Visualization and Exploration of 3D Toponome Data

Abstract

The toponome of a cell describes the location and topological distribution of proteins across the cell. In toponomics, the toponome is imaged and its inner structure and its semantics are investigated in order to understand how cells encode different functionalities both in health and disease. Toponome imaging results in complex multiparameter data composed of a 3D volume per protein affinity reagent. After imaging, the data is binarized such that 1 encodes protein present and 0 encodes protein absent. Biologists are particularly interested in the clustering of these binary protein patterns and in the distribution of clusters across the cell. We present a volume rendering approach for visualizing all unique protein patterns in 3D. A unique color is dynamically assigned to each pattern such that a sufficient perceptual difference between colors in the current view is guaranteed. We further present techniques for interacting with the view in an exploratory analysis. The biologist may for instance ''peel of'' clusters thereby revealing occluded cell structures. The 3D view is integrated in a multiple coordinated view system. Peeling off clusters or brushing protein patterns in the view updates all other views. We demonstrate the utility of the view with a cell sample containing lymphocytes.