Turning complex biomedical data into knowledge

Coordinators: H. Busch (UzL), D. Ellinghaus (CAU), C. Kaleta (CAU), I. König (UzL)

Precision medicine generates enormous amounts of data: from genomes and molecular measurements to medical images and clinical records. The challenge is to combine these different layers of information and extract patterns that are scientifically meaningful and clinically useful.

RTF-VIII develops the data-science methods needed to support this process. We combine machine learning and artificial intelligence with explainable data analysis, visualization, controlled approaches for exploring data availability and aggregate information, and mathematical models that integrate heterogeneous biomedical data, including microbiome-related datasets. These methods aim to identify robust patterns, make relevant data resources more transparently discoverable and queryable, and generate testable hypotheses about disease mechanisms.

RTF-VIII also establishes a Data Science Board as a methodological hub for clinical and applied PMI projects. The Board will collect and discuss data-science and AI/methods questions, match them to relevant expertise within the Cluster, and support the identification or development of suitable analytical approaches.

The goal is to turn complex biomedical data into robust knowledge that supports clinical research and, ultimately, precision medicine.