Should we treat the ionome as a combination of individual elements, or should we be deriving novel combined traits?