Datasets
Standard Dataset
KGGAN-DDI
- Citation Author(s):
- Submitted by:
- Quanao Li
- Last updated:
- Mon, 12/02/2024 - 06:36
- DOI:
- 10.21227/gy6a-ha91
- License:
- Categories:
- Keywords:
Abstract
Create Graph:we collected asymmetric drug-drug interaction (DDI) entries from version 5.1.12 of DrugBank, released on March 14, 2024. After a thorough double-check, we removed drugs with incorrect SMILES strings or those that could not be represented by Morgan fingerprints . This filtering resulted in a dataset containing 1,752 drugs and 508,512 asymmetric interactions. Subsequently, we organized the DDI entries into a directed interaction network, where directed edges represent the asymmetric interactions between drugs.
Create DRKG-DDI:The Drug Repurposing Knowledge Graph (DRKG) is a comprehensive biological knowledge graph that relates entities such as genes, compounds, diseases, biological processes, side effects, and symptoms. DRKG integrates information from six databases, including DrugBank, Hetionet, GNBR, String, IntAct, and DGIdb. It contains 97,238 entities across 13 entity types and 5,874,261 triples spanning 107 edge types . For this study, we extracted triples from DRKG that include the 1,752 drug nodes from the Drug-Drug Directed Interaction Network, constructing a new knowledge graph, termed Drug-DDI. The resulting Drug-DDI graph includes 1,392,973 triples, encapsulating relevant information on diseases, genes, and side effects.
This dataset is divided into two parts: Drug_embeddingdata, which contains DRKG-DDI data, and directed graph data for training and testing. The training data is provided as train_0.5, while the test data is divided into test_0, test_50, and test_100, representing different proportions of reversed positive samples used as negative samples (0%, 50%, and 100%, respectively).