International Mammalian Genome Society

logo18th International Mouse Genome Conference

17-22 October 2004, Seattle, USA


Snoddy J

GST at Oak Ridge National Lab and University of Tennessee, Oak Ridge, United States

There are novel bioinformatics challenges posed by the study of biological systems. Highly networked systems create phenotypes from genotypes and the environment. We now need efforts to understand how those genes, gene products and cells function in these networks.  This will require large data sets whose analysis will further require new bioinformatics. While databases of experimental information are necessary—indeed critical—they are not sufficient to help obtain insight from the analysis of large, complex networks.  We need a mathematical language to describe the interconnections of these networks, i.e., robust tools that can help us get insights from use of that language, and data mining tools to see patterns among these networks.

A number of both experimental and computational collaborators have been working together to develop these needed improvements.  Our current work involves elucidating aspects of regulatory networks and co-expression networks.  We are integrating data from different DNA microarray data sets from gene product function data (e.g. GO), gene families, genetic variation, evolutionary conservation of coding and non-coding parts of the genome, and other similar information.  We are developing methods to find and visualize tightly connected subcomponents in networks that are of interest to researchers after data integration.

We have developed several sophisticated data mining environments and tools. Some of these will be shown as they are now ready for wider use and evaluation. Some biological results will be presented that come from this analysis.

[an error occurred while processing this directive]