KEGG spider implements the Global Network statistical framework to analyze gene list using as reference knowledge a global gene metabolic network inferred from KEGG database. The KEGG database is a collection of chemical structure transformation patterns for substrate-product pairs (reactant pairs). Two genes are connected by edge if they (proteins) catalyze reactions which share the same compound either as product or substrate. The resulting global metabolic gene network links by edges any two genes that are associated with reactions sharing common compounds (classified in KEGG database as main reaction pair). For the human genome, the global metabolic gene network covers in about 1,100 genes involved in approximately 15,000 unique pairwise interactions.
Reference: if you will find the results produced by KEGG spider usefull, please cite:
1. Antonov A.V., Dietmann S., Mewes H.W. KEGG spider: interpretation of genomics data in the context of the global gene metabolic network. Genome Biology 2008, 9:R179.