Spatially informed cell-type deconvolution for spatial transcriptomics

Ma, Ying; Zhou, Xiang

doi:10.1038/s41587-022-01273-7

Cited by 198 publications

(241 citation statements)

References 95 publications

Supporting

Mentioning

189

Contrasting

Order By: Relevance

“…To further investigate the spatial location of single cell clusters in HCC TME, we employed “CARD” package [ 24 ] to deconvolute the spatial transcriptomic data based on our single cell data of cancer state. The spatial transcriptomic data were retrieved from previous study [ 25 ].…”

Section: Methodsmentioning

confidence: 99%

Single-cell transcriptomics reveals the role of Macrophage-Naïve CD4 + T cell interaction in the immunosuppressive microenvironment of primary liver carcinoma

Liu

Chen

et al. 2022

J Transl Med

View full text Add to dashboard Cite

Background Liver carcinoma generally presents as an immunosuppressive microenvironment that promotes tumor evasion. The intercellular crosstalk of immune cells significantly influences the construction of an immunosuppressive microenvironment. This study aimed to investigate the important interactions between immune cells and their targeting drugs in liver carcinoma, by using single-cell and bulk transcriptomic data. Methods Single-cell and bulk transcriptomic data were retrieved from Gene Expression Omnibus (GSE159977, GSE136103, and GSE125449) and The Cancer Genome Atlas (TGCA-LIHC), respectively. Quality control, dimension reduction, clustering, and annotation were performed according to the Scanpy workflow based on Python. Cell–cell interactions were explored using the CellPhone database and CellChat. Trajectory analysis was executed using a partition-based graph abstraction method. The transcriptomic factors (TFs) were predicted using single-cell regulatory network inference and clustering (SCENIC). The target genes from TFs were used to establish a related score based on the TCGA cohort; this score was subsequently validated by survival, gene set enrichment, and immune cell infiltration analyses. Drug prediction was performed based on the Cancer Therapeutics Response Portal and PRISM Repurposing datasets. Results Thirty-one patients at four different states, including health, hepatitis, cirrhosis, and cancer, were enrolled in this study. After dimension reduction and clustering, twenty-two clusters were identified. Cell–cell interaction analyses indicated that macrophage-naive CD4 + T cell interaction significantly affect cancerous state. In brief, macrophages interact with naive CD4 + T cells via different pathways in different states. The results of SCENIC indicated that macrophages present in cancer cells were similar to those present during cirrhosis. A macrophage-naive CD4 + T cell (MNT) score was generated by the SCENIC-derived target genes. Based on the MNT score, five relevant drugs (inhibitor of polo-like kinase 1, inhibitor of kinesin family member 11, dabrafenib, ispinesib, and epothilone-b) were predicted. Conclusions This study reveals the crucial role of macrophage-naive CD4 + T cell interaction in the immunosuppressive microenvironment of liver carcinoma. Tumor-associated macrophages may be derived from cirrhosis and can initiate liver carcinoma. Predictive drugs that target the macrophage-naive CD4 + T cell interaction may help to improve the immunosuppressive microenvironment and prevent immune evasion. The relevant mechanisms need to be further validated in experiments and cohort studies. Graphical Abstract

show abstract

Section: Methodsmentioning

confidence: 99%

Single-cell transcriptomics reveals the role of Macrophage-Naïve CD4 + T cell interaction in the immunosuppressive microenvironment of primary liver carcinoma

Liu

Chen

et al. 2022

J Transl Med

View full text Add to dashboard Cite

show abstract

“…We benchmarked BayesTME against other methods: BayesSpace 11 , cell2location 16 , DestVI 15 , CARD 29 , RCTD 30 , STdeconvolve 14 , stLearn 12 , and Giotto 13 on simulated data based on real single-cell RNA sequencing (scRNA) data. We randomly sampled K * cell types from a previously-clustered scRNA dataset 16 ; we conducted experiments for K * from 3 to 8.…”

Section: Resultsmentioning

confidence: 99%

BayesTME: A unified statistical framework for spatial transcriptomics

Zhang

Hunter

Chou

et al. 2022

Preprint

View full text Add to dashboard Cite

Spatial variation in cellular phenotypes underlies heterogeneity in immune recognition and response to therapy in cancer and many other diseases. Spatial transcriptomics (ST) holds the potential to quantify such variation, but existing analysis methods address only a small part of the analysis challenge, such as spot deconvolution or spatial differential expression. We present BayesTME, an end-to-end Bayesian method for analyzing spatial transcriptomics data. BayesTME unifies several previously distinct analysis goals under a single, holistic generative model. This unified approach enables BayesTME to (i) be entirely reference-free without any need for paired scRNA-seq, (ii) outperform a large suite of methods in quantitative benchmarks, and (iii) uncover a new type of ST signal: spatial differential expression within individual cell types. To achieve the latter, BayesTME models each phenotype as spatially adaptive and discovers statistically significant spatial patterns amongst coordinated subsets of genes within phenotypes, which we term spatial transcriptional programs. On human and zebrafish melanoma tissues, BayesTME identifies spatial transcriptional programs that capture fundamental biological phenomena like bilateral symmetry, differential expression between interior and surface tumor cells, and tumor-associated fibroblast and macrophage reprogramming. Our results demonstrate BayesTME's power in unlocking a new level of insight from spatial transcriptomics data and fostering a deeper understanding of the spatial architecture of the tumor microenvironment. BayesTME is open source and publicly available (https://github.com/tansey-lab/bayestme).

show abstract

“…Most deconvolution methods use annotated reference transcriptome profiles derived from scRNA-seq or bulk RNA-seq datasets, making the accuracy and resolution of the cell type identification highly dependent on the compatibility of the used reference profiles with the cells in the target sample. However, the recently published conditional autoregressive-based deconvolution (CARD) [90] and the latent Dirichlet allocation based STdeconvolve [91] offer reference-free deconvolution methods, which is useful when optimal reference scRNA-seq profiles are not available.…”

Section: Spatial Barcoding Methodsmentioning

confidence: 99%