[Application of drug-target prediction technology in network pharmacology of traditional Chinese medicine]

Abstract

In recent years, network pharmacology has been developed rapidly, and especially, the concept of ″network target″ has brought a new era in the field of traditional Chinese medicine (TCM). The integrity and systematicness emphasized in network pharmacology comply with the characteristics of holistic view and treatment in Chinese medicine. It can provide deeper insights into the underlying mechanisms of TCM theories, including the illustration on action mechanism of Chinese medicine, selection of pharmacodynamic materials and the combination principles of various Chinese herbs, etc. Therefore, this theory is more suitable for TCM academic characteristics and practical conditions. The key problem in network pharmacology is how to efficiently and quickly identify the interactions between large amounts of drugs and target proteins. As an efficient and high throughput way, drug-target prediction technology can reduce costs, quickly predict the component targets, and provide foundation for the application of TCM network pharmacology. In view of the large amount of compounds and target databases, different prediction methods and technologies have been developed, and used to predict the drug-target interactions. Many virtual screening technologies have been successfully applied to network pharmacology. Based on different prediction principles, drug-target prediction technology can be generally divided into four types： ligand-based prediction, receptor-based prediction, machine learning and combined prediction. In this paper, we are going to review the prediction methods of drug-target interactions and give acomprehensive elaboration of their application in network pharmacology of TCM, hoping to provide beneficial references for various Chinese medicine researchers.

Keywords: ligand-based prediction; machine learning; network pharmacology of traditional Chinese medicine; predictions of drug-target interactions; receptor-based prediction.