Determining cell type abundance and expression from bulk tissues with digital cytometry

Abstract

Single-cell RNA-seq (scRNA-seq) has emerged as a powerful technique for characterizing cellular heterogeneity, but it is currently impractical on large sample cohorts and cannot be applied to fixed specimens collected as part of routine clinical care. We previously developed an approach for digital cytometry, called CIBERSORT, that enables estimation of cell type abundances from bulk tissue transcriptomes. We now introduce CIBERSORTx, a machine learning method that extends this framework to infer cell-type-specific gene expression profiles without physical cell isolation. By minimizing platform-specific variation, CIBERSORTx also allows the use of scRNA-seq data for large-scale tissue dissection. We evaluated the utility of CIBERSORTx in multiple tumor types, including melanoma, where single-cell reference profiles were used to dissect bulk clinical specimens, revealing cell type-specific phenotypic states linked to distinct driver mutations and response to immune checkpoint blockade. We anticipate that digital cytometry will augment single-cell profiling efforts, enabling cost-effective, high-throughput tissue characterization without the need for antibodies, disaggregation, or viable cells.

Differential RNA editing between epithelial and mesenchymal tumors impacts mRNA abundance in immune response pathways

“…2A-C, Methods). Supporting the accuracy of this clustering, expression of marker genes was generally highest in their expected cell types when RPKM was calculated from pooled cells and when a signature gene expression matrix was predicted by CIBERSORTx 41 (Supplementary Fig. 2D).…”

ADAPTS: Automated Deconvolution Augmentation of Profiles for Tissue Specific cells

“…Determining cell type enrichment from gene expression data is an step towards determining tumor immune context (Thorsson et al, 2018;Newman et al, 2019). One family of techniques for doing this involves regression using a signature matrix (typically with several hundred genes), where each column represents a cell type and each row contains the average gene expression in that cell type (Erkkilä et al, 2010;Lähdesmäki et al, 2005).…”

DAISM-DNNXMBD: Highly accurate cell type proportion estimation with in silico data augmentation and deep neural networks

“…To bridge the gap, numerous computational methods have been proposed to estimate individual cell type abundance from bulk RNA data of heterogeneous tissues ( Supplementary Table S1). With the bulk gene expression values as input, the abundance of each cell type from the mixed sample can be quantified by aggregating the expressions of the marker genes into an abundance score (MCP-counter 9 ), or by measuring the enrichment level of the marker genes using statistical analysis (xCell 10 ), or by deconvolution algorithms that adopt computational methods, such as least squares (quanTiseq 11 , EPIC 12 ), support vector regression (SVR) (CIBERSORT 13 , CIBERSORTx 14 ), or non-negative matrix factorization (NMF) 15 , to derive an optimal dissection of the original sample based on a set of pre-identified cell type-specific expression signatures. Obviously, regardless of the actual computational methods being used, the adoption of any of these methods as a reliable clinical routine for cell type proportion estimation requires that its underlying assumptions to be held over a large variety of cell types, tissues, and RNA sequencing conditions, which is challenging in practice.…”