Autism in the Blood? Autism Parenting

Trigger Warning: Early Intervention, Possibilities of Eugenics
Leslie, the same person who wrote "How did I know my daughter was autistic?" and possibly "Wholeness and Completion" (unsure, but the daughter's name is the same and the writing has the same extra commas,) also wrote the final article of the December issue, "Autism in the blood?," discussing blood/genetic testing for autism.
For a blood test, the reasoning seems straightforward enough. If we can diagnose earlier, interventions start earlier. Considering what the current interventions tend to look like, that worries me. If the ways that we changed our educational methods for autistic children were ways that worked better for them, instead of making them more convenient for the teachers, parents, and caretakers, being aware that the kid is autistic sooner would be awesome. So this is a mixed bag for me. In an ideal world, this would be really cool. Just as a matter of curiosity, I'd love to know which blood tests currently can and can't tell that I'm autistic. But with the ways that autism is currently handled, I am not comfortable with giving them any information on my genetics that could help them figure this out. Get me a world where knowing you're autistic ASAP is definitively better, not for "acting normal" as most therapies prize, but for navigating a world not designed for you, and we'll talk. In a world where one of the things I am most grateful for is how long it took people to realize that I'm autistic? Not so much.
I still think that the "oh hey, more genes associated with autism means more evidence that autism is genetic" thing is cool, though. The more evidence we have behind genetics for how autism comes about, the better I can smash people who try to tell me that I'm vaccine-injured or something.
The other big worry I have, which wasn't addressed at all despite the fact that the study about it was mentioned, was the issue with prenatally predicting autism. Australian scientists really did develop a genetic autism test using 237 genetic markers that is 70% accurate for, well, white people. (Bayesian inference and an assumed 1% autism prevalence rate gives an actual 2.5% chance that a fetus that tests positive will be autistic. It takes 99% accuracy to make it a 50-50% chance that the kid who tests positive will actually be autistic given a 1% prevalence rate, for reference.) Since the Australian one was designed as a prenatal test, my worry, of course, is selective abortion. The idea of people aborting because the fetus is likely to have a brain wired like mine is terrifying, and it's not something the author is talking about. I envy her innocence, I think.
She's talking about tests and hoping to diagnose toddlers and getting them "closer in relation to their peers by the time they enter school." I'm reading that and wondering: "closer how?" If we're talking about getting whatever forms of communication we can up so that they have similar overall communication abilities (AAC is success here) then I am all in favor. If we're talking about looking normal, I am decidedly not in favor. I don't think we need to abandon all teaching for autistic kids, but I do think we need to be taking a look at what therapies we're using as a higher priority than getting potentially harmful ones to everyone. Take a look at the correlations between ABA and PTSD before we keep calling it the gold standard and getting everyone to cover it?
Increasing social and language skills is great. Make sure you know the cost of the teaching methods, and make sure it's social skills the way they are actually done, not compliance or the way you wish social skills worked. Compliance opens people up for a lifetime of abuse, the way you wish social skills worked isn't going to make them any more friends than the way they think it should work, maybe fewer. Their natural ones will at least work with other autistic people.