A novel 4 immune-related genes as diagnostic markers and correlated with immune infiltrates in major depressive disorder

Abstract

Background Immune response is prevalently related with major depressive disorder (MDD) pathophysiology. However, the study on the relationship between immune-related genes (IRGs) and immune infiltrates of MDD remains scarce. Methods We extracted expression data of 148 MDD patients from 2 cohorts, and systematically characterized differentially expressed IRGs by using limma package in R software. Then, the LASSO and multivariate logistic regression analysis was used to identify the most powerful IRGs. Next, we analyzed the relationship between IRGs and immune infiltrates of MDD. Finally, GSE76826 was used to to verificate of IRGs as a diagnostic markers in MDD. Results 203 different IRGs s in MDD has been identified (P < 0.05). GSEA revealed that the different IRGs was more likely to be enriched in immune-specific pathways. Then, a 9 IRGs was successfully established to predict MDD based on LASSO. Next, 4 IRGs was obtained by multivariate logistic regression analysis, and AUC for CD1C, SPP1, CD3D, CAMKK2, and IRGs model was 0.733, 0.767, 0.816, 0.800, and 0.861, suggesting that they have a good diagnostic performance. Furthermore, the proportion of T cells CD8, T cells γδ, macrophages M0, and NK cells resting in MDD group was lower than that in the healthy controls, suggesting that the immune system in MDD group is impaired. Simultaneously, CD3D was validated a reliable marker in MDD, and was positively correlated with T cells CD8. GSEA revealed high expression CD3D was more likely to be enriched in immune-specific pathways, and low expression CD3D was more likely to be enriched in glucose metabolism metabolism-specific pathways. Conclusions We applied bioinformatics approaches to suggest that a 4 IRGs could serve as diagnostic markers to provide a novel direction to explore the pathogenesis of MDD.

Identification of novel biomarkers linking depressive disorder and Alzheimer’s disease based on an integrative bioinformatics analysis

“…In recent years, some researchers have identified biomarkers associated with immune infiltration for AD or MDD patients by bioinformatics analysis. For example, four immune-related genes were identified as diagnostic biomarkers of MDD and were associated with immune infiltration [ 39 ]. A recent study has identified and verified six immune-related genes in AD patients [ 40 ].…”

Identification of mitophagy-related biomarkers and immune infiltration in major depressive disorder

“…However, the pathophysiology of MDD remains unclear. Recently, high-throughput sequencing technologies have been used in many studies to elucidate the pathophysiological mechanisms of MDD and identify biomarkers for diagnosis [ 40 , 41 ]. However, most of these studies have focused on differences between patients with MDD and healthy controls, while few have examined differences between MDD subtypes.…”

Gene Signatures Associated with Temporal Rhythm as Diagnostic Markers of Major Depressive Disorder and Their Role in Immune Infiltration

“…For instance, using machine learning approaches, Zhao et al [ 20 ] found that classifiers for SVM , RF , CNN , and NB , as well as the AUC for SVM , RF , CNN , and NB were 0.84, 0.81, 0.73, and 0.83, respectively. Linna et al [ 21 ] applied four immune-related genes ( CD1C , SPP1 , CD3D , and CAMKK2 ), and it has shown a good diagnostic value in discriminating MDD from controls based on immune-related genes, with an AUC of 0.861. In contrast, we found four TRRDEGs with an AUC of >0.7 in two datasets.…”