Fig. 3

(A) The PCA plot which represents the potential outlier samples and the boundary of the outlier samples for each group. (B) A heatmap illustrating the effect of batches and covariates on DNA methylation data. Blue tiles represent P-values of 0.05 or lower. Darker blue tiles indicate smaller P-values. The Spearman correlation coefficient is displayed in the center of each tile. (C) A line plot shows the statistical power for given sample sizes and effect sizes generated by the pwrEWAS R package²⁴. (D) Manhattan plot. The x-axis represents the chromosomes, the y-axis represents the -log10 scaled P-value, and the significant P-value threshold was indicated by a horizontal red line. Multiple probe sets could be highlighted in different colors. Significant DMPs are highlighted in green, while the top 20 DMPs are highlighted in blue. Among the top 20 DMPs, those annotated to genes associated with SARS-CoV-2 infection are represented in red. (E) Volcano plot. The x-axis represents Δβ, and the y-axis represents -log10 scaled P-value. The red horizontal line indicates the significance threshold for the P-value, while the dark gray vertical line indicates the threshold for Δβ. The genes associated with the top 20 DMPs are labeled. PCA, principal component analysis; DMP, differentially methylated probe; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; Δβ (delta-beta), the average beta value of patients minus the average beta value of controls.