I guess what I'd like to know is (and I'm specifically asking you because you seem to be knowledgeable):

1. It sounds like de-novo mutations are the best theory for 70%+ of autism cases per this paper, right?

2. Does anybody track the "base rate" of de-novo mutations / generation in humans? If so, how many generations back do we have data for, and how many regions?

3. Aren't we seeing an increase in autism is developed nations even within poor uneducated families with young parents?