Bioinformatic characterization of plant networks

Cells and organisms are governed by networks of genetic, physical and metabolic interactions between proteins and other their substrates, such as DNA, RNA, and other biologically important molecules. Large scale experimental studies of interactions between components of biological systems have been performed for a variety of eukaryotic organisms. However, there is a dearth of such data for plants. Computational methods for prediction of relationships between proteins, primarily based on comparative genomics, i.e., using homology to transfer information between and across organisms, provide useful molecular and systems level views of cellular function and can be used to integrate and extend information available about other eukaryotes to plants. We have predicted proteinprotein and protein-DNA networks for six proteomes of Oryza sativa, Arabidopsis thaliana, and several plant pathogens using the Bioverse framework (http://bioverse.compbio.washington.edu), that relates three-dimensional atomic level detail of information regarding single molecules to systems, cellular, and organismal biology. We show that our predictions are similar to experimentally derived interactions, but provide greater coverage for a larger number of proteins. Predicted interaction networks for plants can also be used to provide novel functional annotations and predictions about plant biochemical pathways to aid in rational engineering, either by genetic modification and/or marker-assisted breeding, to improve human heath and quality of life.