Analysis Methods for Diagnosing Rare Neurodevelopmental Diseases with Episignatures: A Systematic Review of the Literature

Alegret-García, Albert; Cáceres, Alejandro; Sevilla-Porras, Marta; Pérez-Jurado, Luís A.; González, Juan R.

doi:10.3390/biomedicines13123043

Open AccessSystematic Review

Analysis Methods for Diagnosing Rare Neurodevelopmental Diseases with Episignatures: A Systematic Review of the Literature

by

Albert Alegret-García

^1,2

,

Alejandro Cáceres

^1,2

,

Marta Sevilla-Porras

^3,4

,

Luís A. Pérez-Jurado

^3,4,5,†

and

Juan R. González

^1,2,*,†

¹

Barcelona Institute for Global Health (ISGlobal), 08003 Barcelona, Spain

²

Centre for Biomedical Network Research on Epidemiology and Public Health (CIBERESP), 08003 Barcelona, Spain

³

Genetics Unit, Department of Medicine and Life Sciences, Universitat Pompeu Fabra, 08003 Barcelona, Spain

⁴

Centre for Biomedical Network Research on Rare Diseases (CIBERER), 08003 Barcelona, Spain

⁵

Genetics Service, Hospital del Mar and Institut Hospital del Mar d’Investigacions Mèdiques (IMIM), 08003 Barcelona, Spain

^*

Author to whom correspondence should be addressed.

^†

Co-supervised this work.

Biomedicines 2025, 13(12), 3043; https://doi.org/10.3390/biomedicines13123043

Submission received: 26 October 2025 / Revised: 27 November 2025 / Accepted: 5 December 2025 / Published: 11 December 2025

(This article belongs to the Special Issue Unraveling the Epigenetic Blueprint: Exploring DNA Methylation and Its Integration with Omics in Health and Disease)

Download

Browse Figures

Review Reports Versions Notes

Abstract

Background: Rare diseases (RDs) and neurodevelopmental disorders (NDDs) remain under-researched due to their low prevalence, leaving significant gaps in diagnostic strategies. Beyond next-generation sequencing, epigenetic profiling and particularly episignatures have emerged as a promising complementary diagnostic tool and for reclassifying variants of uncertain significance (VUS). However, clinical implementation remains limited, hindered by non-standardized methodologies and restricted data sharing that impede the development of sufficiently large datasets for robust episignature development. Methods: We conducted a systematic literature review following PRISMA 2020 guidelines to identify all studies reporting episignatures published between 2014 and 2025. The review summarizes methodological approaches used for episignature detection and implementation, as well as reports of epimutations. Results: A total of 108 studies met the inclusion criteria. All but three employed Illumina methylation arrays, mostly 450 K and EPIC versions for patient sample analysis. Three main methodological phases were identified: data quality control, episignature detection, and classification model training. Despite methodological variability across these stages, most studies demonstrated high predictive capabilities, often relying on methodologies developed by a small number of leading groups. Conclusions: Epigenetic screening has significant potential to improve diagnostic yield in RDs and NDDs. Continued methodological refinement and collaborative standardization efforts will be crucial for its successful integration into clinical practice. Nevertheless, key challenges persist, including the need for secure and ethical data-sharing frameworks, external validation, and methodological standardization.

Keywords:

rare disease; neurodevelopmental disorders; DNA methylation; episignatures; DMPs; DMRs; VUS; machine learning

1. Introduction

A rare disease (RD) is by definition a disorder that affects less than 1 in 2000 individuals (according to the European Union [1]), or fewer than 200,000 people in total (according to the USA [2]). Despite their individual rarity, over 7000 distinct RDs have been described [1,3], collectively affecting about 5% of the population [4].

Neurodevelopmental disorders (NDDs), many of which also qualify as RDs, encompass a diverse group of conditions characterized by cognitive, behavioral, motor, and adaptive impairments [5]. These disorders arise due to abnormal genetic and/or environmental influences that affect neurodevelopment [6].

Diagnosing NDDs and other RDs is inherently challenging because of overlapping clinical phenotypes, shared biomarkers or genetic variants of uncertain significance (VUS), and the high cost of advanced diagnostic techniques. As a result, it can take years of testing and clinical consultations before reaching a reliable diagnosis, if one is achieved at all [7,8]. This prolonged uncertainty, often referred to as the “diagnostic odyssey,” imposes significant emotional and psychological burdens on patients and their families. Despite extensive efforts, it is estimated that up to 50% of patients remain undiagnosed [9], highlighting the urgent need for innovative diagnostic strategies [10].

When aberrant phenotypes emerge, such as congenital anomalies or developmental delay, there is a need for an early etiological diagnosis, so the focus typically shifts toward identifying the potential causal chromosomal or genetic mutation. Modern diagnostic pipelines rely on genetic and/or genomic testing, either targeting specific genes or regions (if a known syndrome or condition is suspected) or using whole-exome sequencing (WES), which is commonly employed to search for pathogenic variants affecting the coding part of the genome, sometimes supplemented by validation studies in the family and/or gene expression profiling [11]. As whole-genome sequencing (WGS) becomes more accessible and high-fidelity long-read sequencing technologies gain traction, an overall improvement in diagnostic rates is anticipated. These technologies enable more accurate detection of structural variants and genomic rearrangements, often missed by traditional short-read sequencing platforms [12].

However, in a substantial number of cases, these latest procedures still yield inconclusive results. Either no pathogenic variant is detected, the identified variants remain unannotated and unlinked to any known syndrome and deemed VUS [13], or no second hit is found in autosomal recessive disorders even though clinical suspicion exists. As a result, it is not possible to confidently correlate the observed phenotype with a definitive genetic cause. Further progress can be made if the VUS is found to be fully associated with the condition through functional assays such as protein modeling [14] and gene expression analysis. Consequently, while functional genomics-informed diagnostics have the potential to transform the rare disease diagnostic landscape, additional validation on individual approaches is still required. This common testing battery is illustrated in Figure 1.

One of the most promising avenues in functional genomic diagnostics of RDs is epigenetic analysis based on the detection of so-called episignatures (condition-specific position-based methylation differences in a group of cases compared with a reference cohort) and epimutations (sample-specific region-based methylation difference in a sample with respect to a reference cohort). DNA methylation is the addition of a methyl group to the 5-carbon of the cytosine ring, typically in the context of CpG dinucleotides, resulting in 5-methylcytosine [15,16]. Methylation is an absolute measure of the percentage of DNA sequences with a methyl group in a CpG site, making it more robust than other functional quantifications (transcripts or proteins) thanks to its comparability among all marks and samples without having to standardize across datasets. Methylation regulates gene expression, and if observed at an unexpected region (epimutation) may constitute a cause for an undiagnosed NDD. However, methylation at particular sites may also result from an underlying mutation, genetic alteration, or uncommon fetal exposure that might be the root of the RD, leading to global alterations in the epigenome [5,17,18]. These changes give rise to episignatures that enable the identification of differential epigenetic patterns between affected individuals and controls. These patterns have been associated with a wide array of RDs and NDDs [17,19,20] and can provide critical insights when conventional genetic testing falls short or is not easily accessible. Although DNA methylation can vary among different cells and tissue of an individual, the analysis of whole blood and/or other accessible tissue has proven useful for defining clinically relevant episignatures and epimutations.

Despite the significant potential of episignatures and epimutations to enhance RD diagnostics, several challenges still remain. These include: (1) assessing their true diagnostic value, (2) understanding their biological origins, and (3) standardizing methodologies for their identification and analysis. Additionally, the sensitive nature of genetic data presents privacy and security concerns, especially in clinical contexts involving undiagnosed cases, despite existing efforts within the scientific community to promote secure data sharing through data anonymization protocols and the creation of controlled repositories [21,22].

Experimental design for episignature detection in RDs has been established and well documented [23]. By contrast, analysis workflows are diverse and remain to be standardized. We therefore aimed to provide a comprehensive framework for episignature-based diagnosis of NDDs based on the recent literature to guide clinical practitioners and researchers in the field by addressing the following most current and salient issues. (1) How to streamline the development of new episignatures using established methodologies from the current literature? (2) How to process and analyze patient DNA methylation data to test episignatures using available models? (3) How to interpret episignature-based results with clinical relevance? And (4) what gaps still exist in the field, and how can these be overcome?

In this systematic review of the literature, we identify existing gaps in the field and discuss how addressing them will help in the implementation of epigenetic diagnostics for NDDs in clinical practice.

2. Materials and Methods

To describe current methods in episignature detection, we first reviewed studies that conducted epigenome-wide association studies (eWASs) in RDs and NDDs following the PRISMA guidelines [24]. Figure 2 illustrates the PRISMA flowchart with the included and excluded articles based on proposed filters. Specifically, we aimed to identify studies that measured the epigenetic profiles of patients diagnosed with RDs or NDDs, using blood samples, and that conducted differential DNA methylation analysis. Additionally, we included studies that developed classification algorithms based on episignatures and those that also studied epimutations. The article was registered on OSF and the registration code is fs9aj.

To ensure broad coverage, we included all reported types of DNA methylation profiling and methodologies. Only studies published between 2014 and 2025 were considered.

The articles for this review were sourced through the following strategies.

PubMed general search: Using the keywords “episignatures” OR “epimutation” filtered for publications between 2014 and 2025. Broad search parameters were applied due to the lack of standardized terminology in the field. The query was not restricted to RDs or NDDs to prevent missing out relevant articles. We retrieved 667 results. Titles and abstracts were screened to exclude case reports, general reviews without novel algorithmic or diagnostic contributions, and studies not directly related to NDDs or rare diseases. Filtering was guided by specific keywords such as “DNA methylation,” “signature,” and “profile.” Only 94 articles passed the proposed filters. In addition, we tracked new releases throughout 2025 using the same keywords in PubMed and added 7 extra studies.

Additional articles identified from the literature: 59 studies were retrieved from the literature focused on NDDs and RDs, as well as key large-scale studies citing references [25,26,27,28,29,30,31,32,33,34].

Overall, 160 articles were retained for full-text assessment. Particular attention was given to the Methods and Results sections to determine compliance with the following criteria: new episignatures or refinement of multiple episignatures with a new pipeline; blood-based to promote inter-study comparison; and exclusion of case studies and full-text availability. Ultimately, 108 studies that developed models or pipelines for one or more syndromes were included in the review.

3. Results

As of 2025, episignatures have been developed for well over 100 distinct genetic conditions (according to EpiSign v5 catalogue, Table S1). In total, 106 different disorders were identified, and all studies retained from the literature are listed in Table 1, indicating for which disease an episignature was derived, the causal gene (if any), and the array type. The different paths for episignature testing are represented in Figure 3. Each step will be further explained in the sections below.

3.1. Testing Episignatures

3.1.1. Testing Episignatures: External Resources

For users who want to test a given patient with the available episignatures, but do not have the bioinformatics background to apply the pipelines, multiple external resources are currently available for episignature testing: EpigenCentral, EpiSign, NSBEpi and MethaDory. The first two options offer comprehensive pipelines that process raw methylation array data (IDAT files) and return prediction scores indicating the likelihood of a sample being associated with a specific epigenetic disorder. The latter require some data processing.

EpigenCentral [25] is a publicly available web-based platform that enables users to analyze DNA methylation data and assess its association with episignature obtained by the developers. The platform currently supports autism spectrum disorder (CHD8/16p11.2 del), CHARGE syndrome (CHD7), Down syndrome (21 trisomy), Dup7 syndrome (7q11.23 dup), Kabuki syndrome (KMT2D), Nicolaides–Baraitser syndrome (SMARCA2), Sotos syndrome (NSD1), Weaver syndrome (EZH2), Williams–Beuren syndrome (7q11.23 del) (Table S2). All steps are automatically performed, but are limited to available methods. After processing, the platform computes probabilistic scores indicating the likelihood of a sample belonging to a specific disease category using random forests, penalized logistic regression, or support vector machines.

For a complete overview of its functionalities, users are referred to the user guide found in their web application [137].

To our knowledge, only Awamleh et al. [138] has used the tool to generate an episignature anew from a Weaver syndrome cohort, yielding good predictions and case–control clustering when plotted using unsupervised clustering techniques. Further, the article thoroughly details which steps were followed in the platform to detect the episignature, explaining all needed files. Nevertheless, no external study has published results derived from the platform.

EpiSign [139] is a commercially available service that offers clinical-grade episignature diagnostics based on blood-derived DNA methylation profiling. Upon submission of a blood sample, the laboratory performs a methylation array analysis and applies their proprietary classification algorithm. Different pricing tiers are available depending on the scope of analysis requested (e.g., targeted panels vs. whole-platform screening). EpiSign is the main author in the episignatures field and possesses the most comprehensive algorithm developed so far, with more than 100 identifiable conditions on their version 5 platform (according to their web information, accessed on 1 October 2025). In Table S1, a list is included of the offered episignatures.

The tool has been tested by the authors in multiple published studies [20,26,27,36,40], in which they provide excellent results of their predictors in most cases, including the resolution of cases with VUS. Nevertheless, since most of their data are not open-access due to ethical and corporative restrictions in genetic data sharing (as stated by the authors), external validation has not been extensively conducted.

Husson et al. [140] tested some of the episignatures from a past version of the EpiSign classifier [26], and other independent studies conducted in the same disease [49,61,126,133,141]: ATRX, ASD18 (CHD8), Cornelia de Lange syndrome (NIPLB), CHARGE (CHD7), Kabuki (KMT2D), Claes–Jensen (KDM5C), Rubinstein-Taybi (CREBBP), Sotos (NSD1), Tatton-Brown–Rahman (DNMT3A), and Wiedemann–Steiner (KMT2A). The authors describe three types of episignatures: robust, which allow a clear identification of cases and the top PC explains most of the detected variance (the case for ATRX, Sotos, TBRS and Kabuki derived episignatures); unstable, which offers high sensitivity though intermediate profiles, thus not allowing perfect distinction of all cases or certain VUS (the case for Cornelia de Lange, CHARGE, ASD18, Claes–Jensen- and Wiedemann–Steiner-derived episignatures); and weak, which did not work as expected, and the authors advise against their usage (the case for Rubinstein–Taybi and a second AUTS18-derived episignature).

The NSBEpi (nanopore sequencing-based episignature detection) pipeline was developed to test long-read sequencing episignature discovery [142]. The authors demonstrated the feasibility of training classification models using only one positive case against other disease cases and controls averaged by their median methylation value at episignature-defined sites. Their classification models are available on their GitHub repository (JorisVermeeschLab/NSBEpi, version: 21 January 2025), trained over the episignatures developed by Aref-Eshghi et al. [26]. Although developed for sequencing, the tool can also be adapted for array data [34].

MethaDory is a recent open-access shiny application (in beta testing) which incorporates publicly available episignatures and classification models even for those syndromes without open-access data available [34] (GitHub repository: f-ferraro/MethaDory, version: 19 October 2025). The authors designed a method to create synthetic profiles with which they were able to train classification models based on past episignatures using the median-beta approach reported by Geysens et al. [142]. Since the software is in its first stages, not all data are readily available, but the authors offer their models upon request (Table S3).

Although full validation of some existing episignatures is still pending, results are promising, with many allowing highly specific and sensitive identification of cases, even when VUS are involved.

3.1.2. Testing Episignatures: Public Episignatures

In addition to the previously mentioned tools, an episignature can be tested on a set of samples using two common strategies in methylation analysis:

Unsupervised clustering: Techniques such as principal component analysis (PCA), multidimensional scaling (MDS), or hierarchical clustering can be applied to methylation data at disease-relevant CpG sites, using the sites defining the episignature. This step is performed in all studies developing or testing episignatures to assess how well cases and controls are separated. While this heuristic approach does not yield a direct probability score, it can highlight whether an unknown case clusters with known affected individuals, thereby supporting or guiding a diagnostic hypothesis. Nevertheless, it requires confirmed cases of the suspected disease for comparison.

Machine learning-based prediction: When a trained classification model is available, individual samples can be scored to estimate the likelihood of belonging to a specific disease class. This approach does not require comparison with other cases or controls. Currently, EpigenCentral and NSBEpi offer its models open-access (and MethaDory but upon request), so this technique is limited to these tools. This approach is exemplified in the study conducted by Hildonen et al. [143], who used the CpGs describing the KMT2D episignature [26] (for Kabuki syndrome) to test a set of undiagnosed carriers of KMT2D patients, demonstrating the utility of robust episignatures for diagnostic purposes. Lastly, Niceta et al. [144] tested another population of putative Kabuki patients with KMT2D VUS and mosaicism using three different public episignatures [26,61,75] after training independent classification models with confirmed Kabuki patients, which proved fruitful. These studies highlight the potential of episignatures for elusive cases and variant reclassification.

3.2. Episignature Development

If data are available (either private data or open-access from databases such as Gene Expression Omnibus (GEO), ArrayExpress or the European Genome-Phenome archive (EGA)), researchers and clinicians can derive their own episignatures. The process consists of three main computational steps: raw data preprocessing (a general methylation processing step), differential methylation analysis (a general methylation analysis step with specific considerations for RDs and NDDs), and model training (specific to NDDs). While the third step is not essential for episignature identification, it allows the assignment of confidence values to individual samples based on the identified significant sites.

A complete description of the pipeline is shown in Figure 4, and the main approaches used in each study are summarized in Table S4. The following sections provide a detailed explanation of each step.

3.2.1. Episignature Development: Data Formats

Methylation can be routinely measured using both methylation arrays and genome sequencing [145]. Methylation arrays screen specific predefined CpG sites clustered mostly in regions of interest (i.e., gene promoters, gene bodies, regulatory regions, etc.), with newer versions covering up to 930,000 different sites. Genome sequencing in the form of bisulfite sequencing or Oxford Nanopore/PacBio long reads offers more comprehensive methylation detection, though requires more computational capabilities.

To date, most episignatures have been developed using Illumina methylation arrays, specifically with versions 450 K or EPICv1, with a few exceptions where bisulfite sequencing via Illumina short read was employed [76,92,107]. However, the most recent episignatures already use the newest EPICv2 version.

Illumina arrays store raw data in binary .”idat” (intensity data) files, with each subject represented by two files—green and red channels—corresponding to array chemistry. Raw reads, on the other hand, are stored as .”fastq” files, and a series of steps are required to call methylation data per CpG before any downstream analysis.

The following preprocessing options are not specific to RD settings, but represent a common approach for all methylation analysis projects. Specific changes might be introduced due to the characteristics of studies using RDs, which will be specified in each study (Table S4). In addition, Table 2 includes a summary of all packages and tools in the reviewed articles to conduct the analysis.

3.2.2. Episignature Development: DNA Methylation Array Processing

For studies screening patients using arrays, raw data require several preprocessing steps. Data analysis is predominantly conducted in R with open-access Bioconductor packages used in the literature—minfi [146], RnBeads [152,153], ChAMP [147,148], lumi [149], meffil [151], wateRmelon [157], and SeSAMe [150]—in addition to Illumina Genome Studio software. The choice of package often depends on the research group’s experience since each one implements similar algorithms, although minfi has been the most frequently used.

.idat files are uploaded into R at the probe level, utilizing both color channels. Background correction is applied to distinguish true signal from background noise [177], typically based on negative control probes present in the array. Normalization is then performed to ensure comparability across samples and arrays within an experiment [178]. Several normalization methods exist for Illumina arrays, including SWAN [179], quantile normalization [180], noob normalization [177], functional normalization [181], Illumina Genome Studio default normalization and beta mixture quantile dilation (BMIQ) [182]. For further discussion about the topic, see [101,183], who tested the different processing methods using an RD dataset and in general samples, respectively.

For 450 K arrays, MethylAid [156] was developed as a Shiny application that uses minfi-associated methods for interactive array processing and methylation analysis. The package was used to perform basic quality control forf two fetal alcohol spectrum disorder-focused studies [80,81].

For RDs, no consensus was found in the selected articles about processing methodologies, though Illumina, functional, and noob normalizations were the most used and considered the most reliable ones [101], albeit the final selection may depend on cohort size.

Probe-level intensities are then merged at the CpG level, generating methylated and unmethylated signals, or an aggregated cumulative signal. CpG methylated/unmethylated signals can be transformed into quantitative aggregated values that summarize the methylation status of each cytosine: beta values and M-values. For an extensive description of the parameters, we refer to [184,185]. In short, each value represents:

Beta_CpGi = max(CpG_i_meth_intensity, 0)/(max(CpG_i_meth_intensity, 0) + max(CpG_i_unmeth_intensity, 0) + alpha)

(1)

Beta values (1) range from 0 to 1, with values > 0.8 and <0.2 indicating high and low methylation, respectively, but are not optimal for formal statistical testing due to their non-normal distribution [184].

M_CpGi = log2(max(CpG_i_meth_intensity, 0)/max(CpG_i_unmeth_intensity, 0))

(2)

M-values (2) range from −∞ (fully unmethylated) to +∞ (fully methylated) and are better suited for statistical modeling because they approximate a normal distribution, although less intuitive to interpret [184].

To ensure data reliability, several QC metrics are applied at both probe and sample levels.

At the probe level, common QC metrics consider: (1) Probes near single-nucleotide variants (SNVs) with minor allele frequencies (MAFs) > 1%. (2) Cross-reactive probes. (3) Probes with raw beta values of 0 or 1 in >0.25% of samples. (4) Non-CpG probes. (5) Sex chromosome probes (X and Y) are often excluded. (6) Probes with detection p-value > 0.05 or 0.01.

Cross-reactive probes and those near SNVs are commonly removed using standardized probe lists available in R packages, such as minfi, ChAMP, SeSAMe, or maxprobes, which offer automated filtering functions using standard lists of CpGs.

At the sample level: (1) Methylation values should follow a bimodal distribution, typically centered on ~0.2 and ~0.8 (beta values), reflecting the tendency of CpGs to be either almost fully unmethylated or methylated, respectively. (2) Samples with a high percentage of unreliable probes (e.g., with detection p-value >0.05 or >0.01).

The ultimate choice of QC thresholds and filters can vary depending on the specific objectives of the analysis.

3.2.3. Episignature Development: Multi-Cohort Studies

Merging datasets is often required to achieve statistical power for episignature detection in RD studies. When working with merged datasets, and as a normal QC step, batch inspection and correction is critical. Tools like ComBat (from the sva package [158] or its implementation in ChAMP) are used to correct for technical and biological variation across batches/datasets, while preserving key phenotypic differences (i.e., ensuring that correction does not obscure real biological signals).

Several studies used ComBat to correct batch effects (i.e., technical sources of variation) prior to episignature detection, though in all instances technical issues were identified through a PCA or MDS. Different batch sources and explanatory covariates were considered when modeling batches. For instance, Kagami et al. [98] only identified technical batches (array ID) and not clinical covariates while analyzing two imprinting syndromes (Temple and Kagami–Ogata syndromes); the same setting was repeated by Carmel et al. when studying the association of 22q11.2 deletion and schizophrenia [42]; a study that derived an episignature for autism [33] included sex as a potential confounding factor, even though X/Y probes were discarded; in a Williams syndrome study [134], HNRNPU-related neurodevelopmental disorder [68] and Aicardi–Goutières [43] identified both sex and age as potential batch effects. Lastly, two studies deriving the RNU4-2 episignature [123] and PURA-related neurodevelopmental disorders [118] also used ComBat, but did not specify which covariates of interest were accounted for to maintain biological differences.

