Bristol-Myers Squibb (BMS) is a global pharmaceutical company with drug discovery and development programs in several therapeutic areas. As part of its enterprise information governance, there is an ongoing effort to unlock research data from siloed systems, by building a knowledge graph that transcends research domains and departments.
The Hyve, a Dutch IT services provider for the biomedical domain, assists in this effort by creating semantic models of particular areas, such as computational genomics or clinical trials.
In this case-study, we will share our experience modelling research data in Immuno-Oncology & Cell Therapy (IOCT). Here, traceability of research data through the development pipeline poses a particular problem because most data is currently stored in particular vendor solutions and departments. We will first introduce the semantic model that was created to model the IOCT research domain and data, and how it builds on public domain ontologies and existing BMS ontologies. We will then present our efforts to build a knowledge graph by instantiating this model using a selected set of research data. We will evaluate the strategies and tools we explored to unlock as much of the data as possible, and the end-to-end use cases that can now be answered using the knowledge graph.
#knowledgegraphs #knowledgegraphconference #knowledgegraphusecasesinhealthcare #knowledgegraphdatamodel