Fig. 5: Genes in the core inflammation program are predisposed to be upregulated with maturation and activation.
a Shown is the mean log2 fold change of gene expressions across all comparisons versus the regulatory activity (inverse logarithm of ChEA3 Scores per species). Colors and sizes indicate −log10(P), using the Benjamini-Hochberg adjusted P-values from a Fisher’s combined test (Fig. 3). b–d ATAC-sequencing from the air pouch model36. b Members of the core inflammation program show increasing chromatin accessibility (blood vs. bone marrow; membrane and air pouch vs. bone marrow or blood). Genes part of the inflammatory response program were compared with the list of genes that showed increased/decreased accessibility in the depicted comparisons. Asterisks indicate significance versus 1000 repeats of an accessibility analysis of random RNA-expression-matched background genes (two-sided studentized bootstrap: P1: 7.26 × 10−8; P2: 6.86 × 10−17; P3: 2.09 × 10−21; P4: 1.56 × 10−18; P5: 3.07 × 10−22). Bottom-right: Two-sided Fisher’s exact test on the number of core inflammatory genes with increased accessibility versus no increase. FDR-adjusted P-values. c Core inflammation program gene accessibility for each comparison. Rows ordered by nested decreasing rank of peaks associated with an increase, with both, with none, and with a decrease. d A subset of transcription factors shows increased motif enrichment and increased gene expression in inflamed neutrophils. Motif enrichment analysis was performed using HOMER and is compared to the mean log2 fold change across species. A one-sided motif enrichment analysis was run separately for increased and decreased accessibility to calculate \(l{{{{{\rm{n}}}}}}(P)\). ρ, Spearman’s rank correlation coefficient with its respective P-value (two-sided). e Analysis of HoxB8 cells that were treated with zymosan36. Core inflammation genes show upregulation versus random genes in zymosan-stimulated HoxB8 cells and are downregulated in JunB−/−. Core inflammation genes in HoxB8 neutrophil RNA-Seq data36. One row per comparison. Left, experimental conditions for each comparison and respective statistics. Middle, volcano plots for each comparison. Red, Core inflammation program members. Members with the highest combined significance and effect sizes are labeled. Right, histograms showing the percentage of expression-matched background genes (equally sized gene sets, 1000 simulations) up/downregulated (see x-axes) in each comparison. Red arrow indicates observed percentage for core inflammation program members, annotated with its Poverrepresentation (one-sided, Wallenius method99). Source data are provided as a Source Data file.