Before applying any batch correction, researchers should first evaluate whether such correction is necessary. This can be accomplished using dimensionality-reduction and clustering techniques such as PCA, MDS, or hierarchical clustering to detect potential batch effects or sample outliers. If a batch effect is identified and corrected (array ID, processing site if multiple laboratories were involved, or processing date), then the result must subsequently be evaluated again to ensure that it was applied appropriately. In particular, investigators should verify that the adjustment eliminated the batch and did not introduce artifacts, such as inflated differences or excessive shrinkage that undermines confidence in the corrected methylation values. If batch correction proves unsuitable or introduces undesirable distortions, an alternative approach is to work with the uncorrected data while statistically modeling the batch as a covariate and/or incorporating surrogate variables to account for latent sources of heterogeneity [158,186].

3.2.4. Episignature Development: Methylation Screening Using Sequencing Methods

Bisulfite sequencing is another DNA methylation screening usable for episignatures; however, it is not a common choice, with only three reported studies using it [76,92,107]. Raw bisulfite data consist of .fastq files storing the sequence read and the quality of each base (i.e., Phred scores). Three basic steps are required before downstream methylation analysis is conducted: preprocessing, genome alignment, and methylation calling.

All three studies used trim-galore [187] to preprocess raw reads, eliminating adaptors and applying QC filters based on phred scores. Bisulfite sequencing requires specific alignment algorithms due to reduced read complexity while increasing mismatch mappings with the reference genome [188]. In all cases, alignment was conducted using bismark [189], which allows the detection of methylated sites by adding the option “bismark methylation extractor.” The output is stored in .”bismarkCov” files, indicating the genomic position and the ratio of methylated vs. total reads, akin to beta values. .”bismarkCov” files can be read and loaded into R using RnBeads for downstream analysis. Quality control metrics include exclusion of CpGs found in X and Y chromosomes and a minimum 10× coverage.

Long reads (PacBio and Oxford Nanopore), despite screening all genomic and methylomic variants at the same time, have not been systematically used for episignature detection yet [190]. This might be due to their added analysis complexity, requiring higher storage capacity and computer capabilities, and increased cost, limiting their full implementation. Interestingly, a proof-of-concept study demonstrating that nanopore sequencing enables episignature detection for up to 13 different disorders has recently been published, illustrating the possibility to use episignatures developed using arrays in sequencing profiles [142], although variant detection was reported to be not as accurate as using short-read sequencing. Moreover, epimutations have been studied using PacBio-HiFi in another RD cohort [190], proving the possibility to detect allele-specific hypermethylation in regulatory regions (i.e., epimutations).

3.2.5. Episignature Development: Differentially Methylated Position (DMP) Detection

Episignatures are sets of CpGs that consistently exhibit differential methylation between conditions, making them valuable predictive biomarkers. Typically, episignatures are identified by fitting a linear model across the epigenome and selecting CpGs that surpass predefined thresholds of statistical significance and methylation differences.

The limma R package is the most widely used tool, modeling M values per CpG while incorporating relevant covariates to control for confounding factors and derive a list of significant DMPs. Some commonly used R packages (minfi, meffil, RnBeads, ChAMP) offer built-in functions in their pipelines, though these are rarely used. Most studies account for at least immune cell composition, while others also include variables such as age, sex, and known batches to minimize confounding. These covariates represent key sources of biological and technical variation that, if not considered, can lead to spurious associations. However, the selection of appropriate covariates is not always straightforward. The epigenetics community widely acknowledges blood heterogeneity as a significant source of biological variation [191,192,193]. This is particularly relevant for episignature development, as cohort variability in age [193], inflammation, and immune-deficiency-related diseases [194,195,196] can significantly alter blood composition and consequently the epigenetic landscape.

A clear example of this challenge is exemplified during the development of the Nicolaides–Baraitser syndrome episignature [113], where high correlations among blood cells were detected. The authors concluded that only monocytes should be considered a confounding variable in DMP analysis.

Among the available deconvolution methods, the algorithm developed by Houseman et al. [197] remains the most commonly used, although there is no clear consensus on which method is best for rare-disease patients. For instance, some studies used the Houseman algorithm [61,113], while most recent manuscripts [48,130] used FlowSorted.Blood.EPIC [161]. A recent benchmarking of current algorithms and their performance [198] concluded that EpiDISH [162] outperforms other available methods, although none of the reviewed studies used it.

Disagreement also extends to immune cells types that should be included in the models. For instance, Butcher et al. [61] and Velasco et al. [94] did not include cell heterogeneity for HARGE and Kabuki studies or for multiple disorders caused by alterations in the epigenetic machinery, respectively. Chater-Diehl et al. [113] only adjusted for monocyte proportion for a Nicolaides–Baraitser study, while Siu et al. and Choufani et al. [49,130] considered six common cell types (B cells, natural killer, CD4+ T cells, CD8+ T cells, monocytes and neutrophils) using FlowSorted.Blood.EPIC and Houseman’s algorithm for EZH2 variants and CHD8/16p11.2del studies, respectively. To address uncertainty in cell composition modeling, it has been suggested that the top surrogate variables [199] be included, although only two of the reviewed studies used surrogate variables to account for heterogeneity, in addition to blood cells [52,102].

Those studies that screened methylation by bisulfite sequencing [76,92,107] also derived episignatures fitting a linear Nicolaides–Baraitser model, though through a limma implementation in RnBeads.

Despite M values being more statistically suitable for DMP detection, several studies used beta values instead of M values for their analysis and obtained positive results [32,37,38,57,60,66,81,87,88,90,106,110,134,200].

To summarize the methodologies in latest releases, methods have already diverged: both RNU4-2 episignatures [122,123] used meffil and functional normalization for data processing. A new version of the fetal alcohol spectrum disorder episignature [82] and one derived for PTBP1 carriers [117] used SeSAMe without specifying which normalization method was employed. Lastly, two studies used minfi to process MORC2 variants in multiple diseases [60] coupled with quantile normalization and in Sifrim–Hitz–Weiss patients (CHD4 variants) [125], though without specifying the method. Except in Nava et al.’s study [122], who used the limma wrapper from meffil, all other episignatures were derived using the standard limma linear model with commonly chosen covariates: chronological or epigenetic age, sex, immune cell proportions, and/or array ID.

3.2.6. Episignature Development: Other Approaches to Linear Models for Episignatures

Beyond linear models, various alternative strategies have been applied to identify significant DMPs, though not used in recent studies. Prickett et al. [32] compared methylated and unmethylated intensities between cases and controls to find hyper- and hypomethylated sites in Silver–Russell patients. Choufani et al. [126] used a nonparametric Mann–Whitney U test in Sotos syndrome patients adjusting for multiple comparisons using the Bonferroni method. Kagami et al. [98] applied a rule-based method, selecting CpGs with at least a 0.05 beta-value difference and more than three standard deviations between groups to define significant sites in Kagami–Ogata syndrome.

The Partek Genomics Suite provided by Illumina https://help.partek.illumina.com/partek-genomics-suite (accessed on 3 October 2025) was used by four studies [44,59,84,86] to perform DMP analysis. These studies used ANOVA tests with differing thresholds to detect differentially methylated regions (DMRs), applying different criteria to study alpha thalassemia, adult-onset autosomal dominant cerebellar ataxia, floating-harbor syndrome, and fragile X syndrome, respectively.

Finally, Siu et al. [49] developed two pairs of episignatures for ASD patients with 16p11.2 deletions and CHD8 variants. Their strategy involved initial filtering of the most variable sites across batches using Qlucore Omics Explorer https://qlucore.com/ (accessed on 3 October 2025), followed by the intersection of significant results using limma and Mann–Whitney U tests to develop the final episignature. The most recent study that did not use a linear model built in R was conducted by Oussalah et al. [78] in 2022, who used Golden Helix’s SNP and Variation Suite https://www.goldenhelix.com/products/SNP-Variation/ (accessed on 3 October 2025) and MedCalc to analyze a cohort with a vitamin B12 metabolism disorder. These platforms allowed PCA to assess sample distribution and outlier detection, followed by DMP identification using pairwise t-tests.

3.2.7. Episignature Development: DMRs and Epimutations

A different strategy involves identifying differentially methylated regions (DMRs). Biologically speaking, CpGs clustered within a DMR are more likely to regulate nearby genes collectively, making regional methylation patterns more informative than specific CpG sites [201]. Two primary well-established methods for identifying DMRs in the reviewed literature are bumphunter [165], implemented in minfi, and DMRcate [163]. Both can also be implemented directly using the ChAMP package and are based on linear models on multiple CpG sites in a genomic locus. Bumphunter fits a linear model to identify CpGs differentially methylated associated with the trait of interest, and then merges together into significant regions using permutation tests. DMRcate again fits a linear model to find differentially methylated CpGs, though it then scales the CpGs using a Gaussian kernel, assigning higher weights to spatially close CpGs.

Some studies designed custom algorithms to test more restrictive approaches. In 2019, Garg et al. [50] applied a 1 kb sliding window approach, defining a DMR as a region where at least three probes met stringent methylation difference thresholds (≥99.9 or ≤0.01 percentile of control distribution, with absolute differences of ≥0.15) for an ASD and schizophrenia cohort. A more recent approach conducted by LaFlamme et al. [67] in 2024 used a similar approach, but with different thresholds (≥99.25th or ≤0.75th percentile), and accounted for sex-specific X chromosome methylation patterns rather than excluding them altogether in pediatric epilepsies and patients with CHD2 variants.

Besides searching for consensus DMRs, other authors focused on detecting epimutations per sample, offering a patient-specific analysis. In 2018, Barbosa et al. [129] designed a custom sliding window to compare each patient’s reference profile, differentiating for autosomal and X CpGs, with a reference cohort panel to find hypermethylation and hypomethylation events. Their cohort consisted of a mix of congenital disorders, often without any known diagnosis. For this reason, common DMR and DMP approaches could not be conducted due to their high variability.

In 2023, Lee et al. [92] screened a cohort of patients with pathogenic variants in HNRNPU using NGS and detected differentially methylated regions by clustering physically close significant DMPs using a two-sided Welch test or by a linear model implemented in limma. Only CpGs functionally similar (i.e., all found in the same CpG island) were combined.

Despite their potential, DMR-based approaches are not the primary focus of episignature research, as episignatures are typically defined at the CpG-site level. CpGs within a significant region tend to be highly correlated, rendering them redundant if used as separate features in a classification model. However, recent discussions [101] have explored the feasibility of using DMRs for episignature testing, particularly as a means to ensure compatibility with future array versions, since they showed DMR-derived episignatures are less affected by missing data.

Regardless of the approach, once DMPs or DMRs are identified, they can be used to investigate gene dysregulation, cluster cases and controls based on methylation profiles, and provide insights into genotype-phenotype associations. In addition, an extra step is usually taken to filter which CpGs to include to represent the true episignature from all statistically significant ones (i.e., with an adjusted p-value < 0.05). The different approaches will be described in depth below in the section Episignature Development: Classification Model Pre-Filtering.

3.2.8. Episignature Development: Annotation and Enrichment

Another key step is the annotation of significant loci to determine any possible association with the patients’ phenotype. This step allows the identification of whether the episignature is biologically explanatory.

The most straightforward method to annotate DMPs is using the annotation files provided by Illumina and easily accessible through specific Bioconductor packages (IlluminaHumanMethylation450kanno.ilmn12.hg19, IlluminaHumanMethylationEPICanno.ilm10b4.hg19, and IlluminaHumanMethylationEPICv2anno.20a1.hg38), which contain the mapped gene (if any) and the genomic context (i.e., if the CpG is part of a CpG island), among others. In the case of DMRs, the most interesting aspect is finding which genes fall within each region and if they encompass any regulatory element of interest. These can be annotated using packages such as annotatr [169] or LOLA [170].

Enrichment analysis can be conducted to find CpGs associated with known pathways. When analyzing methylation data, enrichment should take into account the positional correlation of each CpG and multiple CpG mapping the same gene such as missMethyl [171] or methylGSA [173], though common enrichment techniques have also been employed using GSEA [174].

Another approach was applied by Kimura et al. [134], who used the WGCNA R package [164] to detect co-methylation correlation patterns among different genes. While the idea might be similar to detect DMRs, their approach tested correlations among independent CpGs throughout the genome without the constraint of being physically close, finding gene networks that might be involved in causal pathways. Kimura et al. tested a cohort of Williams syndrome patients, finding genes associated with ASD and Alzheimer’s disease, among other neuropsychiatric phenotypes related to Williams syndrome.

3.2.9. Episignature Development: Classification Model Introduction

An important objective in episignature detection is the classification of new individuals. Different machine learning (ML) methods can be applied to methylation-based diagnosis of RD. The goal of ML methods is to build an algorithm capable of predicting whether a given epigenetic pattern originates from a patient affected by a specific disease. In other words, it aims to assign a confidence score indicating the likelihood that a sample belongs to a patient with a given disease versus an unaffected individual.

3.2.10. Episignature Development: Classification Model Pre-Filtering

A first key step is the filtering of informative features derived from any of the aforementioned methods. Filtering of candidate DMPs is crucial to prevent overfitting, particularly in cases where low-effect CpGs with minimal predictive power are retained [184]. There is large variability in filtering strategies, reflecting either disease-specific epigenetic alterations that require tailored approaches or the iterative tuning needed to obtain a usable set of CpGs that best discriminates cases and controls given the limited sample size. This variability hinders episignature comparisons and model standardization, since filtering decisions are rarely reported.

Hence, common approaches in this case include applying thresholds to both the adjusted p-value (typically ranging from 0.1 to 0.01 under Benjamini–Hochberg correction) and the absolute methylation difference (Δβ), which generally needs to exceed 0.05 (5% methylation difference). The exact thresholds vary, as exemplified in the study conducted by Aref-Eshghi et al. [36], in which 14 different RDs were tested. Sotos syndrome had the most stringent threshold (Δβ > 0.25), while BAFopathies, Cornelia de Lange syndrome, and CHARGE syndrome had the less restrictive ones (Δβ > 0.05).

Many different studies refined this approach by ranking features based on the product of the Δβ and the −log(adjusted p-value) and selecting the top-ranked probes [26,29,35,40,41,51,53,56,58,60,62,64,66,68,69,72,74,79,82,88,89,96,97,103,109,110,111,112,114,117,120,121,123,125,133,135]. Additional filtering was applied based on probe–probe correlation (typically using Spearman’s correlation with thresholds below 0.7 and 0.9) and individual probe performance metrics such as AUC to assess disease discrimination, keeping a given number of CpGs, which again changed from study to study or within a given study. For instance, in a Menke–Hennekam study [109] of patients with CREBBP or EP300 mutations, the authors tried to identify three different episignatures, which depended on in which domain the variant was found (ZZ, TAZ2, or ID4). For each variant, the authors decided independent thresholds of significance and the number of probes to keep. Only patients with ID4 mutations resulted in a clinically significant episignature, though mild alterations were found for ZZ- and TAZ2-variant carriers.

A different implementation was used by Oexle et al. [75] to account for probe redundancy. A minimum redundancy–maximum relevance (mRMR) ensemble algorithm was used, implemented in the mRMRe R package [167], to leverage importance and correlation between features, allowing a reduction in the gross episignature length while maintaining the signature. In a multi-class study, Levy et al. [38], after filtering by Δβ and adjusted p-value, merged all significant CpGs from all cases and trained a multi-class random forest to select the most informative features for each class.

In other studies, minimal filtering was performed beyond statistical significance and methylation difference, especially when feature selection was integrated within the model itself, allowing the algorithm to choose the most informative features during training [85,101,106,119].

3.2.11. Episignature Development: Classification Model Design

While training a machine learning (ML) model, several aspects must be considered to ensure reliable results.

Validation strategy: If additional samples are not available for testing (ideally external cohorts), some cases and controls should be set aside for model testing and evaluation (usually 25% of the cohort, usually fewer than five samples; Table 3). Nevertheless, if the cohort is small, withholding some data for validation may reduce the model’s robustness.

Variant-level specificity: Should the model be trained on all individuals collectively or should it focus on specific variants? A given disease might be caused by myriad mutations and/or causal genes. For instance, Rubinstein–Taybi syndrome, mainly caused by mutations in CREBBP or EP300 [202], has gene-specific episignatures rather than syndrome-specific., or even at gene-domain level, as reported by Haghshenas et al. [109] in Menke–Hennekam patients, caused by variants in the same two genes.

Class imbalance: large differences in group sizes (e.g., many more controls than cases) can bias the model towards predicting the majority class. For example, if controls outnumber cases 10:1, the model might favor “control” predictions by default, since most predictions would be correct. Several strategies exist to control class imbalance: upsampling, creating synthetic data by subsampling with replacement features from the minor class; downsampling, by subsampling the majority class and training the model iteratively on different subsets of the same size as the minority class; the synthetic minority oversampling technique (SMOTE), a mixed approach that downsamples the majority class and creates new synthetic data for the minority one [203]; or random oversampling example (ROSE), which generates synthetic data of the minority class by adding noise to existing data points [204].

Cross-validation (CV) and hyperparameter tuning: Model performance depends heavily on the choice of hyperparameters (i.e., parameters set before training the ML model, which will influence the final model quality and generalizability [205]), which should be optimized using cross-validation (CV) loops (Figure 5). Evaluation metrics such as precision, AUC, or F1 score guide feature selection [206,207] (Table 3). A common strategy is X-fold CV (Figure 5 exemplifies fivefold CV with three repetitions) or leave-one-out CV (holding out a single sample per fold).

Table 3. Model evaluation metrics used during CV and to assess model performance. In estimation column: TPs (true positives), TNs (true negatives), FPs (false positives), FNs (false negatives), log L(

\hat{θ}

) (log-likelihood).

Table 3. Model evaluation metrics used during CV and to assess model performance. In estimation column: TPs (true positives), TNs (true negatives), FPs (false positives), FNs (false negatives), log L(

\hat{θ}

) (log-likelihood).

Metric	Definition	Estimation	Ref.
Recall or sensitivity	Proportion of real positives correctly identified.	(TPs)/(TPs + FNs)	[206]
Specificity	Proportion of real negatives correctly identified.	(TNs)/(TNs + FPs)	[206]
Accuracy	Proportion of correctly classified instances.	(TPs + TNs)/(TPs + TNs + FPs + FNs)	[206]
Precision	Proportion of predicted positives that are actually positive.	(TPs)/(TPs + FPs)	[206]
AUC	Area under the ROC curve; represents trade-off between sensitivity and specificity.	Computed from ROC curve	[207]
F1 score	Harmonic mean of precision and sensitivity.	(2 × precision × sensitivity)/(precision + sensitivity)	[206]
Deviance	Comparison between trained model and perfect model.	−2 log L $\hat{θ}$	[208]
Cohen’s kappa	Comparison between model predictions (Pr(a)) and random guessing (Pr(e)).	(Pr(a) − Pr(e))/(1 − Pr(e))	[209]

Most recent models have been developed using the caret R package [210], which implements various classification algorithms under a single tool, providing integrated solutions for managing validation strategy, class imbalance, and cross-validation. Specifically, caret offers encapsulation and common function call for all commonly used algorithms: support vector machines (SVMs), random forests (RFs), penalized logistic regression (PLR), elastic net, and coarse approximation linear function. However, ML models are available from specific software packages for each classification model type.

The most comprehensive predictor so far, EpiSign, has been developed using SVMs (with linear kernel). SVMs are a class of supervised machine learning algorithms used for classification and regression. Their core principle is to find the optimal hyperplane that best separates the data into distinct classes. In two dimensions, this hyperplane is simply a line, but in higher dimensions, it corresponds to a more complex topology. The optimal hyperplane is defined as the one that maximizes the margin, or distance between the nearest data points of each class, known as support vectors. In Refs. [211,212], SVM model training is implemented in e1071 and caret R packages.

SVMs are trained based on a kernel function, indicating how the hyperplane should be built: Linear kernel is suitable when classes are linearly separable. The main hyperparameter is the cost, a regularization parameter that prevents overfitting by balancing between low training error and large margin. Radial basis kernel maps data into nonlinear dimensional space. Besides the cost, it is also defined with the sigma hyperparameter, defining the influence each training sample has, leveraging overall model complexity.

Random forests are ensemble learning models in which many classifiers (trees) are generated and the final decision accomplished by aggregating their results. RFs are based on growing several independent and identically distributed decision trees based on different bootstrap subsets of the training data, which offer discriminatory cutoffs to decide an outcome based on the most informative features selected from an initial random subset of them [213,214,215]. RFs are based on two hyperparameters: number of variables per tree and number of trees in the forest. Multiple strategies to train RFs are seen in the literature. Ciolfi et al. [119] trained a model in caret for Rahman syndrome patients with C-terminal frameshift mutations, and Kimura et al. [33] used the randomForest R package [215] to train a classifier for adults with high-functioning ASD. In addition, other Python-based tools were also used for RF training: Giuili et al. [101] and Turinsky et al. [25] used scikit-learn [216] and GenPipes [176], respectively.

Other popular ML methods in methylation analysis, but less exploited in the context of rare diseases are penalized logistic regression (PLR) and elastic net. Penalized methods prevent model overfitting thanks to adding shrinkage parameters to less informative features and controlling the overall importance of each feature using a regularization parameter λ [217,218]. Two methods exist: Lasso (or L₁ penalization) shrinks coefficients of non-informative variables to zero, providing effective variable selection when inter-feature correlation is low. Ridge (or L₂ penalization) retains all features with non-zero coefficients, performing better with highly correlated features and low overall noise.

Elastic net leverages both methods according to an α parameter (ridge: α = 0; lasso: α = 1; in between, elastic net). In the literature, logistic regression approaches have only been implemented using scikit-learn and GenPipes [25,101], despite the availability of packages in R such as caret and glmnet [218].

Another method seen in the literature is the coarse approximation linear function using the CALF R package, a greedy algorithm that builds a classification metric by adding each feature one at a time based on Student t-test p-values until a newly added feature does not improve the prediction capabilities anymore. CALF has only been applied once, in a high-functioning ASD cohort [33], accompanied by ROC as evaluation metric.

In addition to formal ML models, a simpler approach was used in the past consisting of a median-based decision algorithm [48,49,113,130]. Classification was accomplished by estimating the median difference of episignature CpGs between cases and controls.

Overall, the studies reported strong predictive performance, regardless of the analytical method (most did not report balancing strategy or which evaluation metric was used), disease type, or screening approach. Most validations were internal, typically performed by predicting the status of individuals left out before model training. Despite this lack of external validation, it is important to emphasize that such validation is essential for establishing the true generalizability of a model. When discovery and validation samples are processed together, observed case–control differences may reflect not only genuine biological variation but also technical or batch effects. This can substantially diminish discriminatory performance in an independent cohort, as was the case in some episignatures discussed in earlier sections. Despite this, Ferraro et al.’s approach to training independent models based on synthetic data using known episignatures proved that most of them could be validated (though not using the same model as the original article) [34].

