GSEA of our array data reveals GPR84-mediated enrichment of gene expression signatures for Wnt/β-catenin signaling. (A) The plot of β-catenin (CTNNB1) target GSEA²²  for GPR84-overexpressing KLSA9M pre-LSCs. Normalized gene expression data were used for GSEA (http://www.broad.mit.edu/gsea). The green curve of the plot showing the running enrichment score (ES) for the TARGET_GENES_CTNNB1 gene set, which was generated using Wnt/β-catenin target genes identified in various tumors^38-42  and normal tissues^43-45  (listed at http://www.stanford.edu/group/nusselab/cgi-bin/wnt/target_genes). The rectangle indicating the “leading edge” subset (a subgroup of the “most upregulated” genes by GPR84 overexpression) in their ranked order. The list on the right shows the top 14 genes in the leading edge subset. (B) Real time RT-PCR (N = 3) confirmed the expression of the top 7 genes in EV or GPR84-overexpressing KLSA9M pre-LSCs. *P < .05, **P < .01, ****P < .0001. (C) Enrichment plots for statistically significant gene sets identified by GSEA from the molecular signature database (MSigDB)⁴⁶  exhibiting upregulation of Wnt-associated gene signatures upon GPR84 overexpression. False discovery rate (FDR) gives the estimated probability that a gene set with a given normalized ES (NES) represents a false-positive finding; FDR <0.25 is a widely accepted cutoff for the identification of biologically significant gene sets.²²