Identification and validation of biomarkers in gastric cancer-associated membranous nephropathy: Insights from comprehensive bioinformatics analysis and machine learning.
This study explores the genetic basis of membranous nephropathy (MN) in gastric adenocarcinoma (GC) through bioinformatics and machine learning analyses.
Gene expression profiles from MN (GSE108109) and GC (GSE54129) datasets were obtained from the Gene Expression Omnibus. Common differentially expressed genes (DEGs) were identified using the limma R package. Biological functions were analyzed via Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways with the Cluster Profiler package. LASSO regression and Random Forest algorithms were used to identify hub genes associated with GC-related MN. The area under the curve (AUC) of ROC analysis validated these genes for their diagnostic potential. Gene Set Enrichment Analysis (GSEA) and immune cell infiltration analysis were conducted, with hub genes validated through immunohistochemistry on renal and gastric cancer tissues.
We identified 40 common DEGs between GC and MN datasets. Using protein-protein interaction networks, 20 significant hub genes were selected, primarily involved in inflammatory and immune response regulation. Key hub genes identified were CCND1, CEBPD, COL10A1, and BMP2, which demonstrated high accuracy in discriminating MN. Notably, CCND1, CEBPD, and BMP2 were significantly overexpressed in glomerular and gastric cancer tissues.
Our findings highlight the crucial roles of CCND1, CEBPD, and BMP2 in the pathogenesis of GC-associated MN, providing insights for future research and potential therapeutic strategies.
Gene expression profiles from MN (GSE108109) and GC (GSE54129) datasets were obtained from the Gene Expression Omnibus. Common differentially expressed genes (DEGs) were identified using the limma R package. Biological functions were analyzed via Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways with the Cluster Profiler package. LASSO regression and Random Forest algorithms were used to identify hub genes associated with GC-related MN. The area under the curve (AUC) of ROC analysis validated these genes for their diagnostic potential. Gene Set Enrichment Analysis (GSEA) and immune cell infiltration analysis were conducted, with hub genes validated through immunohistochemistry on renal and gastric cancer tissues.
We identified 40 common DEGs between GC and MN datasets. Using protein-protein interaction networks, 20 significant hub genes were selected, primarily involved in inflammatory and immune response regulation. Key hub genes identified were CCND1, CEBPD, COL10A1, and BMP2, which demonstrated high accuracy in discriminating MN. Notably, CCND1, CEBPD, and BMP2 were significantly overexpressed in glomerular and gastric cancer tissues.
Our findings highlight the crucial roles of CCND1, CEBPD, and BMP2 in the pathogenesis of GC-associated MN, providing insights for future research and potential therapeutic strategies.