4. Discussion

We have presented the current state of the art regarding episignatures and epimutation analysis for RD diagnosis after a thorough systematic revision of the literature. Episignature-based diagnosis is likely to become a reference tool in the functional evaluation of RDs. Nevertheless, several challenges must be addressed to standardize their usage in clinical practice.

(1) Limited case availability and data sharing: Clinical application limited by the scarcity of available cases, hindering external validations. There is a notable lack of open-access data in public repositories. In RD research, sharing genetic data is even more difficult due to the uniqueness of patient’s profiles. Even though early studies had published the raw data in open-access repositories, data are not commonly shared anymore, as stated by the authors, owing to ethical or legal concerns. In recent years, data sharing has been hindered due to concerns of patient re-identification [219,220], which could partially explain the reduced number of public releases, but since many studies are conducted by private initiatives that offers episignatures as a service, the underlying decision could be influenced by economic reasons.

Regardless of data-sharing issues, episignature inference requires as many cases as possible to reliably capture disease-specific epigenomic patterns. New approaches have been proposed to comply with data protection policies [21,221], enabling controlled access for research centers to refine existing episignatures or validate new ones. Repositories with controlled access such as EGA could be used so public institutions can decide which authors can access the data to conduct specific analysis. DataSHIELD is a framework that allows research centers and hospitals to securely connect either database through a federated, non-disclosive approach, ensuring data confidentiality [222,223]. DataSHIELD, or other federated data networks, are particularly suitable for RD research [224,225], since they could offer the possibility of using more data to call episignatures or train classification models without actually sharing or disclosing them. Briefly, federated data networks are structured databases under certain data harmonization rules, which are fully controlled by the research center in question, offering non-disclosive access to other users. This way, data governance and patient confidentiality is not compromised [226], though the data are accessible to foster new developments. Regardless, recently developed tools (NSBEpi [142] and MethaDory [34]) demonstrate the possibility of working with a reduced number of cases, offering a good alternative to create custom classification models from known episignatures if no more cases are available.

(2) Array version updates: Microarray versions are regularly updated, causing older arrays to become deprecated. Certain CpGs may no longer be measured, limiting the applicability of episignatures developed on legacy platforms like 450K and EPIC due to missing CpGs [101] if not imputed. The EPICv2 array, for instance, uses updated annotations, complicating direct translation [227]. Two different strategies exist to address this issue:

DMRs represented by a single methylation statistic (mean, median, or quantile metric) [101,201]. Under this approach, all CpGs found in a region take part in the metric, regardless of which specific CpGs are taken into account.
Sequencing via genome bisulfite sequencing or third-generation “five-base” callers. The latter offers promising results, since third-generation sequencing technologies could offer the detection of all available genetic variability, from intronic/exonic variants to whole- genome methylation [190].

However, there exists cross-platform probe mapping to promote compatibility. Some Bioconductor R packages (SeSAMe and MethylCallR [155]) offer this conversion, which is even more essential for EPICv2, since multiple probes can map the same CpG.

(3) Methodological diversity: Heterogeneous analysis pipelines hinder diagnostic standardization. This variability prevents objective comparisons and limits the development of unified workflows, especially since original cohorts are often inaccessible for peer review. Confounding factors in DMP calling are particularly challenging. Blood cell deconvolution methods were trained in the general population or in cancer, hence using them in RDs with inflammatory phenotypes or global immune defects may misestimate cell proportions. Likewise, post-DMP CpG filtering lacks standardized guidelines. Although many approaches are functionally similar and differ only in their software implementation, the overall lack of consistency can make pipelines appear tailored to individual datasets rather than to the disease itself. This in turn increases the risk of model overfitting and reduces reproducibility in external cohorts.

(4) Tissue specificity: Methylation is tissue-specific. Blood-derived episignatures may not reflect changes in affected tissue, such as the brain in NDDs. For that reason, the obtained CpGs might not have any biological meaning, but may work as a biomarker likewise. Alternative tissue (e.g., cerebrospinal fluid or saliva) could be explored, as they may be more biologically relevant or easier to obtain. Mosaicism may also yield inconsistent episignatures depending on the affected tissue and mosaicism extent.

(5) Longitudinal stability: To date, episignatures have been assessed cross-sectionally at a single time point, without follow-up studies evaluating the stability of the profiles. However, methylation studies outside episignature research have tested this stability. Komaki et al. [228] screened the epigenome at 24 time points throughout three months in two control adults and did not detect any major longitudinal differences. In a lupus study, Coit et al. [229] screened 54 patients during 4 years, and again the authors detected a stable profile. However, in a third study using a control children cohort [230], variation was detected throughout the first 5 years of life. Hence, since many RDs and NDDs are studied in children, extra focus should be placed on the area on the grounds of developing meaningful episignatures regardless of age.

(6) Chromosomal sex stratification: Sex is often ignored as a confounding variable. Methylome differences between sexes are well documented during puberty [231] and in the general population [232], potentially affecting episignatures. In addition, many studies discard XY probes, disregarding the epigenetic effect on the gonosomes. X probes can be informative, either through non-sex-specific deviations or disease-related interactions. For instance, Radio et al. [35] studied patients with 1p36 deletion syndrome spanning the SPEN gene, which was reported as a modulator in X chromosome inactivation [233], and found 122 significant CpGs in the X chromosome, but this only worked in females.

(7) Lack of integration among developed episignatures for the same disease: Independent episignatures for the same disease are rarely compared or standardized [140]. Niceta et al. [144] found that three different episignatures for Kabuki syndrome could differentiate most cases and classify them correctly. While each episignature was based on different CpGs, they correctly classified patients with KMT2D variants. A different approach was taken by Davide et al. [28], who tried to integrate multiple episignatures by conducting a meta-analysis using open-access datasets of ASD patients. The pooled significant sites could thus be used as a consensus episignature for the disease.

(8) Clinical utility and VUS interpretation: Episignatures are valuable for functional diagnosis, in unresolved cases, or for clinical confirmation. Episignatures can add crucial information on VUS, particularly when other tests are costly or unavailable. Notably, Ciolfi et al. [119] developed an RF classifier for Rahman syndrome. The authors ran the classifier against known and unresolved cases and identified a patient with a positive episignature, later confirmed via exome sequencing. The same approach has worked for other syndromes [143,144] to confirm cases with VUS.

(9) Study of imprinting disorders: Some RDs can be directly caused by abnormal methylation of imprinted genes and regions [234]. The molecular diagnosis of these syndromes might be better targeted at the analysis of these causal regions rather than using whole episignatures, as discussed by Hara-Isono et al., in which no common episignature could be found in Silver–Russell patients with different molecular etiologies [235]. Another study focused on four different imprinting disorders (Angelman, Prader–Willi, Beckwith–Wiedemann and Silver–Russell syndrome) also had the same approach of analyzing specific regions, which demonstrated the utility of this approach as a diagnostic tool [236]. Variant-specific episignatures may be achievable [235], but focusing on epimutation events at targeted loci could yield more consistent diagnoses.

Finally, this review has some limitations that should be pointed out. Although we aimed to capture the full breadth of episignature research, our search strategy may have missed relevant studies, particularly because terminology varies and some reports do not explicitly use the term “episignature.” The field is largely driven by a relatively small number of groups, introducing the possibility of publication bias and limiting the visibility of alternative methodologies. Moreover, most original datasets and trained classifiers are not publicly accessible, preventing direct, head-to-head performance comparisons across approaches and constraining us to the descriptions provided by the original authors. Negative results are rarely published, so the number of disorders without a confirmed episignature is likely underreported. Until such non-confirmatory findings become more widely available, some conditions may continue to be tested repeatedly even if a robust episignature does not exist.

5. Conclusions

Episignatures represent a promising advancement for improving the diagnosis of RDs and NDDs. They offer a powerful framework for developing new clinical tools that can support and refine diagnostic decision-making. To date, an expanding number of disease-specific episignatures have been identified, highlighting their diagnostic potential across diverse syndromes. Although formal studies in routine clinical settings are still limited, the available evidence suggests that incorporating validated episignatures and epimutations into standard diagnostic pipelines could substantially increase diagnostic yield. Nevertheless, further efforts are required to ensure methodological robustness, promote standardization across analytical workflows, and facilitate their effective integration into clinical practice.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/biomedicines13123043/s1. Table S1: EpiSign syndromes; Table S2: EpigenCentral syndromes; Table S3: MethaDory syndromes; Table S4: Episignature pipelines per article; File S1: PRISMA 2020 Checklist; File S2: Supplementary methods.

Author Contributions

Conceptualization: A.A.-G. and A.C.; methodology: A.A.-G. and A.C.; writing—original draft preparation: A.A.-G.; writing—review and editing: A.C., M.S.-P., L.A.P.-J. and J.R.G.; supervision: L.A.P.-J. and J.R.G.; project administration: J.R.G.; funding acquisition: J.R.G., A.C. and L.A.P.-J. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by MCIN/AEI/10.13039/501100011033 (grant CEX2023-0001290-S) with support from the Generalitat de Catalunya through the CERCA program, by the Spanish Ministry of Science and Innovation MCIN/FIS-ISCIII, Ref PI21/00050, Fondo Europeo de Desarrollo Regional (FEDER), and the Generalitat de Catalunya through the Consolidated Research Group (2017SGR01974), and by the Spanish National Investigation Agency (AEI) (DOI:10.13039/501100011033) through the Unidad de Excelencia María de Maeztu (grant CEX2018-000792-MDM).

Data Availability Statement

No new data were created or analyzed in this study. Data sharing is not applicable to this article.

Acknowledgments

Conflicts of Interest

L.P.J. is a member of the scientific board of qGenomics. All other authors declare that they have no competing interests.

Abbreviations

The following abbreviations are used in this manuscript.

CV	cross-validation
DMP	differentially methylated position
DMR	differentially methylated region
EGA	European Genome-Phenome Archive
eWASs	epigenome-wide association studies
GEO	Gene Expression Omnibus
MAF	minor allele frequency
ML	machine learning
NDD	neurodevelopmental disorder
RD	rare disease
RF	random forest
PLR	penalized logistic regression
QC	quality control
SNVs	single-nucleotide variants
SVM	support vector machine
VUS	variants of uncertain significance
WES	whole-exome sequencing
WGS	whole-genome sequencing

