Skip to main content
. 2020 Aug 24;9:e55792. doi: 10.7554/eLife.55792

Figure 6. The V4.2 annotation improves cluster resolution in scRNA-seq data.

Figure 6.

(A) tSNE plot of all clusters from Ensembl (Ens95)-quantified scRNA-seq of zebrafish embryos at 5 days post fertilization with one erythroid and two leukocyte clusters indicated. (A–E, G) Circled clusters denote sla2pos leukocyte cells. Asterisk marks cells spuriously assigned to sla2pos leukocytes, but identified as erythroid two when using the V4.2 annotation. (B–E) tSNE plots showing only sla2pos leukocyte (circled) and erythroid clusters with expression of (B) known leukocyte markers, coro1a, and rac2, (C) erythroid marker hbae1.1, and (D) sla2. (E) tSNE plots showing expression of pcna and myb in sla2pos and erythrocyte clusters. (B–E) Legends indicate log-transformed and normalized expression values. (F) Erythroid and sla2pos (leukocyte) clusters identified using the same scRNA-seq dataset as in (A) but quantified using V4.2. A unique erythroid-like cluster (erythroid 2) is circled. (G) tSNE plot of erythroid and sla2pos leukocyte clusters identified with Ens95 quantified data. Dark green denotes cells with identical barcodes as those in the erythroid two cluster identified using V4.2 shown in (F). (H) Three-way Venn diagram intersecting cluster-specific genes of indicated clusters by gene symbol. Selected genes are shown.

Figure 6—source data 1. List of genes from 3-way Venn diagram output shown in Figure 6H.
First column is gene symbol, second column is assigned class shown in the Venn diagram.