The Connective Network Construction And Analysis Of Drug Adverse Effects

To pursue a systematic approach to the discovery and understanding of connection among drug adverse efffects and new approach for discovering new side effects of drugs, we construct a connective map of drug- side effects based on the SIDER2 database and introducing link prediction methods in the network.By integrating and organizing the data from SIDER2, we provide an insight into structure and connection among the map, the topological property and community structure were also given. In the significantly connected network, we found some strong connected side effect pairs, who have no correlation and belongs to separate SOC class.We also develop and evaluate the framework for drug side effect prediction based on link prediction in bipartite network. Integrating side effect appearance frequency has also been proved to be of value in improving the performance of drug side effect prediction.