Skip to main content
Fig. 3 | Genome Medicine

Fig. 3

From: Transfer learning between preclinical models and human tumors identifies a conserved NK cell activation signature in anti-CTLA-4 responsive tumors

Fig. 3

CoGAPS and pseudotime analysis reveals a dynamic state change in NK cells during ICI exposure in mouse scRNA-seq data. A UMAP-dimension reduction colored by CoGAPS pattern 7 weights across all cells (left) and magnified view (right) showing that pattern 7 marks a population of NK cells delineated in Fig. 2A. B Boxplot of pattern 7 weights across each immune cell type. Cells with high pattern 7 weights are observed only in NK cells. C Boxplot of pattern 7 weights in individual NK cells faceted by treatment group. Anti-CTLA-4-treated NK cells have increased pattern 7 weights compared to NK cells treated with other immunotherapies. Significant differences in mean pattern 7 weight between treatment groups are indicated by asterisks where p-values < 0.05 = *, < 0.01 = **, and < 0.001 = ***. D Pseudotemporal trajectory of anti-CTLA-4-treated NK cells colored by CoGAPS pattern 7 weight suggesting that anti-CTLA-4 treatment results in NK cell activation. E Heatmap of gene expression for 148 pattern markers with variable expression as a function of pseudotime. Columns are individual cells, and column annotation designates pattern 7 weight in each cell. Rows are differentially expressed pattern markers. F Gene expression of selected NK cell activation genes that are upregulated across pseudotime. Each dot represents a different cell and is colored by CoGAPS pattern 7 weight

Back to article page