Machine learning identified ADM and IL2RB as shared oxidative stress-related hub genes with high diagnostic accuracy for atherosclerosis (AUC 0.787 and 0.887) and NAFLD (AUC 0.979 and 0.931).
Bioinformatics and machine learning identified ADM and IL2RB as shared oxidative stress-related hub genes with high diagnostic potential for both NAFLD and atherosclerosis.
Effect estimate: AUC > 0.9 for nomogram model
Non-alcoholic fatty liver disease (NAFLD) is a chronic hepatic illness linked with metabolic dysfunction that is distinguished by aberrant lipid accumulation in hepatocytes. Emerging evidence indicates an elevated risk of atherosclerosis (AS) in NAFLD patients, but the fundamental biological processes are not well comprehended. The present research aims to uncover co-expressed oxidative stress-related hub genes as well as potential mechanisms shared between NAFLD and AS through bioinformatics and machine learning approaches. Gene expression profiles of NAFLD (GSE89632, Allard JP et al.) and AS (GSE100927, Steenman M et al.) were retrieved from the Gene Expression Omnibus (GEO) database. Oxidative stress-related genes were extracted from the GeneCards database. Differential expression analysis identified shared differentially expressed genes between both diseases. Functional enrichment analysis elucidated associated biological processes. Machine learning approaches were employed to screen for oxidative stress-related hub genes. Differential expression analysis revealed 20 co-expressed oxidative stress-related genes (8 upregulated, 12 downregulated) in NAFLD and AS. Functional enrichment analysis revealed that these genes are predominantly associated with biological processes, including the MAPK as well as the FoxO signaling pathway. Machine learning methods identified ADM and IL2RB as co-expressed oxidative stress-related hub genes with significant diagnostic potential, with Area Under Curve (AUC) values of 0.787 and 0.887 for AS, and 0.979 and 0.931 for NAFLD. The nomogram model utilizing the two hub genes demonstrated excellent predictive accuracy, with AUC values exceeding 0.9 in both diseases. The present study investigated potential important oxidative stress-related molecules as well as pathogenic processes among NAFLD and AS, offering fresh insights for finding potential biomarkers and treating both diseases in future studies.
Qing et al. (Mon,) conducted a other in Non-alcoholic fatty liver disease (NAFLD) and Atherosclerosis (AS) (n=167). ADM and IL2RB gene expression vs. Healthy controls was evaluated on Diagnostic accuracy (Area Under the Curve) (AUC > 0.9 for nomogram model). Machine learning identified ADM and IL2RB as shared oxidative stress-related hub genes with high diagnostic accuracy for atherosclerosis (AUC 0.787 and 0.887) and NAFLD (AUC 0.979 and 0.931).