Fig. 3: Single-cell multi-omics show molecular and phenotypic heterogeneity of HCV-specific CD8+ T cells. | Nature Communications

Fig. 3: Single-cell multi-omics show molecular and phenotypic heterogeneity of HCV-specific CD8+ T cells.

From: Identification of human progenitors of exhausted CD8+ T cells associated with elevated IFN-γ response in early phase of viral infection

Fig. 3

a Experimental design with single-cell multi-omics analysis (created with biorender.com). b Dimensionality reduction (UMAP) and clustering of scRNA-seq data. c Distribution of T cell phenotypes in each UMAP cluster. Single-cell phenotypes are obtained via index sorting protein expression data. d Comparison of protein (left) and corresponding genes (right) expression levels between cells grouped by their phenotypes (N = 1603 cells from 26 biologically independent sample time points across 7 epitope specificities). Pairwise group comparisons were performed with two-sided Wilcoxon Rank Sum Tests (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001). Data are presented as median and 75% quantile. e Upset plot showing the number of commonly differentially expressed genes between T cell phenotypes. Insert shows the proportion of genes upregulated between indicated T cell phenotypes. f Selected genes identified from differential expression analysis between cells grouped by phenotype (pairwise comparisons with a two-sided hurdle-model (MAST72). g Loess curve fit of selected genes as a function of the magnitude of IFN-γ response, shaded regions represent 95% confidence intervals. h Selected genes identified from differential expression analysis between cells grouped by disease outcome and phase of infection (early: ≤120 DPI, late >120 DPI). CL: Clearers, CH: Chronic progressors. i Distribution of the clone size and j in each T cell phenotypic subset. Shown are clones found only in the early (≤120 DPI), late (>120 DPI), or both phases (persistent) of infection. N represents cell numbers, and in brackets is the unique number of clones. k Group comparison of T cell receptor (TCR) diversity measured by Shannon evenness between high and low magnitude of IFN-γ responses (N = 22 sample timepoints, 9 epitope specificities). Statistical comparison is performed with a two-sided Wilcoxon Rank Sum Test. Data are presented as median and 75% quantile. l Linear regression (drawn in red) of the Shannon evenness as a function of the magnitude of IFN-γ.

Back to article page