Differential detection workflows for multi-sample single-cell RNA-seq data

In single-cell transcriptomics, differential expression (DE) tools compare average expression between cell types or conditions. However, scRNA-seq technologies hold the promise to infer differences in other aspects of the expression distribution. We therefore developed workflows for differential detection (DD) to infer differences in the average fraction of cells in which expression is detected. After benchmarking eight DD strategies, we present a unified workflow for jointly assessing DE and DD. Through simulations and two case studies, we demonstrate that our joint analyses provide complementary information, both in terms of the individual genes they report and in their functional interpretation.