〔Abstract〕 Objective To screen the core genes of diabetic kidney disease(DKD) based on bioinformatics, explore the therapeutic targets of DKD, and discuss its possible regulatory mechanism. Methods The expression data matrix of glomerular transcriptome in patients with DKD in GEO database(GSE30528, GSE47183) was extracted, and the differentially expressed genes(DEGs) were screened by bioinformatics methods to identify the core differential genes, and then gene expression and enrichment analysis(GSEA) were conducted to predict effective targets. Results By screening and identifying the mRNA expression matrix of DKD, five core genes were screened out. Among them, C1orf21 and NPHS1 were significantly down regulated, and CD48, COL1A2, and TGFBI were up regulated. NPHS1 and CD48 were significantly related to immune differences, intercellular communication, and cell surface interaction. Through receiver operating characteristic curve(ROC) analysis and GSEA analysis and drug target prediction, it might be of great significance for the treatment of DKD. Conclusion The core genes screened in this study have significant correlation with DKD, which may be used as effective markers for the treatment of diabetes. And then, this study provides a theoretical basis for the treatment and identification of DKD.