Deciphering the Non-Coding Genome in Autism Spectrum Disorders (ASD): A Study of De Novo and Rare Inherited Variants through Targeted Sequencing in Regulatory Regions

ASD (Autism Spectrum Disorders) are NDDs (Neurodevelopmental Disorders) with complex etiology including multiple genetic and environmental factors. Non-coding mutations contribute to the multifactorial etiology of ASD by influencing gene activity through various regulatory mechanisms. Advances in genomic technologies, such as whole-genome sequencing (WGS) and chromatin interaction studies, have highlighted the role of non-coding regions in ASD genetics. Identifying these non-coding variants enhances our understanding of the underlying complex genetic landscape ASD.

This study aims to analyze the impact of non-coding mutations within regulatory regions in Autism Spectrum Disorder (ASD). The research builds upon a cohort of 360 Spanish ASD trios, from which 200 trios were selected after excluding cases with known copy number variants (CNVs) and whole-exome sequencing (WES) mutations. The selection process intentionally enhanced the sample for undiscovered non-coding risk variants by excluding cases with de novo loss-of-function mutations or large de novo CNVs. To identify regulatory regions of interest, the study employed targeting sequencing of a selection of candidate cis-regulatory elements (cCREs) from ENCODE v2. De novo variation and rare inherited variation were studied using different bioinformatic pipelines and their impact on regulatory activities was assessed using a deep-learning approach (Sei framework). Additional analysis including candidate gene elucidation using ATAC-seq and PLAC-seq data in neuronal cells, variant prioritization, protein-protein interaction (PPI) network, Transcription Factor (TF) enrichment, presence in topologically associated domains (TADs) were also carried out. Sex bias in regulatory variation within ASD was also explored in our analysis.

We discovered that 28% of de novo variants and 25% of inherited variants with high regulatory potential were found in patients with negative results from whole-exome sequencing (WES) and microarray analyses, as assessed by Sei. By integrating PLAC-Seq data, we functionally annotated approximately 80% of de novo variants and 85% of inherited variants. While resources like ENCODE provide valuable insights into genomic regulatory elements, it is crucial to be cautious when prioritizing specific regulatory elements based on initial hypotheses regarding their impact on gene regulation: many sequence classes associated with ASD in this study did not show significant enrichment in any particular cCRE signature. Notably, the most important observation in this study is the implication of a global dysregulation of CTCF suggesting a potential mechanistic impact on the chromatin architecture. In addition,We have found that the most high-impact regulatory variants—whether de novo or inherited—are linked to genes not previously associated with ASD. Nevertheless, gene ontology (GO) enrichments indicate that both coding and non-coding variations likely interact within already characterized ASD pathways.