Fig. 2: ICP22 is required for the induction of dOCRs.

a, b Scatter plots correlating downstream FPKM against dOCR length (average of two replicates) in total RNA for WT strain F (a) and in 4sU-RNA for WT strain 17 (b) for all analyzed genes with a downstream FPKM ≥ 0.05. Colors indicate the density of points from high (red) to low (blue). The red line indicates a linear fit of log10(dOCR length) against log10(downstream FPKM). The slope of the fit and p-values for the slope of the linear regression estimate being ≠ 0 (two-sided test) were calculated using the lm function in R and are indicated on top of each figure. The error bands around the red line indicate the 95% confidence level interval for predictions from the lm linear model. Example genes with high induction of dOCRs in HSV-1 infection are highlighted. c, d Number of genes in Cluster 5 from Fig. 1b for which dOCRs reach at least a length greater than the value indicated on the x-axis for mock, WT strain 17 (c), WT strain F (d), ΔICP0 (c), ΔICP22 (c, d), ΔICP27 (c) and Δvhs infection (c). To avoid having to define a threshold on whether or not a particular dOCR length for a gene is considered dOCR induction, we visualize dOCR lengths in each condition for all 305 Cluster 5 genes (excluding only those with a dOCR length = 0 in a particular condition). This depicts whether the number of genes with longer dOCRs was generally increased or not in the respective experimental condition. Results are shown separately for two biological replicates. All infections in d were performed with and without PAA, while in c PAA treatment was only performed for WT strain 17 and ΔICP22 infection (as indicated by solid (no PAA) or dashed (+PAA) lines). e, f Scatter plots as in (a, b) correlating downstream FPKM against dOCR length in total RNA for 12 h p.i. ΔICP22 infection +PAA (e) and in 4sU-RNA for ΔICP27 infection (f). Scatter plots for other analyzed conditions are shown in Supplementary Figs. 3 and 7. Source data are provided as a Source Data file.