ENCODE Project Research Articles

Extracting regulatory active chromatin footprint from cell-free DNA

Cell-free DNA (cfDNA) has emerged as a pivotal player in precision medicine, revolutionizing the diagnostic and therapeutic landscape. While its clinical applications have significantly increased in recent years, current cfDNA assays have limited ability to identify the active transcriptional programs that govern complex disease phenotypes and capture the heterogeneity of the disease. To address these limitations, we have developed a non-invasive platform to enrich and examine the active chromatin fragments (cfDNAac) in peripheral blood. The deconvolution of the cfDNAac signal from traditional nucleosomal chromatin fragments (cfDNAnuc) yields a catalog of features linking these circulating chromatin signals in blood to specific regulatory elements across the genome, including enhancers, promoters, and highly transcribed genes, mirroring the epigenetic data from the ENCODE project. Notably, these cfDNAac counts correlate strongly with RNA polymerase II activity and exhibit distinct expression patterns for known circadian genes. Additionally, cfDNAac signals across gene bodies and promoters show strong correlations with whole blood gene expression levels defined by GTEx. This study illustrates the utility of cfDNAac analysis for investigating epigenomics and gene expression, underscoring its potential for a wide range of clinical applications in precision medicine.

Interdependent regulation of alternative splicing by SR and hnRNP proteins.

Alternative pre-mRNA splicing is a combinatorial process involving SR and hnRNP splicing factors. These proteins can silence or enhance splicing based on their expression levels and binding positions. To better understand their combinatorial and interdependent regulation, computational analyses were performed using HepG2 and K562 cell knockdown and binding datasets from the ENCODE Project. Analyses of diMerential splicing for 6 SR proteins and 13 hnRNP knockdowns revealed statistically significant exon overlap among most RBP combinations, albeit at diMerent levels. Neither SR proteins nor hnRNPs showed strong preferences for collaborating with specific RBP classes in mediating exon inclusion. While SRSF1, hnRNPK, and hnRNPC stand out as major influencers of alternative splicing, they do so predominantly independent of other RBPs. Meanwhile, minor influencers of alternative splicing such as hnRNPAB and hnRNPA0 predominantly regulate exon inclusion in concert with other RBPs, indicating that inclusion can be mediated by both single and multiple RBPs. Interestingly, the higher the number of RBPs that regulate the inclusion of an exon, the more variable exon inclusion preferences become. Interdependently regulated exons are more modular and have diMerent physical characteristics such as reduced exon length compared to their independent counterparts. A comparison of RBP interdependence between HepG2 and K562 cells provides the framework that explains cell-type-specific alternative splicing. Our study highlights the importance of the interdependent regulation of alternative exons and identifies characteristics of interdependently regulated exons that diMer from independently regulated exons.

Transposable elements contribute to tissue-specific gene regulation in humans.

Transposable elements (TEs) contribute to approximately half of the human genome, and along with many other functions, they have been known to play a role in gene regulation in the genome. With TEs' active/repressed states varying across tissue and cell types, they have the potential to regulate gene expression in a tissue-specific manner. To provide a systematic analysis of TEs' contribution in tissue-specific gene regulation, we examined the regulatory elements and genes in association with TE-derived regulatory sequences in 14 human cell lines belonging to 10 different tissue types using the functional genomics data from the ENCODE project. Specifically, we separately analyzed regulatory regions identified by three different approaches (DNase hypersensitive sites (DHS), histone active sites (HA), and histone repressive sites (HR)). These regulatory regions showed to be distinct from each other by sharing less than 2.5% among all three types and more than 95% showed to be cell line-specific. Despite a lower total TE content overall than the genome average, each regulatory sequence type showed enrichment for one or two specific TE type(s): DHS for long terminal repeats (LTRs) and DNA transposons, HA for short interspersed nucleotide elements (SINEs), and HR for LTRs. In contrast, SINE was shown to be overrepresented in all three types of regulatory sequences located in gene-neighboring regions. TE-regulated genes were mostly shown to have cell line specific pattern, and tissue-specific genes (TSGs) showed higher usage of TE regulatory sequences in the tissue of their expression. While TEs in the regulatory sequences showed to be older than their genome-wide counterparts, younger TEs were shown to be more likely used in cell line specific regulatory sequences. Collectively, our study provided further evidence enforcing an important contribution of TEs to tissue-specific gene regulation in humans.

