Identification of novel immune-related biomarker and therapeutic drugs in Parkinson disease via integrated bioinformatics analysis

Abstract

Background: The present study was designed to identify immune-related biomarker and candidate drugs for Parkinson disease (PD) by weighted gene co-expression network analysis. Methods: Differentially expressed genes were identified in PD and healthy samples in the Gene Expression Omnibus (GEO) database. Besides, immune-related genes were obtained from the immunology database. Then, a co-expression network was constructed by the weighted gene co-expression network analysis package. Diagnostic model for PD was constructed by Lasso and multivariate Cox regression. Furthermore, differentially expressed genes (DEGs) were used to establish PPI and competing endogenous RNA (ceRNA) networks. Functional enrichment and pathway analysis were performed. Drug-hub gene interaction analysis was performed via DGIdb database. Results: PD samples and normal samples were found to have 220 upregulated genes and 216 downregulated genes in the GSE6613 dataset. The differentially expressed genes contained 50 immune-related genes, with 40 upregulated genes and 10 downregulated genes. We obtained 7 hub genes by intersecting the DEGs and candidate hub genes. As potential diagnostic markers, 2 immune-related DEGs were identified among the 7 hub genes. According to functional enrichment analysis, these DEGs were mainly enriched in immune response, inflammatory response, and cytokine-cytokine receptor interactions. Totally, we obtained 182 drug-gene interaction pairs in Drug-Gene Interaction database (DGIdb). Conclusion: Our results revealed crucial genes and candidate drugs for PD patients and deepen our understanding of the molecular mechanisms involved in PD.