References

The Lancet Global Health. The Landscape for Rare Diseases in 2024. Lancet Glob. Health 2024, 12, e341. [CrossRef] [PubMed]
Melnikova, I. Rare Diseases and Orphan Drugs. Nat. Rev. Drug Discov. 2012, 11, 267–268. [Google Scholar] [CrossRef] [PubMed]
Haendel, M.; Vasilevsky, N.; Unni, D.; Bologa, C.; Harris, N.; Rehm, H.; Hamosh, A.; Baynam, G.; Groza, T.; McMurry, J.; et al. How Many Rare Diseases Are There? Nat. Rev. Drug Discov. 2020, 19, 77. [Google Scholar] [CrossRef] [PubMed]
Nguengang Wakap, S.; Lambert, D.M.; Olry, A.; Rodwell, C.; Gueydan, C.; Lanneau, V.; Murphy, D.; Le Cam, Y.; Rath, A. Estimating Cumulative Point Prevalence of Rare Diseases: Analysis of the Orphanet Database. Eur. J. Hum. Genet. 2020, 28, 165–173. [Google Scholar] [CrossRef]
Reichard, J.; Zimmer-Bensch, G. The Epigenome in Neurodevelopmental Disorders. Front. Neurosci. 2021, 15, 776809. [Google Scholar] [CrossRef]
Khodosevich, K.; Sellgren, C.M. Neurodevelopmental Disorders—High-Resolution Rethinking of Disease Modeling. Mol. Psychiatry 2022, 28, 34–43. [Google Scholar] [CrossRef]
Michaels-Igbokwe, C.; McInnes, B.; MacDonald, K.V.; Currie, G.R.; Omar, F.; Shewchuk, B.; Bernier, F.P.; Marshall, D.A. (Un)Standardized Testing: The Diagnostic Odyssey of Children with Rare Genetic Disorders in Alberta, Canada. Genet. Med. 2021, 23, 272–279. [Google Scholar] [CrossRef]
Thevenon, J.; Duffourd, Y.; Masurel-Paulet, A.; Lefebvre, M.; Feillet, F.; El Chehadeh-Djebbar, S.; St-Onge, J.; Steinmetz, A.; Huet, F.; Chouchane, M.; et al. Diagnostic Odyssey in Severe Neurodevelopmental Disorders: Toward Clinical Whole-Exome Sequencing as a First-Line Diagnostic Test. Clin. Genet. 2016, 89, 700–707. [Google Scholar] [CrossRef]
van der Sanden, B.P.G.H.; Schobers, G.; Corominas Galbany, J.; Koolen, D.A.; Sinnema, M.; van Reeuwijk, J.; Stumpel, C.T.R.M.; Kleefstra, T.; de Vries, B.B.A.; Ruiterkamp-Versteeg, M.; et al. The Performance of Genome Sequencing as a First-Tier Test for Neurodevelopmental Disorders. Eur. J. Hum. Genet. 2022, 31, 81–88. [Google Scholar] [CrossRef]
Hartley, T.; Lemire, G.; Kernohan, K.D.; Howley, H.E.; Adams, D.R.; Boycott, K.M. New Diagnostic Approaches for Undiagnosed Rare Genetic Diseases. Annu. Rev. Genom. Hum. Genet. 2020, 21, 351–372. [Google Scholar] [CrossRef]
Marwaha, S.; Knowles, J.W.; Ashley, E.A. A Guide for the Diagnosis of Rare and Undiagnosed Disease: Beyond the Exome. Genome Med. 2022, 14, 23. [Google Scholar] [CrossRef]
Rooney, K.; Sadikovic, B. DNA Methylation Episignatures in Neurodevelopmental Disorders Associated with Large Structural Copy Number Variants: Clinical Implications. Int. J. Mol. Sci. 2022, 23, 7862. [Google Scholar] [CrossRef] [PubMed]
Joynt, A.C.M.; Axford, M.M.; Chad, L.; Costain, G. Understanding Genetic Variants of Uncertain Significance. Paediatr. Child. Health 2021, 27, 10. [Google Scholar] [CrossRef] [PubMed]
Ferrer, A.; Mangaonkar, A.A.; Stroik, S.; Zimmermann, M.T.; Sigafoos, A.N.; Kamath, P.S.; Simonetto, D.A.; Wylam, M.E.; Carmona, E.M.; Lazaridis, K.N.; et al. Functional Validation of TERT and TERC Variants of Uncertain Significance in Patients with Short Telomere Syndromes. Blood Cancer J. 2020, 10, 120. [Google Scholar] [CrossRef] [PubMed]
Hamilton, J.P. Epigenetics: Principles and Practice. Dig. Dis. 2011, 29, 130. [Google Scholar] [CrossRef]
Pinel, C.; Prainsack, B.; McKevitt, C. Markers as Mediators: A Review and Synthesis of Epigenetics Literature. BioSocieties 2017, 13, 276–303. [Google Scholar] [CrossRef]
Fu, M.P.; Merrill, S.M.; Sharma, M.; Gibson, W.T.; Turvey, S.E.; Kobor, M.S. Rare Diseases of Epigenetic Origin: Challenges and Opportunities. Front. Genet. 2023, 14, 1113086. [Google Scholar] [CrossRef]
Bjornsson, H.T. The Mendelian Disorders of the Epigenetic Machinery. Genome Res. 2015, 25, 1473–1481. [Google Scholar] [CrossRef]
Benn, P. Uniparental Disomy: Origin, Frequency, and Clinical Significance. Prenat. Diagn. 2021, 41, 564–572. [Google Scholar] [CrossRef]
Kerkhof, J.; Rastin, C.; Levy, M.A.; Relator, R.; McConkey, H.; Demain, L.; Dominguez-Garrido, E.; Kaat, L.D.; Houge, S.D.; DuPont, B.R.; et al. Diagnostic Utility and Reporting Recommendations for Clinical DNA Methylation Episignature Testing in Genetically Undiagnosed Rare Diseases. Genet. Med. 2024, 26, 101075. [Google Scholar] [CrossRef]
Bonomi, L.; Huang, Y.; Ohno-Machado, L. Privacy Challenges and Research Opportunities for Genomic Data Sharing. Nat. Genet. 2020, 52, 646–654. [Google Scholar] [CrossRef] [PubMed]
Mittos, A.; Malin, B.; De Cristofaro, E. Systematizing Genome Privacy Research: A Privacy-Enhancing Technologies Perspective. Proc. Priv. Enhancing Technol. 2017, 2019, 87–107. [Google Scholar] [CrossRef]
Chater-Diehl, E.; Goodman, S.J.; Cytrynbaum, C.; Turinsky, A.L.; Choufani, S.; Weksberg, R. Anatomy of DNA Methylation Signatures: Emerging Insights and Applications. Am. J. Hum. Genet. 2021, 108, 1359–1366. [Google Scholar] [CrossRef] [PubMed]
Page, M.J.; McKenzie, J.E.; Bossuyt, P.M.; Boutron, I.; Hoffmann, T.C.; Mulrow, C.D.; Shamseer, L.; Tetzlaff, J.M.; Akl, E.A.; Brennan, S.E.; et al. The PRISMA 2020 Statement: An Updated Guideline for Reporting Systematic Reviews. BMJ 2021, 372, 71. [Google Scholar] [CrossRef]
Turinsky, A.L.; Choufani, S.; Lu, K.; Liu, D.; Mashouri, P.; Min, D.; Weksberg, R.; Brudno, M. EpigenCentral: Portal for DNA Methylation Data Analysis and Classification in Rare Diseases. Hum. Mutat. 2020, 41, 1722–1733. [Google Scholar] [CrossRef]
Aref-Eshghi, E.; Kerkhof, J.; Pedro, V.P.; Barat-Houari, M.; Ruiz-Pallares, N.; Andrau, J.C.; Lacombe, D.; Van-Gils, J.; Fergelot, P.; Dubourg, C.; et al. Evaluation of DNA Methylation Episignatures for Diagnosis and Phenotype Correlations in 42 Mendelian Neurodevelopmental Disorders. Am. J. Hum. Genet. 2020, 106, 356–370, Erratum in Am. J. Hum. Genet. 2021, 108, 1161–1163. [Google Scholar] [CrossRef]
Haghshenas, S.; Bhai, P.; Aref-Eshghi, E.; Sadikovic, B. Diagnostic Utility of Genome-Wide DNA Methylation Analysis in Mendelian Neurodevelopmental Disorders. Int. J. Mol. Sci. 2020, 21, 9303. [Google Scholar] [CrossRef]
Davide, G.; Rebecca, C.; Irene, P.; Luciano, C.; Francesco, R.; Marta, N.; Miriam, O.; Natascia, B.; Pierluigi, P. Epigenetics of Autism Spectrum Disorders: A Multi-Level Analysis Combining Epi-Signature, Age Acceleration, Epigenetic Drift and Rare Epivariations Using Public Datasets. Curr. Neuropharmacol. 2023, 21, 2362–2373. [Google Scholar] [CrossRef]
Haghshenas, S.; Levy, M.A.; Kerkhof, J.; Aref-Eshghi, E.; McConkey, H.; Balci, T.; Siu, V.M.; Skinner, C.D.; Stevenson, R.E.; Sadikovic, B.; et al. Detection of a Dna Methylation Signature for the Intellectual Developmental Disorder, x-Linked, Syndromic, Armfield Type. Int. J. Mol. Sci. 2021, 22, 1111. [Google Scholar] [CrossRef]
Cavicchi, C.; Oussalah, A.; Falliano, S.; Ferri, L.; Gozzini, A.; Gasperini, S.; Motta, S.; Rigoldi, M.; Parenti, G.; Tummolo, A.; et al. PRDX1 Gene-Related Epi-CblC Disease Is a Common Type of Inborn Error of Cobalamin Metabolism with Mono- or Bi-Allelic MMACHC Epimutations. Clin. Epigenet. 2021, 13, 137. [Google Scholar] [CrossRef]
Garg, P.; Jadhav, B.; Rodriguez, O.L.; Patel, N.; Martin-Trujillo, A.; Jain, M.; Metsu, S.; Olsen, H.; Paten, B.; Ritz, B.; et al. A Survey of Rare Epigenetic Variation in 23,116 Human Genomes Identifies Disease-Relevant Epivariations and CGG Expansions. Am. J. Hum. Genet. 2020, 107, 654–669. [Google Scholar] [CrossRef] [PubMed]
Prickett, A.R.; Ishida, M.; Böhm, S.; Frost, J.M.; Puszyk, W.; Abu-Amero, S.; Stanier, P.; Schulz, R.; Moore, G.E.; Oakey, R.J. Genome-Wide Methylation Analysis in Silver–Russell Syndrome Patients. Hum. Genet. 2015, 134, 317–332. [Google Scholar] [CrossRef] [PubMed]
Kimura, R.; Nakata, M.; Funabiki, Y.; Suzuki, S.; Awaya, T.; Murai, T.; Hagiwara, M. An Epigenetic Biomarker for Adult High-Functioning Autism Spectrum Disorder. Sci. Rep. 2019, 9, 13662. [Google Scholar] [CrossRef] [PubMed]
Ferraro, F.; Drost, M.; van der Linde, H.; Bardina, L.; Smits, D.; de Graaf, B.M.; Schot, R.; van Bever, Y.; Brooks, A.; Kaat, L.D.; et al. Training with Synthetic Data Provides Accurate and Openly-Available DNA Methylation Classifiers for Developmental Disorders and Congenital Anomalies via MethaDory. medRxiv 2025. [Google Scholar] [CrossRef]
Radio, F.C.; Pang, K.; Ciolfi, A.; Levy, M.A.; Hernández-García, A.; Pedace, L.; Pantaleoni, F.; Liu, Z.; de Boer, E.; Jackson, A.; et al. SPEN Haploinsufficiency Causes a Neurodevelopmental Disorder Overlapping Proximal 1p36 Deletion Syndrome with an Episignature of X Chromosomes in Females. Am. J. Hum. Genet. 2021, 108, 502–516. [Google Scholar] [CrossRef]
Aref-Eshghi, E.; Bend, E.G.; Colaiacovo, S.; Caudle, M.; Chakrabarti, R.; Napier, M.; Brick, L.; Brady, L.; Carere, D.A.; Levy, M.A.; et al. Diagnostic Utility of Genome-Wide DNA Methylation Testing in Genetically Unsolved Individuals with Suspected Hereditary Conditions. Am. J. Hum. Genet. 2019, 104, 685–700. [Google Scholar] [CrossRef]
Strong, E.; Butcher, D.T.; Singhania, R.; Mervis, C.B.; Morris, C.A.; De Carvalho, D.; Weksberg, R.; Osborne, L.R. Symmetrical Dose-Dependent DNA-Methylation Profiles in Children with Deletion or Duplication of 7q11.23. Am. J. Hum. Genet. 2015, 97, 216–227. [Google Scholar] [CrossRef]
Levy, M.A.; Relator, R.; McConkey, H.; Pranckeviciene, E.; Kerkhof, J.; Barat-Houari, M.; Bargiacchi, S.; Biamino, E.; Palomares Bralo, M.; Cappuccio, G.; et al. Functional Correlation of Genome-Wide DNA Methylation Profiles in Genetic Neurodevelopmental Disorders. Hum. Mutat. 2022, 43, 1609–1628. [Google Scholar] [CrossRef]
Rots, D.; Rooney, K.; Relator, R.; Kerkhof, J.; McConkey, H.; Pfundt, R.; Marcelis, C.; Willemsen, M.H.; van Hagen, J.M.; Zwijnenburg, P.; et al. Refining the 9q34.3 Microduplication Syndrome Reveals Mild Neurodevelopmental Features Associated with a Distinct Global DNA Methylation Profile. Clin. Genet. 2024, 105, 655–660. [Google Scholar] [CrossRef]
Levy, M.A.; McConkey, H.; Kerkhof, J.; Barat-Houari, M.; Bargiacchi, S.; Biamino, E.; Bralo, M.P.; Cappuccio, G.; Ciolfi, A.; Clarke, A.; et al. Novel Diagnostic DNA Methylation Episignatures Expand and Refine the Epigenetic Landscapes of Mendelian Disorders. Hum. Genet. Genom. Adv. 2022, 3, 100075. [Google Scholar] [CrossRef]
Rooney, K.; Levy, M.A.; Haghshenas, S.; Kerkhof, J.; Rogaia, D.; Tedesco, M.G.; Imperatore, V.; Mencarelli, A.; Squeo, G.M.; Di Venere, E.; et al. Identification of a Dna Methylation Episignature in the 22q11.2 Deletion Syndrome. Int. J. Mol. Sci. 2021, 22, 8611. [Google Scholar] [CrossRef] [PubMed]
Carmel, M.; Michaelovsky, E.; Weinberger, R.; Frisch, A.; Mekori-Domachevsky, E.; Gothelf, D.; Weizman, A. Differential Methylation of Imprinting Genes and MHC Locus in 22q11.2 Deletion Syndrome-Related Schizophrenia Spectrum Disorders. World J. Biol. Psychiatry 2021, 22, 46–57. [Google Scholar] [CrossRef] [PubMed]
Garau, J.; Charras, A.; Varesio, C.; Orcesi, S.; Dragoni, F.; Galli, J.; Fazzi, E.; Gagliardi, S.; Pansarasa, O.; Cereda, C.; et al. Altered DNA Methylation and Gene Expression Predict Disease Severity in Patients with Aicardi-Goutières Syndrome. Clin. Immunol. 2023, 249, 109299. [Google Scholar] [CrossRef] [PubMed]
Schenkel, L.C.; Kernohan, K.D.; McBride, A.; Reina, D.; Hodge, A.; Ainsworth, P.J.; Rodenhiser, D.I.; Pare, G.; Bérubé, N.G.; Skinner, C.; et al. Identification of Epigenetic Signature Associated with Alpha Thalassemia/Mental Retardation X-Linked Syndrome. Epigenetics Chromatin 2017, 10, 10. [Google Scholar] [CrossRef]
Aref-Eshghi, E.; Rodenhiser, D.I.; Schenkel, L.C.; Lin, H.; Skinner, C.; Ainsworth, P.; Paré, G.; Hood, R.L.; Bulman, D.E.; Kernohan, K.D.; et al. Genomic DNA Methylation Signatures Enable Concurrent Diagnosis and Clinical Genetic Variant Classification in Neurodevelopmental Syndromes. Am. J. Hum. Genet. 2018, 102, 156–174. [Google Scholar] [CrossRef]
Houdayer, C.; Rooney, K.; van der Laan, L.; Bris, C.; Alders, M.; Bahr, A.; Barcia, G.; Battault, C.; Begemann, A.; Bonneau, D.; et al. ARID2-Related Disorder: Further Delineation of the Clinical Phenotype of 27 Novel Individuals and Description of an Epigenetic Signature. Eur. J. Hum. Genet. 2025, 10, 1422–1431. [Google Scholar] [CrossRef]
Carvalho, G.F.d.S.; Costa, T.V.M.M.; Nascimento, A.M.; Wolff, B.M.; Damasceno, J.G.; Vieira, L.L.; Almeida, V.T.; de Oliveira, Y.G.; de Mello, C.B.; Muszkat, M.; et al. DNA Methylation Epi-Signature and Biological Age in Attention Deficit Hyperactivity Disorder Patients. Clin. Neurol. Neurosurg. 2023, 228, 107714. [Google Scholar] [CrossRef]
Choufani, S.; McNiven, V.; Cytrynbaum, C.; Jangjoo, M.; Adam, M.P.; Bjornsson, H.T.; Harris, J.; Dyment, D.A.; Graham, G.E.; Nezarati, M.M.; et al. An HNRNPK-Specific DNA Methylation Signature Makes Sense of Missense Variants and Expands the Phenotypic Spectrum of Au-Kline Syndrome. Am. J. Hum. Genet. 2022, 109, 1867–1884, Erratum in Am. J. Hum. Genet. 2025, 112, 1979. [Google Scholar] [CrossRef]
Siu, M.T.; Butcher, D.T.; Turinsky, A.L.; Cytrynbaum, C.; Stavropoulos, D.J.; Walker, S.; Caluseriu, O.; Carter, M.; Lou, Y.; Nicolson, R.; et al. Functional DNA Methylation Signatures for Autism Spectrum Disorder Genomic Risk Loci: 16p11.2 Deletions and CHD8 Variants. Clin. Epigenet. 2019, 11, 103. [Google Scholar] [CrossRef]
Garg, P.; Sharp, A.J. Screening for Rare Epigenetic Variations in Autism and Schizophrenia. Hum. Mutat. 2019, 40, 952–961. [Google Scholar] [CrossRef]
van der Sluijs, P.J.; Moutton, S.; Dingemans, A.J.M.; Weis, D.; Levy, M.A.; Boycott, K.M.; Arberas, C.; Baldassarri, M.; Beneteau, C.; Brusco, A.; et al. Microduplications of ARID1A and ARID1B Cause a Novel Clinical and Epigenetic Distinct BAFopathy. Genet. Med. 2025, 27, 101283. [Google Scholar] [CrossRef] [PubMed]
Qannan, A.; Bejaoui, Y.; Izadi, M.; Yousri, N.A.; Razzaq, A.; Christiansen, C.; Martin, G.M.; Bell, J.T.; Horvath, S.; Oshima, J.; et al. Accelerated Epigenetic Aging and DNA Methylation Alterations in Berardinelli-Seip Congenital Lipodystrophy. Hum. Mol. Genet. 2023, 32, 1826–1835. [Google Scholar] [CrossRef] [PubMed]
Sabbagh, Q.; Haghshenas, S.; Piard, J.; Trouvé, C.; Amiel, J.; Attié-Bitach, T.; Balci, T.; Barat-Houari, M.; Belonis, A.; Boute, O.; et al. Clinico-Biological Refinement of BCL11B-Related Disorder and Identification of an Episignature: A Series of 20 Unreported Individuals. Genet. Med. 2024, 26, 101007. [Google Scholar] [CrossRef] [PubMed]
Levy, M.A.; Beck, D.B.; Metcalfe, K.; Douzgou, S.; Sithambaram, S.; Cottrell, T.; Ansar, M.; Kerkhof, J.; Mignot, C.; Nougues, M.C.; et al. Deficiency of TET3 Leads to a Genome-Wide DNA Hypermethylation Episignature in Human Whole Blood. NPJ Genom. Med. 2021, 6, 92, Erratum in NPJ Genom. Med. 2021, 6, 100. [Google Scholar] [CrossRef]
Cappuccio, G.; Sayou, C.; Tanno, P.L.; Tisserant, E.; Bruel, A.L.; Kennani, S.E.; Sá, J.; Low, K.J.; Dias, C.; Havlovicová, M.; et al. De Novo SMARCA2 Variants Clustered Outside the Helicase Domain Cause a New Recognizable Syndrome with Intellectual Disability and Blepharophimosis Distinct from Nicolaides–Baraitser Syndrome. Genet. Med. 2020, 22, 1838–1850. [Google Scholar] [CrossRef]
Sarli, C.; van der Laan, L.; Reilly, J.; Trajkova, S.; Carli, D.; Brusco, A.; Levy, M.A.; Relator, R.; Kerkhof, J.; McConkey, H.; et al. Blepharophimosis with Intellectual Disability and Helsmoortel-Van Der Aa Syndrome Share Episignature and Phenotype. Am. J. Med. Genet. C Semin. Med. Genet. 2024, 196, e32089. [Google Scholar] [CrossRef]
Awamleh, Z.; Chater-Diehl, E.; Choufani, S.; Wei, E.; Kianmahd, R.R.; Yu, A.; Chad, L.; Costain, G.; Tan, W.H.; Scherer, S.W.; et al. DNA Methylation Signature Associated with Bohring-Opitz Syndrome: A New Tool for Functional Classification of Variants in ASXL Genes. Eur. J. Hum. Genet. 2022, 30, 695–702. [Google Scholar] [CrossRef]
Rouxel, F.; Relator, R.; Kerkhof, J.; McConkey, H.; Levy, M.; Dias, P.; Barat-Houari, M.; Bednarek, N.; Boute, O.; Chatron, N.; et al. CDK13-Related Disorder: Report of a Series of 18 Previously Unpublished Individuals and Description of an Epigenetic Signature. Genet. Med. 2022, 24, 1096–1107. [Google Scholar] [CrossRef]
Kernohan, K.D.; Cigana Schenkel, L.; Huang, L.; Smith, A.; Pare, G.; Ainsworth, P.; Care4Rare Canada Consortium; Boycott, K.M.; Warman-Chardon, J.; Sadikovic, B. Identification of a Methylation Profile for DNMT1-Associated Autosomal Dominant Cerebellar Ataxia, Deafness, and Narcolepsy. Clin. Epigenet. 2016, 8, 91. [Google Scholar] [CrossRef]
Peymani, F.; Ebihara, T.; Smirnov, D.; Kopajtich, R.; Ando, M.; Bertini, E.; Carrozzo, R.; Diodato, D.; Distelmaier, F.; Fang, F.; et al. Pleiotropic Effects of MORC2 Derive from Its Epigenetic Signature. Brain 2025, 9, awaf159. [Google Scholar] [CrossRef]
Butcher, D.T.; Cytrynbaum, C.; Turinsky, A.L.; Siu, M.T.; Inbar-Feigenberg, M.; Mendoza-Londono, R.; Chitayat, D.; Walker, S.; Machado, J.; Caluseriu, O.; et al. CHARGE and Kabuki Syndromes: Gene-Specific DNA Methylation Signatures Identify Epigenetic Mechanisms Linking These Clinically Overlapping Conditions. Am. J. Hum. Genet. 2017, 100, 773–788. [Google Scholar] [CrossRef]
Vos, N.; Haghshenas, S.; van der Laan, L.; Russel, P.K.M.; Rooney, K.; Levy, M.A.; Relator, R.; Kerkhof, J.; McConkey, H.; Maas, S.M.; et al. The Detection of a Strong Episignature for Chung–Jansen Syndrome, Partially Overlapping with Börjeson–Forssman–Lehmann and White–Kernohan Syndromes. Hum. Genet. 2024, 143, 761–773. [Google Scholar] [CrossRef] [PubMed]
Schenkel, L.C.; Aref-Eshghi, E.; Skinner, C.; Ainsworth, P.; Lin, H.; Paré, G.; Rodenhiser, D.I.; Schwartz, C.; Sadikovic, B. Peripheral Blood Epi-Signature of Claes-Jensen Syndrome Enables Sensitive and Specific Identification of Patients and Healthy Carriers with Pathogenic Mutations in KDM5C. Clin. Epigenet. 2018, 10, 21. [Google Scholar] [CrossRef]
van der Laan, L.; Rooney, K.; Alders, M.; Relator, R.; McConkey, H.; Kerkhof, J.; Levy, M.A.; Lauffer, P.; Aerden, M.; Theunis, M.; et al. Episignature Mapping of TRIP12 Provides Functional Insight into Clark-Baraitser Syndrome. Int. J. Mol. Sci. 2022, 23, 13664. [Google Scholar] [CrossRef] [PubMed]
Aref-Eshghi, E.; Bend, E.G.; Hood, R.L.; Schenkel, L.C.; Carere, D.A.; Chakrabarti, R.; Nagamani, S.C.S.; Cheung, S.W.; Campeau, P.M.; Prasad, C.; et al. BAFopathies’ DNA Methylation Epi-Signatures Demonstrate Diagnostic Utility and Functional Continuum of Coffin–Siris and Nicolaides–Baraitser Syndromes. Nat. Commun. 2018, 9, 4885. [Google Scholar] [CrossRef] [PubMed]
Al-Jawahiri, R.; Foroutan, A.; Kerkhof, J.; McConkey, H.; Levy, M.; Haghshenas, S.; Rooney, K.; Turner, J.; Shears, D.; Holder, M.; et al. SOX11 Variants Cause a Neurodevelopmental Disorder with Infrequent Ocular Malformations and Hypogonadotropic Hypogonadism and with Distinct DNA Methylation Profile. Genet. Med. 2022, 24, 1261–1273. [Google Scholar] [CrossRef]
LaFlamme, C.W.; Rastin, C.; Sengupta, S.; Pennington, H.E.; Russ-Hall, S.J.; Schneider, A.L.; Bonkowski, E.S.; Almanza Fuerte, E.P.; Allan, T.J.; Zalusky, M.P.G.; et al. Diagnostic Utility of DNA Methylation Analysis in Genetically Unsolved Pediatric Epilepsies and CHD2 Episignature Refinement. Nat. Commun. 2024, 15, 6524. [Google Scholar] [CrossRef]
Rooney, K.; van der Laan, L.; Trajkova, S.; Haghshenas, S.; Relator, R.; Lauffer, P.; Vos, N.; Levy, M.A.; Brunetti-Pierri, N.; Terrone, G.; et al. DNA Methylation Episignature and Comparative Epigenomic Profiling of HNRNPU-Related Neurodevelopmental Disorder. Genet. Med. 2023, 25, 100871. [Google Scholar] [CrossRef]
Karimi, K.; Weis, D.; Aukrust, I.; Hsieh, T.C.; Horackova, M.; Paulsen, J.; Mendoza Londono, R.; Dupuis, L.; Dickson, M.; Lesman, H.; et al. Epigenomic and Phenotypic Characterization of DEGCAGS Syndrome. Eur. J. Hum. Genet. 2024, 32, 1574–1582. [Google Scholar] [CrossRef]
Nil, Z.; Deshwar, A.R.; Huang, Y.; Barish, S.; Zhang, X.; Choufani, S.; Le Quesne Stabej, P.; Hayes, I.; Yap, P.; Haldeman-Englert, C.; et al. Rare de Novo Gain-of-Function Missense Variants in DOT1L Are Associated with Developmental Delay and Congenital Anomalies. Am. J. Hum. Genet. 2023, 110, 1919. [Google Scholar] [CrossRef]
Bacalini, M.G.; Gentilini, D.; Boattini, A.; Giampieri, E.; Pirazzini, C.; Giuliani, C.; Fontanesi, E.; Scurti, M.; Remondini, D.; Capri, M.; et al. Identification of a DNA Methylation Signature in Blood Cells from Persons with Down Syndrome. Aging 2015, 7, 82–96. [Google Scholar] [CrossRef] [PubMed]
Schreyer, L.; Reilly, J.; McConkey, H.; Kerkhof, J.; Levy, M.A.; Hu, J.; Hnaini, M.; Sadikovic, B.; Campbell, C. The Discovery of the DNA Methylation Episignature for Duchenne Muscular Dystrophy. Neuromuscul. Disord. 2023, 33, 5–14. [Google Scholar] [CrossRef] [PubMed]
Courraud, J.; Chater-Diehl, E.; Durand, B.; Vincent, M.; del Mar Muniz Moreno, M.; Boujelbene, I.; Drouot, N.; Genschik, L.; Schaefer, E.; Nizon, M.; et al. Integrative Approach to Interpret DYRK1A Variants, Leading to a Frequent Neurodevelopmental Disorder. Genet. Med. 2021, 23, 2150–2159. [Google Scholar] [CrossRef] [PubMed]
