Fig. 3: scRNA-seq defined NK cell state in severe trauma patients.

a Cell death analysis pipeline for NK cells. NK cell states were identified via two-step clustering, and cluster difference thereof compared to find death-specific states of NK cells. Further cell death signatures were derived from these states using gene set enrichment analysis (GSEA). b T-distribution stochastic neighbor embedding (t-SNE) plots for each cluster. Each point was colored by the cluster to which it belonged. c Ten cell death pathways in the differentially accessible genes of each cluster. Sizes of circles are proportional to the normalized enrichment score (NES). Color represents the cell death pathways, and alpha represents the log10 p value. p value was calculated by empirical phenotype-based permutation test procedure according to the Kolmogorov-Smirnov-like statistic. d t-SNE plots for each cell state (Autophagy NK cell, apoptosis NK cell, ferroptosis NK cell, immunogenic cell death NK cell, pyroptosis NK cell, Necrotic NK cell, and healthy NK cell), and each point was colored by its cell state. e Pie charts displaying the fraction of cells in six cell death states in healthy controls and different post-trauma time point trauma patients. f The line graphs illustrate variations in the percentage of NK cells with different death paradigms at various post-trauma time points. Colors distinguish between the different paradigms of death while p values were computed utilizing one-way ANOVA (*p < 0.05, **p < 0.01, ***p < 0.001). g Stacked bar plot showing the fraction of each patient in each of the 6 cell death states. Different colored blocks represent the diverse methods of NK cell death, and the colors convey the same meaning as in the illustration displayed in (e).