Fig. 4

A Number of case and control individuals and the total number of cells across cell types and datasets. The x-axis denotes the total number of cells on a log scale, while the partitioning of case and control cells denotes the number of cells from each individual relative to the total number of cells. B, C, G, H Volcano plot, where the set of genes with the largest test statistics (having the same size as the original author’s set) is depicted in orange. The genes reported by the respective authors or housekeeping genes are in purple and green, respectively. The Fisher exact test’s p-value between the enrichment of eSVD-DE’s DE genes and the author’s reported DE genes is also reported. D Upset plot showing the intersection between pairs of DE genes, either the reported DE genes in the Habermann and Adams dataset and the estimated DE genes using eSVD-DE or DESeq2. E, I Hexplots showing the correlation between genes’ eSVD-DE test statistics across datasets, either the originally reported DE genes in either dataset or the housekeeping genes. F, J Similar to E, I but showing the gene’s DESeq2 test statistics