Fig. 2 |. Macrophage regulatory architecture is rewired after proinflammatory response.

a, Schematic of LPS+INFγ-treatment to generate proinflammatory macrophages and genomic data analyzed (top). Pie charts showing the number of differential loops, ATAC peaks, H3K27ac peaks, and genes between resting and LPS+INFγ-treated macrophages (bottom). Differential loops were considered significant with DESeq2 absolute LFC > 1 and FDR < 0.05. Differential genes and peaks were considered significant with DESeq2 absolute LFC > 2 and FDR < 0.05. b, Differential chromatin structure and gene expression at the MARCKS locus. 0, resting macrophages; 24, LPS+INFγ-treated macrophages for 24 hours. c, ATAC-seq peaks in resting and LPS+INFγ-treated macrophages are enriched for AD genetic risk factors. P-values calculated by s-LDSC. d, Box plots show the RNA LFC for LPS+INFγ-treated vs resting macrophages for genes in 12 scRNA-seq microglial clusters from AD postmortem brains. Box plots show the median and IQR with whiskers extending to the most extreme non-outliers. Asterisks (*) represent significance as calculated by a two-sided Wilcoxon rank-sum test (FDR < 0.05). e, TF motif enrichment for differential ATAC peaks in LPS+INFγ-treated macrophages and the expression of the top three most highly expressed members of those TF families. f, Schematic of the ABC model. g, Venn diagram showing the number of ABC-prioritized enhancer-gene pairs (ABC score > 0.05) identified in resting and LPS+INFγ-treated macrophages. h, Density plot showing the LFC in ABC score for genes highly expressed in resting macrophages (peach), genes highly expressed in LPS+INFγ-treated macrophages (salmon), or expression-matched static genes (gray). Asterisks represent significance as calculated by a two-sided Wilcoxon rank-sum test, proinflammatory vs static p = 1.3 × 10⁻¹⁹⁴; resting vs static p = 2.2 × 10⁻¹¹⁶. i, GO terms enriched for DEGs at the anchors of resting- or LPS+INFγ-specific ABC pairs.