singleCellHaystack - A Universal Differential Expression Prediction Tool for Single-Cell and Spatial Genomics Data

One key exploratory analysis step in single-cell genomics data analysis is the prediction of features with different activity levels. For example, we want to predict differentially expressed genes (DEGs) in single-cell RNA-seq data, spatial DEGs in spatial transcriptomics data, or differentially accessible regions (DARs) in single-cell ATAC-seq data. 'singleCellHaystack' predicts differentially active features in single cell omics datasets without relying on the clustering of cells into arbitrary clusters. 'singleCellHaystack' uses Kullback-Leibler divergence to find features (e.g., genes, genomic regions, etc) that are active in subsets of cells that are non-randomly positioned inside an input space (such as 1D trajectories, 2D tissue sections, multi-dimensional embeddings, etc). For the theoretical background of 'singleCellHaystack' we refer to our original paper Vandenbon and Diez (Nature Communications, 2020) <doi:10.1038/s41467-020-17900-3> and our update Vandenbon and Diez (Scientific Reports, 2023) <doi:10.1038/s41598-023-38965-2>.