, 6 neuronal clusters, 6 oligodendrocyte 53 clusters, 4 OPC and 2 endothelial clusters, each enriched for specific ontological 54 gene categories 55 • Our analyses found manifold AD GWAS genes specifically associated with one 56 cell-type, and sets of AD GWAS genes co-ordinately and differentially regulated 57 between different brain cell-types in AD sub-cellular clusters 58 • We mapped the regulatory landscape driving transcriptional changes in AD brain, 59 and identified transcription factor networks which we predict to control cell fate 60 transitions between control and AD sub-cellular clusters 61 • Finally, we provide an interactive web-resource that allows the user to further 62 visua...

show abstract

Section: Discussionmentioning

confidence: 99%

A single cell brain atlas in human Alzheimer’s disease

Grubman

Chew

Ouyang

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…To test the jDR approaches on single-cell omics, we fetched scRNA-seq and scATAC-seq, simultaneously measuring gene expression and chromatin accessibility on three cancer cell lines (HTC, Hela and K562) for a total of 206 cells, and reported in the study of Liu and colleagues 32 . As these cells have been obtained from three different cancer cell lines, we expect that the first two factors of the various jDR approaches would cluster single-cells according to their cancer cell line of origin.…”

Section: Resultsmentioning

confidence: 99%

Benchmarking joint multi-omics dimensionality reduction approaches for the study of cancer

et al. 2021

View full text Add to dashboard Cite

High-dimensional multi-omics data are now standard in biology. They can greatly enhance our understanding of biological systems when effectively integrated. To achieve proper integration, joint Dimensionality Reduction (jDR) methods are among the most efficient approaches. However, several jDR methods are available, urging the need for a comprehensive benchmark with practical guidelines. We perform a systematic evaluation of nine representative jDR methods using three complementary benchmarks. First, we evaluate their performances in retrieving ground-truth sample clustering from simulated multi-omics datasets. Second, we use TCGA cancer data to assess their strengths in predicting survival, clinical annotations and known pathways/biological processes. Finally, we assess their classification of multi-omics single-cell data. From these in-depth comparisons, we observe that intNMF performs best in clustering, while MCIA offers an effective behavior across many contexts. The code developed for this benchmark study is implemented in a Jupyter notebook—multi-omics mix (momix)—to foster reproducibility, and support users and future developers.

show abstract

“…In conclusion, our benchmarking results highlight strengths and weaknesses of latent factor models applied to scRNA-seq data for different tasks. We focused our work on the application of these methods for the purpose of signature discovery, yet latent factor models have been implemented for a variety of purposes (such as denoising 38,39 and multi-omics integration 40,41 ). We propose a framework that makes use of the full spectrum of latent variables and allows to directly define cell subtypes based on their function or cellular phenotype.…”

Section: Discussionmentioning

confidence: 99%

Latent factor modelling of scRNA-seq data uncovers novel pathways dysregulated in cell subsets of autoimmune disease patients

Palla

Ferrero

2019

Preprint

View full text Add to dashboard Cite

Latent factor modelling applied to single-cell RNA-sequencing (scRNA-seq) data is a useful approach to discover gene signatures associated with cell states. However, it is often unclear what method is best suited for specific tasks and how latent factors should be interpreted from a biological perspective.Here, we compare four state-of-the-art methods and explore their stability, predictive power and coverage of known biology. We then propose an approach that leverages the derived latent factors to directly assign pathway activities to specific cell subsets. By applying this framework to scRNA-seq datasets from biopsies of rheumatoid arthritis and systemic lupus erythematosus patients, we discover both known and novel disease-relevant gene signatures in specific cellular subsets in a fully unsupervised way. Focusing on rheumatoid arthritis, we identify an inflammatory Oncostatin M receptor signalling signature active in a subset of synovial fibroblasts and dysregulation of the GAS6 -MERTK axis in a subset of synovial monocytes with efferocytic function.Overall, we provide insights into strengths and weaknesses of latent factors models for the analysis of scRNA-seq data, we develop a framework to identify cell subtypes in a function-or phenotype-driven way and use it to identify novel pathways dysregulated in rheumatoid arthritis.

show abstract

Deconvolution of single-cell multi-omics layers reveals regulatory heterogeneity

Cited by 27 publications

References 35 publications

A single cell brain atlas in human Alzheimer’s disease

A single cell brain atlas in human Alzheimer’s disease

Benchmarking joint multi-omics dimensionality reduction approaches for the study of cancer

Latent factor modelling of scRNA-seq data uncovers novel pathways dysregulated in cell subsets of autoimmune disease patients

Contact Info

Product

Resources

About