Correlation Network Analysis Provides Important Modules and Pathways for Human Hyperlipidemia

Hyperlipidemia casts great threats to humans around the world. The systemic co-expression and function enrichment analysis for this disease is limited to date. This study was to identify co-expression modules to explore hyperlipidemia-associated functional pathways. Gene expression data of human hyperlipidemia (GSE17170) were downloaded from the Gene Expression Omnibus (GEO) database. We evaluated the top 3,000 genes with the highest average expression, with which the co-expression modules were constructed in weighted correlation network analysis (WGC-NA).Cluster analysis was then applied to visualize the interaction relationships of these modules. By gene ontology (GO) and KEGG functional enrichment analysis, we finally investigated the function enrichment of co-expression genes from important modules in the Database for Annotation, Visualization, and Integrated Discovery (DAVID) database (https:// david.ncifcrf.gov/summary.jsp).15 Thirteen co-expression modules were constructed for 3,000 genes in the 70samples. Interaction relationships of hub genes between pairwise modules showed high confidence. In functional enrichments of the co-expression modules, genes in Modules 3 and 4 were significantly enriched in biological processes and pathways that are associated with ubiquitination−for example, G0:0016567 (protein ubiquitination) and hsa04120 (ubiquitin-mediated proteolysis). We inferred these two modules as key modules associated with hyperlipidemia. Additionally, G0:0098609 (cell-cell adhesion) was enriched in four modules, suggesting an important function in hyperlipidemia. In conclusion, Protein ubiquitination may play important roles in human hyperlipidemia. All the discoveries made in this study enrich understanding of the pathogenesis of hyperlipidemia and might contribute much to the development of diagnosis and outcome evaluation of this disease.