Bioinformatics Analysis Of Differentially Expressed Genes In Melanoma

Background and Objectives Melanoma is one of the most aggressive skin tumors,and due to its strong metastatic ability,death from melanoma is common in clinical practice,which is one of the reasons for the urgent need to treat this disease.Many research strategies,such as targeted therapy,immunotherapy,gene therapy,have made some progress in clinical trials of melanoma patients,but there is still a lack of targeted specific drugs with high efficacy.Therefore,screening therapeutic targets or key molecules related to the occurrence and development of melanoma and clarifying the molecular mechanism of their action are of important theoretical and clinical significance for prevention,early screening,clear diagnosis and treatment of melanoma patients.More than 30,000 microarray datasets for human skin melanoma are currently available in the GEO database,but few articles have reported on prognostic biomarkers and therapeutic targets for melanoma.Using computational biology techniques,reanalyzing and integrating the data available in these public databases may provide new clues to understanding of melanoma.Materials and Methods Gene microarray containing human melanoma samples was retrieved from GEO database,and melanoma data microarray used to study the causes of tumor was selected and differential genes were screened.The up-regulated and down-regulated genes screened were input into DAVID database respectively,and the corresponding species were selected.The up-regulated and down-regulated genes were enriched in pathway and functional annotation of genes respectively according to GO and KEGG analysis.Multiple Proteins in the STRING database were used to construct a multi-protein network,which was then imported into the c Bio Portal data analysis platform for visual analysis of protein abundance and co-expression networks.Expression of key genes in melanoma and normal tissue was verified using the GEPIA online database,and correlations between genes were observed.Based on univariate Cox regression analysis,the association between key genes and overall and disease-free survival was analyzed,and hazard ratios were calculated to look for genes that were not only differentially expressed between melanoma and normal skin tissue,but were also associated with patient survival.Results In this study,three gene-chip datasets——GSE130244,GSE31879 and GSE83583——were used to screen out 142 differential genes in melanoma.The 50up-regulated genes and 92 down-regulated genes were conducted by GO and KEGG analysis respectively for pathway enrichment and gene functional annotation.It was found that the up-regulated genes were mainly enriched in nucleic acid binding,zinc ion binding,RNA binding,organic cyclic compound binding,poly(A)RNA binding,DNA binding,which were involved in gene transcription,translation process and post-transcriptional regulation,and affected the expression of functional proteins in melanoma cells.The down-regulated genes are enrichment in Rho guanyl-nucleotide exchange factor activity,chemokine signaling pathway,metabolic pathways,lysosome,apoptosis and other signaling pathways,which are consistent with the pathophysiology of melanoma.Multiple Proteins pattern was used to construct a network in the STRING database,and the correlations among the screened genes were visually shown.Finally,five central genes were screened from the PPI network,namely ADAM10,CCNA2,EBP,GABBR2,and TRIM32,and their expressions were verified in GEPIA and Human Protein Atlas databases.Among them,ADAM10,CCNA2 and TRIM32 were all upregulated in melanoma in GEO and GEPIA databases,with statistical significance.In addition,increased expression of CCNA2 and TRIM32 was observed at the protein level during immunohistochemistry.In the survival analysis,high expression of CCNA2 significantly increased the overall survival risk of melanoma patients.High expression of TRIM32 is related to poor survival outcomes in patients.In melanoma,the gene transcription and protein expression of CCNA2 and TRIM32 are significantly different from those in normal skin tissue,and are related to the prognosis of patients.Therefore,CCNA2 and TRIM32 are expected to be biomarkers for the screening,diagnosis and treatment of melanoma.Conclusions In this study,bioinformatics analysis of high-throughput melanoma gene microarray was performed to screen out biological targets that are both differentially expressed and prognostic in melanoma.CCNA2 and TRIM32 were screened out from a number of differential genes.To our knowledge,there are still few reports on CCNA2 and TRIM32 in melanoma,and their differential expression in melanoma and possible mechanisms are unknown.In this paper,the study found that the two genes in melanoma and normal skin tissues exist significant differences in expression,and related to the prognosis of patients.This may provide a new direction for further understanding of the occurrence,development and metastasis of melanoma,and further studies on CCNA2 and TRIM32 may play a certain guiding role in the early screening and treatment of melanoma.