Personalized Drug Knowledge Graph (CPMKG) enhanced the traditional relationships and used conditional events as the basic knowledge data structure so that it is more in line with the actual needs of users which is providing suggestions for personalized disease treatment and prevention based on specific genetic background, environment and lifestyle. CPMKG discovered and normalized drug-related conditional knowledge including side effect events, drug sensitivity events, drug molecular mechanism events, and variation biological function events. All this knowledge had been carefully curated using our dedicated knowledge annotation system. CPMKG has been extended with a new "explore" feature, which integrates all conditional knowledge by using graph database technology to provide a bigger picture for different users in numerous scenarios. For the researchers, gene expression patterns caused by the interaction between the same drug and different variations can be shown on a single graph. While for doctors, CPMKG can help them provide medical advice to patients who have been prescribed numerous times based on their genetic background.

Publication

Yang J, Zhuang X, Li Z, Xiong G, Xu P, Ling Y, Zhang G.
CPMKG: a condition-based knowledge graph for precision medicine.
Database (Oxford). 2024 Sep 27. DOI: 10.1093/database/baae102. PMID: 39331730