Volume visualization for exploration of population trends in two-dimensional gel electrophoresis protein data

We propose a visualization method for exploring and analyzing data from two-dimensional (2D) gel electrophoresis, a frequently used laboratory technique in proteomics for characterizing and studying protein content in biological samples from patients or cell cultures. We show how 2D gel electrophoresis data from a population can be treated as a volume with no preferred slice ordering, and present a visual analysis approach to investigate the relation between an external biological variable such as subject age and the protein patterns based on dynamic slice sorting. We calculate the Spearman rank correlation as a measure of the relation, and use this together with standard deviation as statistical transfer functions to let the explorer filter the data. We also show a method for embedding the selected external variable distribution and the statistical measures in the volume, providing a statistical context for the pattern exploration. Finally we show the applicability of our method in an example of analysis of data from real populations.