Integrative genomic and epigenomic analysis of autism spectrum disorder

Autism Spectrum Disorder (ASD) is a complex neurodevelopmental disorder. ASD affects 1-1.5% of individuals and its risk is influenced by a combination of genetic and environmental factors. Both genetic and environmental factors can alter epigenetic regulation. DNA methylation (DNAm), an epigenetic modification, plays a pivotal role in neuronal development, proper brain functioning, and thereby mental health. The aim of this study was to identify DNAm signatures of ASD at the time of birth and specifically investigate the impact of common genetic ASD risk variants on the newborn epigenome. DNAm levels were quantified at >850,000 loci in the human genome in 2,065 neonatal dried blood spots from the Danish iPSYCH 2012 cohort, for whom genotyping data was also available. Three statistical methodologies were applied to study the role of DNAm in ASD: 1) epigenome-wide association study (EWAS) of ASD to identify loci with differential DNAm between 843 cases and 688 controls and meta-EWAS with additional ASD cohorts; 2) EWAS of ASD polygenic risk score (PRS) to identify DNAm changes associated with polygenic burden of the disorder; and 3) methylation Quantiative Trait Loci (mQTL) analysis to determine if common ASD risk variants impact epigenetic regulation. ASD diagnosis was associated with DNAm changes at 9 loci (p-value < 10-5), including MYT1L, MAD1L1 and PIWIL1 genes. EWAS of ASD PRS replicated previous findings where differential DNAm at birth was associated with elevated ASD polygenic burden. Further mQTL analyses confirmed the ASD risk SNPs to impact DNAm levels across the genome. Moreover, employing the mQTL approach improved mapping of impact of ASD risk SNPs on epigenomic regulation and identified novel loci associated with the disorder. This study highlights the importance and advantage of combining genomic and epigenomic approaches in studying ASD to provide deeper understanding of its molecular pathophysiology.