Fig. 4: PCA plots of innate cell transcriptomes during the course of treatment.

PCA for all samples were performed using log₂ transformed, quantile and CPM-normalized, and filtered counts of annotated protein-coding genes (A–C). The first two principal components are shown on the X and Y-axes, with the proportion of total variance attributable to that PC indicated. D–F show correlation plots of the log₂FC values for DEGs identified using DESeq2 with the default model including cure/TF as factors with SVA-adjustment (x-axis), and SVA-adjusted data where cure/TF and visit (Pre-Tx, Mid-Tx, and End-Tx) were considered as factors. Pearson correlation coefficient (r) is shown for DEGs and Spearman correlation coefficient (ρ) for associated P-values. (G–I) Correlation plots of the log₂FC values for DEGs using Dream (x-axis) and DESeq2 using SVA-adjusted data. Pearson correlation coefficient (r) is shown for DEGs and R² of the linear regression.