Fig. 2: Pathway approaches prioritized IBD-associated genes derived from case-control studies, and differential expression of IBD-associated genes in CD, UC vs. controls.

A A Circos plot summarizes the process of identifying IBD-associated genes and variants. The outer layer includes all 127 SKAT-O-derived IBD candidate genes with P < 0.01. The intermediate four layers represent the top genes identified by four different pathway enrichment and biological relatedness analyses (green: ToppGene; purple: Human Gene Connectome; blue: GIANT; yellow: Ingenuity Pathway Analysis). The 11 genes listed between the intermediate layers and the inner green layer are significant gene loci identified in common by all four pathway approaches (the three known IBD genes are in green, the 8 novel genes in orange). The inner layer displays all 46 high-impact rare variants in the 11 orange genes. Of the 46 variants, 14 variants (highlighted in red inside the inner green layer) are associated with IBD (P < 0.05). We also prioritized the candidate IBD genes resulting from the SKATO test by combining all pathway and functional module analysis results (“Methods”, Supplementary Data 6 and Supplementary Fig. 7). EGR2, ICAM1, IL33, INPP5D, ITK, LRRK2, NOD2, TLR4, VDR are more significant as they remained among the top 10% IBD-associated genes for both the SKATO test and biological function prioritization. B Log-fold changes of the top 50 genes derived from IBD SKAT-O analysis in CD, UC vs. controls bulk RNA-seq analyses (*P < 0.05; **P < 0.01; ***P < 0.001, same levels for C). C Log-fold changes of the top 51–100 genes derived from IBD SKAT-O analysis in CD, UC vs. controls bulk RNA-seq analyses. All statistical tests are two-sided. Exact P values are provided in Supplementary Data 12.