Deciphering the Molecular Landscape of Schizophrenia: A Multi-Brain-Region Bioinformatics Analysis Identifies Key Hub Genes and miRNA Regulatory Networks

Abstract

Schizophrenia is a complex neuropsychiatric disorder characterized by aberrant brain function and structural alterations. Elucidating the molecular mechanisms underlying SZ is essential for the development of targeted diagnostic and therapeutic approaches. In this study, we conducted a comprehensive bioinformatics analysis of gene expression data from three distinct brain regions: Brodmann Area 46, the Hippocampus, and the Associative Striatum—using the GEO database (GSE53987). Differentially expressed genes (DEGs) were identified with filtering criteria (P-value < 0.05, |log FC| > 0.2), revealing 163 common DEGs across all regions. Gene Ontology (GO) and KEGG pathway analyses highlighted enrichment in processes related to detoxification of copper ion, cellular zinc ion homeostasis, stress response to copper ion, zinc ion homeostasis, and detoxification of inorganic compound, and several signaling pathways, including Mineral absorption, HIF−1 signaling pathway, PI3K−Akt signaling pathway, Adipocytokine signaling pathway. A protein-protein interaction (PPI) network identified two hub genes(STAT3 and CDKN1A), which showed significant diagnostic potential, with ROC analysis yielding AUC values consistently above 0.7 across three brain regions. Additionally, miRNA-gene regulatory networks were constructed for the hub genes, revealing six key miRNAs, (hsa-miR-6825-5p, hsa-miR-665, hsa-miR-6886-3p, hsa-miR-106a-5p, hsa-miR-3135b and hsa-miR-17-5p), that may modulate SZ pathogenesis. Our findings offer valuable insights into the molecular mechanisms of SZ and identify potential biomarkers for early diagnosis and therapeutic intervention.