This is interesting. Can you provide me an example?
I'll try.
Right now, I am interested in finding a virus. The virus is present in many reptiles and mammals, and around a hundred variants of the virus exist in humans. Evolutionary theory tells me that parts of the virus change more than other parts, according to their function within the virus. If the part has a structural function, then it changes a lot (because quite a bit of amino acid variation can occur without drastically changing the structure). But, if the part has a mechanistic function--like an enzyme--then it is resistant to change (because a single amino acid change in the active site can disrupt or even abolish the function of an enzyme). So, since I want to find variants of the virus, some of which may not yet have been found--I am designing an assay to detect the active portions of the virus enzymes, not the structural areas.