Figure 1

Integration schema. Schema used for virulence identification. The records whose weights were converted into features for classification were derived from PDB structures[29], GO terms (from both AmiGO and GenNav)[30, 31], InterPro domains and families[32], TIGRFAM families[33], BioCyc pathways[34], KEGG terms and pathways[35], and CDD domains[36]. Note that GenNav is a recursive source - that is, it may re-query itself to recreate the GO hierarchy within the query graph.