Dataset Information

Evaluation of single-cell classifiers for single-cell RNA sequencing data sets.

ABSTRACT: Single-cell RNA sequencing (scRNA-seq) has been rapidly developing and widely applied in biological and medical research. Identification of cell types in scRNA-seq data sets is an essential step before in-depth investigations of their functional and pathological roles. However, the conventional workflow based on clustering and marker genes is not scalable for an increasingly large number of scRNA-seq data sets due to complicated procedures and manual annotation. Therefore, a number of tools have been developed recently to predict cell types in new data sets using reference data sets. These methods have not been generally adapted due to a lack of tool benchmarking and user guidance. In this article, we performed a comprehensive and impartial evaluation of nine classification software tools specifically designed for scRNA-seq data sets. Results showed that Seurat based on random forest, SingleR based on correlation analysis and CaSTLe based on XGBoost performed better than others. A simple ensemble voting of all tools can improve the predictive accuracy. Under nonideal situations, such as small-sized and class-imbalanced reference data sets, tools based on cluster-level similarities have superior performance. However, even with the function of assigning 'unassigned' labels, it is still challenging to catch novel cell types by solely using any of the single-cell classifiers. This article provides a guideline for researchers to select and apply suitable classification tools in their analysis workflows and sheds some lights on potential direction of future improvement on classification tools.

SUBMITTER: Zhao X

PROVIDER: S-EPMC7947964 | biostudies-literature | 2020 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Evaluation of single-cell classifiers for single-cell RNA sequencing data sets.

Zhao Xinlei X Wu Shuang S Fang Nan N Sun Xiao X Fan Jue J

Briefings in bioinformatics 20200901 5

Single-cell RNA sequencing (scRNA-seq) has been rapidly developing and widely applied in biological and medical research. Identification of cell types in scRNA-seq data sets is an essential step before in-depth investigations of their functional and pathological roles. However, the conventional workflow based on clustering and marker genes is not scalable for an increasingly large number of scRNA-seq data sets due to complicated procedures and manual annotation. Therefore, a number of tools have ...[more]

PMID: 31675098

Similar Datasets

Project description:Background: Identification of cell type subpopulations from complex cell mixtures using single-cell RNA-sequencing (scRNA-seq) data includes automated steps from normalization to cell clustering. However, assigning cell type labels to cell clusters is often conducted manually, resulting in limited documentation, low reproducibility and uncontrolled vocabularies. This is partially due to the scarcity of reference cell type signatures and because some methods support limited cell type signatures. Methods: In this study, we benchmarked five methods representing first-generation enrichment analysis (ORA), second-generation approaches (GSEA and GSVA), machine learning tools (CIBERSORT) and network-based neighbor voting (METANEIGHBOR), for the task of assigning cell type labels to cell clusters from scRNA-seq data. We used five scRNA-seq datasets: human liver, 11 Tabula Muris mouse tissues, two human peripheral blood mononuclear cell datasets, and mouse retinal neurons, for which reference cell type signatures were available. The datasets span Drop-seq, 10X Chromium and Seq-Well technologies and range in size from ~3,700 to ~68,000 cells. Results: Our results show that, in general, all five methods perform well in the task as evaluated by receiver operating characteristic curve analysis (average area under the curve (AUC) = 0.91, sd = 0.06), whereas precision-recall analyses show a wide variation depending on the method and dataset (average AUC = 0.53, sd = 0.24). We observed an influence of the number of genes in cell type signatures on performance, with smaller signatures leading more frequently to incorrect results. Conclusions: GSVA was the overall top performer and was more robust in cell type signature subsampling simulations, although different methods performed well using different datasets. METANEIGHBOR and GSVA were the fastest methods. CIBERSORT and METANEIGHBOR were more influenced than the other methods by analyses including only expected cell types. We provide an extensible framework that can be used to evaluate other methods and datasets at https://github.com/jdime/scRNAseq_cell_cluster_labeling.

Dataset Information

Evaluation of single-cell classifiers for single-cell RNA sequencing data sets.

Publications

Evaluation of single-cell classifiers for single-cell RNA sequencing data sets.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets