I do believe that the genetics is going to continue at a very, very amazing pace. More tools are available, more diagnoses will be made, and the diagnosis will expand and won't be anything called narrow autism, or rarely will it be. There's a lot of overlap with autism and certain anxiety disorders, with different monogenic disorders, but the thing that's still missing and which is desperately needed to make sense of the genetics, is the phenotype. It's how well people have characterized the behavior. Whether it's purely behavior or whether devices are used, such as wrist watches or monitors for heart rate or eye direction gaze, more devices will be used to chart the behavior. But the behavior is still key if we're going to make any sense of the genetics. Frankly, I think that's still a little bit missing or lacking in the SPARK collection. Where that collection is by telephone interview, and I just hope it turns out to be reproducible in every way.

The autism more and more occupies my mind because it does deal with important issues of development and synapses. Someone called autism a synaptopathy. What do they mean by that? Why does it manifest and is there anything unique about it? The group in North Carolina is trying to phenotype infants before they can speak and that may lead to more rigorous segregation and diagnosis. Rare, but still some reproducibility.

That's been the biggest problem in the clinical trials, why the pharmaceutical companies are running away from these type of behavioral disorders; they don't want to deal with them. Because not a large enough percentage of each cohort is affected by the therapy and they're not going to make a lot of money with the subpopulation. I think that's the thinking anyhow. We have to track them back, refine the diagnostic criteria so not everyone's admitted to the same study. That's so different from anything I'd ever dealt with, but I'd be eager to learn more about it, if I was just beginning in that area. I think Simons Foundation is reaching in that direction with a drug called arbaclofen (which is a GABA-mimetic, it mimics GABA). One of the old saws in autism is that there's an imbalance between excitation and inhibition. It does seem as though children on the spectrum are hyperexcitable in one way or another, so why not try and enhance GABA? There's some promising hints that it may be effective, not just in autism, but in certain hyper-anxiety states.

So I think we're getting close to a new era, a new threshold, where the genetics are going to more and more inform clinical trials. But it's the kind of thing that SPARK is doing, not just collecting more genetic information, that's the easy part. The difficult part is characterizing the phenotype. I look forward in my degenerative years, to stay in touch with that, learning more about it.