A review of digital cytometry methods: estimating the relative abundance of cell types in a bulk of cells.

Due to the high cost of flow and mass cytometry, there has been a recent surge in the development of computational methods for estimating the relative distributions of cell types from the gene expression profile of a bulk of cells. Here, we review the five common 'digital cytometry' methods: deconvolution of RNA-Seq, cell-type identification by estimating relative subsets of RNA transcripts (CIBERSORT), CIBERSORTx, single sample gene set enrichment analysis and single-sample scoring of molecular phenotypes deconvolution method. The results show that CIBERSORTx B-mode, which uses batch correction to adjust the gene expression profile of the bulk of cells ('mixture data') to eliminate possible cross-platform variations between the mixture data and the gene expression data of single cells ('signature matrix'), outperforms other methods, especially when signature matrix and mixture data come from different platforms. However, in our tests, CIBERSORTx S-mode, which uses batch correction for adjusting the signature matrix instead of mixture data, did not perform better than the original CIBERSORT method, which does not use any batch correction method. This result suggests the need for further investigations into how to utilize batch correction in deconvolution methods.

Single sample scoring of molecular phenotypes.

BACKGROUND: Gene set scoring provides a useful approach for quantifying concordance between sample transcriptomes and selected molecular signatures. Most methods use information from all samples to score an individual sample, leading to unstable scores in small data sets and introducing biases from sample composition (e.g. varying numbers of samples for different cancer subtypes). To address these issues, we have developed a truly single sample scoring method, and associated R/Bioconductor package singscore ( https://bioconductor.org/packages/singscore ). RESULTS: We use multiple cancer data sets to compare singscore against widely-used methods, including GSVA, z-score, PLAGE, and ssGSEA. Our approach does not depend upon background samples and scores are thus stable regardless of the composition and number of samples being scored. In contrast, scores obtained by GSVA, z-score, PLAGE and ssGSEA can be unstable when less data are available (NS < 25). The singscore method performs as well as the best performing methods in terms of power, recall, false positive rate and computational time, and provides consistently high and balanced performance across all these criteria. To enhance the impact and utility of our method, we have also included a set of functions implementing visual analysis and diagnostics to support the exploration of molecular phenotypes in single samples and across populations of data. CONCLUSIONS: The singscore method described here functions independent of sample composition in gene expression data and thus it provides stable scores, which are particularly useful for small data sets or data integration. Singscore performs well across all performance criteria, and includes a suite of powerful visualization functions to assist in the interpretation of results. This method performs as well as or better than other scoring approaches in terms of its power to distinguish samples with distinct biology and its ability to call true differential gene sets between two conditions. These scores can be used for dimensional reduction of transcriptomic data and the phenotypic landscapes obtained by scoring samples against multiple molecular signatures may provide insights for sample stratification.

TumorDecon: A digital cytometry software.

There are many experimental methods for characterizing immune profiles of tumors, such as flow and mass cytometry. However, these approaches are time and resource intensive. Thus, several "digital cytometry" methods have been developed to extract cell frequencies from RNA-seq data. Here, we introduce TumorDecon, named for its potential to deconvolve the distribution of cells from the gene expression levels of a bulk of cells, such as a tumor. The Python package provides an accessible way of applying these methods. It includes four deconvolution methods as well as several gene sets, signature matrices, and functions for generating custom signature matrices.

SLUG-related partial epithelial-to-mesenchymal transition is a transcriptomic prognosticator of head and neck cancer survival.

Partial epithelial-to-mesenchymal transition (pEMT) contributes to cellular heterogeneity that is associated with nodal metastases and unfavorable clinical parameters in head and neck squamous cell carcinomas (HNSCCs). We developed a single-cell RNA sequencing signature-based pEMT quantification through cell type-dependent deconvolution of bulk RNA sequencing and microarray data combined with single-sample scoring of molecular phenotypes (Singscoring). Clinical pEMT-Singscores served as molecular classifiers in multivariable Cox proportional hazard models and high scores prognosticated poor overall survival and reduced response to irradiation as independent parameters in large HNSCC cohorts [The Cancer Genome Atlas (TCGA), MD Anderson Cancer Centre (MDACC), Fred Hutchinson Cancer Research Center (FHCRC)]. Differentially expressed genes confirmed enhanced cell motility and reduced oxidative phosphorylation and epithelial differentiation in pEMThigh patients. In patients and cell lines, the EMT transcription factor SLUG correlated most strongly with pEMT-Singscores and promoted pEMT, enhanced invasion, and resistance to irradiation in vitro. SLUG protein levels in HNSCC predicted disease-free survival, and its peripheral expression at the interphase to the tumor microenvironment was significantly increased in relapsing patients. Hence, pEMT-Singscores represent a novel risk predictor for HNSCC stratification regarding clinical outcome and therapy response that is partly controlled by SLUG.