Predicting adverse drug‒drug interactions with a deep learning model
2022
Gwangju Institute of Science and Technology (GIST), Gwangju, South Korea
Adverse drug-drug interactions (DDIs) are a major concern due to their unexpected side effects and need to be detected at an early stage of drug discovery and development. Although understanding the mechanisms of DDIs at the molecular level is critical for predicting their adverse effects, current models rely on drug structures and properties, with a prediction range limited to previously observed interactions. They do not consider the effects of DDIs on genes and cellular functions.
Here, a deep learning-based model for predicting DDIs based on signatures of drug-induced gene expression is presented. The DeSIDE DDI model consists of two parts: a feature generation model and a DDI prediction model. The feature generation model predicts the effects of a drug on gene expression by considering both the structure and the properties of the drug, whereas the DDI prediction model predicts various side effects of drug combinations.
This model can detect potentially dangerous drug pairs and serve as a drug safety monitoring system. It can also help determine the proper use of a drug in the development phase.
DeSIDE-DDI: interpretable prediction of drug-drug interactions using drug-induced gene expressions
Hojung Nam
Added on: 05-13-2022
[1] https://jcheminf.biomedcentral.com/articles/10.1186/s13321-022-00589-5[2] https://github.com/GIST-CSBL/DeSIDE-DDI[3] https://www.technologynetworks.com/drug-discovery/news/predicting-adverse-drugdrug-interactions-with-deep-learning-model-361271