Recently, there has been a controversy regarding the possibility of a test for autism during pregnancy. With this information in hand, the parent(s) can make an informed decision on whether to continue the pregnancy, or else prepare for a challenging life as (a) parent(s) of a child with autism.
My problem with this approach is that it presumes that autism most definitely has a genetic marker, and that set of genes alone is responsible. Did they arrive at this conclusion only by analyzing the DNA of people with autism, or did they have a control group of randomly-sampled neurotypical people? Perhaps the genetic marker is really a way to tell whether a person is more susceptible to heavy metal poisioning or gluten/casein allergies, and the cause is partially environmental. The evidence that caused me to form my hypothesis that both are to blame involves a recent study that demonstrates a correlation between rainfall and the number of cases of autism in a given area. The more rainfall there is, the higher the number of cases of autism there are. To me, this correlation (though correlation is not necessarily causation) leads to the possibility that some pollutant (mercury from coal-fired power plants?) is getting into the ground water, then into our children's bodies. Some children can process this pollutant; some cannot.
Parents keep bringing up vaccines as a possible cause for autism. Many "know" that it's the cause. The contradictory evidence comes from a study done on children who have received the MMR combination vaccine. This evidence ignores the fact that there is no safe recommended dose of mercury for children under six months, and many vaccinations are given long before MMR. In the first six months of my daughter's life, she received twenty-six vaccines. Did any of them contain thimersol, a mercury-based preservative, or another, aluminum-based one? I don't know. It's worth studying. I'm betting on the atmospheric mercury from coal-fired power plants combined with moderate to heavy rainfall, but vaccines could still play a role. The way to tell is first demonstrate that cases of autism occur more frequently in places affected by pollution from coal (follow the acid rain path for this data; sulfuric acid raid is directly caused by coal's SO2 pollution, and mercury would follow the same direction), then follow it up by demonstrating that even more children have autism who have had vaccinations in these areas.
The bottom line for me is that no test for autism during pregnancy is valid until someone can show me that the sole cause is genetic.