The differential expression of a featured set of genes may serve as a diagnostic biomarker in hepatocellular carcinoma (HCC) patients. The aim of this study was to identify prognostic biomarkers for the diagnosis and survival of HCC based on the analysis of a large cohort of patients. Clinical and RNA‑seq data were obtained from The Cancer Genome Atlas (TCGA) database. A transcriptomics analysis was conducted to detect differentially expressed genes (DEGs). Samples from 53 tumors and 20 normal tissues of HCC patients were obtained to further analyze the connection between overall survival (OS) and DEG levels. Based on the OS and progression‑free survival (PFS), 4 DEGs (GABRR1, SOX11, COL24A1 and MYLK2) were identified from the TCGA dataset. Using gene ontology (GO) analysis, it was demonstrated that the DEGs were associated with several biological processes, including multicellular organismal and single‑multicellular organism processes, which are involved in the development and migration of HCC. In addition, the four genes were significantly upregulated in tumor tissues. Notably, the mRNA expression of the four genes had a negative association with OS and PFS in HCC patients determined using a Kaplan‑Meir analysis. The four‑gene signature is a potential novel biomarker for the prediction of HCC patient survival.