The genetics of autism

I was lucky enough to attend several scientific meetings this year, thanks to the Advocate Partnership program of the Genetic Alliance, and attended the session where one of the scientists involved presented the information being discussed in many media outlets yesterday and today about the genetics of autism. As you read, listen to, or watch the news presentations, note that the "large, non-inherited chromosomal deletion" being discussed is extremely rare and in no way accounts for all cases of autism. Also, the parents of affected children were themselves unaffected.

By thw way, those who have been following the controversy about a suggested connection between immunization vaccines--specifically, the mercury in these vaccines--and the recent increases in number of children with autism probably have already learned that studies of large numbers of children have not seen any link between the two.

One complicating issue recently in autism is that many diagnoses of mental retardation, slow childhood development, and other concerns have been added to the new definitions of autism. As a result of more clear definitions of autism--I can write about these more when I return to my other computer--genomic studies are now more helpful.

In a genomic study, the full genetic complement of an individual is compared with the results from those with no indication of autism. Researchers look at the printouts of SNP (single nucleotide polymorphism) analysis for "signals". These are places where the peaks on the graph exceed a predetermined value.

As with many presentations of scientific data, SNP patterns are incredibly complex. It took me a while to understand what presenters were trying to show us on these slides that looked like the hair on a person just getting out of bed. I also found that one set of researchers often has a different set of interpretative standards from another. And, finally, there's also a strong tendency to assume that because 5 samples have the same "unique" feature and all 5 samples came from people with condition A that the feature is the cause of the condition. Like storks and babies.

Well, back to genetics and autism. The best article I found explaining the newest discovery of a possible genetic link to autism comes from Science Daily and is a report based on the New England Journal of Medicine publication. The article begins as follows:

A multi-institutional study involving Massachusetts General Hospital (MGH) researchers has identified a chromosomal abnormality that appears to increase susceptibility to autism. The investigators -- most of whom are associated with the Boston-based Autism Consortium -- report that a segment of chromosome 16 is either missing or duplicated in about 1 percent of individuals with autism or related disorders, a frequency that is comparable to other genetic syndromes associated with the disorder.

The article goes on to explain the work:

Population studies indicate that up to 90 percent of cases of autism and what are referred to as autism spectrum disorders have some genetic component, but only 10 percent of cases can be attributed to known genetic and chromosomal syndromes. Since several of those conditions involve deletions or duplications of chromosomal segments -- including an inherited deletion of a region of chromosome 15 -- the investigators conducted a complete genome scan of samples from the Autism Genome Research Exchange, which contains DNA from families in which at least one child has autism or a related disorder.

The scan of more than 1,400 affected individuals and a similar number of their unaffected parents revealed that an identical region of chromosome 16 was deleted in 5 individuals with an autism spectrum disorder but not in any of the parents, implying that the deletion had occurred spontaneously and was not inherited. To confirm this observation, clinical testing data from almost 1,000 patients from Children's Hospital Boston -- about half of whom had been diagnosed with autism or a related developmental delay -- was evaluated. Among those with a developmental disorder, 5 children had the same deletion, and in another 4 the chromosome segment was duplicated. Again, no abnormalities were seen in DNA from children without autism or developmental delay.

If you're interested in this topic, I suggest you look at the webpage for the Science Daily article. In addition to the article, many links are provided for investigating autism, SNPs, and other aspects of the story.

Marie Godfrey, PhD





| mgodfrey39's blog