Extended Data Fig. 5: Presence of all cytokine conditions best correlates with early in vivo MCMV infection.

For in vitro studies, RNA-seq and ATAC-seq was performed as described in Figs. 2 and 3, respectively. H3K4me3 ChIP-seq was performed on sorted NK cells (CD3ε⁻ TCRβ⁻ CD19⁻ F4/80⁻ NK1.1⁺) cultured for 3 h with IFN-α, IL-12/IL-18, IL-2/IL-15, or a combination of all three. For in vivo studies, RNA-seq and ATAC-seq was performed on sorted Ly49H⁺ NK cells (TCRβ^–CD19^–CD3ε^–F4/80^–NK1.1⁺Ly49H⁺) at indicated timepoints after MCMV infection. H3K4me3 ChIP-seq was performed on sorted Ly49H⁺ NK cells (TCRβ^–CD19^–CD3ε^–F4/80^–NK1.1⁺Ly49H⁺) at indicated time points after MCMV infection from either WT mice or NKp46-CreERT2 Rosa26-tdTomato reporter mice (gated additionally on TdTom⁺). Heatmaps show matrices of Spearman correlation coefficients using log₂ FC comparing cytokine-stimulated versus unstimulated conditions or d2 versus d0 post-infection using (a) RNA-seq data, (b) ATAC-seq, or (c) H3K4me3 ChIP-seq data. Circle sizes are proportional to coefficients. X-ed out values indicate p > 0.05. For in vitro studies, n = 2-8, and for in vivo studies, n = 2-4. (d) Scatter plots of ChIP-seq log₂ FC comparing cytokine-stimulated versus unstimulated (x-axis) conditions to d2 versus d0 post-infection (y-axis). (e) Center dots indicate Spearman correlations calculated from ChIP-seq values depicted in (c). Ranges show 95% confidence interval calculated by Fisher’s z-transformation.