Common Risk Variants Identified in Autism Spectrum Disorder

Background: Autism Spectrum Disorder (ASD) is a complex disorder, and the aetiology is largely unknown. The heritability is as high as 80% according to some estimates and common genetic variation is estimated to explain half of the genetic risk. Recently, iPSYCH and Broad/MGH conducted the largest GWAS of ASD to date comprised of 13,076 cases and 22,664 controls from Denmark and meta-analyzed with 10,610 samples from the Psychiatric Genomics Consortium (PGC).

Objectives: To identify genetic loci associated with autism, derive insight into the underlying biology, and to dissect the polygenic architecture of autism and its clinical subtypes.

Methods: The iPSYCH sample is a population sample comprised of cases with one or more psychiatric diagnoses identified in the Danish Psychiatric Central Research Register and a control sample consisting of a 2% random sample from the corresponding birth cohort. The Danish Neonatal Screening Biobank supplied archived dried blood spot samples and DNA was extracted, whole-genome amplified and genotyped on the PsychChip, a customized HumanCoreExome chip. Quality control, imputation and principal component analyses were conducted using the Ricopili pipeline of PGC, as was the meta-analysis with the summary statistics from the PGC. ASD signals were followed up in a European sample consisting of 2,119 cases and 142,379 mainly from Iceland. Top hits were integrated with data of the 3d structure of the chromatin using the method Hi-C. SNP heritability and genetic correlations was estimated using LD score regression and GCTA, and polygenic risk scores were generated and analyzed in an in-house pipeline. Phenotypes with significantly overlapping genetic architectures (schizophrenia, major depression, and educational attainment) were employed to identify additional putative loci using the recent MTAG method.

Results: We identify 3 genome-wide significant loci in the iPSYCH-PGC sample and additional 2 when including the follow-up sample. Seven extra putative ASD loci were highlighted in the MTAG analysis leveraging data from the related phenotypes of schizophrenia, major depression, and educational attainment. Novel genetic correlations were estimated for eg. ADHD, rG=0.360 (SE=0.051) and major depression, rG=0.412 (SE=0.039). Indications of differences between diagnostic subtyped in genetic architecture was found in both estimates of SNP heritability (ranging from 0.03 for other pervasive developmental disorders to 0.10 for Asperger’s syndrome, P = 0.001) and to a lesser extent in genetic correlations between hierarchical subtypes (rG ranging from 0.71 to 1). The differences were confirmed in a novel polygenic risk score (PRS) analysis comparing levels of PRS for various related phenotypes across subtypes.

Conclusions: We identified novel risk loci for ASD highlighting biological insights, particularly relating to neuronal function and corticogenesis. Interrogating the polygenic architecture through different means, we found heterogeneity across ASD subtypes. Overall, these results indicate that GWAS performed at scale will be much more productive in the near term in ASD.