An image-based data-driven analysis of cellular architecture in a developing tissue

A data-driven analysis of cell morphology and intracellular organization in the developing zebrafish posterior lateral line primordium, a model tissue for the study of self-organized morphogenesis. 3D image stacks were acquired using AiryScan FAST-mode confocal fluorescence microscopy. Automated single-cell segmentation and point cloud-based morphometry were developed to extract numerical features representing cell morphology and intracellular protein distributions. Machine learning was used with the extracted numerical features to perform data integration across experiments and context-guided data visualization. The resulting data was analyzed to discover biologically meaningful patterns at the cell and tissue scale.