Disease target discovery method and system based on multi-agent architecture
By employing a disease target discovery method based on a multi-agent architecture, which combines the collaborative work of a central agent, multi-omics agents, and structural agents, this approach addresses the issues of long cycles, high costs, and high technical barriers to data fusion in traditional target discovery methods. It achieves efficient and accurate target discovery, supporting new drug development.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA RESOURCES PHARM RES INST (SHENZHEN) CO LTD
- Filing Date
- 2026-05-14
- Publication Date
- 2026-07-24
AI Technical Summary
Traditional target discovery methods are time-consuming and costly. Multi-omics and structural computation methods have limitations in target discovery, such as insufficient drugability assessment, high dependence on structural information, and high technical barriers to data fusion, making it difficult to achieve efficient collaboration and large-scale application.
A disease target discovery method based on a multi-agent architecture is adopted, including a central agent, a multi-omics agent, and a structural agent, which achieve target discovery through division of labor and cooperation. The central agent is responsible for task decomposition, result integration, and arbitration, while the multi-omics agent and the structural agent perform autonomous reasoning and verification, respectively. The system identifies targets through multiple rounds of autonomous exploration.
It lowers the threshold for analyzing complex multimodal data, improves the accuracy and comprehensiveness of target discovery, achieves more comprehensive coverage and accurate identification of disease targets, and significantly improves the efficiency and reliability of new drug development.
Smart Images

Figure CN122201415B_ABST