Ciolfi, A.; Foroutan, A.; Capuano, A.; Pedace, L.; Travaglini, L.; Pizzi, S.; Andreani, M.; Miele, E.; Invernizzi, F.; Reale, C.; et al. Childhood-Onset Dystonia-Causing KMT2B Variants Result in a Distinctive Genomic Hypermethylation Profile. Clin. Epigenet. 2021, 13, 157. [Google Scholar] [CrossRef]
Oexle, K.; Zech, M.; Stühn, L.G.; Siegert, S.; Brunet, T.; Schmidt, W.M.; Wagner, M.; Schmidt, A.; Engels, H.; Tilch, E.; et al. Episignature Analysis of Moderate Effects and Mosaics. Eur. J. Hum. Genet. 2023, 31, 1032. [Google Scholar] [CrossRef]
Lee, S.; Ochoa, E.; Barwick, K.; Cif, L.; Rodger, F.; Docquier, F.; Pérez-Dueñas, B.; Clark, G.; Martin, E.; Banka, S.; et al. Comparison of Methylation Episignatures in KMT2B- and KMT2D-Related Human Disorders. Epigenomics 2022, 14, 537–547. [Google Scholar] [CrossRef]
Mirza-Schreiber, N.; Zech, M.; Wilson, R.; Brunet, T.; Wagner, M.; Jech, R.; Boesch, S.; Škorvánek, M.; Necpál, J.; Weise, D.; et al. Blood DNA Methylation Provides an Accurate Biomarker of KMT2B-Related Dystonia and Predicts Onset. Brain 2022, 145, 644–654. [Google Scholar] [CrossRef]
Oussalah, A.; Siblini, Y.; Hergalant, S.; Chéry, C.; Rouyer, P.; Cavicchi, C.; Guerrini, R.; Morange, P.E.; Trégouët, D.; Pupavac, M.; et al. Epimutations in Both the TESK2 and MMACHC Promoters in the Epi-CblC Inherited Disorder of Intracellular Metabolism of Vitamin B12. Clin. Epigenet. 2022, 14, 52. [Google Scholar] [CrossRef]
Pagliara, D.; Ciolfi, A.; Pedace, L.; Haghshenas, S.; Ferilli, M.; Levy, M.A.; Miele, E.; Nardini, C.; Cappelletti, C.; Relator, R.; et al. Identification of a Robust DNA Methylation Signature for Fanconi Anemia. Am. J. Hum. Genet. 2023, 110, 1938. [Google Scholar] [CrossRef]
Cobben, J.M.; Krzyzewska, I.M.; Venema, A.; Mul, A.N.; Polstra, A.; Postma, A.V.; Smigiel, R.; Pesz, K.; Niklinski, J.; Chomczyk, M.A.; et al. DNA Methylation Abundantly Associates with Fetal Alcohol Spectrum Disorder and Its Subphenotypes. Epigenomics 2019, 11, 767–785. [Google Scholar] [CrossRef]
Krzyzewska, I.M.; Lauffer, P.; Mul, A.N.; van der Laan, L.; Yim, A.Y.F.L.; Cobben, J.M.; Niklinski, J.; Chomczyk, M.A.; Smigiel, R.; Mannens, M.M.A.M.; et al. Expression Quantitative Trait Methylation Analysis Identifies Whole Blood Molecular Footprint in Fetal Alcohol Spectrum Disorder (FASD). Int. J. Mol. Sci. 2023, 24, 6601. [Google Scholar] [CrossRef] [PubMed]
van der Laan, L.; Relator, R.; Valenzuela, I.; Mul, A.N.; Alders, M.; Levy, M.A.; Kerkhof, J.; Rzasa, J.; Cueto-González, A.M.; Lasa-Aranzasti, A.; et al. Discovery of a DNA Methylation Episignature as a Molecular Biomarker for Fetal Alcohol Syndrome. Genet. Med. 2025, 27, 101586. [Google Scholar] [CrossRef] [PubMed]
Haghshenas, S.; Putoux, A.; Reilly, J.; Levy, M.A.; Relator, R.; Ghosh, S.; Kerkhof, J.; McConkey, H.; Edery, P.; Lesca, G.; et al. Discovery of DNA Methylation Signature in the Peripheral Blood of Individuals with History of Antenatal Exposure to Valproic Acid. Genet. Med. 2024, 26, 101226. [Google Scholar] [CrossRef] [PubMed]
Hood, R.L.; Schenkel, L.C.; Nikkel, S.M.; Ainsworth, P.J.; Pare, G.; Boycott, K.M.; Bulman, D.E.; Sadikovic, B. The Defining DNA Methylation Signature of Floating-Harbor Syndrome. Sci. Rep. 2016, 6, 38803. [Google Scholar] [CrossRef]
Rots, D.; Chater-Diehl, E.; Dingemans, A.J.M.; Goodman, S.J.; Siu, M.T.; Cytrynbaum, C.; Choufani, S.; Hoang, N.; Walker, S.; Awamleh, Z.; et al. Truncating SRCAP Variants Outside the Floating-Harbor Syndrome Locus Cause a Distinct Neurodevelopmental Disorder with a Specific DNA Methylation Signature. Am. J. Hum. Genet. 2021, 108, 1053–1068. [Google Scholar] [CrossRef]
Schenkel, L.C.; Schwartz, C.; Skinner, C.; Rodenhiser, D.I.; Ainsworth, P.J.; Pare, G.; Sadikovic, B. Clinical Validation of Fragile X Syndrome Screening by DNA Methylation Array. J. Mol. Diagn. 2016, 18, 834–841. [Google Scholar] [CrossRef]
Li, Y.; Chen, J.A.; Sears, R.L.; Gao, F.; Klein, E.D.; Karydas, A.; Geschwind, M.D.; Rosen, H.J.; Boxer, A.L.; Guo, W.; et al. An Epigenetic Signature in Peripheral Blood Associated with the Haplotype on 17q21.31, a Risk Factor for Neurodegenerative Tauopathy. PLoS Genet. 2014, 10, e1004211. [Google Scholar] [CrossRef]
Cherik, F.; Reilly, J.; Kerkhof, J.; Levy, M.; McConkey, H.; Barat-Houari, M.; Butler, K.M.; Coubes, C.; Lee, J.A.; Le Guyader, G.; et al. DNA Methylation Episignature in Gabriele-de Vries Syndrome. Genet. Med. 2022, 24, 905–914. [Google Scholar] [CrossRef]
van der Laan, L.; Karimi, K.; Rooney, K.; Lauffer, P.; McConkey, H.; Caro, P.; Relator, R.; Levy, M.A.; Bhai, P.; Mignot, C.; et al. DNA Methylation Episignature, Extension of the Clinical Features, and Comparative Epigenomic Profiling of Hao-Fountain Syndrome Caused by Variants in USP7. Genet. Med. 2024, 26, 101050. [Google Scholar] [CrossRef]
Breen, M.S.; Garg, P.; Tang, L.; Mendonca, D.; Levy, T.; Barbosa, M.; Arnett, A.B.; Kurtz-Nelson, E.; Agolini, E.; Battaglia, A.; et al. Episignatures Stratifying Helsmoortel-Van Der Aa Syndrome Show Modest Correlation with Phenotype. Am. J. Hum. Genet. 2020, 107, 555–563. [Google Scholar] [CrossRef]
Bend, E.G.; Aref-Eshghi, E.; Everman, D.B.; Rogers, R.C.; Cathey, S.S.; Prijoles, E.J.; Lyons, M.J.; Davis, H.; Clarkson, K.; Gripp, K.W.; et al. Gene Domain-Specific DNA Methylation Episignatures Highlight Distinct Molecular Entities of ADNP Syndrome. Clin. Epigenet. 2019, 11, 64. [Google Scholar] [CrossRef] [PubMed]
Lee, S.; Ochoa, E.; Badura-Stronka, M.; Donnelly, D.; Lederer, D.; Lynch, S.A.; Gardham, A.; Morton, J.; Stewart, H.; Docquier, F.; et al. Germline Pathogenic Variants in HNRNPU Are Associated with Alterations in Blood Methylome. Eur. J. Hum. Genet. 2023, 31, 1040–1047. [Google Scholar] [CrossRef] [PubMed]
Velasco, G.; Grillo, G.; Touleimat, N.; Ferry, L.; Ivkovic, I.; Ribierre, F.; Deleuze, J.F.; Chantalat, S.; Picard, C.; Francastel, C. Comparative Methylome Analysis of ICF Patients Identifies Heterochromatin Loci That Require ZBTB24, CDCA7 and HELLS for Their Methylated State. Hum. Mol. Genet. 2018, 27, 2409–2424. [Google Scholar] [CrossRef]
Velasco, G.; Ulveling, D.; Rondeau, S.; Marzin, P.; Unoki, M.; Cormier-Daire, V.; Francastel, C. Interplay between Histone and Dna Methylation Seen through Comparative Methylomes in Rare Mendelian Disorders. Int. J. Mol. Sci. 2021, 22, 3735. [Google Scholar] [CrossRef] [PubMed]
Karimi, K.; Mol, M.O.; Haghshenas, S.; Relator, R.; Levy, M.A.; Kerkhof, J.; McConkey, H.; Brooks, A.; Zonneveld-Huijssoon, E.; Gerkes, E.H.; et al. Identification of DNA Methylation Episignature for the Intellectual Developmental Disorder, Autosomal Dominant 21 Syndrome, Caused by Variants in the CTCF Gene. Genet. Med. 2024, 26, 101041. [Google Scholar] [CrossRef]
Verberne, E.A.; van der Laan, L.; Haghshenas, S.; Rooney, K.; Levy, M.A.; Alders, M.; Maas, S.M.; Jansen, S.; Lieden, A.; Anderlid, B.M.; et al. DNA Methylation Signature for JARID2-Neurodevelopmental Syndrome. Int. J. Mol. Sci. 2022, 23, 8001. [Google Scholar] [CrossRef]
van der Laan, L.; Rooney, K.; Haghshenas, S.; Silva, A.; McConkey, H.; Relator, R.; Levy, M.A.; Valenzuela, I.; Trujillano, L.; Lasa-Aranzasti, A.; et al. Functional Insight into and Refinement of the Genomic Boundaries of the JARID2-Neurodevelopmental Disorder Episignature. Int. J. Mol. Sci. 2023, 24, 14240. [Google Scholar] [CrossRef]
Kagami, M.; Matsubara, K.; Nakabayashi, K.; Nakamura, A.; Sano, S.; Okamura, K.; Hata, K.; Fukami, M.; Ogata, T. Genome-Wide Multilocus Imprinting Disturbance Analysis in Temple Syndrome and Kagami-Ogata Syndrome. Genet. Med. 2017, 19, 476–482. [Google Scholar] [CrossRef]
Aref-Eshghi, E.; Schenkel, L.C.; Lin, H.; Skinner, C.; Ainsworth, P.; Paré, G.; Rodenhiser, D.; Schwartz, C.; Sadikovic, B. The Defining DNA Methylation Signature of Kabuki Syndrome Enables Functional Assessment of Genetic Variants of Unknown Clinical Significance. Epigenetics 2017, 12, 923–933. [Google Scholar] [CrossRef]
Sobreira, N.; Brucato, M.; Zhang, L.; Ladd-Acosta, C.; Ongaco, C.; Romm, J.; Doheny, K.F.; Mingroni-Netto, R.C.; Bertola, D.; Kim, C.A.; et al. Patients with a Kabuki Syndrome Phenotype Demonstrate DNA Methylation Abnormalities. Eur. J. Hum. Genet. 2017, 25, 1335–1344. [Google Scholar] [CrossRef]
Giuili, E.; Grolaux, R.; Macedo, C.Z.N.M.; Desmyter, L.; Pichon, B.; Neuens, S.; Vilain, C.; Olsen, C.; Van Dooren, S.; Smits, G.; et al. Comprehensive Evaluation of the Implementation of Episignatures for Diagnosis of Neurodevelopmental Disorders (NDDs). Hum. Genet. 2023, 142, 1721. [Google Scholar] [CrossRef] [PubMed]
Awamleh, Z.; Choufani, S.; Cytrynbaum, C.; Alkuraya, F.S.; Scherer, S.; Fernandes, S.; Rosas, C.; Louro, P.; Dias, P.; Neves, M.T.; et al. ANKRD11 Pathogenic Variants and 16q24.3 Microdeletions Share an Altered DNA Methylation Signature in Patients with KBG Syndrome. Hum. Mol. Genet. 2023, 32, 1429–1438. [Google Scholar] [CrossRef]
van Jaarsveld, R.H.; Reilly, J.; Cornips, M.C.; Hadders, M.A.; Agolini, E.; Ahimaz, P.; Anyane-Yeboa, K.; Bellanger, S.A.; van Binsbergen, E.; van den Boogaard, M.J.; et al. Delineation of a KDM2B-Related Neurodevelopmental Disorder and Its Associated DNA Methylation Signature. Genet. Med. 2023, 25, 49–62. [Google Scholar] [CrossRef] [PubMed]
Goodman, S.J.; Cytrynbaum, C.; Chung, B.H.Y.; Chater-Diehl, E.; Aziz, C.; Turinsky, A.L.; Kellam, B.; Keller, M.; Ko, J.M.; Caluseriu, O.; et al. EHMT1 Pathogenic Variants and 9q34.3 Microdeletions Share Altered DNA Methylation Patterns in Patients with Kleefstra Syndrome. J. Transl. Genet. Genom. 2020, 4, 144–158. [Google Scholar] [CrossRef]
Rots, D.; Choufani, S.; Faundes, V.; Dingemans, A.J.M.; Joss, S.; Foulds, N.; Jones, E.A.; Stewart, S.; Vasudevan, P.; Dabir, T.; et al. Pathogenic Variants in KMT2C Result in a Neurodevelopmental Disorder Distinct from Kleefstra and Kabuki Syndromes. Am. J. Hum. Genet. 2024, 111, 1626. [Google Scholar] [CrossRef]
Awamleh, Z.; Choufani, S.; Wu, W.; Rots, D.; Dingemans, A.J.M.; Nadif Kasri, N.; Boronat, S.; Ibañez-Mico, S.; Cuesta Herraiz, L.; Ferrer, I.; et al. A New Blood DNA Methylation Signature for Koolen-de Vries Syndrome: Classification of Missense KANSL1 Variants and Comparison to Fibroblast Cells. Eur. J. Hum. Genet. 2024, 32, 324–332, Erratum in Eur. J. Hum. Genet. 2024, 32, 366. [Google Scholar] [CrossRef]
Lee, S.; Menzies, L.; Hay, E.; Ochoa, E.; Docquier, F.; Rodger, F.; Deshpande, C.; Foulds, N.C.; Jacquemont, S.; Jizi, K.; et al. Epigenotype-Genotype-Phenotype Correlations in SETD1A and SETD2 Chromatin Disorders. Hum. Mol. Genet. 2023, 32, 3123–3134. [Google Scholar] [CrossRef]
Dámaso, E.; Castillejo, A.; Arias, M.d.M.; Canet-Hermida, J.; Navarro, M.; del Valle, J.; Campos, O.; Fernández, A.; Marín, F.; Turchetti, D.; et al. Primary Constitutional MLH1 Epimutations: A Focal Epigenetic Event. Br. J. Cancer 2018, 119, 978–987. [Google Scholar] [CrossRef]
Haghshenas, S.; Bout, H.J.; Schijns, J.M.; Levy, M.A.; Kerkhof, J.; Bhai, P.; McConkey, H.; Jenkins, Z.A.; Williams, E.M.; Halliday, B.J.; et al. Menke-Hennekam Syndrome; Delineation of Domain-Specific Subtypes with Distinct Clinical and DNA Methylation Profiles. Hum. Genet. Genom. Adv. 2024, 5, 100287, Erratum in Hum. Genet. Genom. Adv. 2024, 5, 100337. [Google Scholar] [CrossRef]
Caraffi, S.G.; van der Laan, L.; Rooney, K.; Trajkova, S.; Zuntini, R.; Relator, R.; Haghshenas, S.; Levy, M.A.; Baldo, C.; Mandrile, G.; et al. Identification of the DNA Methylation Signature of Mowat-Wilson Syndrome. Eur. J. Hum. Genet. 2024, 32, 619–629. [Google Scholar] [CrossRef]
Karayol, R.; Borroto, M.C.; Haghshenas, S.; Namasivayam, A.; Reilly, J.; Levy, M.A.; Relator, R.; Kerkhof, J.; McConkey, H.; Shvedunova, M.; et al. MSL2 Variants Lead to a Neurodevelopmental Syndrome with Lack of Coordination, Epilepsy, Specific Dysmorphisms, and a Distinct Episignature. Am. J. Hum. Genet. 2024, 111, 1330–1351. [Google Scholar] [CrossRef]
van der Laan, L.; Silva, A.; Kleinendorst, L.; Rooney, K.; Haghshenas, S.; Lauffer, P.; Alanay, Y.; Bhai, P.; Brusco, A.; de Munnik, S.; et al. CUL3-Related Neurodevelopmental Disorder: Clinical Phenotype of 20 New Individuals and Identification of a Potential Phenotype-Associated Episignature. Hum. Genet. Genom. Adv. 2025, 6, 100380. [Google Scholar] [CrossRef] [PubMed]
Chater-Diehl, E.; Ejaz, R.; Cytrynbaum, C.; Siu, M.T.; Turinsky, A.; Choufani, S.; Goodman, S.J.; Abdul-Rahman, O.; Bedford, M.; Dorrani, N.; et al. New Insights into DNA Methylation Signatures: SMARCA2 Variants in Nicolaides-Baraitser Syndrome. BMC Med. Genom. 2019, 12, 105. [Google Scholar] [CrossRef] [PubMed]
Schenkel, L.C.; Aref-Eshghi, E.; Rooney, K.; Kerkhof, J.; Levy, M.A.; McConkey, H.; Rogers, R.C.; Phelan, K.; Sarasua, S.M.; Jain, L.; et al. DNA Methylation Epi-Signature Is Associated with Two Molecularly and Phenotypically Distinct Clinical Subtypes of Phelan-McDermid Syndrome. Clin. Epigenet. 2021, 13, 2. [Google Scholar] [CrossRef] [PubMed]
van der Laan, L.; Lauffer, P.; Rooney, K.; Silva, A.; Haghshenas, S.; Relator, R.; Levy, M.A.; Trajkova, S.; Huisman, S.A.; Bijlsma, E.K.; et al. DNA Methylation Episignature and Comparative Epigenomic Profiling for Pitt-Hopkins Syndrome Caused by TCF4 Variants. Hum. Genet. Genom. Adv. 2024, 5, 100289. [Google Scholar] [CrossRef]
Hara-Isono, K.; Matsubara, K.; Fuke, T.; Yamazawa, K.; Satou, K.; Murakami, N.; Saitoh, S.; Nakabayashi, K.; Hata, K.; Ogata, T.; et al. Genome-Wide Methylation Analysis in Silver–Russell Syndrome, Temple Syndrome, and Prader–Willi Syndrome. Clin. Epigenet. 2020, 12, 159. [Google Scholar] [CrossRef]
Masson, A.; Paccaud, J.; Orefice, M.; Colin, E.; Mäkitie, O.; Cormier-Daire, V.; Relator, R.; Ghosh, S.; Strub, J.-M.; Schaeffer-Reiss, C.; et al. PTBP1 Variants Displaying Altered Nucleocytoplasmic Distribution Are Responsible for a Neurodevelopmental Disorder with Skeletal Dysplasia. J. Clin. Investig. 2025, 40, e182100. [Google Scholar] [CrossRef]
Xiao, B.; Dai, W.; Zhan, Y.; Qiu, W.; Zhang, H.; Liu, D.P.; Xu, N.; Yu, Y. Genome-Wide Epigenetic Signatures Facilitated the Variant Classification of the PURA Gene and Uncovered the Pathomechanism of PURA-Related Neurodevelopmental Disorders. Genet. Med. 2024, 26, 101167. [Google Scholar] [CrossRef]
Ciolfi, A.; Aref-Eshghi, E.; Pizzi, S.; Pedace, L.; Miele, E.; Kerkhof, J.; Flex, E.; Martinelli, S.; Radio, F.C.; Ruivenkamp, C.A.L.; et al. Frameshift Mutations at the C-Terminus of HIST1H1E Result in a Specific DNA Hypomethylation Signature. Clin. Epigenet. 2020, 12, 7. [Google Scholar] [CrossRef]
Haghshenas, S.; Karimi, K.; Stevenson, R.E.; Levy, M.A.; Relator, R.; Kerkhof, J.; Rzasa, J.; McConkey, H.; Lauzon-Young, C.; Balci, T.B.; et al. Identification of a DNA Methylation Episignature for Recurrent Constellations of Embryonic Malformations. Am. J. Hum. Genet. 2024, 111, 1643. [Google Scholar] [CrossRef]
Haghshenas, S.; Foroutan, A.; Bhai, P.; Levy, M.A.; Relator, R.; Kerkhof, J.; McConkey, H.; Skinner, C.D.; Caylor, R.C.; Tedder, M.L.; et al. Identification of a DNA Methylation Signature for Renpenning Syndrome (RENS1), a Spliceopathy. Eur. J. Hum. Genet. 2023, 31, 879–886. [Google Scholar] [CrossRef] [PubMed]
Nava, C.; Cogne, B.; Santini, A.; Leitão, E.; Lecoquierre, F.; Chen, Y.; Stenton, S.L.; Besnard, T.; Heide, S.; Baer, S.; et al. Dominant Variants in Major Spliceosome U4 and U5 Small Nuclear RNA Genes Cause Neurodevelopmental Disorders through Splicing Disruption. Nat. Genet. 2025, 57, 1374–1388. [Google Scholar] [CrossRef] [PubMed]
Sevilla-Porras, M.; Nieto-Molina, E.; Medina, Z.; Damián, A.; Tejedor, J.R.; Ruiz-Arenas, C.; Cazalla, M.; Carmona, R.; Amigo, J.; Goicoechea, I.; et al. Characterization of SnRNA-Related Neurodevelopmental Disorders through the Spanish Undiagnosed Rare Disease Programs. medRxiv 2025. [Google Scholar] [CrossRef]
Krzyzewska, I.M.; Maas, S.M.; Henneman, P.; Lip, K.V.D.; Venema, A.; Baranano, K.; Chassevent, A.; Aref-Eshghi, E.; Van Essen, A.J.; Fukuda, T.; et al. A Genome-Wide DNA Methylation Signature for SETD1B-Related Syndrome. Clin. Epigenet. 2019, 11, 156. [Google Scholar] [CrossRef]
Karimi, K.; Lichtenstein, Y.; Reilly, J.; McConkey, H.; Relator, R.; Levy, M.A.; Kerkhof, J.; Bouman, A.; Symonds, J.D.; Ghoumid, J.; et al. Discovery of a DNA Methylation Profile in Individuals with Sifrim-Hitz-Weiss Syndrome. Am. J. Hum. Genet. 2025, 112, 414–427. [Google Scholar] [CrossRef]
Choufani, S.; Cytrynbaum, C.; Chung, B.H.Y.; Turinsky, A.L.; Grafodatskaya, D.; Chen, Y.A.; Cohen, A.S.A.; Dupuis, L.; Butcher, D.T.; Siu, M.T.; et al. NSD1 Mutations Generate a Genome-Wide DNA Methylation Signature. Nat. Commun. 2015, 6, 10207. [Google Scholar] [CrossRef]
Bogaert, E.; Garde, A.; Gautier, T.; Rooney, K.; Duffourd, Y.; LeBlanc, P.; van Reempts, E.; Tran Mau-Them, F.; Wentzensen, I.M.; Au, K.S.; et al. SRSF1 Haploinsufficiency Is Responsible for a Syndromic Developmental Disorder Associated with Intellectual Disability. Am. J. Hum. Genet. 2023, 110, 790. [Google Scholar] [CrossRef]
Jeffries, A.R.; Maroofian, R.; Salter, C.G.; Chioza, B.A.; Cross, H.E.; Patton, M.A.; Dempster, E.; Karen Temple, I.; Mackay, D.J.G.; Rezwan, F.I.; et al. Growth Disrupting Mutations in Epigenetic Regulatory Molecules Are Associated with Abnormalities of Epigenetic Aging. Genome Res. 2019, 29, 1057–1066. [Google Scholar] [CrossRef]
Barbosa, M.; Joshi, R.S.; Garg, P.; Martin-Trujillo, A.; Patel, N.; Jadhav, B.; Watson, C.T.; Gibson, W.; Chetnik, K.; Tessereau, C.; et al. Identification of Rare de Novo Epigenetic Variations in Congenital Disorders. Nat. Commun. 2018, 9, 2064. [Google Scholar] [CrossRef]
Choufani, S.; Gibson, W.T.; Turinsky, A.L.; Chung, B.H.Y.; Wang, T.; Garg, K.; Vitriolo, A.; Cohen, A.S.A.; Cyrus, S.; Goodman, S.; et al. DNA Methylation Signature for EZH2 Functionally Classifies Sequence Variants in Three PRC2 Complex Genes. Am. J. Hum. Genet. 2020, 106, 596–610. [Google Scholar] [CrossRef]
Guastafierro, T.; Bacalini, M.G.; Marcoccia, A.; Gentilini, D.; Pisoni, S.; Di Blasio, A.M.; Corsi, A.; Franceschi, C.; Raimondo, D.; Spanò, A.; et al. Genome-Wide DNA Methylation Analysis in Blood Cells from Patients with Werner Syndrome. Clin. Epigenet. 2017, 9, 92. [Google Scholar] [CrossRef] [PubMed]
Maierhofer, A.; Flunkert, J.; Oshima, J.; Martin, G.M.; Poot, M.; Nanda, I.; Dittrich, M.; Müller, T.; Haaf, T. Epigenetic Signatures of Werner Syndrome Occur Early in Life and Are Distinct from Normal Epigenetic Aging Processes. Aging Cell 2019, 18, e12995. [Google Scholar] [CrossRef] [PubMed]
Foroutan, A.; Haghshenas, S.; Bhai, P.; Levy, M.A.; Kerkhof, J.; McConkey, H.; Niceta, M.; Ciolfi, A.; Pedace, L.; Miele, E.; et al. Clinical Utility of a Unique Genome-Wide DNA Methylation Signature for KMT2A-Related Syndrome. Int. J. Mol. Sci. 2022, 23, 1815. [Google Scholar] [CrossRef] [PubMed]
Kimura, R.; Lardenoije, R.; Tomiwa, K.; Funabiki, Y.; Nakata, M.; Suzuki, S.; Awaya, T.; Kato, T.; Okazaki, S.; Murai, T.; et al. Integrated DNA Methylation Analysis Reveals a Potential Role for ANKRD30B in Williams Syndrome. Neuropsychopharmacology 2020, 45, 1627–1636. [Google Scholar] [CrossRef]
Coenen-van der Spek, J.; Relator, R.; Kerkhof, J.; McConkey, H.; Levy, M.A.; Tedder, M.L.; Louie, R.J.; Fletcher, R.S.; Moore, H.W.; Childers, A.; et al. DNA Methylation Episignature for Witteveen-Kolk Syndrome Due to SIN3A Haploinsufficiency. Genet. Med. 2023, 25, 63–75. [Google Scholar] [CrossRef]
Kawai, T.; Kinoshita, S.; Takayama, Y.; Ohnishi, E.; Kamura, H.; Kojima, K.; Kikuchi, H.; Terao, M.; Sugawara, T.; Migita, O.; et al. Loss of Function in NSD2 Causes DNA Methylation Signature Similar to That in Wolf-Hirschhorn Syndrome. Genet. Med. Open 2024, 2, 101838. [Google Scholar] [CrossRef]
EpigenCentral. Available online: https://epigen.ccm.sickkids.ca/ (accessed on 15 October 2025).
Awamleh, Z.; Goodman, S.; Kallurkar, P.; Wu, W.; Lu, K.; Choufani, S.; Turinsky, A.L.; Weksberg, R. Generation of DNA Methylation Signatures and Classification of Variants in Rare Neurodevelopmental Disorders Using EpigenCentral. Curr. Protoc. 2022, 2, e597. [Google Scholar] [CrossRef]
Episign—Providing Answers from Beyond the Genome The First Clinically Validated Diagnostic Test Utilizing Peripheral Blood to Assess Whole Genome Methylation. Available online: https://episign.com/#our-technology (accessed on 15 October 2025).
Husson, T.; Lecoquierre, F.; Nicolas, G.; Richard, A.C.; Afenjar, A.; Audebert-Bellanger, S.; Badens, C.; Bilan, F.; Bizaoui, V.; Boland, A.; et al. Episignatures in Practice: Independent Evaluation of Published Episignatures for the Molecular Diagnostics of Ten Neurodevelopmental Disorders. Eur. J. Hum. Genet. 2023, 32, 190–199. [Google Scholar] [CrossRef]
Grafodatskaya, D.; Chung, B.H.; Butcher, D.T.; Turinsky, A.L.; Goodman, S.J.; Choufani, S.; Chen, Y.A.; Lou, Y.; Zhao, C.; Rajendram, R.; et al. Multilocus Loss of DNA Methylation in Individuals with Mutations in the Histone H3 Lysine 4 Demethylase KDM5C. BMC Med. Genom. 2013, 6, 1. [Google Scholar] [CrossRef]
Geysens, M.; Huremagic, B.; Souche, E.; Breckpot, J.; Devriendt, K.; Peeters, H.; Van Buggenhout, G.; Van Esch, H.; Van Den Bogaert, K.; Vermeesch, J.R. Clinical Evaluation of Long-Read Sequencing-Based Episignature Detection in Developmental Disorders. Genome Med. 2025, 17, 1. [Google Scholar] [CrossRef]
