a MA plots of differentially expressed enhancers (False Discovery Rate (FDR) < 0.0001) in synovial fluid (SF) versus peripheral blood (PB) Treg cells for H3K27ac ChIP-seq with the number of SF- and PB-specific enhancers indicated (top). Gene set enrichment analysis (GSEA) of the top 250 upregulated (middle) and downregulated (bottom) enhancers in pairwise comparisons involving transcriptome data of SF Treg cells and healthy adult PB Treg cells, represented by the normalized enrichment score (NES) and FDR statistical value (FDRq, multiple hypothesis testing using sample permutation). b Same as in a but for super-enhancers (FDR < 0.05). c Selection of super-enhancers up- and downregulated in SF Treg cells versus healthy adult PB Treg cells for H3K27ac and H3K4me1 ChIP-seq (FDR < 0.05; bold = up/down for both and differentially expressed (DE) in SF versus PB Treg cells on transcriptome level, see also Supplementary Data 2). d Motifs, known and de novo, for transcription factor binding sites predicted using HOMER, enriched in the upregulated (super-)enhancers in SF Treg cells (n = 3) compared to healthy adult PB Treg cells (n = 3) for H3K27ac and H3K4me1 ChIP-seq. p-values: cumulative binomial distribution to calculate enrichment in target versus background sequences; BATF H3K27ac known p = 1e−15, de novo p = 1e−22, and VDR H3K4me1 p = 1e−19. e Gene tracks for VDR and BATF (H3K27ac) displaying ChIP-seq signals, with the super-enhancer region highlighted in blue, in healthy adult PB Treg cells, SF Treg cells, VDR-specific (GSE89431), and a BATF-specific (GSE32465) ChIP-seq. RNA-seq signals for both VDR and BATF in child PB Treg cells and SF Treg cells are displayed as well. Representative samples for adult PB (n = 3), child PB (n = 3), and SF (n = 3 for ChIP-seq and n = 4 for RNA-seq) Treg cells are shown. For all panels, FDR values were calculated using the Benjamini Hochberg method unless otherwise indicated. Source data are deposited under GSE161426 and GSE156426.