Epigenetics of Autism Spectrum Disorders: A Multi-level Analysis Combining Epi-signature, Age Acceleration, Epigenetic Drift and Rare Epivariations Using Public Datasets

Background: Epigenetics of Autism Spectrum Disorders (ASD) is still an understudied field. The majority of the studies on the topic used an approach based on mere classification of cases and controls. Objective: The present study aimed at providing a multi-level approach in which different types of epigenetic analysis (epigenetic drift, age acceleration) are combined. Methods: We used publicly available datasets from blood (n = 3) and brain tissues (n = 3), separately. Firstly, we evaluated for each dataset and meta-analyzed the differential methylation profile between cases and controls. Secondly, we analyzed age acceleration, epigenetic drift and rare epigenetic variations. Results: We observed a significant epi-signature of ASD in blood but not in brain specimens. We did not observe significant age acceleration in ASD, while epigenetic drift was significantly higher com-pared to controls. We reported the presence of significant rare epigenetic variations in 41 genes, 35 of which were never associated with ASD. Almost all genes were involved in pathways linked to ASD etiopathogenesis (i.e., neuronal development, mitochondrial metabolism, lipid biosynthesis and anti-gen presentation). Conclusion: Our data support the hypothesis of the use of blood epi-signature as a potential tool for diagnosis and prognosis of ASD. The presence of an enhanced epigenetic drift, especially in brain, which is linked to cellular replication, may suggest that alteration in epigenetics may occur at a very early developmental stage (i.e., fetal) when neuronal replication is still high.