SmileGNN: Drug–Drug Interaction Prediction Based on the SMILES and Graph Neural Network

Abstract

Concurrent use of multiple drugs can lead to unexpected adverse drug reactions. The interaction between drugs can be confirmed by routine in vitro and clinical trials. However, it is difficult to test the drug–drug interactions widely and effectively before the drugs enter the market. Therefore, the prediction of drug–drug interactions has become one of the research priorities in the biomedical field. In recent years, researchers have been using deep learning to predict drug–drug interactions by exploiting drug structural features and graph theory, and have achieved a series of achievements. A drug–drug interaction prediction model SmileGNN is proposed in this paper, which can be characterized by aggregating the structural features of drugs constructed by SMILES data and the topological features of drugs in knowledge graphs obtained by graph neural networks. The experimental results show that the model proposed in this paper combines a variety of data sources and has a better prediction performance compared with existing prediction models of drug–drug interactions. Five out of the top ten predicted new drug–drug interactions are verified from the latest database, which proves the credibility of SmileGNN.

Applying precision medicine principles to the management of multimorbidity: the utility of comorbidity networks, graph machine learning, and knowledge graphs

“…Finally, the new overall graph is fed into the GNN. The HSGNN outperformed other baseline GNN models when applied to more than 46,000 patients in the MIMIC-III dataset, improving the AUC for ICD-9 classification disease diagnosis at both the patient level and at the visit level (46).…”

“…can then be used for downstream tasks, including risk prediction. Alternatively, separate PSNs can be used as input for heterogeneous graph neural networks (46). The development of a PSN was used for the prediction of future diabetes, with 38 communities detected from a weighted unipartite projection with edge weights inversely proportional to the degree (number of connections) of each node (62).…”

“…The ability to predict separate ICD-9 disease diagnosis codes in ICU patients was examined using a heterogeneous graph similarity neural network (HSGNN) (46). A heterogenous graph consisting of multiple medical entities is first transformed into multiple similarity subgraphs using the different meta-paths (visit-disease-visit, medication-visit-patient, disease-visitmedication) contained within the initial overall heterogeneous graph.…”

“…Graph embedding techniques capture the latent topology of a graph, including node similarity in the form of vectors for use in prediction models (20), and graph ML enables fully end-to-end models with graph data as input and node, edge, or subgraph prediction as output for data outside of the existing network (44,45). The use of graph neural networks (GNNs) to date includes predicting drug-drug interactions (46), modeling polypharmacy side effects, and learning the temporal patterns of disease development in comorbidity networks (47). Finally, knowledge graphs can be used to make use of the known connections between drugs, proteins, and genes to uncover new associations between different entities for use in areas such as drug repurposing and disease diagnosis.…”

Healthcare knowledge graph construction: A systematic review of the state-of-the-art, open issues, and opportunities

“…The aim of the KG is to discover ADRs of antitumor drugs as well as provide explanations why they occur. Predicting and discovering ADRs have been further reported in [ 69 – 74 ]. Zhao et al [ 75 ] designed their drug action mechanism KG after extracting information from 770,000 abstracts of medical papers.…”