Fig. 1: Ashkenazi Jewish population-specific case-control study using high-impact variants revealed IBD-associated genes.

A Flowchart of present work. Firstly, AJ samples from all WES participants that passed quality control were genetically identified. Then, high impact rare variants from exomes were filtered using cutting-edge mutation filtering approaches; high impact rare variants were aggregated into gene sets to perform SKAT-O gene burden analyses on IBD cases and controls of AJs. IBD associations were validated and prioritized at the pathway level, gene level, and variant level using multiple methods. Next, we replicated the top candidate genes in an independent cohort and identified their relatedness to diseases other than IBD using gene-level PheWAS. Then, gene expression was tested in both bulk RNA-seq and single-cell RNA-seq. Lastly, we built polygenic risk score models to predict IBD patients using high-impact variants. B We filtered a set of LD-pruned independent sites to perform a fastSTRUCTURE admixture analysis by comparing individual samples with 36 known AJ reference samples. The lowest AJ fraction (0.645) in the AJ reference panel was used as the threshold, above which a WES sample was deemed to be genetically AJ and retained for further analyses. Genetically identified Ashkenazi Jewish samples are displayed on a PCA plot compared to the Jewish and European reference panels. The genetically identified AJs constitute an independent cluster, which overlapped with the AJ reference panel but was distinct from the European cluster. C Distributions of filtered high impact rare variants by molecular function. D SKAT-O analysis on 1734 AJ IBD cases and 2719 AJ controls. The red dashed line indicates the Bonferroni-adjusted P values of genome-wide significance. All dots represent negative log unadjusted P values.