Prognosis Model Of Breast Cancer Immune-related LncRNAs Based On Bioinformatics Database

PurposeUse the biological information database to find the differentially expressed immune-related long non-coding RNA(lncRNA)that can predict the prognosis of breast cancer,use the screened immune-related lncRNA to construct a risk model,and use the model for the prognosis of breast cancer patients Make predictions and analyze the prediction results to find suitable prognostic models and potential therapeutic targets.Materials and MethodsDownload the expression data of cancer tissues and normal tissues of breast patients from the Cancer Genome Atlas(TCGA)database,set the data format standard as:HTSeq-FPKM,and then download the clinical data corresponding to the patient separately,using biological information Scientific method,screening out mRNA and lncRNA in cancer tissues,combined with the immunology database and analysis portal(The Immunology Database and Analysis Portal,ImmPort)database,will take the intersection to screen out the immune-related mRNA and the lncRNA selected in the previous screening Perform co-expression analysis,and then screen and analyze the differential expression of these lncRNAs that are co-expressed with immune-related mRNAs,and screen out immune-related lncRNAs with differential expression.Combining the patient’s clinical information,and then using single factor COX analysis to screen out the immune-related lncRNAs related to the patient’s overall survival in the training set,and then using Lasso regression analysis to analyze the overall survival-related lncRNAs to reduce the effect of overfitting.Then these lncRNAs are subjected to multi-factor COX regression analysis,lncRNAs are screened and a risk scoring model is constructed,and the risk model is used to predict prognosis.Use survival and model data to draw survival curves,3-year and 5-year survival ROC curves and evaluate the predictive ability of the model.In addition,the clinical information and risk scores were simultaneously subjected to single-factor and single-factor COX regression analysis to verify the independence of risk scores,and analyzed and verified in the validation set.ResultA total of 197 differentially expressed immune-related lncRNAs were obtained through screening,including 145 up-regulated lncRNAs and 52 down-regulated lncRNAs.The differentially expressed lncRNAs were subjected to single-factor COX analysis.As a result,12 prognostic-related lncRNAs were screened.After 12 lncRNAs were subjected to Lasso regression analysis and multi-factor COX regression,the results further screened out 8 differentially expressed immune-related lncRNAs of the constructed models,namely TFAP2A-AS 1,U62317.4,AL513283.1,AC 147067.2,and AL645608.7.AL031316.1,AL121832.2,AC110995.1.The p value of the survival curve in the training set was 1.716e-05,and the p value of the survival curve in the validation set was 1.146e-02.The results showed statistical significance.The ROC curve results showed that the area under the ROC curve(AUC)of the training set for 3-year and 5-year survival were 0.747 and 0.665,respectively,and the area under the ROC curve(AUC)for the 3-year and 5-year survival of the validation set were 0.603 and 0.666,respectively.The independence test showed that the risk scores were independent prognostic factors in the training set and the validation set.ConclusionFrom the results of the TCGA database analysis,there are many expression differences in the expression of breast cancer tissue lncRNA and immune-related lncRNA,and these lncRNA and patient survival prognostication is very closely related,the use of these lncRNA built risk scoring model can effectively predict the patient prognostic,but also for the clinical group of breast cancer research and molecular targeted treatment related research.