Read the most advanced science of autism causes here. Bypass the commenterati and go direct to the science. Don't waste your time at the sites which pretend "no-one" knows what causes (or what sometimes cures) autism.
This is a website relating to the unchallenged theory of autism, IQ and genius, Personality and Individual Differences 14:459-482 (1993) by Robin P Clarke (the antiinnatia theory). An update review paper is being prepared for publication. Meanwhile you can download the original 1993 publication (presentationally revised) here, and the original 1993 publication (author's reprint) here . (the journal site version is here: http://dx.doi.org/10.1016/0191-8869(93)90316-U, but without added charts of social class and you may have to pay Elsevier $31.)

Copy number variations are at best only marginal to autism causation (which is in accordance with antiinnatia theory)

The main importance of this study in Nature lies in what it has not found. It has involved a huge amount of time and effort expended on seeking for pathological genetics underlying this so-called 'disease' or 'disorder' (ASD), and yet such pathology as it has found can account for only about three percent of all the autism-related cases under study. That doesn't surprise me at all. It is entirely in accordance with the still-unchallenged antiinnatia theory. Therein, I agreed with those who reckoned that autism was a polygenetic condition just as IQ genetic variance was polygenetic. The vast majority of genetic variations genuinely associated with autism would be the very same "normal", non-pathological, variations that contribute to IQ differences -- for the reasons the theory explained.

These authors are scraping the bottom of the wrong barrel, looking for the wrong things, and as a result they find the wrong things which then lead them away from the main story of autism causation. So-called ASD is a very rough and broad empirical diagnosis which is liable to include quite a number of cases who do not really belong in the true autism broad syndrome but just have some genetic twirks that make them behave rather similarly (like a bee misleadingly resembles a wasp). All the vast media/discussion hype then goes on to assume, falsely, that the study results really are about autism per se rather than such false trails. Their study could have had more hope if it had used narrowly-defined core autism as its cases.

But the people there are working to a pre-set paradigm, within what we might call the disease model. This presumes that something has "gone wrong" with these individuals and seeks out that "wrong". Furthermore they assume it is genetic! It may be significant that their 90% genetic stat is supported only by a reference that is 15 years old. In my 1993-published theory I indicated conditions in which autism would change from being mainly genetic to being mainly environmental. And exactly that has now happened in the last 2 decades.

Their pre-set paradigm is to find what has "gone wrong" and then find a drug to "put it right" and voila! a profitable patent emerges (they hope).

Huge publicity is being given to these marginal fringe findings, while the real central ressearch is all but completely hidden by this hype. For instance the Autism Research Institute's evidence that DMSA chelation to remove mercury can cure about 70% of cases, as documented by online videos of the cured children. But that doesn't generate profits for big corporations, or jobs for the 170 researchers listed here. So it is ignored, despised and even persecuted just as was Semmelweiss's immense discovery, inter many alias. Or they could have just spent a little time more enlighteningly reading my 1993 paper. But then Wegener's continental drift had to wait through 50 years of professional derision so perhaps I'm calling time a little prematurely here.

There's reason to suspect there's already a fairly good genetic test for autism, namely whether both parents have Mensa-level IQs. An even better test is how many non-gamma-2 dental amalgams the parents have, and how little outdoor air. But why follow the facts when you can follow the crowd that's following the money instead?

Edit: There's now appeared a much more detailed (but far longer) critique by Mark Blaxill. My only reservation about it is that the claim in the title that it "proves that inherited genes don't cause autism" is absolutely unfounded.

P.S.: Prof Alexandre Raymond explains that the genome is more complicated than geneticists have assumed, and so we should be wary of imagining we have more knowledge and understanding than is actually the case.