Type 2 diabetes (T2D) is a complex and clinically heterogeneous disease. Although clustering approaches have defined clinical subtypes, their genetic and molecular architectures remain poorly understood. Efforts using genome-wide association studies (GWAS) have been constrained by modest sample sizes and conservative significance thresholds, limiting subgroup resolution. Leveraging recent expansions in cohort scale and diversity, we developed a new analytical framework that integrates GWAS with Machine Learning and eXplainable Artificial Intelligence. Applied to the UK Biobank, this strategy enabled the first comprehensive genetic and molecular characterization of T2D subtypes, identifying 184 genes that define each subgroup’s molecular landscape and associated tissue profiles. In addition, the framework improved detection of T2D risk compared to current Polygenic Risk Scores. The findings underscore the importance of clinical stratification to uncover the complex pathophysiology of T2D and to pave the way for more precise prevention and treatment strategies.
Common Data Models (CDMs) enhance data exchange and integration across diverse sources, preserving semantics and context. Transforming local data into CDMs is typically cumbersome and resource-intensive, with limited reusability. This article compares OntoBridge, an ontology-based tool designed to streamline the conversion of local datasets into CDMs, with traditional ETL methods in adopting the OMOP CDM. We examine flexibility and scalability in the management of new data sources, CDM updates, and the adoption of new CDMs. OntoBridge showed greater flexibility in integrating new data sources and adapting to CDM updates. It was also more scalable, facilitating the adoption of various CDMs like i2b2, unlike traditional methods reliant on OMOP-specific tools developed by OHDSI. In summary, while traditional ETL provides a structured approach to data integration, OntoBridge offers a more flexible, scalable, and maintenance-efficient alternative.
Transforming the population based biomedical cohort into the Common Data Model (OMOP-CDM) empowers researchers to access direct sources of information, enabling a deeper understanding of how genetic profiles relate to clinical outcomes and providing new knowledge that can significantly influence health care practices around the world.