| According to the 2020 global cancer statistics,breast cancer(BRCA)now accounts for 11.7%of the overall cancer incidence,becoming the most common cancer in the world,and its incidence and death have been ranked first among female cancers.With the development of information technology,the concept of medical data has gradually been widely used.Through the application of data mining technology to the field of BRCA,the prognostic molecular mechanism of breast cancer is deeply explored,and new therapeutic targets are found for its patients.There.are of great practical significance for the prognosis and survival of BRCA patients.Firstly,this paper uses Cox regression based on gene expression data for the prognostic analysis.The clinical data of BRCA patients were filled and digitized,and differentially expressed genes were initially screened.Then,through the use of Cox regression and Lasso regression,the key factors affecting the overall survival rate of BRCA patients were screened from the training set,that is nine genes that are significantly related to survival,and use the test set for model validation.The results showed that the low-risk group and the high-risk group were significantly different,and the model classification effect was good.Secondly,in order to further search for new prognostic targets for BRCA patients,this paper introduces alternative splicing data,and constructs a breast cancer prognostic model based on alternative splicing events.Extract splicing factors from BRCA gene expression data,fill in and filter the corresponding alternative splicing events,and screen out 10 alternative splicing events that are significantly related to survival by using Cox regression and Lasso regression.Established a Cox model to classify BRCA patients.The results show that there are significant differences between the low-risk group and the high-risk group.The model has an excellent classification effect,confirming the relevance of alternative splicing events and the prognosis of breast cancer and its potential value.Finally,the 9 genes and 10 alternative splicing events that affect the survival of BRCA were screened by the prognostic model,and 40 splicing factors that regulate several splicing events were extracted through the correlation test of alternative splicing events and splicing factors.Some of which have been confirmed to be closely related to the prognosis of BRCA,and some have been confirmed to be closely related to the prognosis of other cancers,and others current related research are relatively scarce,the role played in the occurrence and development of BRCA remains to be studied and explored.This provides researchers with a deeper understanding of the prognostic characteristics of breast cancer and provides theoretical support and data support for finding new prognostic targets,as well as key targets for related medical experiments. |
