Bioinformatics Analysis of the Diagnostic Value of Copper and Zinc Metabolism‐Related Genes in Major Depressive Disorder: An In Silico Multi‐Cohort Study

This study aimed to identify copper and zinc metabolism-related genes as potential diagnostic biomarkers for major depressive disorder through an integrated multi-cohort bioinformatics analysis. GSE98793 (whole blood) served as the training set, with GSE39653 and GSE52790 (both PBMC) as external validation sets. Platform differences were corrected with ComBat using disease status as a covariate; batch parameters were estimated on the training set alone and applied to the validation cohorts by a frozen reference-batch approach, preventing data leakage. Copper and zinc metabolism-related genes (CZMRGs) were compiled from GeneCards and KEGG, then intersected with differentially expressed genes and WGCNA module genes. Candidates were prioritized by protein-protein interaction analysis and three machine learning algorithms (LASSO, SVM-RFE, random forest), and a logistic regression model was evaluated by ROC, calibration, and decision curve analysis. Of the dual-metal genes shared by the copper and zinc lists, 22.5% fell within the MDD-associated co-expression module, against 17.1% of single-metal genes (Fisher's exact test, OR = 1.41, p = 0.012), indicating modest but significant coupling of the two metal programs. Three genes were selected concordantly by all three algorithms: MT1A, MT2A, and SLC30A1, each significant after false discovery rate correction. The model achieved an AUC of 0.88 (training), 0.81 (Validation-1), and 0.78 (Validation-2), outperforming every individual gene. All three genes remained independently associated with MDD after adjustment for immune cell composition (adjusted AUC 0.84), and all were re-selected under stricter gene-universe thresholds. MT1A, MT2A, and SLC30A1 form a validated, externally reproducible diagnostic panel for MDD, pointing to a coordinated disturbance of copper-zinc buffering, although the specific evidence for copper-zinc coupling beyond general metal-gene enrichment remains modest.