On the limits of inferring biophysical parameters of RBP-RNA interactions from in vitro RNA Bind’n Seq data

We develop a thermodynamic model describing the binding of RNA binding proteins (RBP) to oligomers in vitro. We apply expectation-maximization to infer the specificity of RBPs, represented as position-specific weight matrices (PWMs), by maximizing the likelihood of RNA Bind’n Seq data from the ENCODE project. Analyzing these public data we find sequence motifs that can partly explain the data for more than half of the studied 111 RBPs, and for 48 of the proteins these motifs are consistent with the known specificity. Our code is publicly available, facilitating analysis of RBP binding data.

Transcriptional Hubs Within Cliques in Ensemble Hi-C Chromatin Interaction Networks.

Chromatin conformation capture technologies permit the study of chromatin spatial organization on a genome-wide scale at a variety of resolutions. Despite the increasing precision and resolution of high-throughput chromatin conformation capture (Hi-C) methods, it remains challenging to conclusively link transcriptional activity to spatial organizational phenomena. We have developed a clique-based approach for analyzing Hi-C data that helps identify chromosomal hotspots that feature considerable enrichment of chromatin annotations for transcriptional start sites and, building on previously published work, show that these chromosomal hotspots are not only significantly enriched in RNA polymerase II binding sites as identified by the ENCODE project, but also identify a noticeable increase in FANTOM5 and GTEx transcription within our identified cliques across a variety of tissue types. From the obtained data, we surmise that our cliques are a suitable method for identifying transcription factories in Hi-C data, and outline further extensions to the method that may make it useful for locating regions of increased transcriptional activity in datasets where in-depth expression or polymerase data may not be available.

Pan-evolutionary and regulatory genome architecture delineated by an integrated macro- and microsynteny approach.

The forthcoming massive genome data generated by the Earth BioGenome Project will open up a new era of comparative genomics, for which genome synteny analysis provides an important framework. Profiling genome synteny represents an essential step in elucidating genome architecture, regulatory blocks/elements and their evolutionary history. Here we describe PanSyn, ( https://github.com/yhw320/PanSyn ), the most comprehensive and up-to-date genome synteny pipeline, providing step-by-step instructions and application examples to demonstrate its usage. PanSyn inherits both basic and advanced functions from existing popular tools, offering a user-friendly, highly customized approach for genome macrosynteny analysis and integrated pan-evolutionary and regulatory analysis of genome architecture, which are not yet available in public synteny software or tools. The advantages of PanSyn include: (i) advanced microsynteny analysis by functional profiling of microsynteny genes and associated regulatory elements; (ii) comprehensive macrosynteny analysis, including the inference of karyotype evolution from ancestors to extant species; and (iii) functional integration of microsynteny and macrosynteny for pan-evolutionary profiling of genome architecture and regulatory blocks, as well as integration with external functional genomics datasets from three- or four-dimensional genome and ENCODE projects. PanSyn requires basic knowledge of the Linux environment and Perl programming language and the ability to access a computer cluster, especially for large-scale genomic comparisons. Our protocol can be easily implemented by a competent graduate student or postdoc and takes several days to weeks to execute for dozens to hundreds of genomes. PanSyn provides yet the most comprehensive and powerful tool for integrated evolutionary and functional genomics.

Gender-specific genetic and epigenetic signatures in cardiovascular disease.

Cardiac sex differences represent a pertinent focus in pursuit of the long-awaited goal of personalized medicine. Despite evident disparities in the onset and progression of cardiac pathology between sexes, historical oversight has led to the neglect of gender-specific considerations in the treatment of patients. This oversight is attributed to a predominant focus on male samples and a lack of sex-based segregation in patient studies. Recognizing these sex differences is not only relevant to the treatment of cisgender individuals; it also holds paramount importance in addressing the healthcare needs of transgender patients, a demographic that is increasingly prominent in contemporary society. In response to these challenges, various agencies, including the National Institutes of Health, have actively directed their efforts toward advancing our comprehension of this phenomenon. Epigenetics has proven to play a crucial role in understanding sex differences in both healthy and disease states within the heart. This review presents a comprehensive overview of the physiological distinctions between males and females during the development of various cardiac pathologies, specifically focusing on unraveling the genetic and epigenetic mechanisms at play. Current findings related to distinct sex-chromosome compositions, the emergence of gender-biased genetic variations, and variations in hormonal profiles between sexes are highlighted. Additionally, the roles of DNA methylation, histone marks, and chromatin structure in mediating pathological sex differences are explored. To inspire further investigation into this crucial subject, we have conducted global analyses of various epigenetic features, leveraging data previously generated by the ENCODE project.

