| More than 90%of Head and neck cancer are head and neck squamous cell carcinoma(HNSCC).Approximately more than 500,000 new cases occur,and nearly 200,000 patients are dying from HNSCC annually[1,2].The treatment of HNSCC is mainly surgery,supplemented with radiotherapy,chemotherapy and targeted molecular therapy and so on[3].Although advanced in the option of treatment methods,the 5-year overall survival of HNSCC patients continue to be below 50%,and the 5-year overall survival of the patient who develop distant metastasis could reduce to about 20%[4],which poses a significant threat to the health and life of the patient.Therefore,this paper uses bioinformatics database to study the development of HNSCC at the molecular level and find effective measures to prevent and suppress the metastasis of HNSCC.Bioinformatics is an interdisciplinary subject and widely involves in multiple fields.Core components of which is to study the natural mechanism of gene expression and reveal the essential laws of human diseases to create conditions for the accurate,rapid diagnosis and effective treatment of illness.This study aims to find disease-associated genes and potential mechanisms in HNSCC through bioinformatics analysis of the DNA microarrays.First,the gene expression profiles of GSE6791[5]were downloaded from the Gene Expression Omnibus(GEO)database,which included 42 squamous cell squamous cell carcinomas and their adjacent normal tissues.Compared with the normal tissue,the Affy package and the Affy chip probe annotation package in Bioconductor were used to preprocess the probe level data,and differentially expressed genes(DEGs)were obtained in the pretreatment of expression matrix with limma in R language(|logFC|>1,p<0.05).Then we uploaded the upregulated and downregulated DEGs to the online software DAVID to identify represented GO categories and KEGG pathways,separately.The online STRING database constructed protein-protein interaction(PPI)network of the DEGs with combined score>0.8.It was visualized by Cytoscape software,and module analysis of the PPI network was performed by Molecular Complex Detection plugin(MCODE).Next,function and pathway analysis were carried out for different modules.Finally,the hub proteins in the PPI network were selected according to the network topological analysis(node degree),and the TCGA HNSCC cohort was used to verify whether the expression levels of hub genes were significantly correlated with the overall survival rate of HNSCC.Through the above methods,a total of 811 DEGs were obtained,including 550 up-regulated and 261 down-regulated DEGs.Function and pathway analysis showed that they were mainly enriched in the terms related to the proteasome and ECM-receptor interaction pathway.Three modules in the PPI network were obtained by MCODE plug-in,and they were mainly enriched in the proteasome and ECM-receptor interaction pathway,using the STRING database to construct the standard differential gene into a protein-interacting network.According to the size of node degree,15 hub genes in the PPI network were obtained,and most of hub genes are in the modules.In addition,High expression of four genes of 15 hub genes were negatively correlated with poor overall survival of patients in TCGA HNSCC cohort,including PSMA7,ITGA6,ITGB4,and APP.In conclusion,we propose these hub genes may play a major role in the development of HNSCC through the above pathway,which provides a direction for the further study of the underlying mechanisms of HNSCC,and could be further explored as potential biomarkers to aid HNSCC diagnosis and treatment.However,further experiments are needed to verify our conclusion. |
PDF Full Text Request