The volume of biomedical research data stored in various databases has grown immensely in recent years due to the proliferation of high-throughput biomedical ‘-omics’ technologies. Nearly all of respective databases, or ‘knowledge sources’ (KSs), address a particular area of biomedical research, leading to natural diversity but also growing disintegration between individual KSs, which generates downstream inefficiencies when mining diverse databases for knowledge discovery. Expanding efforts, both in academia and industry, are focused on the development of methods and tools to enable semantic integration and concurrent exploration of disparate biomedical KSs, using specially constructed biomedical ‘graph knowledgebases’ (GKBs) that support the generation of new knowledge through the application of reasoning tools and algorithms. Our group has contributed to these efforts by initiating the development of a GKB-based question-answering system termed Reasoning Over Biomedical Objects linked in Knowledge-Oriented Pathways (ROBOKOP) 1,2. ROBOKOP’s publicly accessible user interface (UI) 3 allows users to address both relatively simple questions such as “what genes are associated with drug-induced liver injury?” and more complex ones such as “What biological mechanisms are responsible for drug action?” I will discuss the development of ROBOKOP and provide examples of applications including the elucidation of Adverse or Clinical Outcome Pathways of drug action, development of domain-specific KG such as COVID-KOP 4, and drug repurposing including methodologies relying on KG embeddings 5 and machine learning.

Biomedical knowledge graphs and their application to drug discovery

Fall 2022 abstracts, posters, and presentations will launch in mid-September.

ACS Fall 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.