Malignant melanoma is uncommon worldwide, accounting for only ∼1% of all malignant tumors. However, it has the fastest growth rate among all malignant tumors, 3%–7% annually. Overall, malignant melanoma is an insidious disease with a high degree of malignancy. Surgery, radiotherapy, and chemotherapy have long been the main methods for the treatment of malignant melanoma. However, because malignant melanoma exhibits poor sensitivity to adjuvant chemotherapy, the clinical prognosis is poor, and the mortality rate is high. In recent years, it has been found that malignant melanoma is a type of cancer with high immunogenicity. Many research strategies, such as cellular immunotherapy and cytokine and monoclonal antibody therapies, have been applied in the treatment of malignant melanoma. Studying the pathogenesis of melanoma, with the constant discovery of new targets, is still a hot research topic in medicine (Abbas et al., 2014; Chattopadhyay et al., 2016; Wong et al., 2017; Cabrera and Recule, 2018; Kastelan et al., 2018).

The development of high-throughput gene chip and sequencing technology has enabled rapid analysis of the gene expression profile of melanoma, revealing the expression levels of genes related to melanoma and changes in key genes. With the help of gene chip research, bioinformatics allows for elucidating valuable information from complex data through sequence alignment, statistical analysis, visualization mapping, biological clustering, and biomolecular network and pathway analyses. Furthermore, bioinformatics data mining provides a means of comprehensively and systematically studying diseases.

Using ONCOMINE and Gene Expression Omnibus (GEO) databases, Ci et al. (2019) verified that CDCA8 mRNA expression and CDCA8 in cutaneous melanoma tissues are potential therapeutic targets. In the study of Wei (2018) three gene expression data sets were downloaded from the GEO and The Cancer Genome Atlas (TCGA) databases, and 110 characteristic genes were found through meta-analysis to be significantly overexpressed in metastatic melanoma. Wei et al. (2019) selected GSE3189 and GSE31879 profiles and suggested that MFSD12 and SLC45A2 affected the overall survival (OS) and disease-free survival (DFS) of melanoma patients. At present, there are >30,000 microarray data sets for human cutaneous melanoma in the GEO database, although there are few articles about utilizing gene chips from GEO to screen biomarkers that affect the prognosis of melanoma. Through computational biology technology, reanalyzing and integrating the data available in these public databases may provide new clues for understanding cancer.

The expression profiles of the GSE130244, GSE31879, and GSE83583 data sets were downloaded from the GEO database and reanalyzed, including 110 melanoma and 16 normal skin tissue samples. Differentially expressed genes (DEGs) were screened between melanoma and normal tissues. Venn diagrams were used to reveal genes differentially expressed in all three data sets, and possible hub genes affecting melanoma development were identified through a PPI (protein–protein interaction) network. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed to identify pathways that might affect the occurrence of melanoma, and Gene Expression Profiling Interactive Analysis (GEPIA) was applied to detect the expression levels of these hub genes and analyze their interactions. The protein expression of hub genes in melanoma tissue was assessed by immunohistochemistry, and survival analysis was performed to determine hub genes affecting OS and DFS of melanoma. This study constitutes a preliminary exploration through bioinformatics tools to screen out relevant functional genes in an effort to obtain more biological information about the molecular mechanisms involved in the development of melanoma (Chiriboga et al., 2016).

In summary, this study mined melanoma gene chip data and screened CCNA2 and TRIM32 from among five hub genes through bioinformatics methods. To our knowledge, there is a lack of research on CCNA2 and TRIM32 in melanoma. Significant differences in the expression of these two genes in melanoma were observed, as were correlations with the prognosis of patients, which may provide a new direction for further understanding the occurrence, development, and metastasis of melanoma. These findings are expected to provide potential targets and biomarkers for clinical detection and treatment. However, the lack of further molecular biology experiments to confirm the function of TRIM32 in melanoma limits such analyses. Nevertheless, the findings of this study contribute to a complete understanding the underlying molecular mechanisms of melanoma and provide guidance for subsequent experimental studies.

 

For the complete journal submission and to view related graphs and tables click here.

 

This article was originally published in the March 4, 2021 issue of the Journal of Computational Biology, published by Mary Ann Liebert, Inc.

Previous articleDelivering Your New Diagnostic Test into a Clinical Laboratory
Next articleAnnovis Bio Plans to Build on Alzheimer’s Clinical Trial Positive Results