Skip to main content
Fig. 2 | Genome Biology

Fig. 2

From: CB2 improves power of cell detection in droplet-based single-cell RNA sequencing data

Fig. 2

Results from the Alzheimer dataset. a t-SNE plot of cells identified by CB2 and ED. High-count barcodes exceeding an upper threshold are identified as real cells by both methods without a statistical test (dark pink); barcodes identified as cells by both methods following statistical test are shown in pink. Cells identified uniquely by CB2 (yellow) and ED (black) are also shown. CB2 identifies an increased number of cells in existing subpopulations (Subpop1–Subpop4) and also identifies a novel subpopulation (Subpop5). b Distribution plots of the 100 genes having highest average expression in Subpop1 are shown for cells identified by both CB2 and ED (upper) and identified uniquely by CB2 (middle). The estimated background distribution is also shown (lower). Cells uniquely identified by CB2 in Subpop1 have a distribution similar to other Subpop1 cells and differ from the background. c Heatmap of log transformed raw UMI counts for the same 100 genes for barcodes identified by CB2 and ED (left) and barcodes uniquely identified by CB2 (right). d t-SNE plots of cells colored by neuron marker genes SYT1, SNAP25, and GRIN1 in all cells (upper) and those identified uniquely by CB2 (lower)

Back to article page