Researchers have developed THBKG, a Temporal Heterogeneous Biomedical Knowledge Graph designed to predict the advancement of therapeutic programs in clinical trials. This graph accounts for evidence available at specific past decision points, addressing a critical gap in existing biomedical knowledge graphs. THBKG successfully predicts Phase III advancement for target-disease pairs, outperforming direct-evidence models, particularly for cases lacking immediate supporting data. The system's ability to propagate information over intervening biological connections allows it to identify promising hypotheses even when direct evidence is scarce, offering explainable predictions. AI
IMPACT This temporal knowledge graph could improve the efficiency of drug development by better predicting clinical trial success.
RANK_REASON The cluster contains a research paper detailing a new knowledge graph and its application. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