Hildonen, M.; Ferilli, M.; Hjortshøj, T.D.; Dunø, M.; Risom, L.; Bak, M.; Ek, J.; Møller, R.S.; Ciolfi, A.; Tartaglia, M.; et al. DNA Methylation Signature Classification of Rare Disorders Using Publicly Available Methylation Data. Clin. Genet. 2023, 103, 688–692. [Google Scholar] [CrossRef]
Niceta, M.; Ciolfi, A.; Ferilli, M.; Pedace, L.; Cappelletti, C.; Nardini, C.; Hildonen, M.; Chiriatti, L.; Miele, E.; Dentici, M.L.; et al. DNA Methylation Profiling in Kabuki Syndrome: Reclassification of Germline KMT2D VUS and Sensitivity in Validating Postzygotic Mosaicism. Eur. J. Hum. Genet. 2024, 32, 819–826. [Google Scholar] [CrossRef] [PubMed]
Mehrmohamadi, M.; Sepehri, M.H.; Nazer, N.; Norouzi, M.R. A Comparative Overview of Epigenomic Profiling Methods. Front. Cell Dev. Biol. 2021, 9, 714687. [Google Scholar] [CrossRef] [PubMed]
Aryee, M.J.; Jaffe, A.E.; Corrada-Bravo, H.; Ladd-Acosta, C.; Feinberg, A.P.; Hansen, K.D.; Irizarry, R.A. Minfi: A Flexible and Comprehensive Bioconductor Package for the Analysis of Infinium DNA Methylation Microarrays. Bioinformatics 2014, 30, 1363–1369. [Google Scholar] [CrossRef] [PubMed]
Tian, Y.; Morris, T.J.; Webster, A.P.; Yang, Z.; Beck, S.; Feber, A.; Teschendorff, A.E. ChAMP: Updated Methylation Analysis Pipeline for Illumina BeadChips. Bioinformatics 2017, 33, 3982–3984. [Google Scholar] [CrossRef]
Morris, T.J.; Butcher, L.M.; Feber, A.; Teschendorff, A.E.; Chakravarthy, A.R.; Wojdacz, T.K.; Beck, S. ChAMP: 450k Chip Analysis Methylation Pipeline. Bioinformatics 2014, 30, 428–430. [Google Scholar] [CrossRef]
Du, P.; Kibbe, W.A.; Lin, S.M. Lumi: A Pipeline for Processing Illumina Microarray. Bioinformatics 2008, 24, 1547–1548. [Google Scholar] [CrossRef]
Zhou, W.; Triche, T.J.; Laird, P.W.; Shen, H. SeSAMe: Reducing Artifactual Detection of DNA Methylation by Infinium BeadChips in Genomic Deletions. Nucleic Acids Res. 2018, 46, e123. [Google Scholar] [CrossRef]
Min, J.L.; Hemani, G.; Smith, G.D.; Relton, C.; Suderman, M. Meffil: Efficient Normalization and Analysis of Very Large DNA Methylation Datasets. Bioinformatics 2018, 34, 3983–3989. [Google Scholar] [CrossRef]
Müller, F.; Scherer, M.; Assenov, Y.; Lutsik, P.; Walter, J.; Lengauer, T.; Bock, C. RnBeads 2.0: Comprehensive Analysis of DNA Methylation Data. Genome Biol. 2019, 20, 55. [Google Scholar] [CrossRef]
Assenov, Y.; Müller, F.; Lutsik, P.; Walter, J.; Lengauer, T.; Bock, C. Comprehensive Analysis of DNA Methylation Data with RnBeads. Nat. Methods 2014, 11, 1138–1140. [Google Scholar] [CrossRef] [PubMed]
Wang, D.; Yan, L.; Hu, Q.; Sucheston, L.E.; Higgins, M.J.; Ambrosone, C.B.; Johnson, C.S.; Smiraglia, D.J.; Liu, S. IMA: An R Package for High-Throughput Analysis of Illumina’s 450K Infinium Methylation Data. Bioinformatics 2012, 28, 729. [Google Scholar] [CrossRef] [PubMed]
Yang, H.H.; Han, M.R. MethylCallR: A Comprehensive Analysis Framework for Illumina Methylation Beadchip. Sci. Rep. 2024, 14, 27026. [Google Scholar] [CrossRef] [PubMed]
Van Iterson, M.; Tobi, E.W.; Slieker, R.C.; Den Hollander, W.; Luijk, R.; Slagboom, P.E.; Heijmans, B.T. MethylAid: Visual and Interactive Quality Control of Large Illumina 450k Datasets. Bioinformatics 2014, 30, 3435–3437. [Google Scholar] [CrossRef]
Pidsley, R.; Wong, C.C.Y.; Volta, M.; Lunnon, K.; Mill, J.; Schalkwyk, L.C. A Data-Driven Approach to Preprocessing Illumina 450K Methylation Array Data. BMC Genom. 2013, 14, 293. [Google Scholar] [CrossRef]
Leek, J.T.; Johnson, W.E.; Parker, H.S.; Jaffe, A.E.; Storey, J.D. The Sva Package for Removing Batch Effects and Other Unwanted Variation in High-Throughput Experiments. Bioinformatics 2012, 28, 882. [Google Scholar] [CrossRef]
Smyth, G.K.; Ritchie, M.; Thorne, N.; Wettenhall, J.; Shi, W.; Hu, Y. Limma: Linear Models for Microarray and RNA-Seq Data User’s Guide; The Bioconductor Project: Parkville, Australia, 2017. [Google Scholar]
Ritchie, M.E.; Phipson, B.; Wu, D.; Hu, Y.; Law, C.W.; Shi, W.; Smyth, G.K. Limma Powers Differential Expression Analyses for RNA-Sequencing and Microarray Studies. Nucleic Acids Res. 2015, 43, e47. [Google Scholar] [CrossRef]
Salas, L.A.; Koestler, D.C.; Butler, R.A.; Hansen, H.M.; Wiencke, J.K.; Kelsey, K.T.; Christensen, B.C. An Optimized Library for Reference-Based Deconvolution of Whole-Blood Biospecimens Assayed Using the Illumina HumanMethylationEPIC BeadArray. Genome Biol. 2018, 19, 64. [Google Scholar] [CrossRef]
Teschendorff, A.E.; Breeze, C.E.; Zheng, S.C.; Beck, S. A Comparison of Reference-Based Algorithms for Correcting Cell-Type Heterogeneity in Epigenome-Wide Association Studies. BMC Bioinform. 2017, 18, 105. [Google Scholar] [CrossRef]
Peters, T.J.; Buckley, M.J.; Statham, A.L.; Pidsley, R.; Samaras, K.; Lord, R.V.; Clark, S.J.; Molloy, P.L. De Novo Identification of Differentially Methylated Regions in the Human Genome. Epigenetics Chromatin 2015, 8, 6. [Google Scholar] [CrossRef]
Langfelder, P.; Horvath, S. WGCNA: An R Package for Weighted Correlation Network Analysis. BMC Bioinform. 2008, 9, 559. [Google Scholar] [CrossRef]
Jaffe, A.E.; Murakami, P.; Lee, H.; Leek, J.T.; Fallin, M.D.; Feinberg, A.P.; Irizarry, R.A. Bump Hunting to Identify Differentially Methylated Regions in Epigenetic Epidemiology Studies. Int. J. Epidemiol. 2012, 41, 200–209. [Google Scholar] [CrossRef] [PubMed]
Pedersen, B.S.; Schwartz, D.A.; Yang, I.V.; Kechris, K.J. Comb-p: Software for Combining, Analyzing, Grouping and Correcting Spatially Correlated P-Values. Bioinformatics 2012, 28, 2986. [Google Scholar] [CrossRef] [PubMed]
De Jay, N.; Papillon-Cavanagh, S.; Olsen, C.; El-Hachem, N.; Bontempi, G.; Haibe-Kains, B. MRMRe: An R Package for Parallelized MRMR Ensemble Feature Selection. Bioinformatics 2013, 29, 2365–2368. [Google Scholar] [CrossRef] [PubMed]
Pelegi-Siso, D.; De Prado, P.; Ronkainen, J.; Bustamante, M.; Gonzalez, J.R. Methylclock: A Bioconductor Package to Estimate DNA Methylation Age. Bioinformatics 2021, 37, 1759–1760. [Google Scholar] [CrossRef]
Cavalcante, R.G.; Sartor, M.A. Annotatr: Genomic Regions in Context. Bioinformatics 2017, 33, 2381. [Google Scholar] [CrossRef]
Sheffield, N.C.; Bock, C. LOLA: Enrichment Analysis for Genomic Region Sets and Regulatory Elements in R and Bioconductor. Bioinformatics 2016, 32, 587–589. [Google Scholar] [CrossRef]
Phipson, B.; Maksimovic, J.; Oshlack, A. MissMethyl: An R Package for Analyzing Data from Illumina’s HumanMethylation450 Platform. Bioinformatics 2016, 32, 286–288. [Google Scholar] [CrossRef]
Maksimovic, J.; Oshlack, A.; Phipson, B. Gene Set Enrichment Analysis for Genome-Wide DNA Methylation Data. Genome Biol. 2021, 22, 173. [Google Scholar] [CrossRef]
Ren, X.; Kuan, P.F. MethylGSA: A Bioconductor Package and Shiny App for DNA Methylation Data Length Bias Adjustment in Gene Set Testing. Bioinformatics 2019, 35, 1958–1959. [Google Scholar] [CrossRef]
Subramanian, A.; Tamayo, P.; Mootha, V.K.; Mukherjee, S.; Ebert, B.L.; Gillette, M.A.; Paulovich, A.; Pomeroy, S.L.; Golub, T.R.; Lander, E.S.; et al. Gene Set Enrichment Analysis: A Knowledge-Based Approach for Interpreting Genome-Wide Expression Profiles. Proc. Natl. Acad. Sci. USA 2005, 102, 15545–15550. [Google Scholar] [CrossRef] [PubMed]
Karatzoglou, A.; Hornik, K.; Smola, A.; Zeileis, A. Kernlab—An S4 Package for Kernel Methods in R. J. Stat. Softw. 2004, 11, 1–20. [Google Scholar] [CrossRef]
Bourgey, M.; Dali, R.; Eveleigh, R.; Chen, K.C.; Letourneau, L.; Fillon, J.; Michaud, M.; Caron, M.; Sandoval, J.; Lefebvre, F.; et al. GenPipes: An Open-Source Framework for Distributed and Scalable Genomic Analyses. Gigascience 2019, 8, giz037. [Google Scholar] [CrossRef] [PubMed]
Triche, T.J.; Weisenberger, D.J.; Van Den Berg, D.; Laird, P.W.; Siegmund, K.D. Low-Level Processing of Illumina Infinium DNA Methylation BeadArrays. Nucleic Acids Res. 2013, 41, e90. [Google Scholar] [CrossRef]
Xu, Z.; Taylor, J.A. Reliability of DNA Methylation Measures Using Illumina Methylation BeadChip. Epigenetics 2021, 16, 495–502. [Google Scholar] [CrossRef]
Maksimovic, J.; Gordon, L.; Oshlack, A. SWAN: Subset-Quantile within Array Normalization for Illumina Infinium HumanMethylation450 BeadChips. Genome Biol. 2012, 13, R44. [Google Scholar] [CrossRef]
Touleimat, N.; Tost, J. Complete Pipeline for Infinium^® Human Methylation 450K BeadChip Data Processing Using Subset Quantile Normalization for Accurate DNA Methylation Estimation. Epigenomics 2012, 4, 325–341. [Google Scholar] [CrossRef]
Fortin, J.P.; Labbe, A.; Lemire, M.; Zanke, B.W.; Hudson, T.J.; Fertig, E.J.; Greenwood, C.M.T.; Hansen, K.D. Functional Normalization of 450k Methylation Array Data Improves Replication in Large Cancer Studies. Genome Biol. 2014, 15, 503. [Google Scholar] [CrossRef]
Teschendorff, A.E.; Marabita, F.; Lechner, M.; Bartlett, T.; Tegner, J.; Gomez-Cabrero, D.; Beck, S. A Beta-Mixture Quantile Normalization Method for Correcting Probe Design Bias in Illumina Infinium 450 k DNA Methylation Data. Bioinformatics 2013, 29, 189–196. [Google Scholar] [CrossRef]
Welsh, H.; Batalha, C.M.P.F.; Li, W.; Mpye, K.L.; Souza-Pinto, N.C.; Naslavsky, M.S.; Parra, E.J. A Systematic Evaluation of Normalization Methods and Probe Replicability Using Infinium EPIC Methylation Data. Clin. Epigenet. 2023, 15, 41. [Google Scholar] [CrossRef]
Du, P.; Zhang, X.; Huang, C.C.; Jafari, N.; Kibbe, W.A.; Hou, L.; Lin, S.M. Comparison of Beta-Value and M-Value Methods for Quantifying Methylation Levels by Microarray Analysis. BMC Bioinform. 2010, 11, 587. [Google Scholar] [CrossRef] [PubMed]
Mansell, G.; Gorrie-Stone, T.J.; Bao, Y.; Kumari, M.; Schalkwyk, L.S.; Mill, J.; Hannon, E. Guidance for DNA Methylation Studies: Statistical Insights from the Illumina EPIC Array. BMC Genom. 2019, 20, 366. [Google Scholar] [CrossRef] [PubMed]
Chen, J.; Behnam, E.; Huang, J.; Moffatt, M.F.; Schaid, D.J.; Liang, L.; Lin, X. Fast and Robust Adjustment of Cell Mixtures in Epigenome-Wide Association Studies with SmartSVA. BMC Genom. 2017, 18, 413. [Google Scholar] [CrossRef] [PubMed]
Babraham Bioinformatics—Trim Galore! Available online: https://www.bioinformatics.babraham.ac.uk/projects/trim_galore/ (accessed on 26 September 2025).
Gong, W.; Pan, X.; Xu, D.; Ji, G.; Wang, Y.; Tian, Y.; Cai, J.; Li, J.; Zhang, Z.; Yuan, X. Benchmarking DNA Methylation Analysis of 14 Alignment Algorithms for Whole Genome Bisulfite Sequencing in Mammals. Comput. Struct. Biotechnol. J. 2022, 20, 4704. [Google Scholar] [CrossRef]
Krueger, F.; Andrews, S.R. Bismark: A Flexible Aligner and Methylation Caller for Bisulfite-Seq Applications. Bioinformatics 2011, 27, 1571–1572. [Google Scholar] [CrossRef]
Cheung, W.A.; Johnson, A.F.; Rowell, W.J.; Farrow, E.; Hall, R.; Cohen, A.S.A.; Means, J.C.; Zion, T.N.; Portik, D.M.; Saunders, C.T.; et al. Direct Haplotype-Resolved 5-Base HiFi Sequencing for Genome-Wide Profiling of Hypermethylation Outliers in a Rare Disease Cohort. Nat. Commun. 2023, 14, 3090. [Google Scholar] [CrossRef]
Teschendorff, A.E.; Zheng, S.C. Cell-Type Deconvolution in Epigenome-Wide Association Studies: A Review and Recommendations. Epigenomics 2017, 9, 757–768. [Google Scholar] [CrossRef]
Reinius, L.E.; Acevedo, N.; Joerink, M.; Pershagen, G.; Dahlén, S.E.; Greco, D.; Söderhäll, C.; Scheynius, A.; Kere, J. Differential DNA Methylation in Purified Human Blood Cells: Implications for Cell Lineage and Studies on Disease Susceptibility. PLoS ONE 2012, 7, e41361. [Google Scholar] [CrossRef]
Jaffe, A.E.; Irizarry, R.A. Accounting for Cellular Heterogeneity Is Critical in Epigenome-Wide Association Studies. Genome Biol. 2014, 15, R31. [Google Scholar] [CrossRef]
Carnero-Montoro, E.; Alarcón-Riquelme, M.E. Epigenome-Wide Association Studies for Systemic Autoimmune Diseases: The Road behind and the Road Ahead. Clin. Immunol. 2018, 196, 21–33. [Google Scholar] [CrossRef]
Surace, A.E.A.; Hedrich, C.M. The Role of Epigenetics in Autoimmune/Inflammatory Disease. Front. Immunol. 2019, 10, 466419. [Google Scholar] [CrossRef] [PubMed]
Bergstedt, J.; Azzou, S.A.K.; Tsuo, K.; Jaquaniello, A.; Urrutia, A.; Rotival, M.; Lin, D.T.S.; MacIsaac, J.L.; Kobor, M.S.; Albert, M.L.; et al. The Immune Factors Driving DNA Methylation Variation in Human Blood. Nat. Commun. 2022, 13, 5895. [Google Scholar] [CrossRef] [PubMed]
Houseman, E.A.; Accomando, W.P.; Koestler, D.C.; Christensen, B.C.; Marsit, C.J.; Nelson, H.H.; Wiencke, J.K.; Kelsey, K.T. DNA Methylation Arrays as Surrogate Measures of Cell Mixture Distribution. BMC Bioinform. 2012, 13, 86. [Google Scholar] [CrossRef] [PubMed]
De Ridder, K.; Che, H.; Leroy, K.; Thienpont, B. Benchmarking of Methods for DNA Methylome Deconvolution. Nat. Commun. 2024, 15, 4134. [Google Scholar] [CrossRef]
McGregor, K.; Bernatsky, S.; Colmegna, I.; Hudson, M.; Pastinen, T.; Labbe, A.; Greenwood, C.M.T. An Evaluation of Methods Correcting for Cell-Type Heterogeneity in DNA Methylation Studies. Genome Biol. 2016, 17, 84. [Google Scholar] [CrossRef]
Li, W.; Mills, A.A. Architects of the Genome: CHD Dysfunction in Cancer, Developmental Disorders and Neurological Syndromes. Epigenomics 2014, 6, 381–395. [Google Scholar] [CrossRef]
Yan, Q.; Forno, E.; Celedón, J.C.; Chen, W. A Region-Based Method for Causal Mediation Analysis of DNA Methylation Data. Epigenetics 2021, 17, 286. [Google Scholar] [CrossRef]
Van Gils, J.; Magdinier, F.; Fergelot, P.; Lacombe, D. Rubinstein-Taybi Syndrome: A Model of Epigenetic Disorder. Genes 2021, 12, 968. [Google Scholar] [CrossRef]
Blagus, R.; Lusa, L. SMOTE for High-Dimensional Class-Imbalanced Data. BMC Bioinform. 2013, 14, 106. [Google Scholar] [CrossRef]
Menardi, G.; Torelli, N. Training and Assessing Classification Rules with Imbalanced Data. Data Min. Knowl. Discov. 2014, 28, 92–122. [Google Scholar] [CrossRef]
Ilemobayo, J.A.; Durodola, O.; Alade, O.; Awotunde, O.J.; Olanrewaju, A.T.; Falana, O.; Ogungbire, A.; Osinuga, A.; Ogunbiyi, D.; Ifeanyi, A.; et al. Hyperparameter Tuning in Machine Learning: A Comprehensive Review. J. Eng. Res. Rep. 2024, 26, 388–395. [Google Scholar] [CrossRef]
Rainio, O.; Teuho, J.; Klén, R. Evaluation Metrics and Statistical Tests for Machine Learning. Sci. Rep. 2024, 14, 6086. [Google Scholar] [CrossRef] [PubMed]
Santafe, G.; Inza, I.; Lozano, J.A. Dealing with the Evaluation of Supervised Classification Algorithms. Artif. Intell. Rev. 2015, 44, 467–508. [Google Scholar] [CrossRef]
Dobson, A.J.; Barnett, A.G. An Introduction to Generalized Linear Models, 4th ed.; Chapman and Hall/CRC: New York, NY, USA, 2018. [Google Scholar] [CrossRef]
McHugh, M.L. Interrater Reliability: The Kappa Statistic. Biochem. Med. 2012, 22, 276. [Google Scholar] [CrossRef]
Kuhn, M. Building Predictive Models in R Using the Caret Package. J. Stat. Softw. 2008, 28, 1–26. [Google Scholar] [CrossRef]
Guido, R.; Ferrisi, S.; Lofaro, D.; Conforti, D. An Overview on the Advancements of Support Vector Machine Models in Healthcare Applications: A Review. Information 2024, 15, 235. [Google Scholar] [CrossRef]
Cervantes, J.; Garcia-Lamont, F.; Rodríguez-Mazahua, L.; Lopez, A. A Comprehensive Survey on Support Vector Machine Classification: Applications, Challenges and Trends. Neurocomputing 2020, 408, 189–215. [Google Scholar] [CrossRef]
Breiman, L. Random Forests. Mach. Learn. 2001, 45, 5–32. [Google Scholar] [CrossRef]
Aria, M.; Cuccurullo, C.; Gnasso, A. A Comparison among Interpretative Proposals for Random Forests. Mach. Learn. Appl. 2021, 6, 100094, Erratum in Mach. Learn. Appl. 2022, 10, 100438. [Google Scholar] [CrossRef]
Liaw, A.; Wiener, M. Classification and Regression by RandomForest. R News 2002, 2, 18–22. [Google Scholar]
Pedregosa, F.; Michel, V.; Grisel, O.; Blondel, M.; Prettenhofer, P.; Weiss, R.; Vanderplas, J.; Cournapeau, D.; Pedregosa, F.; Varoquaux, G.; et al. Scikit-Learn: Machine Learning in Python. J. Mach. Learn. Res. 2011, 12, 2825–2830. [Google Scholar]
Pavlou, M.; Ambler, G.; Seaman, S.; De Iorio, M.; Omar, R.Z. Review and Evaluation of Penalised Regression Methods for Risk Prediction in Low-dimensional Data with Few Events. Stat. Med. 2015, 35, 1159–1177. [Google Scholar] [CrossRef]
Tay, J.K.; Narasimhan, B.; Hastie, T. Elastic Net Regularization Paths for All Generalized Linear Models. J. Stat. Softw. 2023, 106, 1. [Google Scholar] [CrossRef] [PubMed]
Brauneck, A.; Schmalhorst, L.; Weiss, S.; Baumbach, L.; Völker, U.; Ellinghaus, D.; Baumbach, J.; Buchholtz, G. Legal Aspects of Privacy-Enhancing Technologies in Genome-Wide Association Studies and Their Impact on Performance and Feasibility. Genome Biol. 2024, 25, 154, Erratum in Genome Biol. 2024, 25, 160. [Google Scholar] [CrossRef] [PubMed]
Dyke, S.O.M.; Cheung, W.A.; Joly, Y.; Ammerpohl, O.; Lutsik, P.; Rothstein, M.A.; Caron, M.; Busche, S.; Bourque, G.; Rönnblom, L.; et al. Epigenome Data Release: A Participant-Centered Approach to Privacy Protection. Genome Biol. 2015, 16, 142. [Google Scholar] [CrossRef] [PubMed]
Duhm-Harbeck, P.; Köbler, J. Rare Diseases and Data Protection (Part I). In GDPR Requirements for Biobanking Activities Across Europe; Springer: Cham, Switzerland, 2023; pp. 305–315. [Google Scholar] [CrossRef]
Gaye, A.; Marcon, Y.; Isaeva, J.; Laflamme, P.; Turner, A.; Jones, E.M.; Minion, J.; Boyd, A.W.; Newby, C.J.; Nuotio, M.L.; et al. DataSHIELD: Taking the Analysis to the Data, Not the Data to the Analysis. Int. J. Epidemiol. 2014, 43, 1929–1944. [Google Scholar] [CrossRef]
Alvarellos, M.; Sheppard, H.E.; Knarston, I.; Davison, C.; Raine, N.; Seeger, T.; Prieto Barja, P.; Chatzou Dunford, M. Democratizing Clinical-Genomic Data: How Federated Platforms Can Promote Benefits Sharing in Genomics. Front. Genet. 2023, 13, 1045450. [Google Scholar] [CrossRef]
van Baalen, V.; Didden, E.M.; Rosenberg, D.; Bardenheuer, K.; van Speybroeck, M.; Brand, M. Increase Transparency and Reproducibility of Real-World Evidence in Rare Diseases through Disease-Specific Federated Data Networks. Pharmacoepidemiol. Drug Saf. 2024, 33, e5778. [Google Scholar] [CrossRef]
Hunger, M.; Bardenheuer, K.; Passey, A.; Schade, R.; Sharma, R.; Hague, C. The Value of Federated Data Networks in Oncology: What Research Questions Do They Answer? Outcomes From a Systematic Literature Review. Value Health 2022, 25, 855–868. [Google Scholar] [CrossRef]
Mandl, K.D.; Kohane, I.S. Federalist Principles for Healthcare Data Networks. Nat. Biotechnol. 2015, 33, 360–363. [Google Scholar] [CrossRef]
Kaur, D.; Lee, S.M.; Goldberg, D.; Spix, N.J.; Hinoue, T.; Li, H.-T.; Dwaraka, V.B.; Smith, R.; Shen, H.; Liang, G.; et al. Comprehensive Evaluation of the Infinium Human MethylationEPIC v2 BeadChip. Epigenetics Commun. 2023, 3, 6. [Google Scholar] [CrossRef] [PubMed]
Komaki, S.; Ohmomo, H.; Hachiya, T.; Sutoh, Y.; Ono, K.; Furukawa, R.; Umekage, S.; Otsuka-Yamasaki, Y.; Tanno, K.; Sasaki, M.; et al. Longitudinal DNA Methylation Dynamics as a Practical Indicator in Clinical Epigenetics. Clin. Epigenet. 2021, 13, 219. [Google Scholar] [CrossRef] [PubMed]
Coit, P.; Ortiz-Fernandez, L.; Lewis, E.E.; McCune, W.J.; Maksimowicz-McKinnon, K.; Sawalha, A.H. A Longitudinal and Transancestral Analysis of DNA Methylation Patterns and Disease Activity in Lupus Patients. JCI Insight 2020, 5, e143654. [Google Scholar] [CrossRef] [PubMed]
Urdinguio, R.G.; Torró, M.I.; Bayón, G.F.; Álvarez-Pitti, J.; Fernández, A.F.; Redon, P.; Fraga, M.F.; Lurbe, E. Longitudinal Study of DNA Methylation during the First 5 Years of Life. J. Transl. Med. 2016, 14, 160. [Google Scholar] [CrossRef]
Thompson, E.E.; Nicodemus-Johnson, J.; Kim, K.W.; Gern, J.E.; Jackson, D.J.; Lemanske, R.F.; Ober, C. Global DNA Methylation Changes Spanning Puberty Are near Predicted Estrogen-Responsive Genes and Enriched for Genes Involved in Endocrine and Immune Processes. Clin. Epigenet. 2018, 10, 62. [Google Scholar] [CrossRef]
Grant, O.A.; Wang, Y.; Kumari, M.; Zabet, N.R.; Schalkwyk, L. Characterising Sex Differences of Autosomal DNA Methylation in Whole Blood Using the Illumina EPIC Array. Clin. Epigenet. 2022, 14, 62. [Google Scholar] [CrossRef]
Dossin, F.; Pinheiro, I.; Żylicz, J.J.; Roensch, J.; Collombet, S.; Le Saux, A.; Chelmicki, T.; Attia, M.; Kapoor, V.; Zhan, Y.; et al. SPEN Integrates Transcriptional and Epigenetic Control of X-Inactivation. Nature 2020, 578, 455–460. [Google Scholar] [CrossRef]
Eggermann, T.; Perez de Nanclares, G.; Maher, E.R.; Temple, I.K.; Tümer, Z.; Monk, D.; Mackay, D.J.G.; Grønskov, K.; Riccio, A.; Linglart, A.; et al. Imprinting Disorders: A Group of Congenital Disorders with Overlapping Patterns of Molecular Changes Affecting Imprinted Loci. Clin. Epigenet. 2015, 7, 123, Erratum in Clin. Epigenet. 2016, 8, 27. [Google Scholar] [CrossRef]
Hara-Isono, K.; Inoue, T.; Nakamura, A.; Fuke, T.; Yamazawa, K.; Matsubara, K.; Fukami, M.; Ogata, T.; Kawai, T.; Kagami, M. Investigation of Methylation Profiles in Silver–Russell Syndrome to Explore Episignatures. Clin. Epigenet. 2025, 17, 184. [Google Scholar] [CrossRef]
Aref-Eshghi, E.; Schenkel, L.C.; Lin, H.; Skinner, C.; Ainsworth, P.; Paré, G.; Siu, V.; Rodenhiser, D.; Schwartz, C.; Sadikovic, B. Clinical Validation of a Genome-Wide DNA Methylation Assay for Molecular Diagnosis of Imprinting Disorders. J. Mol. Diagn. 2017, 19, 848–856. [Google Scholar] [CrossRef]

