An Item-Level Approach to Genome-Wide Association of Autism Spectrum Disorders

Background: In recent years, efforts to understand the genetic basis of autism spectrum disorders (ASDs) have moved forward considerably. However, progress has been hampered somewhat by the genetic complexity and heterogeneity of the disorders. A potentially productive strategy for reducing this complexity is to target endophenotypes, simpler biologically-based measures that may involve fewer genes and constitute a more homogenous sample.

Objectives: In this study, we tackle this complexity by selecting individual items from the Autism Diagnostic Interview-Revised, Autism Diagnostic Observation Schedule, and Social Responsiveness Scale for genome-wide association analysis, each of which is treated as a potential endophenotype.

Methods: We conducted a genome-wide association study (GWAS) of 2,165 participants from the Autism Genetic Resource Exchange, and a replication sample of participants from the 1,231 Autism Genome Project, with the aim of identifying genes that correlate with individual items from these three instruments.

Results: A number of genes known to be involved in neurodevelopment were shown to correlate significantly with selected assessment items. These include NELL1 (faints/fits/blackouts), NOS2A (loss of motor skills), and MPN2 (functional play with objects).

Conclusions: These findings may help prioritize study design and directions for future genomic efforts and suggest that certain discrete variables may help to define potential ASD sub-types.