198. Knowledge Graph based Drug Repositioning Framework
Invited abstract in session MB-56: CBBM: Drug Discovery, stream Computational Biology, Bioinformatics and Medicine.
Monday, 10:30-12:00Room: Liberty 1.11
Authors (first author is the speaker)
| 1. | Se-Jin Chung
|
| Industrial Engineering, Yonsei University | |
| 2. | Yurim Kim
|
| Industrial Engineering, Yonsei University | |
| 3. | Geon Hyeok Chun
|
| Industrial Engineering (Industrial Statistics Lab), Yonsei University | |
| 4. | So Young Sohn
|
| Industrial Engineering, Yonsei university |
Abstract
The increasing expenses and prolonged timelines associated with drug development have sparked interests in alternative strategies, such as in-silico drug repositioning, that can accelerate the drug development process while reducing the cost. Among these alternative strategies, modern research has utilized artificial intelligence to analyze the Drug-Disease Associations (DDA) or Drug-Target Interactions (DTI). However, these approaches still face challenges related with overly optimistic prediction and incomplete integration of biological and chemical database. In this study, we propose an integrated knowledge graph-based drug repositioning framework that encompasses different biological, chemical and clinical entities. The diverse types of interactions within and between different types of entities are derived by utilizing various databases, and diverse knowledge graph embedding methods are compared for their treatment pathway prediction performance. We apply the proposed framework to type 2 diabetes, which is a clinical condition that is globally recognized for its prevalence, mortality and pathological complications.
Keywords
- Artificial Intelligence
- Graphs and Networks
- Computational Biology, Bioinformatics and Medicine
Status: accepted
Back to the list of papers