Figure 1. Global approach in genetic testing for a rare disease (RD). First, either targeted (for suspected cases) or global methods are used to detect variants. Afterwards, the variant must be annotated according to the literature. If the variant can be considered causal and matches the clinical diagnosis, the patient might be diagnosed. If either no variant or a variant of uncertain significance (VUS) is found or the identified variant does not match the clinical phenotype, additional techniques are required to reach a diagnosis. However, additional testing can also be conducted to validate genetic variants. Icons downloaded from BioRender.org. There is no copyright issue.

Figure 2. PRISMA flowchart for article selection and bibliographic filters considered during the first stage of article review. There is no copyright issue.

Figure 3. Proposed pipeline for episignature detection. Two main paths exist depending on whether other methylation profiles are available for comparison. If not, an external facility is required. If so, the user can either develop their own episignature or test any open-access model, searching for the disease of interest by consulting the studies summarized in this review. In the principal component analysis (PCA) and hierarchical clustering colors indicate case (orange) and control (green/blue) samples. Icons downloaded from BioRender.org. There is no copyright issue.

Figure 4. Full episignature detection pipeline using methylation arrays. Continuous dark green lines and boxes indicate steps conducted in most studies or considered key. Dotted light-green lines and boxes indicate processes that sometimes are not conducted, but were considered important to keep in mind. In the PCA, hierarchical clustering, epimutation plot, and discovery/validation plot, colors indicate cases (orange) and controls (green/blue). In the external validation plot, colors represent different diseases to illustrate validation using cases of the studied episignature (positive controls) and controls plus other cases that should not respond to the episignature (negative controls). Icons downloaded from BioRender.org. There is no copyright issue.

Figure 5. Summary of key steps in classification model training. CpGs have already been filtered and curated. Different balancing methods, ML model types, and validation settings are available. Before the final model is trained, the most optimal hyperparameters should be found, with internal CV rounds selecting an evaluation metric. The colors in evaluation metric plots indicate different combination of model-specific hyperparameters. The final model should be tested to assess its performance (in the example, with ROC curves). Icons downloaded from BioRender.org. There is no copyright issue.

Table 1. List of included articles with the studied disease, altered gene or variant, and screening method. “*” indicates a locus erroneously annotated in the article. “---” indicates no specific gene. Genes marked /*/ found no episignature even though other genes causing the disease yielded significant results.

