Brain dysfunction has a detrimental effect on our daily lives. Yet, surprisingly little is known about how microglia, traditionally classified as immune-responsive cells, maintain brain functionality and how deviation from this task will induce disease onset. As highly dynamic cells, the microglia branching tree gives a first intuition on their intrinsic state; massively reduced branching is commonly found in severe pathological conditions and indicates phagocytic activity. However, attempts to morphologically classify microglia beyond this extreme condition have been unsatisfactory. In our line of research, we noticed that strategies to simplify the branching tree are insufficient to handle the dynamic nature and often cause loss of discriminative information.

In this talk, I will introduce our developed algorithm morphOMICs, which is based on principles of topological data analysis, statistics, and machine-learning dimensionality-reduction techniques [1]. I will show how morphOMICs has been robust to quantify subtle yet functionally informative morphological adaptations. The talk will finish with a roadmap to connect microglia morphology to their intrinsic functional state taking advantage of the retina and in this way establish microglia as a sensor for the physiological state of the brain and as an early indicator for environmental changes.