Identification of diagnostic gene signatures and molecular mechanisms for non-alcoholic fatty liver disease and Alzheimer's disease through machine learning algorithms.

Background: Non-alcoholic fatty liver disease (NAFLD) and Alzheimer's disease (AD) pose significant global health challenges. Recent studies have suggested a link between these diseases; however, the underlying mechanisms remain unclear. This study aimed to decode the shared molecular landscapes of NAFLD and AD using bioinformatic approaches.

Methods: We analyzed three datasets for NAFLD and AD from the Gene Expression Omnibus (GEO). This study involved identifying differentially expressed genes (DEGs), using weighted gene co-expression network analysis (WGCNA), and using machine learning for biomarker discovery. The diagnostic biomarkers were validated using expression analysis, receiver operating characteristic (ROC) curves, and nomogram models. Furthermore, Gene Set Enrichment Analysis (GSEA) and CIBERSORT were used to investigate molecular pathways and immune cell distributions related to GADD45G and NUPR1.

Results: This study identified 14 genes that are common to NAFLD and AD. Machine learning identified six biomarkers for NAFLD, four for AD, and two crucial shared biomarkers: GADD45G and NUPR1. Validation confirmed their expression patterns and robust predictive abilities. GSEA revealed the intricate roles of these biomarkers in disease-associated pathways. Immune cell profiling highlighted the importance of macrophages under these conditions.

Conclusion: This study highlights GADD45G and NUPR1 as key biomarkers for NAFLD and AD, and provides novel insights into their molecular connections. These findings revealed potential therapeutic targets, particularly in macrophage-mediated pathways, thus enriching our understanding of these complex diseases.