Applications of Computational Intelligence in Biology

Volume 122 of the series Studies in Computational Intelligence pp 231-255

Visualization and Interactive Exploration of Large, Multidimensional Data Sets

  • John T. LangtonAffiliated withMichtom School of Computer Science, Brandeis University
  • , Elizabeth A. GiffordAffiliated withMichtom School of Computer Science, Brandeis University
  • , Timothy J. HickeyAffiliated withMichtom School of Computer Science, Brandeis University

* Final gross prices may vary according to local VAT.

Get Access


As biologists work with more and more data, there is an increasing need for effective tools to analyze it. Visualization has long been used to communicate experimental results. It is now being used for exploratory analysis where users rapidly determine significant trends and features by working with visual projections of data. A basic workflow is to a) record experimental results or simulate some biological system, b) form hypotheses, c) verify hypotheses with interactive visualizations and statistical methods, d) revise hypotheses, and e) confirm computational results with experiments in wet-lab.

In this chapter we describe a number of visualization methods and tools for investigating large, multidimensional data sets. We focus on approaches that have been used to analyze a model neuron simulation database. These methods are best applied to databases resulting from brute force parameter space exploration or uniform sampling of a biological system; however, their wider applicability is currently under investigation. An example analysis is provided using a generalized tool for interactive visualization called NDVis. We include a summary of NDVis, its plugin architecture, and JavaSim which can be used as a plugin for NDVis or as a stand alone tool for investigating model neuron simulations.