Meta-analysis is a powerful statistical approach for integrating multiple datasets, being increasingly applied in genome research. Transcriptome meta-analysis provides valuable insights into the functions of stress-responsive genes. In an earlier study, we developed the Stress Response score (SRscore), a novel metric designed to identify stress-responsive genes by capturing expression variability across multiple transcriptome datasets. Using Arabidopsis thaliana as a model, we demonstrated that the SRscore effectively quantifies the degree of stress responsiveness under specific stress conditions. However, extending this approach to other species and ensuring reproducible workflows required the development of a dedicated analytical tool. In this study, we introduce SRscore, an R package designed to streamline transcriptome meta-analyses. The package organizes the analytical workflow into three key steps: (1) pairing of comparison groups, (2) calculation of the expression change ratio (SRratio), and (3) computation of the SRscore. By providing curated datasets and metadata handling functions, SRscore enables straightforward, reproducible, and error-minimized analyses. Furthermore, SRscore outputs can be seamlessly integrated with visualization, enrichment analysis, and interpretation using other R/Bioconductor packages. This package not only facilitates the study of stress-responsive genes across species but also broadens the scope of transcriptome meta-analysis. SRscore package is freely available at CRAN (https://cran.r-project.org/package=SRscore).
Fukuda et al. (Sun,) studied this question.