Project Summary Fluorescence microscopy combined with the optically transparent Zebrafish xenograft model has the potential to unravel the functional heterogeneity and adaptation of cancer cells that underlies their metastatic potential. However, this requires bridging disparate scales, as metastatic sites can form across an entire organism, and cancer cells within need to be imaged with sub-micron 3D resolution. Further, low occupancy of specific metastatic niches requires imaging many xenografts. This stretches the capabilities of current microscope technology as they fail to provide high-throughput, high-spatial resolution, and whole organism imaging capability at the same time. To break through this boundary, we will develop a new multi-modal microscope to autonomously image both at the organism and subcellular scale in a high-throughput fashion. An optogenetic module further allows to manipulate pathways or photo-convert selected cancer cells for either tracking or single cell sequencing approaches. A computational pipeline will automatically detect and select recurring metastatic sites across many xenografts for high-resolution imaging. This will allow for the first time the quantitative comparison of functional states such as morphology, survival and proliferation, and signaling between different niches. We anticipate that the combination of multi-modal microscopy and the computational pipelines for autonomous imaging will shed new light on key aspects of functional heterogeneity in cancer metastasis.