RankMap Speeds Cell Type Annotation in Single-Cell and Spatial Transcriptomics

Researchers have developed RankMap, a fast and flexible R package designed for reference-based cell type annotation in single-cell and spatial transcriptomics analysis.

Accurate cell type annotation represents a key step in single-cell and spatial transcriptomics research. However, many current reference-based computational methods are computationally intensive, presenting processing bottlenecks that limit their application to large spatial datasets.

To address these scaling challenges, RankMap represents each cell using rank-based features derived from its top-expressed genes. This rank-based approach reduces sensitivity to platform-specific differences in expression scale across experimental technologies. RankMap then applies elastic net-regularized multinomial regression to predict cell types and calculate associated confidence scores for each prediction.

The authors benchmarked RankMap on five spatial datasets generated with Xenium, MERFISH, and Stereo-seq platforms, along with two single-cell datasets, comparing its performance against SingleR, Azimuth, and RCTD. Across these settings, RankMap achieved competitive annotation accuracy while substantially reducing runtime, with marked efficiency improvements on large spatial datasets.

The RankMap software package has been made publicly accessible through GitHub at https://github.com/jinming-cheng/RankMap, providing an open R implementation for cell type annotation.

Subject of Research: Reference-based cell type annotation in spatial and single-cell transcriptomics

Article Title: RankMap: rank-based reference mapping for fast and robust cell type annotation in spatial and single-cell transcriptomics

Article References: Cheng, J., Li, S., Kim, S., Ang, C. H., Chew, S. C., Chow, P. K. H., & Liu, N. (2026). RankMap: rank-based reference mapping for fast and robust cell type annotation in spatial and single-cell transcriptomics. BMC Bioinformatics. https://doi.org/10.1186/s12859-026-06699-3

Image Credits: AI Generated

DOI: 10.1186/s12859-026-06699-3

Keywords: spatial transcriptomics, single-cell transcriptomics, cell type annotation, RankMap, bioinformatics, multinomial regression, computational biology, benchmarking, rank-based, reference, mapping, fast