Fully-automated and ultra-fast cell-type identification using specific marker combinations from single-cell transcriptomic data

Single-cell transcriptomics enables systematic charting of cellular composition of complex tissues. Identification of cell populations often relies on unsupervised clustering of cells based on the similarity of their scRNA-seq profiles, followed by manual annotation of cell clusters using established marker genes. However, manual selection of marker genes is a time-consuming process that may lead to sub-optimal annotation results as the selected markers must be informative of both the individual cell clusters and various cell types present in the complex samples. Here, we developed a computational platform, termed ScType, which enables data-driven, fully-automated and ultra-fast cell-type identification based solely on given scRNA-seq data, combined with our comprehensive cell marker database as background information. Using a compendium of six scRNA-seq datasets from various human and mouse tissues, we show how ScType provides an unbiased and accurate cell-type annotation by guaranteeing the specificity of positive and negative marker genes both across cell clusters and cell types. We also demonstrate how ScType enables distinguishing between healthy and malignant cell populations, based on single-cell calling of single-nucleotide variants, making it a versatile tool for exploration and use of single-cell transcriptomic data for anticancer applications. The widely-applicable method is deployed both as an interactive web-tool (https://sctype.app), and as an open-source R-package, connected with a comprehensive ScType database of specific markers.