Abstract
Hepatoblastoma (HB) is one of the most common liver malignancies in children, while the molecular basis of the disease is largely unknown. Therefore, this study aims to explore the key genes and molecular mechanisms of the pathogenesis of HB using a bioinformatics approach. The gene expression dataset GSE131329 was used to find differentially expressed genes (DEGs). Functional and enrichment analyses of the DEGs were performed by the EnrichR. Then, the protein-protein interaction (PPI) network of the up-regulated genes was constructed and visualized using STRING database and Cytoscape software, respectively. MCODE was used to detect the significant modules of the PPI network, and cytoHubba was utilized to rank the important nodes (genes) of the PPI modules. Overall, six ranking methods were employed and the results were validated by the Oncopression database. Moreover, the upstream regulatory network and the miRNA-target interactions of the up-regulated DEGs were analyzed by the X2K web and the miRTarBase respectively. A total of 594 DEGs, including 221 up- and 373 down-regulated genes, were obtained, which were enriched in different cellular and metabolic processes, human diseases, and cancer. Furthermore, 15 hub genes were screened, out of which, 11 were validated. Top 10 transcription factors, kinases, and miRNAs were also determined. To the best of our knowledge, the association of RACGAP1, MKI67, FOXM1, SIN3A, miR-193b, and miR-760 with HB was reported for the first time. Our findings may be used to shed light on the underlying mechanisms of HB and provide new insights for better prognosis and therapeutic strategies.