Identification of the core genes KLRB1 and RETN as potential shared diagnostic markers for major depressive disorder and systemic lupus erythematosus through bioinformatics and machine learning methodologies.

This study investigates the shared molecular mechanisms between major depressive disorder (MDD) and systemic lupus erythematosus (SLE) through integrated bioinformatics analysis. Analysis of multiple GEO datasets identified 23 common differentially expressed genes (coDEGs), primarily enriched in immune and metabolic pathways. Among 13 high-confidence hub genes derived from protein-protein interaction networks, two biomarkers-KLRB1 and RETN-were consistently selected by machine learning algorithms (including LASSO, SVM, and Random Forest) and validated across independent cohorts. Both biomarkers demonstrated high diagnostic accuracy, with combined AUC values of 0.929 for MDD and 0.948 for SLE. Immune infiltration analysis revealed common reductions in NK cells and elevated monocyte levels in both disorders. KLRB1 expression was positively correlated with NK cell function, while RETN showed disease-specific immune correlations. Gene Set Enrichment Analysis (GSEA) indicated associations between both genes and neurodegeneration-related pathways. Additionally, a predicted ceRNA network suggested potential upstream non-coding RNA regulators. These results underscore dysregulated innate immunity and inflammatory pathways as key common features of MDD and SLE, providing new insights into their diagnosis and therapeutic targeting.