a. Mean untransformed genome-aligning spacer counts by group (error bars represent standard error of the mean). b. PCA plots for rlog transformed gene-aligning spacer counts for the top 50 variable genes with clusters defined by k-means clustering (k=2). c. Venn diagram showing overlap in significant DE genes (padj < 0.05) detected by the three DE tools (DESeq2, edgeR and baySeq). d. Heatmap showing unsupervised hierarchical clustering of samples based on rlog transformed genome-aligning spacer counts for high-confidence DE genes detected by the three DE tools. For all panels, n=6 independent biological replicates.