A data-driven approach to establishing cell motility patterns as predictors of macrophage subtypes and their relation to cell morphology.

The motility of macrophages in response to microenvironment stimuli is a hallmark of innate immunity, where macrophages play pro-inflammatory or pro-reparatory roles depending on their activation status during wound healing. Cell size and shape have been informative in defining macrophage subtypes. Studies show pro and anti-inflammatory macrophages exhibit distinct migratory behaviors, in vitro, in 3D and in vivo but this link has not been rigorously studied. We apply both morphology and motility-based image processing approaches to analyze live cell images consisting of macrophage phenotypes. Macrophage subtypes are differentiated from primary murine bone marrow derived macrophages using a potent lipopolysaccharide (LPS) or cytokine interleukin-4 (IL-4). We show that morphology is tightly linked to motility, which leads to our hypothesis that motility analysis could be used alone or in conjunction with morphological features for improved prediction of macrophage subtypes. We train a support vector machine (SVM) classifier to predict macrophage subtypes based on morphology alone, motility alone, and both morphology and motility combined. We show that motility has comparable predictive capabilities as morphology. However, using both measures can enhance predictive capabilities. While motility and morphological features can be individually ambiguous identifiers, together they provide significantly improved prediction accuracies (75%) from a training dataset of 1000 cells tracked over time using only phase contrast time-lapse microscopy. Thus, the approach combining cell motility and cell morphology information can lead to methods that accurately assess functionally diverse macrophage phenotypes quickly and efficiently. This can support the development of cost efficient and high through-put methods for screening biochemicals targeting macrophage polarization.