Rare copy number variants highlight candidate genes for autism

14 June 2010

In the UK, around 1 in every 100 people has autism (according to The National Autistic Society). The autism spectrum disorders (ASD) are a group of heterogeneous conditions that cause variable degrees of lifelong developmental disability by affecting the way a person communicates with and relates to others. Although ASD is known to be highly heritable (see previous news), uncovering the genetic basis for this group of conditions has been more tricky (see previous news). A new study published in the journal Nature has investigated rare copy number variants (CNVs) across the genome in individuals with ASD compared with controls.

The study by Pinto et al. report findings from the second phase of the Autism Genome Project (see previous news) [Pinto et al. Nature (2010) doi:10.1038/nature09146]. This involved initial genotyping of 1,275 ASD cases and their parents (from 1,256 families) and 1,981 unaffected controls at 1 million SNPs. Data from 996 ASD cases (from 876 families) and 1,287 controls of European ancestry were chosen for analysis. Rare CNVs were defined as those

1) present at less than 1% frequency in the total sample and;

2) greater than 30kb in size (to maximise the possibility of verification);

3) additional statistical and other quality control measures were also included.

Almost 5,500 CNVs were identified and chosen for analysis. When assessing the impact of rare CNVs between the cases and controls, no difference was observed with the number of CNVs per individual and the estimated size of these CNVs The number of genes affected by the CNVs was found to be higher in cases than controls.

Parent to child transmission of these rare CNVs was also assessed and confirmed 5.7% of cases (50 out of 876) had at least one de novo CNV that could be associated with ASD, not being present in either of the parents. A total of 226 validated de novo (7) and inherited (219) CNVs that affected genes were identified in cases but not controls. Novel candidate genes such as SHANK2 (related to known ASD susceptibility gene SHANK3), SYNGAP1, DLGAP2, and maternally-inherited X-linked deletions at DDX53-PTCHD1 were identified. Further validation of these genes was conducted in an additional 3,677 controls and again, CNVs were not identified.

Comment: Although this, the largest genetic study into ASD, provides additional clues to the underlying genetic basis of ASD, there remains much more research to be done. The candidate genes identified are biologically interesting, although any talk of targets for genetic testing and therapeutic intervention are highly premature. With the continuing use of microarrays for genetic analysis (see previous news) as well as whole-genome approaches (see previous news) identification of rare CNVs involved in disease aetiology is sure to continue.

More from us

Genomics and policy news