Diagnostic immune fingerprints
Research in the Eberl lab aims to identify diagnostically relevant biomarker signatures (immune fingerprints) for point-of-care diagnosis of patients presenting with symptoms of acute infections.
We assess such pathogen-specific local patterns quantitatively and qualitatively in different patient cohorts, long before traditional test results including microbiological cultures become available.
The definition of the best possible biomarker combination and its validation poses considerable computational challenges, with approximately 1.27×1030 possible combinations of a panel of 100 different biomarkers in 100 individual patients.
We develop and apply novel statistical tools to interrogate highly complex datasets and utilise supervised and unsupervised machine learning approaches and exhaustive searches on the RAVEN supercomputer in order to define and cross-validate relevant biomarker signatures.
This cross-disciplinary research has received support from the NIHR Invention for Innovation (i4i) programme, the MRC Confidence in Concept scheme and Health Technology Challenge Wales, and is conducted in collaboration with basic, computational and clinical scientists as well as with industrial partners.
If the concept of pathogen-specific diagnostic fingerprints were to be introduced into clinical practice, targeted therapy could be given immediately, improving outcomes, avoiding the emergence of antibiotic resistant strains, and saving costs for the NHS.