The Impact of Rare Genomic Variants in Autism Spectrum Disorders – Evidence for Converging Pathways

Background: The autism spectrum disorders (ASDs) are a group of early-onset conditions that affect 1% of the general population. ASDs are characterized by impairments in reciprocal social interaction and communication, and the presence of restricted and repetitive behaviours. Individuals with an ASD vary greatly in cognitive development, which can range from above average to intellectual disability. ASDs have a substantial genetic aetiology, but the underlying genetic determinants are still largely unknown. Recent association studies provide only weak evidence for common allele risk effects. In contrast, the role of rare variants in ASDs is being increasingly recognized with recent surveys for copy number variation (CNVs) and emerging sequencing efforts.

Objectives: Evaluation of the genome-wide characteristics of rare variants in ~2,000 ASD trios.

Methods: In Phase II of the Autism Genome Project (AGP), we recently reported on the genome-wide characteristics of rare (<1% frequency) CNV in ~1,000 ASD trios using the Illumina 1M SNP platform (Pinto et al. 2010). Here we report on an additional independent collection of ~1,000 ASD trios screened using a similar analysis pipeline.

Results: The integration of the two CNV sets from up to 2,000 ASD trios highlights several aspects: i) an extremely heterogeneous genomic architecture of ASD; ii) rare de novo CNVs are present in at least 4-5% of subjects with idiopathic ASD, iii)  rare inherited CNVs are seen to disrupt loci previously implicated by de novo CNVs; iv) incomplete penetrance, failure to segregate and significant phenotypic heterogeneity associated with specific CNVs; v) while separate ascertainment delivers different representations of rare CNVs, they can be grouped in a few common functional pathways.

Conclusions: Our study further expands on the extremely heterogeneous nature of ASDs, and reveals additional new genetic and functional targets in ASD that point towards connected pathways in brain function and development. We expect that integration of the two CNV sets, coupled with deeper phenotyping and additional genomic analysis such as sequencing, will further aid in establishing genotype-phenotype correlations.