Drug-drug interactions prediction based on deep learning and knowledge graph: A review
Problem
Detecting drug-drug interactions through biological or pharmacological analysis is time-consuming and expensive, leaving patients at risk of adverse events.
Why it matters
Unpredicted drug-drug interactions can cause adverse events and irreversible damage to patients.
What the AI did
The paper reviews deep learning and knowledge graph methods that extract entity features to predict drug-drug interactions, grouping them into three main classes.
Results
not stated in source
Impact
Consolidates AI-based DDI prediction approaches to guide faster, cheaper drug-safety research.
Record
| Venue | iScience |
|---|---|
| Published | 2024 |
| Work type | article |
| Authors | Huimin Luo, Weijie Yin, Jianlin Wang, Ge Zhang, Wenjuan Liang, Junwei Luo, Chaokun Yan |
| Citations | 102 (OpenAlex, at import time) |
| References | 166 |
| Model or method | deep learning and knowledge graph methods (reviewed, none specific) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1016/j.isci.2024.109148 |
| Open access | gold |
| Keywords | drug-drug interactions, deep learning, knowledge graph, prediction, review |
Importance rating
significant — thousands to millions affected
Abstract states DDIs can cause adverse events with potential for irreversible damage and an urgent need for computational prediction methods; traditional detection is time-consuming and expensive, affecting drug safety broadly.