Single-cell RNA-sequencing (scRNA-seq) has brought the study of the transcriptome to higher resolution and makes it possible for scientists to provide answers with more clarity to the question of ‘differential expression’. However, most computational methods still stick with the old mentality of viewing differential expression as a simple ‘up or down’ phenomenon. We advocate that we should fully embrace the features of single cell data, which allows us to observe binary (from Off to On) as well as continuous (the amount of expression) regulations.
Brown University researchers have developed a method, termed SC2P, that first identifies the phase of expression a gene is in, by taking into account of both cell- and gene-specific contexts, in a model-based and data-driven fashion. They then identified two forms of transcription regulation: phase transition, and magnitude tuning. The researchers demonstrate that compared with existing methods, SC2P provides substantial improvement in sensitivity without sacrificing the control of false discovery, as well as better robustness. Furthermore, the analysis provides better interpretation of the nature of regulation types in different genes.
Comparing DESeq2 with other methods for DE detection in human brain data, astrocytes vs. oligodendrocytes comparison.