The talk features one applet and two applications: Linked Micromap Plots (LM plots), Conditioned Choropleth Maps (CCmaps), and GLISTEN.
LMplots: In an award winning team project, graphics research centered at the National Cancer Institute transformed this static template into an interactive and sometimes dynamic applet. Their purpose is to communicate health related statistics about states and counties to health planners. The graphics design has been refined by usability assessment in NCI's impressive new usability lab. The software and design should be of interest to other government agencies, academia, and industry. Applications can extend to data mining as indicated in a previous publication.
CCmaps: The purpose of this Java shareware application is to promote better hypothesis generation about patterns in geospatially-indexed statistics. Researchers often use classed choropleth (colored) maps for hypothesis generation. However, it is helpful to control for known sources of variation before speculating about new relationships. CCmaps introduces dynamic partitioning sliders to produce a 3 x 3 layout of maps. The regions highlighted in each map are homogenous with respect to two additional variables. The remaining spatial patterns in each map are of interest as are distributional comparisons across maps. CCmaps also features dynamic population-weighted statistics and QQplots.
GLISTEN: Bioinformatics researchers and students please attend. You may have a host of applications for this new software. This C++/openGL application has great generality. The presented applications represent statistics indexed by m-mers of nucleotides or amino acids. While statistical computations in studies of binding and possible binding sites are increasingly sophisticated, typical views of the results are still the top of a sorted table. The examples display 3-D representations of m-mer indexed statistics in the effort to obtain a better overview of results and to possibly obtain new insights. Examples illustrate a few options related to labeling, filtering and alternative views.