Racoon_clip-a complete pipeline for single-nucleotide analyses of iCLIP and eCLIP data.

A vast variety of biological questions connected to RNA-binding proteins can be tackled with UV crosslinking and immunoprecipitation (CLIP) experiments. However, the processing and analysis of CLIP data are rather complex. Moreover, different types of CLIP experiments like iCLIP or eCLIP are often processed in different ways, reducing comparability between multiple experiments. Therefore, we aimed to build an easy-to-use computational tool for the processing of CLIP data that can be used for both iCLIP and eCLIP data, as well as data from other truncation-based CLIP methods. Here, we introduce racoon_clip, a sustainable and fully automated pipeline for the complete processing of iCLIP and eCLIP data to extract RNA binding signal at single-nucleotide resolution. racoon_clip is easy to install and execute, with multiple pre-settings and fully customizable parameters, and outputs a conclusive summary report with visualizations and statistics for all analysis steps. racoon_clip is implemented as a Snakemake-powered command line tool (Snakemake version ≥7.22, Python version ≥3.9). The latest release can be downloaded from GitHub (https://github.com/ZarnackGroup/racoon_clip/tree/main) and installed via pip. A detailed documentation, including installation, usage, and customization, can be found at https://racoon-clip.readthedocs.io/en/latest/. The example datasets can be downloaded from the Short Read Archive (SRA; iCLIP: SRR5646576, SRR5646577, SRR5646578) or the ENCODE Project (eCLIP: ENCSR202BFN).

Association between acetaminophen metabolites and CYP2E1 DNA methylation level in neonate cord blood in the Boston Birth Cohort

BackgroundAcetaminophen is a commonly used medication by pregnant women and is known to cross the placenta. However, little is known about the biological mechanisms that regulate acetaminophen in the developing offspring. Cytochrome 2E1 (CYP2E1) is the primary enzyme responsible for the conversion of acetaminophen to its toxic metabolite. Ex vivo studies have shown that the CYP2E1 gene expression in human fetal liver and placenta is largely controlled by DNA methylation (DNAm) at CpG sites located in the gene body of CYP2E1 at the 5’ end. To date, no population studies have examined the association between acetaminophen metabolite and fetal DNAm of CYP2E1 at birth.MethodsWe utilized data from the Boston Birth Cohort (BBC) which represents an urban, low-income, racially and ethnically diverse population in Boston, Massachusetts. Acetaminophen metabolites were measured in the cord plasma of newborns enrolled in BBC between 2003 and 2013 using liquid chromatography-tandem mass spectrometry. DNAm at 28 CpG sites of CYP2E1 was measured by Illumina Infinium MethylationEPIC BeadChip. We used linear regression to identify differentially methylated CpG sites and the “DiffVar” method to identify differences in methylation variation associated with the detection of acetaminophen, adjusting for cell heterogeneity and batch effects. The false discovery rate (FDR) was calculated to account for multiple comparisons.ResultsAmong the 570 newborns included in this study, 96 (17%) had detectable acetaminophen in cord plasma. We identified 7 differentially methylated CpGs (FDR < 0.05) associated with the detection of acetaminophen and additional 4 CpGs showing a difference in the variation of methylation (FDR < 0.05). These CpGs were all located in the gene body of CYP2E1 at the 5’ end and had a 3–6% lower average methylation level among participants with detectable acetaminophen compared to participants without. The CpG sites we identified overlap with previously identified DNase hypersensitivity and open chromatin regions in the ENCODE project, suggesting potential regulatory functions.ConclusionsIn a US birth cohort, we found detection of cord biomarkers of acetaminophen was associated with DNAm level of CYP2E1 in cord blood. Our findings suggest that DNA methylation of CYP2E1 may be an important regulator of acetaminophen levels in newborns.

The CUT&RUN suspect list of problematic regions of the genome

BackgroundCleavage Under Targets and Release Using Nuclease (CUT&RUN) is an increasingly popular technique to map genome-wide binding profiles of histone modifications, transcription factors, and co-factors. The ENCODE project and others have compiled blacklists for ChIP-seq which have been widely adopted: these lists contain regions of high and unstructured signal, regardless of cell type or protein target, indicating that these are false positives. While CUT&RUN obtains similar results to ChIP-seq, its biochemistry and subsequent data analyses are different. We found that this results in a CUT&RUN-specific set of undesired high-signal regions.ResultsWe compile suspect lists based on CUT&RUN data for the human and mouse genomes, identifying regions consistently called as peaks in negative controls. Using published CUT&RUN data from our and other labs, we show that the CUT&RUN suspect regions can persist even when peak calling is performed with SEACR or MACS2 against a negative control and after ENCODE blacklist removal. Moreover, we experimentally validate the CUT&RUN suspect lists by performing reiterative negative control experiments in which no specific protein is targeted, showing that they capture more than 80% of the peaks identified.ConclusionsWe propose that removing these problematic regions can substantially improve peak calling in CUT&RUN experiments, resulting in more reliable datasets.

