A Gut Feeling: Biomedical Knowledge Graphs for Interrelating the Gut Microbiome and Mental Health

Abstract

The thesis delves into an investigation of the association between gut microbiota and mental health by constructing biomedical knowledge graphs sourced from scientific literature, including text and depicted molecular pathway graphs, and public databases, where the information is retrieved using a combination of automated and manual techniques. We provide an extensive introductory overview of the general architecture of knowledge graph construction and a ten-rule guide for applying Semantic Web technologies effectively when constructing knowledge graphs. The thesis introduces three relevant biomedical knowledge graphs: Microbiota Knowledge Graph (MiKG), Pre-/Probiotics Knowledge Graph (PPKG) and Ketamine Pathway (KetPath) Knowledge Graph. These graphs are built upon the tripartite relationship between gut microbiota, neurotransmitters, and mental disorders. Querying the integrated knowledge graphs revealed novel insights into the intricate interplay among these elements, ranging from the general level, such as potential new overall strategies for using probiotics in treating mental disorders, to the specific level, such as particular expression patterns of the bdnf gene in certain health conditions.