Interaction of Epstein-Barr virus genes with human gastric carcinoma transcriptome

Gastric carcinoma (GC) is a leading cause of mortality. 10% of GC cases are related with EBV (Epstein-Barr virus) infection. The detailed mechanistic roles EBV genes play and especially the interaction between the viral genes and human genes in GC remain unclear. In this study, raw fastq data from 285 GC samples were downloaded from TCGA (The Cancer Genome Atlas), including 25 EBV positive (EBV+) GC samples and 260 EBV negative (EBV−) GC samples. RNA-seq based expression data were generated for both human genes (among all the samples) and for the EBV genes (among the 25 EBV+ samples). Bioinformatics analyses were performed to identify differentially expressed (DEx) human genes and DEx KEGG pathways in EBV+ vs. EBV− samples and co-expressed human gene modules and hub genes among the DEx genes. Within the EBV+ samples, analyses were conducted to find correlation between EBV gene expression and the human gene expression modules, between EBV gene expression and the human hub genes, and between EBV gene expression and the DEx human pathways. EBV genes LMP-1, BALF1 and BALF2 were found to have significant correlation with human hub genes, CNTD2 and VANGL2. EBV genes BALF4 and BALF5 were found to correlate with human pathways, including Jak-STAT signaling and Phosphatidylinositol Signaling System. Our study has revealed the coordinated expression patterns between EBV and human GC transcriptome and identified several key EBV genes that may play an important role in EBV+ GC pathogenesis through their interactions with human genes and pathways.