Research On The Identification Of Key Genes And Pathways In Triple-negative Breast Cancer Based On Weighted Gene Co-expression Analysis (WGCNA)

Background&objective:Triple-negative Breast Cancer(TNBC)is a special type of breast malignancy.The prognosis is usually poor due to its highly invasive biological characteristics and the lack of effective endocrine and targeted therapies.With the development of molecular biology,the diagnosis and treatment of various types of tumors have entered the molecular stage by detecting genes and gene expression products.However,there were no new predictors of clinical guidance found for TNBC,and the molecular mechanism of the occurrence and development of TNBC is still unclear.Therefore,the identification of efficient gene networks and representative biomarkers can help reveal the underlying molecular mechanisms of TNBC and provide new clinical diagnostic markers and therapeutic targets.Methods:The transcriptome and clinical information data of TNBC meeting the study conditions were obtained by downloading,screening,and collating two public database data(The Cancer Genome Atlas,TCGA and Gene Expression Omnibus,GEO).For TCGA data,Trimmed Mean of M values(TMM)standardization and differentially expressed genes(DEGs)screening for TNBC tissues and normal breast RNAs were performed using the "edgeR" package of R language.The thresholds were set as(| log FC |>1 and the adjusted P value,FDR<0.05).For the standardized GSE76250 expression data of TNBC tissues and normal breast RNAs,the DEGs screening was performed directly using the "limma" package of R language,and the thresholds were set as(| log FC |>0.4 and FDR<0.05).Then,WGCNA,kind of an R package,was used to perform weighted co-expression analysis on the expression data of the DEGs in two datasets,in order to find the modules that have the same effect on the biological characteristics of TNBC.The gene of the module was extracted,and the gene co-expression relationship was first visualized in Gephi software,then,combined with online analysis of STRING protein interaction network,the co-expression regulation relationship was further screened and verified,and extracted the hub gene.At the same time,GO analysis and KEGG analysis of each module genes,intersection genes and hub genes were performed using online analytical tools such as DAVID,WikiPathways and GSEA to clarify the functions and pathways of related genes,and the results were visualized by R package.Results:In the study,a total of 166 TCGA and 188 TNBC samples were included.In the TCGA database,4258(2125 up-regulated and 2133 down-regulated)differentially expressed genes(DEGs)were screened.In the GEO database,a total of 2503 DEGs were screened(1265 up and 1238 down).The results of weighted gene co-expression analysis showed that the two data sets have high similarity.The TCGA turquoise module corresponds to the GEO turquoise module and the TCGA green module corresponds to the GEO blue module,which are significantly related to the formation of TNBC tumors.Combined with STRING protein interaction network analysis,11 top hub genes(TOP2A,CCNA2,PLK1,BUB1,NDC80,KIF11,NCAPG,TTK,EXO1,ASPM,DLGAP5)were screened out in the first group of modules,and CD34 was the hub gene of another group.Visual network diagram is constructed for each group of relationships.Gene and pathway enrichment analyses of the regulatory networks involving hub genes suggest that cell division,DNA replication,extracellular matrix interactions and pathways in cancer are crucial determinants of TNBC tumorigenesis.Conclusion:Using bioinformatics data analysis methods,two sets of modules closely related to TNBC tumorigenesis were identified,and the gene network relationship of TNBC tumorigenesis was predicted.Through the analysis of module gene and pathway enrichment,the potential mechanism of TNBC tumorigenesis was revealed.GSEA comprehensively analyze the functional enrichment of gene sets and the contribution of genes to phenotypes,which is the verification and supplement of previous studies.Methods based on co-expression network analysis may be helpful in discovering the biomarkers of TNBC tumorigenesis and development,and as a basis for establishing personalized diagnosis and treatment.