SigBridgeR: An R Toolkit for Standardized Identification and Benchmarking of Phenotype-Associated Cell Subpopulations in Single-Cell Transcriptomics
GigaScience · 2026-11-10
DOI: 10.1093/gigascience/giag100 ↗Abstract
Abstract Background Single-cell RNA sequencing (scRNA-seq) has revolutionized our understanding of cellular heterogeneity, yet linking specific cell subpopulations to clinically relevant phenotypes remains a major challenge. Although multiple computational methods have been developed to integrate bulk transcriptomic and single-cell data for phenotype-associated cell identification, differences in algorithmic strategies, parameter configurations, software implementations, and output formats make it difficult to compare methods fairly, reproduce analyses, and determine which approach is most suitable for a given biological question. Results Here, we present SigBridgeR, an extensible R framework that integrates eight representative phenotype-associated cell subpopulation identification algorithms within a unified workflow. Using paired bulk RNA-seq and scRNA-seq datasets from four cancer types under both binary and survival phenotypes, we performed a comprehensive benchmark combining positive and negative control experiments, accuracy evaluation, Hallmark pathway enrichment analysis, parameter sensitivity analysis, and computational performance profiling. The benchmark revealed substantial differences in algorithm robustness, biological consistency, computational efficiency, and parameter sensitivity. Across multiple evaluation criteria, Scissor and scAB consistently achieved the most balanced overall performance, whereas several methods exhibited important limitations, including poor scalability, high parameter sensitivity, or unstable phenotype detection. Conclusions SigBridgeR provides standardized preprocessing, unified algorithm interfaces, and an extensible registry-based architecture for phenotype-associated cell subpopulation analysis. By integrating systematic benchmarking and optimized implementations into a single platform, SigBridgeR facilitates reliable evaluation of existing methods, simplifies multi-algorithm analyses, and supports the development of improved phenotype-guided single-cell analysis methods for precision medicine.
EUVIMED