Aspies For Freedom

http://content.nejm.org/cgi/content/full...?query=TOC

(Note: really technical paper)

Association between Microdeletion and Microduplication at 16p11.2 and Autism

Lauren A. Weiss, Ph.D., Yiping Shen, Ph.D., Joshua M. Korn, B.S., Dan E. Arking, Ph.D., David T. Miller, M.D., Ph.D., Ragnheidur Fossdal, B.Sc., Evald Saemundsen, B.A., Hreinn Stefansson, Ph.D., Manuel A.R. Ferreira, Ph.D., Todd Green, B.S., Orah S. Platt, M.D., Douglas M. Ruderfer, M.S., Christopher A. Walsh, M.D., Ph.D., David Altshuler, M.D., Ph.D., Aravinda Chakravarti, Ph.D., Rudolph E. Tanzi, Ph.D., Kari Stefansson, M.D., Ph.D., Susan L. Santangelo, Sc.D., James F. Gusella, Ph.D., Pamela Sklar, M.D., Ph.D., Bai-Lin Wu, M.Med., Ph.D., Mark J. Daly, Ph.D., for the Autism Consortium

ABSTRACT

Background Autism spectrum disorder is a heritable developmental disorder in which chromosomal abnormalities are thought to play a role.

Methods As a first component of a genomewide association study of families from the Autism Genetic Resource Exchange (AGRE), we used two novel algorithms to search for recurrent copy-number variations in genotype data from 751 multiplex families with autism. Specific recurrent de novo events were further evaluated in clinical-testing data from Children's Hospital Boston and in a large population study in Iceland.

Results Among the AGRE families, we observed five instances of a de novo deletion of 593 kb on chromosome 16p11.2. Using comparative genomic hybridization, we observed the identical deletion in 5 of 512 children referred to Children's Hospital Boston for developmental delay, mental retardation, or suspected autism spectrum disorder, as well as in 3 of 299 persons with autism in an Icelandic population; the deletion was also carried by 2 of 18,834 unscreened Icelandic control subjects. The reciprocal duplication of this region occurred in 7 affected persons in AGRE families and 4 of the 512 children from Children's Hospital Boston. The duplication also appeared to be a high-penetrance risk factor.

Conclusions We have identified a novel, recurrent microdeletion and a reciprocal microduplication that carry substantial susceptibility to autism and appear to account for approximately 1% of cases. We did not identify other regions with similar aggregations of large de novo mutations.

Just had a quick look through...

Let me know if I'm misreading this, but the way I've interpreted what they're saying is that the microdeletion appears in only 1% of autism cases, and also appears in some non-autistic people. Isn't this a bit too low a result to draw any conclusions from? Especially when the 1% only constitutes 7 people in total.

The only way to make this result meaningful would be to study a larger number of people, and compare it to an equally large study of neurotypical people. The reason for including neurotypical people in the study would be to establish that the 1% wasn't found in everyone, rather than just autistic people.

Of course, I haven't read too many genetic reports, so I could be missing something blindingly obvious in this one...

Okay--here's what I think this paper is saying

1.  Despite the fact that there appears to be a strong genetic component to autism, only in about 10% of people with autism can the cause be attributed to a known genetic variant (e.g. the Prader-Willi regions of duplications on chrom 15).

2.  In light of another study that suggested association of some instances of autism with chromosomal deletions (e.g. in neurexin 1 on chrom 2), the authors decided to do genome scans on multiple families with a significant prevalence of autism to see if they could associate regions of deletions or duplications with autism.  They used a technique that could detect a deletion or duplication if it was > 30,000 bases (DNA building blocks) in length.

3.  In their initial study, they did the chromosome scans on 1441 people with autism, 1420 parents, and 2814 controls (which partly consisted of people with bipolar disorder) (Table 1.).  In other words, they scanned genomes for about 3 times as many presumably non-autistic people as autistic people.  They found that one region of the genome, 16p11.2 was enriched in deletions or duplications in the autistic group over the others.  The actual numbers (in Table 1) are 12 events for the autistic group and 5 events for the non-autistic group. In other words, the autistic group had over twice as many observed changes, relative to the non-autistic group, but the non-autistic group had three times the number of people in it.  So, doing statistical magic, they conclude that there's about a 1 in 10,000 probability that this distribution of instances of the deletion/duplication would happen by chance.

4.  For the 5 children with deletions in 16p11.2 (table 1), NONE of their parents had that deletion, suggesting it was a "de novo" (i.e. new) event.  And NONE of the other parents in the sample had the deletion either. (Some of the bipolor people did, though, given the data in Table 1. These people had not been screened for autism.)

5.  So, the authors wanted to see if the same result would turn up in a different group of people (called "replication.") So they examined genome scan data that had been obtained from patients at Children's Hospital. The subject group of 512 children had been diagnosed with developmental delay, mental retardation or autism.  The control group was 434 other hospital patients who'd had genome scans to check for other possible causes of abnormalities (e.g. seizures).  They found 5 deletions (but includes a pair of identical twins) and 4 duplications in the group of 512 and no duplications or deletions in the group of 434.  This distribution was assigned a statistical probability of happening by chance of about 1 in 7000.

6.  So THEN (and I think this is interesting) they looked at the frequency of occurrence of this deletion in a large group of folks from Iceland--some with various psychiatric or neurologic diagnoses, and they found the 16p11.2 deletion in:
3 out of 299 people with autism
1 out of 648 people with schizophrenia
1 out of 420 people with bipolar disorder
1 out of 203 people with ADHD
1 out of 3000 people with panic disorder, anxiety, depression, or addiction
1 out of 748 people with dyslexia
2 out of 18,834 people with none of the above.

7.  They conclude that maybe about 1% of people with autism might carry this deletion and that it arises spontaneously rather than from inheritance.  In light of the numbers above, they also conclude that this deletion might have an effect on other psychiatric/neurological syndromes.

8.  This study leaves the specific (and presumably mostly heritable) genetic influences on the majority of autism incidences still unexplained.

Hope this helps a bit, Zakkie.

Ah, that makes sense - I missed the significance of the 2 in 18,834 figure...

Thankee muchly. *grins*

Thanks for the superduper NEJM URL. As I messaged you, I wonder, anyone know how much it would cost to do a genetic testing of the chromosome 16? I sure am curious and excited. For some reason, it's great news, kind of real for a change instead of just words.

Thanks.

It means that there are an extremely small number of people who have the genetic anomaly or defect or whatever, and they are completely normal as far as can be determined.

The real significance is why these people are immune from the genetic defects' effects when they have the genetic defects. Not everyone gets the whammy but most due except for these 2 in almost 20,000 or 1 in 10,000 which is a pretty small percentage. 1/100th of 1%?

The other significance is that most of the people who have the genetic defect have something going on, aspergers, ADHD, and so on with aspergers getting most of the action, if you will.

Any great help here or? Is this about right? This is my second post and I am just winging it.