Discovering novel drug-supplement interactions using SuppKG generated from the biomedical literature

Dalton Schutte, Jake Vasilakes, Anusha Bompelli, Yuqi Zhou, Marcelo Fiszman, Hua Xu, Halil Kilicoglu, Jeffrey R. Bishop, Terrence Adam, Rui Zhang

Abstract: Highlights•Developed a novel methodology to create SuppKG, a dietary supplement specialized knowledge graph.•Produced a text mining tool, SemRepDS, using iDISK.•Extracted relations from PubMed abstracts to produce SemMedDB-DS.•Constructed a knowledge graph containing unique dietary supplement relations.•Generated novel drug-supplement interactions using discovery patterns.