Abstract:The development of single-cell RNA sequencing (scRNA-seq) has enabled scientists to catalog and probe the transcriptional heterogeneity of individual cells in unprecedented detail. A common step in the analysis of scRNA-seq data is the selection of so-called marker genes, most commonly to enable annotation of the biological cell types present in the sample. In this paper, we benchmark 59 computational methods for selecting marker genes in scRNA-seq data.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the evaluation and comparison of methods for selecting marker genes in single - cell RNA sequencing (scRNA - seq) data. Specifically, the objectives of the paper include: 1. **Method evaluation**: Benchmark 59 different computational methods for selecting marker genes from scRNA - seq data. These methods cover traditional statistical methods and modern machine - learning methods. 2. **Performance comparison**: Compare the performance of these methods in the following aspects by using 14 real scRNA - seq datasets and over 170 simulated datasets: - **Ability to recover known marker genes**: That is, whether the method can successfully identify marker genes annotated by experts. - **Predictive performance**: The predictive ability of the selected gene sets. - **Memory usage and speed**: The computational efficiency of the method. - **Implementation quality**: The quality of the software implementation of the method. 3. **Analysis of method characteristics**: Analyze the characteristics of marker genes selected by different methods, including gene expression levels, the proportion of up - regulated or down - regulated genes, etc. 4. **Method consistency**: Evaluate the consistency of marker genes selected among different methods. For example, the paper mentions that there is a low overlap between two commonly used methods, Scanpy and Seurat, and even their rankings also differ significantly. 5. **Recommend best practices**: Based on the results of the benchmark tests, provide researchers with best - practice suggestions for selecting marker genes. The paper specifically points out that simple statistical methods such as Wilcoxon rank - sum test, Student's t - test, and logistic regression perform well in selecting marker genes. Overall, this paper aims to provide a scientific basis and practical guidance for marker gene selection in scRNA - seq data analysis through comprehensive benchmark tests.

A comparison of marker gene selection methods for single-cell RNA sequencing data

Optimal marker gene selection for cell type discrimination in single cell analyses

Systematic comparative analysis of single cell RNA-sequencing methods

CellMarkerPipe: cell marker identification and evaluation pipeline in single cell transcriptomes

GeneCover: A Combinatorial Approach for Label-free Marker Gene Selection

scRNA-seq mixology: towards better benchmarking of single cell RNA-seq analysis methods

Comparative Analysis of Commercial Single-Cell RNA Sequencing Technologies

Benchmarking single cell RNA-sequencing analysis pipelines using mixture control experiments

Comparison of Single Cell Transcriptome Sequencing Methods: Of Mice and Men

A comparison of automatic cell identification methods for single-cell RNA sequencing data

Evaluating Imputation Methods for Single-Cell RNA-seq Data

Stabilized marker gene identification and functional annotation from single-cell transcriptomic data

A multicenter study benchmarking single-cell RNA sequencing technologies using reference samples

Comparative Analysis of Single-Cell RNA Sequencing Methods with and without Sample Multiplexing

Benchmarking single-cell RNA-sequencing protocols for cell atlas projects

A machine learning method for the discovery of minimum marker gene combinations for cell type identification from single-cell RNA sequencing

SMaSH: a scalable, general marker gene identification framework for single-cell RNA-sequencing

Automated cell annotation in scRNA-seq data using unique marker gene sets

MarkerMap: nonlinear marker selection for single-cell studies

A natural language processing system for the efficient extraction of cell markers

A systematic performance evaluation of clustering methods for single-cell RNA-seq data