normalizing by number of species (which is a vague concept anyways, most species nomenclature in bacteria is driven by the need to publish - salmonella and e co

normalizing by number of species (which is a vague concept anyways, most species nomenclature in bacteria is driven by the need to publish - salmonella and e coli are basically the same) is completely irrelevant concept for this discussion. When you're looking for 'random interactions' most chemists reach for combichem anyways. Look at all of the drugs that treat CML - (gleevec eg) none of them come from any biotic source.