Yannis Kalaidzidis (bit.ly/q2dm1Y) explains the 'inverse problem' of high-throughput cellular imaging, i.e., how to transform raw images into abstract visualizations that give insight into biological processes. He gives a clear, concise account of several common strategies, especially profile plots and 2D clustering of phenotypic data, and concludes that high-throughput and multidimensional phenotypic data will remain an ongoing visualization challenge. This talk was presented at VIZBI 2011, an international conference series on visualizing biological data (vizbi.org) funded by NIH & EMBO. This video was filmed and distributed with permission under a creative common license.