Functional Annotation and Gene Set Analysis of Gastric Cancer Risk Loci in a Korean Population

Abstract

Purpose We aimed to identify the associated single nucleotide polymorphisms (SNPs) with gastric cancer (GC) risk by genome-wide association study (GWAS) and to explore the pathway enrichment of implicated genes and gene-sets with expression patterns.Materials and Methods The study population was comprised of 1,253 GC cases and 4,827 controls from National Cancer Center and an urban community of the Korean Genome Epidemiology Study and their genotyping was performed. SNPs were annotated, and mapped to genes to prioritize by three mapping approaches by functional mapping and annotation (FUMA). The gene-based analysis and gene-set analysis were conducted with full GWAS summary data using MAGMA. Gene-set pathway enrichment test with those prioritized genes were performed.Results In GWAS, rs2303771, a nonsynonymous variant of <i>KLHDC4</i> gene was top SNP associated significantly with GC (odds ratio, 2.59; p=1.32×10^–83). In post-GWAS, 71 genes were prioritized. In gene-based GWAS, seven genes were under significant p < 3.80×10^–6 (0.05/13,114); <i>DEFB108B</i> had the lowest p=5.94×10^–15, followed by <i>FAM86C1</i> (p=1.74×10^–14), <i>PSCA</i> (p=1.81×10^–14), and <i>KLHDC4</i> (p=5.00×10^–10). In gene prioritizing, <i>KLDHC4</i> was the only gene mapped with all three gene-mapping approaches. In pathway enrichment test with prioritized genes, <i>FOLR2, PSCA, LY6K, LYPD2</i>, and <i>LY6E</i> showed strong enrichment related to cellular component of membrane; a post-translation modification by synthesis of glycosylphosphatidylinositol (GPI)-anchored proteins pathway.Conclusion While 37 SNPs were significantly associated with the risk of GC, genes involved in signaling pathways related to purine metabolism and GPI-anchored protein in cell membrane are pinpointed to be playing important role in GC.