People seem to be ignoring the elephant in the room. Which is ... what are the odds the n assay works and the pretest will work? I don't know if its 90 or 50 10 or 1? This is not a slam-dunk technology-wise right? Nor is it an impossibility.