A Study On The Relationship Between MRNA And Prognosis Of Invasive Breast Cancer By Using Artificial Neural Network

Background and purpose:Breast cancer is one of the most common malignant tumors in women.The incidence of breast cancer is the highest among all malignant tumors in women,and it is increasing year after year.It is also a young trend.Breast invasive cancer accounts for about 80%of breast cancer The 5-year disease-free survival rate of breast cancer patients showed that the stage Ⅰ survival rate was 90%;the stage Ⅱ survival rate was 70-80%;the stage Ⅲ survival rate was 60-70%;the stage Ⅳ survival rate was only 30%.After receiving systemic treatment,approximately 40%of breast cancer patients still have recurrence and metastasis,and most patients die.Therefore,to actively study the pathogenesis of breast cancer and prognostic factors,to find a potential new target for the diagnosis,treatment and prognosis of breast cancer has become an urgent problem to be solved.The data of this study through the clinical correlation of mRNA in TCGA in breast cancer model was established by artificial neural network,analyzed the mRNA expression and the survival of the relationship,to provide the basis for further study of breast cancer occurrence and development mechanism,can also provide basis for subsequent mRNA molecular biology research.Artificial neural network(ANN)is an artificial intelligence model that integrates and processes information by simulating the working process of neurons in the human brain,and establishes a simple model,which can produce different results for each type of connection change.The network can be used to predict some non-linear,non-stationary,and complex problems.Multi-layer perceptron(MLP)is a relatively simple feedforward network,similar to a single biological neuron.Artificial neural network is a tool with great potential for development in the field of bioinformatics and has a wide application prospect.It has been successfully applied in the fields of medicine,biology and economy.According to statistics,the number of bioinformatics papers on artificial neural networks in PubMed has been increasing in recent years.The Cancer Genome Atlas(TCGA)program is sponsored by the US government,by the National Cancer Institute(NCI)and the National Human Genome Research Institute(NHGRI)jointly implemented,trying to through the application of genome analysis technology,especially the use of large-scale genome sequencing,all human cancer(recent goal including 50 kinds of subtype,tumor genome variation map)out,and system analysis.Based on the results of the human genome project(HGP),TCGA has studied the change of cancer genome,which is equivalent to more than 100 HGP.So far,it is the largest genetic engineering in the world.TCGA online for researchers around the world to provide public data resources for free,TCGA’s powerful data driven platform of genome data sharing platform(GDC),can be linked to external analysis tools,such as cBioPortal,Firehose,Firebrowse Website Web site,this study also used these three analytical tools.So far,The TCGA database provides genome,transcriptome,proteome,epigenetic group data and clinical data associated with more than 30 different tumors.The TCGA project has analyzed the expression of mRNA in 825 Cases of Breast Invasive Carcinoma(BRCA)in 2012.This study included the clinical data and mRNA expression data of TCGA about invasive breast cancer,and applied ANN model to study the relationship between mRNA and prognosis of invasive breast cancer.Data and methods:The clinical data of invasive breast cancer were downloaded on the TCGA platform.Clinical data and mRNA expression data were analyzed by log-rank test,descriptive statistical analysis,and binary regression analysis.When p<0.05,consider the test results statistically significant.The expression data in the TCGA platform to download and invasive breast cancer related mRNA,and analysis tools and screening of invasive breast cancer survival in stage mRNA associated with GDC,log-rank test p value 10 mRNA minimum to establish a neural network model for predicting the prognosis of invasive breast cancer.The above steps are mainly completed by the SPSS 22.0 software package.Result:1.The survival period of invasive breast cancer is related to ER expression,PR expression,HER2 expression,and tumor staging,and is not related to age and menstrual state.2.The expression of mRNA in different case samples.3.In the TCGA platform,we used GDC analysis tool to screen 78 mRNA that may be related to the prognosis of invasive breast cancer,and constructed mRNA artificial neural network model to predict the prognosis of invasive breast cancer patients.Conclusion:1.The clinical prognostic factors of patients with invasive breast cancer include ER,PR,HER2,and case staging,which are not related to age and menstrual state.2.The expression of mRNA is associated with the prognosis of patients with invasive breast cancer.3.The artificial neural network of mRNA can be used to predict the prognosis of breast cancer patients.