Multiscale visualisation and analysis of innate immune cell migration at sites of hypoxic inflammation in vivo
Project number: CRC 1450 B04
Project term: 01/2021–12/2024
Hypoxia is an efficient inducer of inflammation, resulting in immune cell invasion into the oxygen-deprived tissue. Accumulation of metabolically highly active immune cells increases oxygen consumption, aggravating hypoxia and tissue damage. The aim of this project is to generate new hypoxia reporters (1) and reporter mouse lines (2) based on the oxygen-independent fluorescent protein UnaG, as well as metabolic reporters (3), which will be functionally tested and applied in preclinical models of myocardial infarction (4), kidney ischemia and intestinal infection.
Using probabilistic machine learning on the imaging data generated, we will attempt to analyse quantitatively tissue hypoxia across multiple imaging scales (5) with a particular focus on the explainability of our deep learning models (6) to ensure trustworthy predictions. These insights will feed back into the sensor design and reporter mouse generation to further optimise our multi-scale analysis of hypoxia, metabolism and cell migration.