Machine Learning Identifies Shared Regulatory Mechanisms of Genes Associated with Ferroptosis in Major Depressive Disorder and Inflammatory Bowel Disease.

backgroundMajor depressive disorder (MDD) and inflammatory bowel disease (IBD) form a "bidirectional vicious cycle" through the gut-brain axis: psychological and emotional abnormalities can induce intestinal inflammation, while intestinal inflammation can in turn exacerbate mental health disorders. Ferroptosis is an iron-dependent form of regulated cell death that is driven by lipid peroxidation. Although this process has been molecularly defined in recent years, its role in the context of IBD and MDD remains insufficiently investigated. This study investigates the molecular roles of ferroptosis-related genes (FRGs) in both conditions and explores potential therapeutic strategies targeting these genes.

methodsWe first identified differentially expressed FRGs (DE-FRGs) by comparing normal and disease samples. Subsequently, we screened for DE-FRGs in both IBD and MDD and named them Co-DEGs. Correlation analyses of these co-FRGs were performed, including comparisons between disease and control groups, as well as associations between Co-DEGs and immune cell infiltrations. Four distinct machine learning algorithms were employed to identify the core Co-DEGs associated with both IBD and MDD. Moreover, analyses of drug sensitivity, molecular docking, and molecular dynamics simulations were carried out to predict potential therapeutic agents for both conditions. Finally, single-cell sequencing analysis was also performed.

resultsWe identified 29 Co-DEGs in both IBD and MDD. Machine learning analysis identified

conclusionsThese discoveries enhance the understanding of the shared and distinct regulatory mechanisms of FRGs in gut-brain axis disorders. We have pinpointed key biomarkers and predicted potential therapeutic agents that may offer dual-targeting strategies for both IBD and MDD.