Cancer Drugs Repositioning Based On Drug-disease Relationship Model

Drug repositioning research not only greatly reduces the period of researching and development(R&D)for new drugs,but also reduces the economic cost.Recently,the economic cost of new drugs.R&D has increased year by year and the risk of failure is increasing.Dig for new indications for existing drugs and reposition them,can effectively avoid the risk,reduce the cost and accelerate the pace of the drugs listing to meet the needs of clinical drug use,so it has got more and more attention from medical world.TCGA,an important data source for cancer researchers,includes clinical data,genome variation and mRNA expression of various human cancers.The Connectivity Map database studies the functional links between drugs,genes and diseases based on changes in gene expression.The database contains gene chip data before and after treatment of 1309 small-molecule drugs and 5 human tumor cell lines.Mutual information is a useful measure of information in information theory because genes are related to each other.The correlation can be retained by the feature gene extraction method based on mutual information algorithm.Fold Change is the differential expression multiple.For gene expression data,the traditional calculation method is to screen feature genes directly by calculating its logFC value.For drugs repositioning,this paper will use prostate cancer and breast cancer gene expression data of gene expression data from TCGA,using R software of mutual information algorithm and Fold Change(FC)algorithm respectively in the fight against prostate cancer and breast cancer drug repositioning of relocation,with scores than Connectivity Map database,Thioridazine(Thioridazine),trichostatin A(trachomectin A),gliclazide(gliclazide)and other drugs that may have therapeuticeffect on prostate cancer were obtained.Gliquidone,levomepromazine,Imatinib and other drugs that may have therapeutic effects on breast cancer have been obtained.The traditional FC algorithm has a complicated calculation process,a large amount of computation and takes a long time,this method is improved on the basis of the mutual information algorithm,and the mutual information algorithm is combined with the FC algorithm to successfully obtain the repositioned drug.In this way,the computing speed is improved,the time to acquire feature genes is greatly shortened,the time cost is saved,the calculation results are more accurate,the selected feature genes are more representative,and the process of drug repositioning is accelerated.