Acquisition Of Sample Specific Biomarkers And Personalized Drugs Recommendation

Cancer is an invisible killer that threaten human health and life,the primary cause of the initiation and progression of cancer is that some important genes mutation arose in the human body,and these mutations result in the growth of cells not under the control of the body.How to accurately identify the key factor of the initiation and progression of cancer,and make an appropriate treatment strategy is the key to cancer treatment.In the traditional methods,the main goal of the research is to obtain these biomarkers from the difference of the normal samples and tumor samples,but these methods can only discover the common features of all the samples.In the real world,under the influence of various factors,such as different living environment,there are big difference among individuals,this also reflect the key factor of the initiation and progression of cancer in individual is unique.Driven by the idea,discovery of the sample specific biomarkers attracts people's attention.Discovering the cause of cancer for each patient and making the corresponding treatment strategies for the patient based on these characteristics,then we can achieve the purpose of cancer treatment.Firstly,we translate the gene expression data use the different distribution between different phenotype,then based on the network information,at the same time combined with the one-dimensional normal distribution and the two-dimensional normal distribution to obtain two different types sample specific biomarkers in this thesis respectively.Next,we use the known drug-target relationship and the cancer subtypes,the drug information of each cancer subtype was found by enrichment analysis.Finally,based on the differences in pathogenesis among different individuals,the applicable drugs are also different.we designed the score weight for each drug based on the sample specific biomarkers,drug-target relationship and drug side effect information.From the drug set with different weight,we select a smallest and more suitable drug combinations for each cancer sample,and the personalized drug recommendation realized.