Disease	Loci	Variant	Array	Article Title	Ref.
1p36 deletion syndrome	1p36, SPEN	Deletion in females	EPIC	SPEN haploinsufficiency causes a neurodevelopmental disorder overlapping proximal 1p36 deletion syndrome with an episignature of X chromosomes in females	[35]
6q24–q25 deletion syndrome	6q24–q25	Deletion	450 k, EPIC	Diagnostic utility of genome-wide DNA methylation testing in genetically unsolved individuals with suspected hereditary conditions	[36]
7q11.23 duplication syndrome	7q11.23	Duplication	450 k	Symmetrical dose-dependent DNA-methylation profiles in children with deletion or duplication of 7q11.23	[37]
	7q11.23	Duplication	450 k, EPIC	Diagnostic utility of genome-wide DNA methylation testing in genetically unsolved individuals with suspected hereditary conditions	[36]
	7q11.23	Duplication	450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	7q11.23	Duplication	450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
9q34.3 microduplication syndrome	9q34.3	Microduplication	EPIC	Refining the 9q34.3 microduplication syndrome reveals mild neurodevelopmental features associated with a distinct global DNA methylation profile	[39]
16p11.2 deletion syndrome	16p11.2	Deletion	450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
16p11.2 deletion syndrome	16p11.2	Deletion	450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
22q11.2 deletion syndrome (velocardiofacial syndrome)	22q11.2	Deletion	EPIC	Identification of a DNA Methylation Episignature in the 22q11.2 Deletion Syndrome	[41]
	22q11.2	Deletion	EPIC	Differential methylation of imprinting genes and MHC locus in 22q11.2 deletion syndrome-related schizophrenia spectrum disorders	[42]
	22q11.2	Deletion	450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	22q11.2	Deletion	450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Aicardi–Goutières syndrome	RNASEH2B		EPIC	Altered DNA methylation and gene expression predict disease severity in patients with Aicardi–Goutières syndrome	[43]
Alpha-thalassemia/impaired intellectual development syndrome, X-linked	ATRX		450 k	Identification of epigenetic signature associated with alpha thalassemia/mental retardation X-linked syndrome	[44]
	ATRX		450 k	Genomic DNA methylation signatures enable concurrent diagnosis and clinical genetic variant classification in neurodevelopmental syndromes	[45]
	ATRX		450 k, EPIC	Diagnostic utility of genome-wide DNA methylation testing in genetically unsolved individuals with suspected hereditary conditions	[36]
	ATRX		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	ATRX		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	ATRX		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Arboleda–Tham syndrome	KAT6A		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
Arboleda–Tham syndrome	KAT6A		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
ARID2-related disorder	ARID2		EPIC	ARID2-related disorder: further delineation of the clinical phenotype of 27 novel individuals and description of an epigenetic signature	[46]
Attention deficit/hyperactivity disorder (ADHD)			450 k	DNA methylation epi-signature and biological age in attention deficit hyperactivity disorder patients	[47]
Au–Kline syndrome	HNRNPK	Missense and loss of function	EPIC	An HNRNPK-specific DNA methylation signature makes sense of missense variants and expands the phenotypic spectrum of Au-Kline syndrome	[48]
Autism spectrum disorder (ASD)	16p11.2 del		450 k, EPIC	Functional DNA methylation signatures for autism spectrum disorder genomic risk loci: 16p11.2 deletions and CHD8 variants	[49]
	CHD8		450 k, EPIC		[49]
			450 k	An epigenetic biomarker for adult high-functioning autism spectrum disorder	[33]
			450 k, EPIC	Epigenetics of autism spectrum disorders: a multi-level analysis combining epi-signature, age acceleration, epigenetic drift and rare epivariations using public datasets	[28]
			450 k	Screening for rare epigenetic variations in autism and schizophrenia	[50]
	CHD8		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	CHD8		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
			450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Autosomal dominant intellectual developmental disorder—65 (MRD65)	KDM4B		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	KDM4B		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
BAFopathy nonsyndromic	ARID1A, ARID1B	Duplications	EPIC	Microduplications of ARID1A and ARID1B cause a novel clinical and epigenetic distinct BAFopathy	[51]
Beck–Fahrner syndrome	TET3		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
Beck–Fahrner syndrome	TET3		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Berardinelli-Seip Congenital Lipodystrophy type 2 (CGL2)	BSCL2		EPIC	Accelerated epigenetic aging and DNA methylation alterations in Berardinelli–Seip congenital lipodystrophy	[52]
BCL11B-related disease (BCL11B-RD)	BCL11B		EPIC	Clinico-biological refinement of BCL11B-related disorder and identification of an episignature: a series of 20 unreported individuals	[53]
BCL11B-related disease (BCL11B-RD)	BCL11B		EPIC		[53]
Beck–Fahrner syndrome	TET3		EPIC	Deficiency of TET3 leads to a genome-wide DNA hypermethylation episignature in human whole blood	[54]
Blepharophimosis intellectual disability syndrome (BIS)	SMARCA2	Exons 8 and 9	EPIC	De novo SMARCA2 variants clustered outside the helicase domain cause a new recognizable syndrome with intellectual disability and blepharophimosis distinct from Nicolaides-Baraitser syndrome	[55]
	SMARCA2		EPIC	Blepharophimosis with intellectual disability and Helsmoortel-Van der Aa syndrome share episignature and phenotype	[56]
	SMARCA2		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	SMARCA2		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Börjeson–Forssman–Lehmann syndrome (BFLS)	PHF6		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	PHF6		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	PHF6		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Bohring–Opitz syndrome (BOS)	ASXL		EPIC	DNA methylation signature associated with Bohring-Opitz syndrome: a new tool for functional classification of variants in ASXL genes	[57]
CDK13-related disorder (CDK13-RD)	CDK13		EPIC	CDK13-related disorder: report of a series of 18 previously unpublished individuals and description of an epigenetic signature	[58]
Cerebellar ataxia, deafness and narcolepsy, autosomal dominant (ADCADN)	DNMT1		450 k	Identification of a methylation profile for DNMT1-associated autosomal dominant cerebellar ataxia, deafness, and narcolepsy	[59]
	DNMT1		450 k	Genomic DNA methylation signatures enable concurrent diagnosis and clinical genetic variant classification in neurodevelopmental syndromes	[45]
	DNMT1		450 k, EPIC	Diagnostic utility of genome-wide DNA methylation testing in genetically unsolved individuals with suspected hereditary conditions	[36]
	DNMT1		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	DNMT1		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	DNMT1		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Charcot–Marie–Tooth disease type 2Z (CMT2Z)	MORC2		EPIC, EPICv2	Pleiotropic effects of MORC2 derive from its epigenetic signature	[60]
CHARGE syndrome	CHD7		450 k	CHARGE and Kabuki syndromes: gene-specific DNA methylation signatures identify epigenetic mechanisms linking these clinically overlapping conditions	[61]
	CHD7		EPIC	Genomic DNA methylation signatures enable concurrent diagnosis and clinical genetic variant classification in neurodevelopmental syndromes	[45]
	CHD7		450 k, EPIC	Diagnostic utility of genome-wide DNA methylation testing in genetically unsolved individuals with suspected hereditary conditions	[36]
	CHD7		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	CHD7		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	CHD7		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Chung–Jansen syndrome	PHIP		EPIC	The detection of a strong episignature for Chung–Jansen syndrome, partially overlapping with Börjeson–Forssman–Lehmann and White–Kernohan syndromes	[62]
Claes–Jensen syndrome	KDM5C		450 k	Peripheral blood epi-signature of Claes-Jensen syndrome enables sensitive and specific identification of patients and healthy carriers with pathogenic mutations in KDM5C	[63]
	KDM5C		450 k	Genomic DNA methylation signatures enable concurrent diagnosis and clinical genetic variant classification in neurodevelopmental syndromes	[45]
	KDM5C		450 k, EPIC	Diagnostic utility of genome-wide DNA methylation testing in genetically unsolved individuals with suspected hereditary conditions	[36]
	KDM5C		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	KDM5C		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Clark–Baraitser syndrome	TRIP12		EPIC	Episignature mapping of TRIP12 provides functional insight into Clark-Baraitser syndrome	[64]
Coffin–Lowry syndrome	RSK2		450 k	Genomic DNA methylation signatures enable concurrent diagnosis and clinical genetic variant classification in neurodevelopmental syndromes (Not significant)	[45]
Coffin–Siris syndrome	ARID1B, SMARCB1		450, EPIC	Diagnostic utility of genome-wide DNA methylation testing in genetically unsolved individuals with suspected hereditary conditions	[36]
	ARID1B		450 k, EPIC	BAFopathies’ DNA methylation epi-signatures demonstrate diagnostic utility and functional continuum of Coffin–Siris and Nicolaides–Baraitser syndromes	[65]
	SMARCB1		450 k, EPIC		[65]
	SOX11		EPIC	SOX11 variants cause a neurodevelopmental disorder with infrequent ocular malformations and hypogonadotropic hypogonadism and with distinct DNA methylation profile	[66]
	ARID1B, SMARCB1		450 k	Genomic DNA methylation signatures enable concurrent diagnosis and clinical genetic variant classification in neurodevelopmental syndromes (Not significant)	[45]
	ARID1A, ARID1B, SMARCB1, SMARCA4		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	ARID1A, ARID1B, SMARCB1, SMARCA4, SOX11		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	ARID1A, ARID1B, SMARCB1, SMARCA4, SOX11		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
	ARID1A, ARID1B	c6200
	SMARCA4	c2650
Cohen–Gibson syndrome	EED		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
Cohen–Gibson syndrome	EED		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Cockayne syndrome (CS)	MORC2		EPIC, EPICv2	Pleiotropic effects of MORC2 derive from its epigenetic signature	[60]
Cornelia de Lange syndrome	NIPBL, RAD21, SMC1A, SMC3		450 k, EPIC	Diagnostic utility of genome-wide DNA methylation testing in genetically unsolved individuals with suspected hereditary conditions	[36]
	NIPBL, RAD21, SMC3, SMC1A		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders (Not significant)	[26]
	HDAC8		450 k, EPIC		[26]
	NIPBL, RAD21, SMC1A, SMC3, HDAC8		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	NIPBL, SMC1A, SMC3, RAD21		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Developmental and epileptic encephalopathy	CHD2, ---		450 k, EPIC	Diagnostic utility of DNA methylation analysis in genetically unsolved pediatric epilepsies and CHD2 episignature refinement	[67]
Developmental and epileptic encephalopathy	HNRNPU		EPIC	DNA methylation episignature and comparative epigenomic profiling of HNRNPU-related neurodevelopmental disorder	[68]
Developmental delay with gastrointestinal, cardiovascular, genitourinary, and skeletal abnormalities syndrome (DEGCAGS)	ZNF699		EPIC	Epigenomic and phenotypic characterization of DEGCAGS syndrome	[69]
DOT1L-associated syndrome	DOT1L	Increase of function	EPIC	Rare de novo gain-of-function missense variants in DOT1L are associated with developmental delay and congenital anomalies	[70]
Down syndrome	+chr21	Trisomy	450 k	Identification of a DNA methylation signature in blood cells from persons with Down syndrome	[71]
	+chr21	Trisomy	450 k, EPIC	Diagnostic utility of genome-wide DNA methylation testing in genetically unsolved individuals with suspected hereditary conditions	[36]
	+chr21	Trisomy	450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	+chr21	Trisomy	450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	+chr21	Trisomy	450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Duchenne muscular dystrophy	DMD		EPIC	The discovery of the DNA methylation episignature for Duchenne muscular dystrophy	[72]
DYRK1A intellectual disability	DYRK1A	Loss of function	EPIC	Integrative approach to interpret DYRK1A variants, leading to a frequent neurodevelopmental disorder	[73]
Dystonia 28, childhood onset	KMT2B		EPIC	Childhood-onset dystonia-causing KMT2B variants result in a distinctive genomic hypermethylation profile	[74]
	KMT2B		EPIC	Episignature analysis of moderate effects and mosaics	[75]
	KMT2B		NGS	Comparison of methylation episignatures in KMT2B- and KMT2D-related human disorders	[76]
	KMT2B		EPIC	Blood DNA methylation provides an accurate biomarker of KMT2B-related dystonia and predicts onset	[77]
	KMT2B		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	KMT2B		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
epi-cblC disease	PRDX1, MMACHC	515-1G > T (MMACHC epimutation)	EPIC	PRDX1 gene-related epi-cblC disease is a common type of inborn error of cobalamin metabolism with mono- or bi-allelic MMACHC epimutations	[30]
epi-cblC disease	TESK2, MMACHC	Epimutation	450 k	Epimutations in both the TESK2 and MMACHC promoters in the Epi-cblC inherited disorder of intracellular metabolism of vitamin B12	[78]
Epileptic encephalopathy, childhood onset	CHD2		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	CHD2		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	CHD2		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Fanconi anemia	FANCA		EPIC	Identification of a robust DNA methylation signature for Fanconi anemia	[79]
Fetal alcohol spectrum disorder (FASD)			450 k	DNA methylation abundantly associates with fetal alcohol spectrum disorder and its subphenotypes	[80]
			450 k	Expression quantitative trait methylation analysis identifies whole blood molecular footprint in fetal alcohol spectrum disorder (FASD)	[81]
			EPICv2	Discovery of a DNA methylation episignature as a molecular biomarker for fetal alcohol syndrome	[82]
Fetal valproate syndrome (SVF)			EPIC	Discovery of DNA methylation signature in the peripheral blood of individuals with history of antenatal exposure to valproic acid	[83]
FG syndrome (also Opitz–Kaveggia syndrome)	MED12		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders (Not significant)	[26]
FG syndrome (also Opitz–Kaveggia syndrome)	MED12		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
Floating-harbor syndrome	SRCAP		450 k	The defining DNA methylation signature of floating-harbor syndrome	[84]
	SRCAP		EPIC	Truncating SRCAP variants outside the floating-harbor syndrome locus cause a distinct neurodevelopmental disorder with a specific DNA methylation signature	[85]
	SRCAP		450 k	Genomic DNA methylation signatures enable concurrent diagnosis and clinical genetic variant classification in neurodevelopmental syndromes	[45]
	SRCAP		450 k, EPIC	Diagnostic utility of genome-wide DNA methylation testing in genetically unsolved individuals with suspected hereditary conditions	[36]
	SRCAP		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	SRCAP		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	SRCAP		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Fragile X	FMR1	GCC repetitions and deletion	450 k	Clinical validation of fragile X syndrome screening by DNA methylation array	[86]
Frontotemporal dementia (FTD)	17q21.31		450 k	An epigenetic signature in peripheral blood associated with the haplotype on 17q21.31, a risk factor for neurodegenerative tauopathy	[87]
Gabriele–de Vries syndrome (GADEVS)	YY1		EPIC	DNA methylation episignature in Gabriele-de Vries syndrome	[88]
	YY1		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	YY1		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Genitopatellar syndrome (GTPTS)	KAT6B		450 k	Genomic DNA methylation signatures enable concurrent diagnosis and clinical genetic variant classification in neurodevelopmental syndromes	[45]
	KAT6B		450 k, EPIC	Diagnostic utility of genome-wide DNA methylation testing in genetically unsolved individuals with suspected hereditary conditions	[36]
	KAT6B		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	KAT6B		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	KAT6B		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Glass syndrome	SATB2		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders (Not significant)	[26]
Glass syndrome	SATB2		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
Hao–Fountain syndrome	USP7		EPIC	DNA methylation episignature, extension of the clinical features, and comparative epigenomic profiling of Hao-Fountain syndrome caused by variants in USP7	[89]
Helsmoortel–Van der Aa Syndrome (HVDAS)	ADNP	Class I (outside 2000–2340 bp)	EPIC	Episignatures sStratifying Helsmoortel-Van der Aa syndrome show modest correlation with phenotype	[90]
	ADNP	Class II (inside 2156 and 2317)	EPIC		[90]
	ADNP		EPIC	Blepharophimosis with intellectual disability and Helsmoortel-Van Der Aa Syndrome share episignature and phenotype	[56]
	ADNP	Central Terminal	EPIC	Gene domain-specific DNA methylation episignatures highlight distinct molecular entities of ADNP syndrome	[91]
	ADNP	Central Terminal	450 k, EPIC	Diagnostic utility of genome-wide DNA methylation testing in genetically unsolved individuals with suspected hereditary conditions	[36]
	ADNP	Central Terminal	EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	ADNP	Central Terminal	450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	ADNP	Central Terminal	450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
HNRNPU-related syndrome	HNRNPU		NGS	Germline pathogenic variants in HNRNPU are associated with alterations in blood methylome	[92]
Hunter–McAlpine syndrome	NSD1, 5q35 *	Duplication	450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	NSD1, 5q35	q terminal duplication	450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	NSD1, 5q35	q terminal duplication	450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Immunodeficiency with centromeric instability and facial anomalies (ICF) syndrome	DNMT3B, ZBTB24, CDCA7, HELLS		450 k	Comparative methylome analysis of ICF patients identifies heterochromatin loci that require ZBTB24, CDCA7 and HELLS for their methylated state	[93]
	DNMT3B, ZBTB24, CDCA7, HELLS		450 k	Interplay between histone and DNA methylation seen through comparative methylomes in rare Mendelian disorders	[94]
	DNMT3B, CDCA7, ZBTB24, HELLS		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	DNMT3B, CDCA7, ZBTB24, HELLS		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[26]
	DNMT3B, CDCA7, ZBTB24, HELLS		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Intellectual developmental disorder with seizures and language delay (IDDSELD)	SETD1B		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	SETD1B		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Intellectual developmental disorder, autosomal dominant 21	CTCF		EPIC	Identification of DNA methylation episignature for the intellectual developmental disorder, autosomal dominant 21 syndrome, caused by variants in the CTCF gene	[95]
Intellectual developmental disorder, X-linked, syndromic, Armfield type (MRXSA)	FAM50A		EPIC	Detection of a DNA methylation signature for the intellectual developmental disorder, X-linked, syndromic, Armfield type	[29]
	FAM50A		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	FAM50A		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
JARID2 neurodevelopmental syndrome	JARID2		EPIC	DNA methylation signature for JARID2-neurodevelopmental syndrome	[96]
JARID2 neurodevelopmental syndrome	JARID2		EPIC	Functional Insight into and Refinement of the Genomic Boundaries of the JARID2-Neurodevelopmental Disorder Episignature	[97]
Kagami–Ogata syndrome (KOS14)	14q32.2	Imprinting	450 k	Genome-wide multilocus imprinting disturbance analysis in Temple syndrome and Kagami-Ogata syndrome	[98]
Kabuki syndrome	KMT2D		450 k	The defining DNA methylation signature of Kabuki syndrome enables functional assessment of genetic variants of unknown clinical significance	[99]
	KMT2D		450 k	CHARGE and Kabuki syndromes: gene-specific DNA methylation signatures identify epigenetic mechanisms linking these clinically overlapping conditions	[61]
	KMT2D		450 k	Interplay between histone and DNA methylation seen through comparative methylomes in rare Mendelian disorders	[94]
	KMT2D		EPIC	Episignature analysis of moderate effects and mosaics	[75]
	KMT2D		NGS	Comparison of methylation episignatures in KMT2B- and KMT2D-related human disorders	[76]
	KMT2D		450 k	Patients with a Kabuki syndrome phenotype demonstrate DNA methylation abnormalities	[100]
	KMT2D		450 k	Comprehensive evaluation of the implementation of episignatures for diagnosis of neurodevelopmental disorders (NDDs)	[101]
	KMT2D, KDM6A		450 k	Genomic DNA methylation signatures enable concurrent diagnosis and clinical genetic variant classification in neurodevelopmental syndromes	[45]
	KMT2D		450 k, EPIC	Diagnostic utility of genome-wide DNA methylation testing in genetically unsolved individuals with suspected hereditary conditions	[36]
	KMT2D, KDM6A		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	KMT2D, KDM6A		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	KMT2D, KDM6A		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
KBG syndrome	ANKRD11, 16p24.3	Variants or deletions	EPIC	ANKRD11 pathogenic variants and 16q24.3 microdeletions share an altered DNA methylation signature in patients with KBG syndrome	[102]
KDM2B-related neurodevelopmental disorder	KDM2B		EPIC	Delineation of a KDM2B-related neurodevelopmental disorder and its associated DNA methylation signature	[103]
	KDM2B		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	KDM2B		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Kleefstra syndrome	EHMT1, 9q34.3	Microdeletions	EPIC	EHMT1 pathogenic variants and 9q34.3 microdeletions share altered DNA methylation patterns in patients with Kleefstra syndrome	[104]
	EHMT1		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	EHMT1		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	EHMT1		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
KMT2C-related syndrome	KMT2C		EPIC	Pathogenic variants in KMT2C result in a neurodevelopmental disorder distinct from Kleefstra and Kabuki syndromes	[105]
	KMT2C		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders (Not significant)	[26]
	KMT2C		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
Koolen–De Vries syndrome	KANSL1, 17q21.31	Deletion	EPIC	A new blood DNA methylation signature for Koolen-de Vries syndrome: Classification of missense KANSL1 variants and comparison to fibroblast cells	[106]
	KANSL1		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	KANSL		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	KANSL1		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Leigh syndrome (LS)	MORC2		EPIC, EPICv2	Pleiotropic effects of MORC2 derive from its epigenetic signature	[60]
Luscan–Lumish syndrome (LLS)	SETD2	1740 codon, Truncating	NGS	Epigenotype-genotype-phenotype correlations in SETD1A and SETD2 chromatin disorders	[107]
	SETD2		EPIC	Interplay between histone and DNA methylation seen through comparative methylomes in rare Mendelian disorders	[94]
	SETD2		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	SETD2		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Lynch syndrome	MLH1	Epimutation	450 k	Primary constitutional MLH1 epimutations: a focal epigenetic event	[108]
Menke–Hennekam syndrome	CREBBP, EP300	ZZ, TAZ2, ID4	EPIC	Menke-Hennekam syndrome delineation of domain-specific subtypes with distinct clinical and DNA methylation profiles	[109]
	CREBBP, EP300		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	CREBBP, EP300	ID4	450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Mental retardation (autosomal dominant 23)	SETD5		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
Mental retardation (autosomal dominant 23)	SETD5		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Mental retardation (autosomal dominant 51)	KMT5B		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	KMT5B		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	KMT5B		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Mental retardation (X-linked)	BRWD3, ZNF711, UBE2A, SMS, PHF8/*/		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	BRWD3, ZNF711, UBE2A, SMS, PHF8/*/		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	BRWD3, ZNF711, UBE2A, SMS		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Mitochondrial disease (MD)	MORC2		EPIC, EPICv2	Pleiotropic effects of MORC2 derive from its epigenetic signature	[60]
Mowat–Wilson	ZEB2		EPIC	Identification of the DNA methylation signature of Mowat-Wilson syndrome	[110]
MSL2-related NDDs	MSL2		EPIC	MSL2 variants lead to a neurodevelopmental syndrome with lack of coordination, epilepsy, specific dysmorphisms, and a distinct episignature	[111]
Myopathy, lactic acidosis and sideroblastic anemia 2 (MLASA2)	YARS2		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	YARS2		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Neurodevelopmental disorder with or without autism or seizures (NEDAUS)	CUL3		EPIC	CUL3-related neurodevelopmental disorder: Clinical phenotype of 20 new individuals and identification of a potential phenotype-associated episignature	[112]
Neurodevelopmental disorder with coarse facies and mild distal skeletal abnormalities (NEDCFSA)	KDM6B		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders (Not significant)	[26]
Nicolaides–Baraitser syndrome	SMARCA2		EPIC	New insights into DNA methylation signatures: SMARCA2 variants in Nicolaides-Baraitser syndrome	[113]
	SMARCA2		450 k, EPIC	BAFopathies’ DNA methylation epi-signatures demonstrate diagnostic utility and functional continuum of Coffin–Siris and Nicolaides–Baraitser syndromes	[65]
	SMARCA2		450 k, EPIC	Diagnostic utility of genome-wide DNA methylation testing in genetically unsolved individuals with suspected hereditary conditions	[36]
	SMARCA2		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	SMARCA2		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	SMARCA2		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Phelan–McDermid syndrome	SHANK3, 22q13.3	Large deletions	EPIC	DNA methylation epi-signature is associated with two molecularly and phenotypically distinct clinical subtypes of Phelan-McDermid syndrome	[114]
	22q13.3	Deletions	450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	22q13.3	Deletions	450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Pitt–Hopkins syndrome	TCF4		EPIC	DNA methylation episignature and comparative epigenomic profiling for Pitt-Hopkins syndrome caused by TCF4 variants	[115]
Prader–Willi syndrome	15q11–q13	Hypermethylated SNURF, UPD (15) maternal, epimutation, deletion	450 k	Genome-wide methylation analysis in Silver-Russell syndrome, Temple syndrome, and Prader-Willi syndrome	[116]
Progressive supranuclear palsy (PSP—tauopathies)	17q21.31		450 k	An epigenetic signature in peripheral blood associated with the haplotype on 17q21.31, a risk factor for neurodegenerative tauopathy	[87]
PTBP1-associated syndrome	PTBP1		EPIC	PTBP1 variants displaying altered nucleocytoplasmic distribution are responsible for a neurodevelopmental disorder with skeletal dysplasia	[117]
PURA-related neurodevelopmental disorder	PURA, 5q31.23	Loss of function, deletion	EPIC	Genome-wide epigenetic signatures facilitated the variant classification of the PURA gene and uncovered the pathomechanism of PURA-related neurodevelopmental disorders	[118]
Rahman syndrome	HIST1H1E		EPIC	Frameshift mutations at the C-terminus of HIST1H1E result in a specific DNA hypomethylation signature	[119]
	HIST1H1E		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	HIST1H1E		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	HIST1H1E		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Recurrent constellations of embryonic malformations (RCEM)			EPIC	Identification of a DNA methylation episignature for recurrent constellations of embryonic malformations	[120]
Renpenning syndrome	PQBP1		EPIC	Identification of a DNA methylation signature for Renpenning syndrome (RENS1), a spliceopathy	[121]
	PQBP1		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	PQBP1		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
ReNU syndrome	RNU4-2		EPIC, EPICv2	Dominant variants in major spliceosome U4 and U5 small nuclear RNA genes cause neurodevelopmental disorders through splicing disruption	[122]
ReNU syndrome	RNU4-2		EPICv2	Characterization of snRNA-related neurodevelopmental disorders through the Spanish Undiagnosed Rare Disease Programs	[123]
Rett syndrome	MECP2		450 k	Genomic DNA methylation signatures enable concurrent diagnosis and clinical genetic variant classification in neurodevelopmental syndromes (Not significant)	[45]
Rett syndrome	MECP2		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders (Not significant)	[26]
Rubinstein–Taybi syndrome	CREBBP, EP300		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	CREBBP, EP300		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	CREBBP, EP300		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Saethre–Chotzen syndrome	TWIST		450 k	Genomic DNA methylation signatures enable concurrent diagnosis and clinical genetic variant classification in neurodevelopmental syndromes (Not significant)	[45]
Say–Barber–Biesecker–Young–Simpson syndrome (SBBYSS, Ohdo syndrome)	KAT6B		450 k	Genomic DNA methylation signatures enable concurrent diagnosis and clinical genetic variant classification in neurodevelopmental syndromes	[45]
	KAT6B		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	KAT6B		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	KAT6B		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Schizophrenia			450 k	Screening for rare epigenetic variations in autism and schizophrenia	[50]
Scrap-related syndrome (non-specified)	SRCAP	Proximal variants	EPIC	Truncating SRCAP variants outside the floating-harbor syndrome locus cause a distinct neurodevelopmental disorder with a specific DNA methylation signature	[85]
SETD1B-related syndrome	SETD1B, KMT2B	Deletion, SNVs	EPIC	A genome-wide DNA methylation signature for SETD1B-related syndrome	[124]
	SETD1B		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	SETD1B		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
Sifrim–Hitz–Weiss syndrome (SIHIWES)	CHD4	Nonsense, ATPase domain, PHD domain	EPICv2	Discovery of a DNA methylation profile in individuals with Sifrim-Hitz-Weiss syndrome	[125]
Silver–Russell syndrome (SRS)	H19/IGF2	Maternal UPD7 and ICR1	450 k	Genome-wide methylation analysis in Silver-Russell syndrome patients	[32]
Silver–Russell syndrome (SRS)	H19/IGF2	Loss of methylation	450 k	Genome-wide methylation analysis in Silver-Russell syndrome, Temple syndrome, and Prader-Willi syndrome	[116]
Smith–Magenis syndrome	RAI1		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders (Not significant)	[26]
Sotos syndrome	NSD1		450 k	NSD1 mutations generate a genome-wide DNA methylation signature	[126]
	NSD1		450 k	Comprehensive evaluation of the implementation of episignatures for diagnosis of neurodevelopmental disorders (NDDs)	[101]
	NSD1		450 k	Interplay between histone and DNA methylation seen through comparative methylomes in rare Mendelian disorders	[94]
	NSD1		450 k	Genomic DNA methylation signatures enable concurrent diagnosis and clinical genetic variant classification in neurodevelopmental syndromes	[45]
	NSD1		450 k, EPIC	Diagnostic utility of genome-wide DNA methylation testing in genetically unsolved individuals with suspected hereditary conditions	[36]
	NSD1		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	NSD1		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	NSD1		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Spinal muscular atrophy (SMA)	MORC2		EPIC, EPICv2	Pleiotropic effects of MORC2 derive from its epigenetic signature	[60]
SRSF1-related syndrome	SRSF1		EPIC	SRSF1 haploinsufficiency is responsible for a syndromic developmental disorder associated with intellectual disability	[127]
Tatton-Brown–Rahman syndrome	DNMT3A		EPIC	Growth disrupting mutations in epigenetic regulatory molecules are associated with abnormalities of epigenetic aging	[128]
	DNMT3A		EPIC	Interplay between histone and DNA methylation seen through comparative methylomes in rare Mendelian disorders	[94]
	DNMT3A		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	DNMT3A		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	DNMT3A		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Temple syndrome	14q32.2	Imprinting	450 k	Genome-wide multilocus imprinting disturbance analysis in Temple syndrome and Kagami-Ogata syndrome	[98]
Temple syndrome	14q32.2	UPD14 maternal Epimutation	450 k	Genome-wide methylation analysis in Silver-Russell syndrome, Temple syndrome, and Prader-Willi syndrome	[116]
Unidentified cases (mixed congenital diseases)			450 k	Identification of rare de novo epigenetic variations in congenital disorders	[129]
Weaver syndrome	EZH2		450 k	DNA methylation signature for EZH2 functionally classifies sequence variants in three PRC2 complex genes	[130]
	EZH2		450 k	Genomic DNA methylation signatures enable concurrent diagnosis and clinical genetic variant classification in neurodevelopmental syndromes (Not significant)	[45]
	EZH2		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	EZH2		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Werner syndrome	WRN		EPIC	Genome-wide DNA methylation analysis in blood cells from patients with Werner syndrome	[131]
Werner syndrome	WRN, LMNA, POLD1		EPIC	Epigenetic signatures of Werner syndrome occur early in life and are distinct from normal epigenetic aging processes	[132]
Wiedemann–Steiner syndrome	KMT2A		EPIC	Clinical utility of a unique genome-wide DNA methylation signature for KMT2A-related syndrome	[133]
	KMT2A		450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	KMT2A		450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	KMT2A		450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Williams syndrome	7q11.23	Deletion	450 k	Symmetrical dose-dependent DNA-methylation profiles in children with deletion or duplication of 7q11.23	[37]
	7q11.23	Deletion	450 k	Integrated DNA methylation analysis reveals a potential role for ANKRD30B in Williams syndrome	[134]
	7q11.23	Deletion	450 k	Diagnostic utility of genome-wide DNA methylation testing in genetically unsolved individuals with suspected hereditary conditions	[36]
	7q11.23	Deletion	450 k, EPIC	Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 Mendelian neurodevelopmental disorders	[26]
	7q11.23	Deletion	450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	7q11.23	Deletion	450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]
Witteveen–Kolk syndrome	SIN3A		EPIC	DNA methylation episignature for Witteveen-Kolk syndrome due to SIN3A haploinsufficiency	[135]
Wolf–Hirschhorn syndrome	4q16.13, NSD2	Deletion	EPIC	Loss of function in NSD2 causes DNA methylation signature similar to that in Wolf-Hirschhorn syndrome	[136]
	4q16.13	Deletion	450 k, EPIC	Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders	[40]
	4q16.13	Deletion	450 k, EPIC	Functional correlation of genome-wide DNA methylation profiles in genetic neurodevelopmental disorders	[38]

Table 2. Computational resources for episignature development used in any step of the pipeline, extracted from the literature.

Step	Package	Source	Usage	Ref.
Array processing	minfi	R Bioconductor	.idat reading and processing.	[146]
	ChAMP	R Bioconductor	.idat reading and processing.	[147,148]
	lumi	R Bioconductor	.idat reading and processing.	[149]
	SeSAMe	R Bioconductor	.idat reading and processing.	[150]
	meffil	R Bioconductor	.idat reading and processing.	[151]
	RnBeads	R Bioconductor	.idat reading and processing.	[152,153]
	Genome Studio	Illumina	.idat reading and processing (official Illumina tool).
	IMA	R	.idat reading.	[154]
	MethylCallR	R Bioconductor	.idat reading and processing.	[155]
	MethAid	R Bioconductor	.idat reading and processing.	[156]
	wateRmelon	R Bioconductor	.idat reading and processing.	[157]
	maxProbes	R	CpG QC filters.
	sva	R	Surrogate variable identification and batch correction.	[158]
NGS processing	Trim Galore!	Babraham Bioinformatics (Bash)	FASTQ quality control.
NGS processing	Bismark	Babraham Bioinformatics (Bash)	Bisulfite sequencing aligner and methylation caller.
Epigenetic analysis and episignature development	RnBeads	R Bioconductor	Methylation processing data. Allow .bismarkCov files.	[152,153]
	limma	R Bioconductor	Array data analysis.	[159,160]
	ChAMP	R Bioconductor	Array data analysis.	[147,148]
	meffil	R Bioconductor	Array data analysis.	[151]
	Qlucore	Web-based	Methylation analysis.
	minfi	R Bioconductor	DMP analysis. Blood cells proportion estimation (Houseman). DMR analysis (bumphunter).	[146]
	FlowSorted. Blood.EPIC	R Bioconductor	Blood cells proportion estimation.	[161]
	EpiDISH	R Bioconductor	Blood cells proportion estimation.	[162]
	DMRcate	R Bioconductor	DMR analysis.	[163]
	Golden Helix—SNP and variation	Independent app.	DMP analysis.
	MedCalc	Independent app	DMP analysis.
	WGCNA	R	Weighted correlation network analysis. Identification of co-methylation system networks.	[164]
	bumphunter	R Bioconductor	DMR analysis.	[165]
	comp-p	R	DMR analysis.	[166]
	mRMRe	R	Feature selection through minimum-redundancy-maximum-relevance ensemble algorithm.	[167]
	DNAmAge	Web-based	Epigenetic age estimation (Horvath clock).
	methylclock	R Bioconductor	Epigenetic age estimation.	[168]
Annotation and enrichment	Illumina annotation packages	R Bioconductor	CpG annotation for all Illumina arrays.
	annotatr	R Bioconductor	Region annotation.	[169]
	LOLA	R Bioconductor	Region annotation.	[170]
	missMethyl	R Bioconductor	CpG enrichment analysis.	[171,172]
	methylGSA	R Bioconductor	CpG enrichment analysis.	[173]
	GSEA	R	Gene enrichment analysis.	[174]
ML model training	Caret	R	Train machine learning models. Offers wide range of customization, balancing, and CV techniques.
	Kernlab	R	Train support vector machines.	[175]
	e1071	R	Train support vector machines
	CALF	R	Train coarse approximation linear function.
	glmnet	R	Train ridge, lasso and elastic net regression.
	scikit-learn	Python	Train machine learning models. Offers wide range of customization, balancing and CV techniques.
	GenPipes	Bash	NGS workflow management.	[176]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

Share and Cite

MDPI and ACS Style

Alegret-García, A.; Cáceres, A.; Sevilla-Porras, M.; Pérez-Jurado, L.A.; González, J.R. Analysis Methods for Diagnosing Rare Neurodevelopmental Diseases with Episignatures: A Systematic Review of the Literature. Biomedicines 2025, 13, 3043. https://doi.org/10.3390/biomedicines13123043

AMA Style

Alegret-García A, Cáceres A, Sevilla-Porras M, Pérez-Jurado LA, González JR. Analysis Methods for Diagnosing Rare Neurodevelopmental Diseases with Episignatures: A Systematic Review of the Literature. Biomedicines. 2025; 13(12):3043. https://doi.org/10.3390/biomedicines13123043

Chicago/Turabian Style

Alegret-García, Albert, Alejandro Cáceres, Marta Sevilla-Porras, Luís A. Pérez-Jurado, and Juan R. González. 2025. "Analysis Methods for Diagnosing Rare Neurodevelopmental Diseases with Episignatures: A Systematic Review of the Literature" Biomedicines 13, no. 12: 3043. https://doi.org/10.3390/biomedicines13123043

APA Style

Alegret-García, A., Cáceres, A., Sevilla-Porras, M., Pérez-Jurado, L. A., & González, J. R. (2025). Analysis Methods for Diagnosing Rare Neurodevelopmental Diseases with Episignatures: A Systematic Review of the Literature. Biomedicines, 13(12), 3043. https://doi.org/10.3390/biomedicines13123043

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Analysis Methods for Diagnosing Rare Neurodevelopmental Diseases with Episignatures: A Systematic Review of the Literature

Abstract

1. Introduction

2. Materials and Methods

3. Results

3.1. Testing Episignatures

3.1.1. Testing Episignatures: External Resources

3.1.2. Testing Episignatures: Public Episignatures

3.2. Episignature Development

3.2.1. Episignature Development: Data Formats

3.2.2. Episignature Development: DNA Methylation Array Processing

3.2.3. Episignature Development: Multi-Cohort Studies

3.2.4. Episignature Development: Methylation Screening Using Sequencing Methods

3.2.5. Episignature Development: Differentially Methylated Position (DMP) Detection

3.2.6. Episignature Development: Other Approaches to Linear Models for Episignatures

3.2.7. Episignature Development: DMRs and Epimutations

3.2.8. Episignature Development: Annotation and Enrichment

3.2.9. Episignature Development: Classification Model Introduction

3.2.10. Episignature Development: Classification Model Pre-Filtering

3.2.11. Episignature Development: Classification Model Design

4. Discussion

5. Conclusions

Supplementary Materials

Author Contributions

Funding

Data Availability Statement

Acknowledgments

Conflicts of Interest

Abbreviations

References

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI