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

Abstract

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.

Data Driven Mathematical Model of Colon Cancer Progression

“…In recent years, several tumor deconvolution methods have been developed to estimate the relative abundance of various cell types in a tumor from its gene expression profile. A review of these methods [97] and an application of CIBERSORTx on renal cancer [98] show a great performance of CIBERSORTx model. To identify the immune profiles of colonic tumors, we applied CIBERSORTx [99] on RNA-seq gene expression profiles of primary tumors of patients with colon cancer from the Cancer Genome Atlas (TCGA) project of Colon Adenocarcinoma (COAD) downloaded from University of California Santa Cruz (UCSC) Xena web portal.…”

Immune classification of osteosarcoma

“…There are 2 modes of this algorithm: B-mode and S-mode, whose difference lies in the way batch correction is applied in the algorithm. Our recent study has shown that CIBERSORTx B-mode gives good estimates of immune abundance in both RNA-Seq and microarray data with the use of LM22, and in fact outperforms CIBERSORT and other tumor deconvolution methods [ 16 ]. We obtain estimated immune fractions by running the algorithm on their website with 100 permutations.…”

“…Several computational methods, alternative ways to immunohistochemistry and flow cytometry [ 15 ], have been recently developed to derive tumor immune infiltrates using gene expression profiles of bulk tumors. In this study, we use the latest popular deconvolution method [ 16 – 20 ], CIBERSORTx [ 21 ], to investigate the immune patterns of tumors and analyze the relationship between immune composition and clinical features of osteosarcoma patients.…”

Data driven mathematical model of colon cancer progression

“…In recent years, several tumor deconvolution methods have been developed to estimate the relative abundance of various cell types in a tumor from its gene expression profile. A review of these methods [79] and an application of CIBERSORTx on renal cancer [80] show a great performance of CIBERSORTx model. To identify the immune profiles of colonic tumors, we applied CIBERSORTx [81] on RNA-seq gene expression profiles of primary tumors of patients with colon cancer from the TCGA project of COAD downloaded from UCSC Xena web portal.…”