Systematic analysis of microarray datasets to identify Parkinson's disease-associated pathways and genes

Abstract

In order to investigate commonly disturbed genes and pathways in various brain regions of patients with Parkinson's disease (PD), microarray datasets from previous studies were collected and systematically analyzed. Different normalization methods were applied to microarray datasets from different platforms. A strategy combining gene co-expression networks and clinical information was adopted, using weighted gene co-expression network analysis (WGCNA) to screen for commonly disturbed genes in different brain regions of patients with PD. Functional enrichment analysis of commonly disturbed genes was performed using the Database for Annotation, Visualization, and Integrated Discovery (DAVID). Co-pathway relationships were identified with Pearson's correlation coefficient tests and a hypergeometric distribution-based test. Common genes in pathway pairs were selected out and regarded as risk genes. A total of 17 microarray datasets from 7 platforms were retained for further analysis. Five gene coexpression modules were identified, containing 9,745, 736, 233, 101 and 93 genes, respectively. One module was significantly correlated with PD samples and thus the 736 genes it contained were considered to be candidate PD-associated genes. Functional enrichment analysis demonstrated that these genes were implicated in oxidative phosphorylation and PD. A total of 44 pathway pairs and 52 risk genes were revealed, and a risk gene pathway relationship network was constructed. Eight modules were identified and were revealed to be associated with PD, cancers and metabolism. A number of disturbed pathways and risk genes were unveiled in PD, and these findings may help advance understanding of PD pathogenesis.

Identification of two novel biomarkers of rectal carcinoma progression and prognosis via co-expression network analysis

“…Given that the biological significance of these modules might be related to the potential clinical manifestations, we performed Cox regression analysis for each module to determine their survival outcomes [ 1 , 11 , 39 , 40 ]. Poor outcomes were found for the high expression group of the salmon module, which was consistent with previous results indicating that the high expression of the salmon module might represent a higher pathological stage and the T stage [ 2 ].…”

“…Accurate staging of rectal cancer (RC) is essential for carrying out precise therapies that increase the survival rate of patients [ 1 ]. Previous studies have found no differences in specific clinicopathological risk factors between patients with a high or low risk of RC progression [ 2 ]. Recent technological breakthroughs in genome-wide sequencing have shed new insights on deregulated mRNAs that have been identified and characterized in the past few years [ 3 ].…”

The Predicted Key Molecules, Functions, and Pathways That Bridge Mild Cognitive Impairment (MCI) and Alzheimer's Disease (AD)

“…The enrichment of AD genes in cancer-related pathways reveal that some degrees of overlap in the potential pathogenesis of AD and cancers. Genes associated with neurodegenerative diseases, such as PD, are often abnormally expressed in cancers and also involved in cell cycle maintenance (12). Therefore, genes involved in cancers may be potential therapeutic targets for AD.…”

“…Weighted gene co-expression network analysis (WGCNA) is a bioinformatic analysis method that can determine the bridge between sample characteristics and gene expression changes (9). Previous studies had reported that WGCNA could be utilized to analyze the relationships between genes and relationships between gene expressions and clinical characteristics of neurodegenerative diseases (10)(11)(12). Six genes were recognized to be involved in the pathological changes of AD (11).…”

“…Six genes were recognized to be involved in the pathological changes of AD (11). It had been reported that 44 pathway pairs and 52 risk genes might participate in the pathological changes of PD (12). Protein-protein interactions (PPIs) are very important in most biological functions and processes (13) and mainly focus on detecting the relationship between protein complexes and functional modules (14).…”