Functional correlation of genome‐wide DNA methylation profiles in genetic neurodevelopmental disorders

Abstract

An expanding range of genetic syndromes are characterized by genome‐wide disruptions in DNA methylation profiles referred to as episignatures. Episignatures are distinct, highly sensitive, and specific biomarkers that have recently been applied in clinical diagnosis of genetic syndromes. Episignatures are contained within the broader disorder‐specific genome‐wide DNA methylation changes, which can share significant overlap among different conditions. In this study, we performed functional genomic assessment and comparison of disorder‐specific and overlapping genome‐wide DNA methylation changes related to 65 genetic syndromes with previously described episignatures. We demonstrate evidence of disorder‐specific and recurring genome‐wide differentially methylated probes (DMPs) and regions (DMRs). The overall distribution of DMPs and DMRs across the majority of the neurodevelopmental genetic syndromes analyzed showed substantial enrichment in gene promoters and CpG islands, and under‐representation of the more variable intergenic regions. Analysis showed significant enrichment of the DMPs and DMRs in gene pathways and processes related to neurodevelopment, including neurogenesis, synaptic signaling and synaptic transmission. This study expands beyond the diagnostic utility of DNA methylation episignatures by demonstrating correlation between the function of the mutated genes and the consequent genomic DNA methylation profiles as a key functional element in the molecular etiology of genetic neurodevelopmental disorders.

Episignature Mapping of TRIP12 Provides Functional Insight into Clark–Baraitser Syndrome

“…Only controls without a known episignature or controls unaffected by a neurodevelopmental phenotype were used in this analysis. We generated a list of differentially methylated probes (DMPs) for the TRIP12 cohort and compared it to the DMP lists for 56 other episignature cohorts on the EpiSign™ v3 clinical classifier previously described by Levy et al [ 25 ] ( Figure 6 ). A list of EpiSign™ disorders and their abbreviations are listed in Supplementary Table S3 .…”

“…A list of EpiSign™ disorders and their abbreviations are listed in Supplementary Table S3 . There were 4813 DMPs for TRIP12 , and the range across all cohorts was 279 to 151,848 DMPs [ 25 ]. The highest percent overlap of the TRIP12 DMPs with other EpiSign™ disorders were for BAFopathy (~10%, including ARID1A , ARID1B , SMARCB1 , SMARCA2 , SMARCA4 ), CHARGE (~8%, CHD7 ) and MRD23 (~8%, SETD5 ) ( Figure 6 a).…”

“…Functional annotation and EpiSign™ cohort comparisons were performed according to our previously published methods [ 25 ]. In short, to assess the percentage of differentially methylated probes (DMPs) shared between the TRIP12 episignature and the 56 other neurodevelopmental conditions on the EpiSign™ clinical classifier, heatmaps and circos plots were produced.…”

“…Promoter, 0–1 kb upstream of the transcription start site (TSS); Promoter+, 1–5 kb upstream of the TSS; CDS, coding sequence; Intergenic, all other regions of the genome. The Probes column in both ( a , b ) represents the background distribution determined in the Levy et al study [ 25 ] of all array probes after initial filtering and used as input for DMP analysis. Abbreviations of all array probes after initial filtering and used as input for DMP analysis are listed in Supplementary Table S3 [ 25 ].…”

“…The Probes column in both ( a , b ) represents the background distribution determined in the Levy et al study [ 25 ] of all array probes after initial filtering and used as input for DMP analysis. Abbreviations of all array probes after initial filtering and used as input for DMP analysis are listed in Supplementary Table S3 [ 25 ].…”