Identification of Immune-Related Genes as Biomarkers for Uremia

• Published in: International Journal of General Medicine
• Main Authors: Lyu D, He G, Zhou K, Xu J, Zeng H, Li T, Tang N
• Format: Article
• Language: English
• Published: Dove Medical Press 2023-11-01
• Subjects: differential expression analysis; wgcna; immune infiltration; nomogram; diagnosis
• Online Access: https://www.dovepress.com/identification-of-immune-related-genes-as-biomarkers-for-uremia-peer-reviewed-fulltext-article-IJGM

Dongning Lyu, Guangyu He, Kan Zhou, Jin Xu, Haifei Zeng, Tongyu Li, Ningbo Tang Department of Nephrology Clinic, Guangxi International Zhuang Medicine Hospital Affiliated to Guangxi University of Chinese Medicine, Nanning, Guangxi, People’s Republic of ChinaCorrespondence: Dongning Lyu, Guangxi International Zhuang Medicine Hospital Affiliated to Guangxi University of Chinese Medicine, No. 8 Qiuyue Road, Liangqing District, Nanning, Guangxi, 530200, People’s Republic of China, Tel [phone], Email [e-mail]: Uremia, which is characterized by immunodeficiency, is associated with the deterioration of kidney function. Immune-related genes (IRGs) are crucial for uremia progression.Methods: The co-expression network was constructed to identify key modular genes associated with uremia. IRGs were intersected with differentially expressed genes (DEGs) between uremia and control groups and key modular genes to obtain differentially expressed IRGs (DEIRGs). DEIRGs were subjected to functional enrichment analysis. The protein-protein interaction (PPI) network was constructed. The candidate genes were identified using the cytoHubba tool. The biomarkers were identified using various machine learning algorithms. The diagnostic value of the biomarkers was evaluated using receiver operating characteristic (ROC) analysis. The immune infiltration analysis was implemented. The biological pathways of biomarkers were identified using gene set enrichment analysis and ingenuity pathway analysis. The mRNA expression of biomarkers was validated using blood samples of patients with uremia and healthy subjects with quantitative real-time polymerase chain reaction (qRT-PCR).Results: In total, four biomarkers (PDCD1, NGF, PDGFRB, and ZAP70) were identified by machine learning methods. ROC analysis demonstrated that the area under the curve values of individual biomarkers were > 0.9, indicating good diagnostic power. The nomogram model of biomarkers exhibited good predictive power. The proportions of six immune cells significantly varied between the uremia and control groups. ZAP70 expression was positively correlated with the proportions of resting natural killer (NK) cells, naïve B cells, and regulatory T cells. Functional enrichment analysis revealed that the biomarkers were mainly associated with translational function and neuroactive ligand-receptor interaction. ZAP70 regulated NK cell signaling. The PDCD1 and NGF expression levels determined using qRT-PCR were consistent with those determined using bioinformatics analysis.Conclusion: PDCD1, NGF, PDGFRB, and ZAP70 were identified as biomarkers for uremia, providing a theoretical foundation for uremia diagnosis. Keywords: differential expression analysis, WGCNA, immune infiltration, nomogram, diagnosis