Network insights into childhood obesity: unveiling methylated-differentially expressed genes and pathways through integrative bioinformatics analysis

Abstract

Background: Childhood obesity, a global epidemic with profound impacts on physical and psychological health, remains a complex challenge with elusive underlying mechanisms. This study aimed to unravel the epigenetic landscape of this disease by identifying methylated-differentially expressed genes (MeDEGs) in childhood obesity through integrated bioinformatics approaches.

Methods: Expression profiling (GSE9624) and methylation profiling (GSE25301, GSE27860, and GSE57484) datasets containing data on children with obesity (cases) and eutrophic children (control group) were obtained from the Gene Expression Omnibus (GEO) repository. Differentially expressed genes (DEGs) and differentially methylated genes (DMGs) between the groups were identified using GEO2R. MeDEGs were identified by superimposing the lists of DEGs and DMGs. The protein-protein interaction (PPI) network was constructed using the STRING database and analyzed using Cytoscape. Topological and modular PPI network analyses were carried out using the CytoHubba and MCODE plugins, respectively. Functional enrichment analyses were performed based on Gene Ontology terms and KEGG pathways.

Results: A total of 70 MeDEGs were identified, including 45 hypomethylated high-expression and 25 hypermethylated low-expression genes. The PPI network highlighted three hub-bottleneck genes (CCL5, STAT1, and GATA3) and two functional modules. Overall, the 70 MeDEGs were associated with KEGG pathways related to cellular differentiation, inflammation, chemokine signaling, lipid and glucose metabolism, insulin resistance, and apoptosis.

Conclusion: This study, employing integrative bioinformatics approaches, provides insights into the methylation-mediated mechanisms contributing to childhood obesity, advancing our understanding of this multifaceted chronic disease.

Keywords: bioinformatics; childhood obesity; differentially methylated genes; methylation-regulated differentially expressed genes (MeDEGs); systems biology.

Figure 1

Flowchart of the identification of methylation-regulated differentially expressed genes (MeDEGs) in childhood obesity. DEG, differentially expressed genes; DMG, differentially methylated genes.

Figure 2

Identification of methylation-regulated differentially expressed genes (MeDEGs) in childhood obesity by superimposition of gene expression (Exp) dataset (GSE9624) and DNA methylation (Met) datasets (GSE25301, GSE27860, and GSE57484). (A) Hypomethylated high-expression genes. (B) Hypermethylated low-expression genes.

Figure 3

PPI network formed by the 29 MeDEGs. Red nodes represent hypomethylated high-expression genes in childhood obesity, while blue nodes represent hypermethylated low-expression genes. Diamond nodes represent hub-bottleneck genes identified through network topological analysis. Nodes represent the proteins that are encoded by each gene and edges represent PPIs.

Figure 4

Significantly enriched KEGG pathways in which each hub-bottleneck gene participates. The color of the circle represents statistical significance according to q-values.

Figure 5

Module analysis showing the genes present in each functional module. (A) Module 1: red nodes represent hypomethylated high-expression genes in childhood obesity, while diamond nodes represent hub-bottleneck genes identified in the PPI network analysis. (B) Module 2: red nodes represent hypomethylated high-expression genes, while blue nodes represent hypermethylated low-expression genes. (C) Functional enrichment analysis demonstrating the enriched KEGG pathways of each functional module; the color of the circle represents statistical significance according to q-values.