Family-Based Association Study of the X Chromosome Reveals ASD Genes

Background: The autism spectrum disorders (ASD) are a group of a complex neurodevelopmental disorders with a strong genetic component. The skewed prevalence toward males compared to females and evidence of linkage to the X chromosome in some studies suggests the presence of X-linked susceptibility genes for ASD.

Objectives: To identify X-linked genes in ASD using genome wide association data (GWAS) and a family based association test.

Methods: We have conducted the first analysis of GWAS data on the X chromosome in ASD families. We analyzed 1497 samples from 488 nuclear families from the MIHG Collaborative Autism Project (CAP), genotyped using the Illumina Human 1M beadchip, and a second independent dataset of 3304 samples from 630 ASD families from the Autism Genetic Resource Exchange (AGRE) 550K SNP dataset.  Markers were analyzed using the XAPL, a family-based test of association, in each dataset separately and jointly in the combined data.

Results: Thirteen regions containing 28 genes were significant (p<0.005) in both datasets. Joint analysis of the combined datasets gave 15 significant (p<0.001) regions containing 15 genes.  Taken together, the results most strongly implicate two genes, dystrophin (DMD) and patched domain containing 1 (PTCHD1). These genes replicate consistently in our two family datasets (p<0.005) and are significant in the combined analysis (p<0.001). Eight other genes were also significant in the datasets separately and in the joint analysis.

Conclusions: Family-based association analysis on the X chromosome has not been routine. It requires proper analytic tools and careful interpretation. Our results in two independent datasets provide strong support for involvement of DMD and PTCHD1 in ASD and suggest several other X-chromosome candidates.