Diabetes mellitus is one of the most widespread diseases worldwide. The strong increase in incidence not only of type 2 diabetes in the aging population, but also of the autoimmune condition of type 1 diabetes will present major challenges to healthcare systems in the years to come. The German Center for Diabetes Research (DZD) seeks to investigate the causes of the disease and, through new scientific findings, to develop effective prevention and treatment measures to halt the emergence or progression of diabetes.
In order to investigate diabetes, we have to combine basic research data sources from genetics, epigenetics, metabolic pathways with data from our clinical studies. Connecting these highly heterogenous types data is a challenge, but today this is necessary to answer biomedical questions across disciplines.
Graph technology enables a new dimension of data analyses to fight diabetes by connecting data from various species, locations and disciplines. Here we present a use case to study prediabetes where our graph includes data from animal models, genetics, metabolomics and literature to deduce causes of prediabetes in human.
Connecting data and applying modern machine learning techniques will help scientists getting closer to understand this complex disease and will hopefully help to care for patients in the future.