Visualization is a fundamental challenge when trees get larger than 50-100 species. The phylogeny of all plants, if printed out on a giant piece of paper with species labels printed at 10-point font size, would be 2116 meters long, which is almost five times the height of the Empire State Building. Imagine visualization and analytical software that can zoom out to see the landscape of five Empire State Buildings end to end, yet still be capable of zooming in to read the text on newspapers in the hands of people sitting by their office windows. Then imagine this vast and gigantic tree used as a giant computational template to infer the historical patterns for any known biological data – for example, detecting orthologous, paralogous, and xenologous genes, tracing the historical sequence of genome reorganizations, correlating genomic changes with patterns of plant innovation and adaptation, and examining patterns of adaptation with geographic vicariance, climatic change, or coevolution. Improving visualization is important not only for surmounting this emerging obstacle to interpreting research results, but also for directly reaching out to a much broader audience, because the metaphor of the tree of life is widely understood by the general public. Several models for scaling tree visualizations have been explored (Munzner et al. 2003; Hughes et al. 2004; Sanderson 2006; Jordan and Piel 2008), but one of the great obstacles to progress in tree visualization is almost more aesthetic than technical. What kinds of information ought visualizations convey? Several independent efforts in the larger phylogenetic community are aimed at addressing the problem of tree visualization. Initially this team will focus on establishing requirements and evaluating available tools.
Working Group Members
Karen Cranston, Team Lead, Duke University
Kris Urie – developer
Naim Matasci – scientific lead