On the limits of inferring biophysical parameters of RBP-RNA interactions from in vitro RNA Bind'n Seq data.

We develop a thermodynamic model describing the binding of RNA binding proteins (RBP) to oligomers in vitro. We apply expectation-maximization to infer the specificity of RBPs, represented as position-specific weight matrices (PWMs), by maximizing the likelihood of RNA Bind'n Seq data from the ENCODE project. We demonstrate that the model can reproduce known specificities for well-studied proteins and that in some cases we predict novel, longer binding motifs. However, the model does not recover all the motifs that are in principle known, indicating that the data is not well explained by a single underlying biophysical model. Our code is publicly available.

Improved analysis of (e)CLIP data with RCRUNCH yields a compendium of RNA-binding protein binding sites and motifs

We present RCRUNCH, an end-to-end solution to CLIP data analysis for identification of binding sites and sequence specificity of RNA-binding proteins. RCRUNCH can analyze not only reads that map uniquely to the genome but also those that map to multiple genome locations or across splice boundaries and can consider various types of background in the estimation of read enrichment. By applying RCRUNCH to the eCLIP data from the ENCODE project, we have constructed a comprehensive and homogeneous resource of in-vivo-bound RBP sequence motifs. RCRUNCH automates the reproducible analysis of CLIP data, enabling studies of post-transcriptional control of gene expression.

Chicken chromatin accessibility atlas accelerates epigenetic annotation of birds and gene fine-mapping associated with growth traits.

The development of epigenetic maps, such as the ENCODE project in humans, provides resources for gene regulation studies and a reference for research of disease-related regulatory elements. However, epigenetic information, such as a bird-specific chromatin accessibility atlas, is currently lacking for the thousands of bird species currently described. The major genomic difference between birds and mammals is their shorter introns and intergenic distances, which seriously hinders the use of humans and mice as a reference for studying the function of important regulatory regions in birds. In this study, using chicken as a model bird species, we systematically compiled a chicken chromatin accessibility atlas using 53 Assay of Transposase Accessible Chromatin sequencing (ATAC-seq) samples across 11 tissues. An average of 50 796 open chromatin regions were identified per sample, cumulatively accounting for 20.36% of the chicken genome. Tissue specificity was largely reflected by differences in intergenic and intronic peaks, with specific functional regulation achieved by two mechanisms: recruitment of several sequence-specific transcription factors and direct regulation of adjacent functional genes. By integrating data from genome-wide association studies, our results suggest that chicken body weight is driven by different regulatory variants active in growth-relevant tissues. We propose CAB39L (active in the duodenum), RCBTB1 (muscle and liver), and novel long non-coding RNA ENSGALG00000053256 (bone) as candidate genes regulating chicken body weight. Overall, this study demonstrates the value of epigenetic data in fine-mapping functional variants and provides a compendium of resources for further research on the epigenetics and evolution of birds and mammals.

Transcriptome - wide bioinformatics analysis of the binding sites of RNA - binding proteins and their putative role in mendelian diseases

Background: Post-transcriptional regulation is the control of gene expression at the RNA level. After produced, the stability and distribution of the different transcripts are regulated by means of RNA-binding proteins (RBPs). Mutations in RNA-binding proteins can cause Mendelian diseases - prominently neuromuscular disorders and cancers. This study determines the interaction between RBPs and target-RNA complexes from public data of the ENCODE project and identifies mutations associated with Mendelian diseases that could disrupt the RBP-RNA interactions. Materials and methods: we performed a transcriptome - wide bioinformatics prediction of the binding sites of RBPs in the human transcriptome from public data of the ENCODE project. Results: The majority (54%) of pathogenic mutation putatively affecting the binding sites of RBPs are located in protein - coding genes and are mainly classified as loss - of - function mutations. Mutations located in the binding sites of RBPs related to RNA processing. For 13 diseases, Familial hypercholesterolemia is the most significant disease with about 40% of mutations in ClinVar database located into the binding sites of RBPs (p=2.3e-65), but congenital hypogonadotropic hypogonadism is the disease with the highest percentage of mutations affecting the binding sites of RBPs (98%, p=2.7e-25). The RBPs most involved in human Mendelian diseases by binding sites-disrupting mutations are YBX3, AQR and PRPF8. Conclusions: A large number of Mendelian diseases are potentially mediated by disease - causing variants that potentially disrupt the binding sites of RBPs. This will provide insight sharper on post - transcriptional mechanisms. Besides, it is useful to know the role of protein - RNA interactome networks in pathologies, thereby serving the treatment of diseases. Key words: bioinformatics analysis, ENCODE project, ClinVar, RNA-binding proteins, Mendelian diseases

