Font Size: a A A

Network Pharmacology Analysis And Prediction Of Drug-Drug Interactions

Posted on:2017-03-25Degree:MasterType:Thesis
Country:ChinaCandidate:P Y ZhaoFull Text:PDF
GTID:2284330485957910Subject:Computer technology
Abstract/Summary:PDF Full Text Request
The arrival of the aging society and coexistence of multiple chronic diseases and underlying diseases, has greatly promoted combinatorial therapy. So drug-drug interactions will become an unavoidable problem in clinical treatment. Combinatorial therapy is a promising strategy for combating complex disorders due to improved efficacy, but in fact, the combinations of drugs often cause adverse reactions or even serious side effects. There are many factors affecting the result of drug-drug interactions, so the mechanism is very complex. Although we have got some achievements in this field, the mechanism of drug interaction is still in the fuzzy state, which needs to be further explored.In this article, we mainly obtain the network pharmacology characteristics of drug-drug interactions by analyzing the topological structure and molecular mechanism, and also propose a computational approach which can predict novel drug-drug interactions. The main work is as follows:First, we analysis the topological characteristics of drug-drug interactions network via several important topological properties, such as degree distribution, betweenness, clustering coefficient, closeness centrality and modularity. The results show that the network is a modular network(the module is 0.523, the average clustering coefficient is 0.226). In addition, the functions of drugs in the same module were similar, but the relationship between the module and drug disease is relatively weak. This study suggests that drugs with functional similarity exists clustering phenomenon, which may indicate molecular network mechanism of drug-drug interaction.Second, through integrating the PPI network, drug targets, drug ATC classification and chemical structure, we qualitatively analyze the molecular network mechanism of drug-drug interactions. The results show that drugs’ targets tend to be in the same molecular network topology module, and drug-drug interactions are related to the ATC structure and chemical structure of the drug. In addition, we also analyze the different types of interactions, and found that these effects are related to their modules, functional properties and chemical structure similarity.Third, the forecast method in this paper is based on hypothesis that similar drugs have similar interactions. From the perspective of the drug ATC structure and module, we can acquire novel drug-drug interactions by integrating a reference standard database of known drug-drug interactions and the similarity. The effect of the prediction method is better, which is evaluated by the ROC curve, and the AUROC value is 0.95 and 0.92. In addition, the method is interpretable in that it generates drug interaction candidates that are traceable to pharmacological or clinical effects.
Keywords/Search Tags:Drug-Drug Interactions, Network Medicine, Topological Structure, Molecular Mechanism, Prediction
PDF Full Text Request
Related items