Analyzing hub genes related to tumorigenesis based on biological big data has recently become a hotspot in biomedicine. Nanoprobes, nanobodies and theranostic molecules targeting hub genes delivered by nanocarriers have been widely applied in tumor theranostics. Hepatocellular carcinoma (HCC) is one of the most common cancers, with a poor prognosis and high mortality. Identifying hub genes according to the gene expression levels and constructing prognostic signatures related to the onset and outcome of HCC will be of great significance. In this study, the expression profiles of HCC and normal tissue were obtained from the GEO database and analyzed by GEO₂R to identify DEGs. GO terms and KEGG pathways were enriched in DAVID software. The STRING database was consulted to find protein-protein interactions between proteins encoded by the DEGs, which were visualized by Cytoscape. Then, overall survival associated with the hub genes was calculated by the Kaplan-Meier plotter online tool, and verification of the results was carried out on TCGA samples and their corresponding clinical information. A total of 603 DEGs were obtained, of which 479 were upregulated and 124 were downregulated. PPI networks including 603 DEGs and 18 clusters were constructed, of which 7 clusters with MCODE score ≥3 and nodes ≥5 were selected. The 5 genes with the highest degrees of connectivity were identified as hub genes, and a prognostic model was constructed. The expression and prognostic potential of this model was validated on TCGA clinical data. In conclusion, a five-gene signature (TOP2A, PCNA, AURKA, CDC20, CCNB2) overexpressed inHCC was identified, and a prognostic model was constructed. This gene signature may act as a prognostic model for HCC and provide potential targets of nanotechnology.