Further Introduction of DNA Methylation (DNAm) Arrays in Regular Diagnostics

Abstract

Methylation tests have been used for decades in regular DNA diagnostics focusing primarily on Imprinting disorders or specific loci annotated to specific disease associated gene promotors. With the introduction of DNA methylation (DNAm) arrays such as the Illumina Infinium HumanMethylation450 Beadchip array or the Illumina Infinium Methylation EPIC Beadchip array (850 k), it has become feasible to study the epigenome in a timely and cost-effective way. This has led to new insights regarding the complexity of well-studied imprinting disorders such as the Beckwith Wiedemann syndrome, but it has also led to the introduction of tests such as EpiSign, implemented as a diagnostic test in which a single array experiment can be compared to databases with known episignatures of multiple genetic disorders, especially neurodevelopmental disorders. The successful use of such DNAm tests is rapidly expanding. More and more disorders are found to be associated with discrete episignatures which enables fast and definite diagnoses, as we have shown. The first examples of environmentally induced clinical disorders characterized by discrete aberrant DNAm are discussed underlining the broad application of DNAm testing in regular diagnostics. Here we discuss exemplary findings in our laboratory covering this broad range of applications and we discuss further use of DNAm tests in the near future.

Keywords: DNAm arrays; epigenome; episign; genome diagnostics; genomic imprinting.

FIGURE 1

Hierarchical clustering heatmap showing different methylation profiles between Sotos syndrome patients (red) and control samples (green) using the top 1,000 most differentially methylated probes between these two groups. Methylation shown from 0 to 1. Patients carrying different types of a variant of unknown significance (grey) map either with controls or with cases, confirming the diagnosis Sotos syndrome in those clustered within the true cases. Samples: publicly available data set GSE74432 (62 samples, Choufani et al., 2015) and an Amsterdam UMC cohort of 15 samples.