A Systems Biology Approach on the Regulatory Footprint of Human Endogenous Retroviruses (HERVs).

Human endogenous retroviruses (HERVs) are a family of endogenous retroviruses that comprise the ~8.93% of the human genome sequence, with a high proportion being human specific. The recent expansion of repeated HERV sequences has offered a framework for genetic and epigenetic innovation. In the current report, a systematic approach is implemented to catalogue regulatory elements within HERVs, as a roadmap to potential functions of HERV sequences in gene networks. ENCODE Project has offered a wealth of epigenetic data based on omics technologies. I analyzed the presence of HERV sequences on consensus cis-regulatory elements (cCREs) from ENCODE data. On the one side, HERVs are in 1 out of 9 cCREs (&gt;100.000 cCREs in total), dispersed within the genome and present in cis-regulatory regions of ~81% of human genes, as calculated following gene enrichment analysis. On the other side, promoter-associated HERV cCREs are present adjacent to (in a 200 bp window) the transcription start sites of 256 human genes. Regulatory network production, followed by centrality analysis led to the discovery of 90 core genes containing HERV-associated promoters. Pathway analysis on the core network genes and their immediate neighbors revealed a regulatory footprint that, among others, is associated with inflammation, chemokine signaling and response to viral infection. Collectively, these results support the concept that the expansion of regulatory sequences derived from HERVs is critical for epigenetic innovation that may have wired together genes into novel transcriptional networks with critical roles in cellular physiology and pathology.

Interpretable deep learning translation of GWAS and multi-omics findings to identify pathobiology and drug repurposing in Alzheimer's disease.

Translating human genetic findings (genome-wide association studies [GWAS]) to pathobiology and therapeutic discovery remains a major challenge for Alzheimer's disease (AD). We present a network topology-based deep learning framework to identify disease-associated genes (NETTAG). We leverage non-coding GWAS loci effects on quantitative trait loci, enhancers and CpG islands, promoter regions, open chromatin, and promoter flanking regions under the protein-protein interactome. Via NETTAG, we identified 156 AD-risk genes enriched in druggable targets. Combining network-based prediction and retrospective case-control observations with 10 million individuals, we identified that usage of four drugs (ibuprofen, gemfibrozil, cholecalciferol, and ceftriaxone) is associated with reduced likelihood of AD incidence. Gemfibrozil (an approved lipid regulator) is significantly associated with 43% reduced risk of AD compared with simvastatin using an active-comparator design (95% confidence interval 0.51-0.63, p<0.0001). In summary, NETTAG offers a deep learning methodology that utilizes GWAS and multi-genomic findings to identify pathobiology and drug repurposing in AD.

Identification of Enriched Regions in ChIP-Seq Data via a Linear-Time Multi-Level Thresholding Algorithm.

Chromatin immunoprecipitation (ChIP-Seq) has emerged as a superior alternative to microarray technology as it provides higher resolution, less noise, greater coverage and wider dynamic range. While ChIP-Seq enables probing of DNA-protein interaction over the entire genome, it requires the use of sophisticated tools to recognize hidden patterns and extract meaningful data. Over the years, various attempts have resulted in several algorithms making use of different heuristics to accurately determine individual peaks corresponding to unique DNA-protein. However, finding all the significant peaks with high accuracy in a reasonable time is still a challenge. In this work, we propose the use of Multi-level thresholding algorithm, which we call LinMLTBS, used to identify the enriched regions on ChIP-Seq data. Although various suboptimal heuristics have been proposed for multi-level thresholding, we emphasize on the use of an algorithm capable of obtaining an optimal solution, while maintaining linear-time complexity. Testing various algorithm on various ENCODE project datasets shows that our approach attains higher accuracy relative to previously proposed peak finders while retaining a reasonable processing speed.

