GenGIS: 3D interactive visualizations of biodiversity.

Michael Porter, Donovan H. Parks, Norman J. MacDonald, and Robert G. Beiko

GenGIS is available under the GNU GPL from http://kiwi.cs.dal.ca/GenGIS, and is supported on Mac OS 10.4 / 10.5 / 10.6 and Windows XP / Vista / 7. 

With an explosion of DNA sequencing, geospatial data collection, and network computing, we are faced with a rapidly growing repository of free data that can be instantaneously combined in new and exciting ways. The key to making sense of large datasets lies in our ability to visualize, explore, and quantitatively analyze their contents. It is no trivial matter to combine geographic, environmental, genetic and other types of data within a software framework that is both easy to use and supports a rich set of interactive analysis tools. Approaches based on virtual globe software such as Google Earth are engaging and easily shared, but commercial restrictions on map data limit the potential downstream uses and analysis is constrained by what is possible in a closed software package. Commercial GIS applications can be powerful and extensible, but require expensive licenses and are typically closed-source.

GenGIS is a new, open-source, three-dimensional geographic information system that builds on other applications including GDAL for geospatial data interpretation, R for statistical analysis, and Python for custom scripting. The graphical core of GenGIS is implemented in C++/OpenGL, and supports the display of digital map data from many sources such as GTOPO30 and SRTM that impose few or no limitations on the use of cartographic data. Apart from maps, input data to GenGIS takes the form of comma-separated attribute files for sequence and location attributes, and Newick-formatted trees. GenGIS will also soon support queries to reference databases via the World Wide Web. The graphical user interface offers a wide range of options for customized data displays within the graphical environment.

Spatial visualization and analysis of biodiversity in GenGIS can follow several paths. Location attributes such as species composition can be visualized using charts and graphs, either natively in the OpenGL window or by using the wide range of data visualizations available in R. Data can also be subjected to appropriate statistical analyses using R and its add-on modules. An important area of focus is the interactive display of hierarchical data, which can represent the phylogeny of a set of sequences or species, or the relative similarity (i.e., a clustering) among sites based on shared attributes. The now-familiar 3D geophylogenies are supplemented with 2D trees which can be laid out along arbitrary axes, allowing the user to test diversity hypotheses along geographic gradients. Finally, custom Python libraries we provide allow the visualization of data as time series, and the creation of fly-through movies. In our presentation we will show how the features of GenGIS can be used to highlight meaningful spatial and temporal patterns in viral, microbial, and eukaryotic data sets.