Construction of immune/Creutzfeldt-Jakob disease-related gene coexpression network to predict biomarkers.

Abstract

Creutzfeldt-Jakob disease (CJD) is a transmissible spongiform encephalopathy characterized by rapid onset and high mortality. Despite considerable progress in the treatment and diagnosis of CJD, patient prognosis remains poor. Many studies have found that the immune response is associated with the pathophysiology of CJD. However, few studies have reported coexpression correlations between genes associated with CJD and the immune response. This study was undertaken to construct a network of coexpressed immune- and CJD-related genes that may reveal new biomarkers and therapeutic targets for CJD. Gene expression data from 11 CJD patients and 10 nonneurological controls were obtained from the Gene Expression Omnibus database. High-confidence protein-protein interaction (PPI) data were downloaded from the Human Protein Reference Database, and gene expression data of immune- and CJD-associated genes were downloaded from the AmiGo16 and DisGeNET databases, respectively. An immune/CJD-related expression network was constructed based on Pearson correlation coefficients and PPI networks, and a CJD-directed neighbour coexpression network was extracted, in which we compared the gene expression patterns and correlations between different groups. The samples were classified using CJD-specific modules, and differentially expressed genes (DEGs) between the CJD and nonneurological controls groups were identified within the CJD-specific modules. Further functional analysis was performed using Kyoto Encyclopaedia of Genes and Genomes (KEGG) enrichment analysis of genes in each CJD-specific module. We constructed an immune/CJD-related coexpression gene network comprising 2007 nodes and 5268 edges, with immune-associated genes occupying important positions in the network. In the CJD-directed neighbour coexpression network, immune-associated genes exhibited the highest coexpression level with their interacting genes. Results from Pearson correlation analysis showed that most of the CJD-associated genes were positively correlated with immune-associated genes. Screening for CJD-specific modules identified MAPK1, CASP3, APP, MAPT, SNCA, and YWHAH, indicating a close connection between CJD and the immune response. Analyses of coexpression status and expression level of CJD-specific genes revealed a very high coexpression pattern for any two genes, with most genes being DEGs. Finally, KEGG enrichment analyses of all CJD-specific genes showed that the pathophysiology of CJD is closely related to infection and the immune response. Our coexpression network analysis revealed a close connection between CJD- and immune-associated genes, and we identified six CJD-specific modules. Biological function analysis of CJD-specific module genes revealed that immune responses are associated with CJD pathophysiology and may provide novel diagnostic and therapeutic biomarkers for this disease.