C-G.40: From DMPs to DMRs: CMEnt, Characterization of Methylation using positional ENTanglement
Authors
Ken Op de Beeck
Center of Medical Genetics, University of Antwerp and Antwerp University Hospital
Joris Vermeesch
Center for Human Genetics Leuven, University Hospital Leuven
Timon Vandamme
Integrated Personalized and Precision Oncology Network (IPPON), Center for Oncological Research (CORE), University of Antwerp, and Antwerp University Hospital
Guy Van Camp
Center of Medical Genetics, University of Antwerp, and Antwerp University Hospital
Joe Ibrahim
Center of Medical Genetics, University of Antwerp, and Antwerp University Hospital
Keywords
epigenetics,differential methylation,microarray,long read sequencing
[sponser-meet-now-chat][/sponser-meet-now-chat]
DNA cytosine methylation is a spatially structured epigenetic modification, with neighboring CpG sites often sharing regulatory context and coordinated methylation behavior. In disease, and particularly in cancer, differential methylation frequently emerges as regional alterations rather than isolated single-site events. However, many DMR detection workflows require method-specific smoothing, windowing, or segmentation assumptions, and do not directly exploit a curated set of differentially methylated positions as input evidence. Here, we present CMEnt, Characterization of Methylation through positional Entanglement, an R/Bioconductor package for seed-guided identification and characterization of differentially methylated regions. CMEnt starts from user-defined seed loci, typically significant DMPs, and constructs DMRs by testing local within-group methylation correlation among proximal sites. Candidate regions are assembled through seed connection, expanded to neighboring methylation sites, and merged into coherent intervals, enabling a transparent transition from position-level differential methylation to region-level interpretation. The package supports Illumina microarrays, BED-like methylation tables, BSseq objects, and disk-backed sequencing-derived methylation data, including Oxford Nanopore bedMethyl inputs. In two cancer methylation case studies, CMEnt converted large DMP sets into interpretable regional methylation landscapes, identifying AML-associated CpG-rich regulatory regions and PDAC tumour-stroma compartment-specific DMRs with enriched adhesion, developmental and transcriptional regulatory signals. Together, these results support CMEnt as a flexible and complementary framework for DMP-guided DMR assembly across array- and sequencing-based methylation studies.
Co-authors: Ken Op de Beek, Joris Vermeesch, Timon Vandamme, Guy Van Camp, Joe Ibrahim
Contact Attendee
Warning: Attempt to read property "user_email" on string in /home/1276969.cloudwaysapps.com/ydbgzhdjeq/public_html/wp-content/plugins/my-conference-now/functions.php on line 2826