Survey of the binding preferences of RNA-binding proteins to RNA editing events

BackgroundAdenosine-to-inosine (A-to-I) editing is an important RNA posttranscriptional process related to a multitude of cellular and molecular activities. However, systematic characterizations of whether and how the events of RNA editing are associated with the binding preferences of RNA sequences to RNA-binding proteins (RBPs) are still lacking.ResultsWith the RNA-seq and RBP eCLIP-seq datasets from the ENCODE project, we quantitatively survey the binding preferences of 150 RBPs to RNA editing events, followed by experimental validations. Such analyses of the RBP-associated RNA editing at nucleotide resolution and genome-wide scale shed light on the involvement of RBPs specifically in RNA editing-related processes, such as RNA splicing, RNA secondary structures, RNA decay, and other posttranscriptional processes.ConclusionsThese results highlight the relevance of RNA editing in the functions of many RBPs and therefore serve as a resource for further characterization of the functional associations between various RNA editing events and RBPs.

Integrated bioinformatics analysis reveals that EZH2-rich domains promote transcriptional repression in cervical cancer.

Cervical cancer is the third female cancer most common worldwide. The carcinogenic process involves an alteration of the mechanisms associated with transcription. Several studies have reported an oncogenic role of the polycomb complex subunit, EZH2. However, the role of EZH2 in cervical cancer is unknown. Hence, the objective of this study was to determine the role of EZH2 in transcriptional regulation in cervical cancer. The EZH2 expression and the methylation status of its promoter were analyzed in The Cancer Genome Atlas. The EZH2 enrichment profile was analyzed using chromatin immunoprecipitation with massively parallel DNA sequencing data provided by ENCODE project. The chromatin compartments were identified in the 4D Nucleome Data Portal. The functional annotation was examined in Enrichr. We report that EZH2 expression is increased in cervical cancer which is associated with hypomethylation of its promoter. EZH2 is enriched at promoter and distal intergenic regions. We identified that EZH2 defines chromatin domains enriched with H3K27me3 within repressive compartments in the HeLa-S3 cell line. Additionally, high EZH2 expression is associated with the repression of the senescent phenotype in cervical cancer patients. Our results suggest the participation of EZH2 in the generation of domains with a silencer function in cervical cancer, which regulate the expression of genes associated with cellular senescence.

Abstract 1936: MethylationToRegulation: A deep-learning approach to infer chromatin properties from DNA methylomes

Abstract Gene regulation is critical for cell identity, and its dysregulation is a defining characteristic of common diseases including cancers. Although promoter activity is a strong predictor of gene expression, activities at other regulatory elements, including enhancers and super-enhancers (SE), are major contributors to gene regulation. For example, distinct transcription factor-regulating SEs define neuroblastoma (NB) subtypes with distinct clinical outcomes. However, technical limitations largely prevent ChIP-seq based enhancer/SE activity profiling in primary patient samples. We previously developed MethylationToActivity (M2A) and demonstrated that high-order DNA methylation (DNAm) features are strong predictors of promoter activity measured by ChIP-seq. This tool is important because genomewide DNAm assays are widely used in clinic to classify tumors and stratify patients on clinical trials. However, it is limited by its reliance on known gene position annotations. Here we present MethylationToRegulation (M2R), a method that utilizes a convolutional neural network (CNN)-based deep learning framework to infer intergenic enhancer/SE activities from DNA methylomes. We obtained paired WGBS and H3K27ac ChIP-seq data for 16 pediatric NB samples profiled in the Pediatric Cancer Genome Project. M2R was trained on 6 samples and tested on the remaining 10 samples. It achieved an average relative prediction accuracy of 84% on test samples when compared to the H3K27ac ChIP-seq replicate consistency from the ENCODE project. We adapted the Rank Ordering of Super-Enhancers algorithm to interpret H3K27ac signals inferred from DNA methylomes to identify SEs. M2R faithfully captured subtype-defining SEs with high specificity, including those associated with master NB transcription factors. Our results demonstrate that M2R accurately quantifies enhancer/SE activities and infers critical epigenetic marks from DNA methylomes. Application of M2R will enable the profiling of epigenetic dysregulation in patient tumor samples, seeking to improve clinical outcomes. Citation Format: Daniel K. Putnam, Brian J. Abraham, Xiang Chen. MethylationToRegulation: A deep-learning approach to infer chromatin properties from DNA methylomes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1936.