Bioinformatics analysis of gene expression profiles in hepatocellular carcinoma

OBJECTIVE: The aim of this study is to identify the gene expression profile specific to Hepatocellular Carcinoma (HCC) by comparing the different expression profiles in cirrhosis, dysplastic nodule (DN) and HCC tissues. MATERIALS AND METHODS: The microarray data were downloaded from Gene Expression Omnibus (GEO) repository, involving 39 samples of normal liver tissues, 33 samples of cirrhosis, 17 samples of DNs and 286 samples of HCCs of different stages. Differential Expressed Genes (DEGs) of cirrhosis, DN and HCC liver tissues were analyzed by BRB-ArrayTool software; besides, the Gene Ontology (GO) analysis, Kyoto encyclopedia of Genes and Genomes (KEGG) and Biocarta pathway enrichment analysis were also performed. A protein-protein interaction (PPI) network was then constructed by STRING software using the genes in significantly different pathways. The resulting network was analyzed by Cytoscape software with CentiScaPe plugin to calculate the topological characteristics of the network and its individual node. Key genes were screened according to betweenness and degree of nodes. RESULTS: few overlaps occurred in the GO categories of DEGs and in the gene sets from pathway analysis between HCCs, cirrhosis and DNs. DEGs in abnormal tissues were shown to be enriched in 29 KEGG pathways and 18 Biocarta pathways; and 43 key genes were identified to be involved in the maintenance of PPI network. In addition, the gene expression profiles were significantly different among cirrhosis, DN and HCC tissues. CONCLUSIONS: The bioinformatic analysis of GEO datasets of HCC identified the functional gene sets associated with the genesis and development of HCC, and the key genes that were playing important roles in the maintenance of the molecular network for biological function specific to HCC. It provides the insights for more precise understandings of pathogenic mechanism, which will further expand the study on biomarker and targeted therapy of HCC.