"Are you saying there's no reason to study the relationship between behavior and genes because there's too much noise" No, I'm trying really hard to express th

"Are you saying there's no reason to study the relationship between behavior and genes because there's too much noise" No, I'm trying really hard to express the idea that the question of whether there is/is not too much noise is a completely meaningless, wrongheaded question to ask. If you did the analysis on a computer game that sucked, maybe you find that GPU related code was correlated? Does that mean less GPU code = less suckage? You didn't respond to the THERAC-25 comment, which may have been the most targeted one to what I'm trying to communicate. What do you think is "proven wrong"?