By analyzing the gene expression of endometrial carcinoma (EC) patients, the key factors in PI3K signaling pathway and its related genes mediating EC were explored. The EC samples and normal endometrial samples were downloaded from TCGA database and GTEx database. The R language "limma" package was used for differential analysis, and the expression level of genes in each tissue was analyzed by "gganatogram" package. Functional enrichment analysis of differential genes was carried out by KOBAS, an online bioinformatics website. The correlation between key genes and differential genes was evaluated using TCGA data and GTEx combined gene expression data. The corresponding clinical data were downloaded from TCGA database and GTEx database, and the R language "survival" package was used to assess the potential of candidate differential genes as a key factor of EC. Based on the combined differential analysis of TCGA and GTEx databases, 299 genes with significant differential in expression were finally got. Functional enrichment analysis revealed that genes were predominantly enriched in the entry of "Pathways in cancer", including RAC2 and PIK3R3 genes which were related with the abnormal PI3K pathway in cancer. PIK3R3, a key gene in the PI3K signaling pathway, was highly-expressed in EC. SPDEF, GCNT2, KIAA1324, C9orf152, MARVELD3, and APEX2 genes were found to be positively correlated with PIK3R3 in EC, all of which were highly expressed in EC. KM survival analysis showed that SPDEF, GCNT2, KIAA1324 and C9orf152 were significantly correlated with patients' survival. ROC analysis showed that SPDEF, GCNT2, KIAA1324 and C9orf152 gene could be used as potential markers for prognosis and survival of EC patients. It was found that PIK3R3, a key gene in the PI3K signaling pathway, was highly expressed in EC. The SPDEF, GCNT2, KIAA1324 and C9orf152 genes were also highly expressed in EC, and were positively correlated with PIK3R3 in EC. Moreover, they are significantly correlated with the patients' survival, suggesting that they may be potential markers for the prognosis of patients with EC.
Keywords: Difference analysis; Endometrial carcinoma; GTEX; ROC analysis; Survival analysis; TCGA.