Screening Of Key Genes In Hepatocellular Carcinoma Based On TCGA Database

Primary liver cancer is one of the most common malignant tumors in the worldwide.In the past,many important biomarkers and signaling pathways closely related to the biological characteristics of tumors have been found.In recent years,with the rapid development of bioinformatics research,using highthroughput data analysis to explore disease-related biomarkers has become an efficient research method.We downloaded the liver cancer samples from TCGA database and constructed gene co-expression module by weighted gene coexpression network analysis(WGCNA)to identify the pivotal genes of the meaningful module in order to find the biomarker genes or therapeutic targets associated with primary liver cancer.OBJECTIVE: The purpose of this study is to obtain the data of primary hepatocellular carcinoma gene chip in TCGA database,and to use WGCNA method to mine the gene set highly related to primary liver cancer,and to find out the pivotal genes from it,so as to provide a new idea for clinical exploration of biomarkers with diagnostic and therapeutic significance of HCC.Methods:We downloaded 423 samples of primary liver cancer patients with complete RNA_SEQ and clinical characteristics and follow-up data from TCGA database using R script,and then standardized the original transcription data of 423 patients.Subsequently,patients were divided into four groups according to TNM stages: Stage I,Stage II,Stage III and Stage IV.Genes with significant difference level of 0.05 were screened to hierarchical clustering to divide gene modules,and core modules with potential biological significance was further excavated by WGCNA.we combine the core module with the clinical feature information of the samples and analyze and mine the core module related to Stage.By calculating the correlation between the genes and the modules,we can find many pivotal genes in the core module,which is the new biomark we are looking for.We consulted relevant literature to find biomark that had not been studied for follow-up study.Finally,we validate the differential expression of the new biomark in tumors and normal tissues.We then divide the samples into high-expression group and low-expression group according to the median expression value,and do further survival analysis to predict whether the gene results we care about can be used as a new prognostic factor for HCC.Results: 1.WGCNA algorithm mine 23 gene modules through the co-expression network.Three of them are significantly correlated with TNM staging in patients with primary liver cancer,namely steelblue,violet,and lightyellow,especially in Stage I and Stage III(p < 0.01).2.Twenty-three key genes(19 lnc RNA,4 m RNA)were identified.The genes ADAM18,CD226,LYRM4 and RUNX1 may be carcinogenic genes of HCC.The overall survival curve of LYRM4 showed that the survival time of LYRM4 high expression group was significantly lower than that of LYRM4 low expression group.CONCLUSION:1.Weighted gene co-expression network analysis as an advanced bioinformatics analysis method is helpful to find theoretical support for the genesis and development of primary liver cancer at the molecular level and guide subsequent experimental basic research;2.The scale-free network analysis of WGCNA revealed that LYRM4 gene could be a potential biological target gene for primary liver cancer and guide individualization.Diagnosis and treatment programs.