Related Experiment Video
Updated: Nov 4, 2025
13:21
Comprehensive DNA Methylation Analysis Using a Methyl-CpG-binding Domain Capture-based Method in Chronic Lymphocytic Leukemia Patients
Published on: June 16, 2017
10.2K
EnMCB: an R/bioconductor package for predicting disease progression based on methylation correlated blocks using
Xin Yu1,2, De-Xin Kong1,2
1State Key Laboratory of Agricultural Microbiology, Huazhong Agricultural University, Wuhan 430070, China.
Bioinformatics (Oxford, England)
|May 29, 2021
Summary
A new R package, EnMCB, identifies cancer by analyzing DNA methylation correlated blocks. This tool aids in predicting patient survival using machine learning models on CpG methylation patterns.
Area of Science:
- Genomics
- Bioinformatics
- Cancer Research
Background:
- DNA methylation patterns, particularly contiguous cytosine-phosphorothioate-guanine (CpG) sites, are crucial in biological processes.
- Co-methylation of CpG sites within blocks suggests coordinated epigenetic regulation.
- Current diagnostic tools lack methods to model these methylation correlated blocks effectively.
Purpose of the Study:
- To develop an R package for building predictive models based on DNA methylation correlated blocks.
- To enable cancer diagnosis and survival prediction using epigenetic signatures.
- To address the absence of computational tools for analyzing methylation correlated blocks.
Main Methods:
- Development of the EnMCB (ensemble of machine learning models for methylation correlated blocks) R package.
- Automated partitioning of the genome into methylation correlated blocks of co-methylated CpG sites.
- Application of an ensemble model combining Cox regression, support vector regression, mboost, and elastic-net.
Main Results:
- The EnMCB package successfully builds signatures from DNA methylation correlated blocks.
- The models demonstrated diagnostic capacity for predicting patient survival using The Cancer Genome Atlas (TCGA) methylation data.
- The package provides a novel approach to leveraging co-methylation patterns for cancer prognostics.
Conclusions:
- EnMCB offers a robust computational framework for analyzing methylation correlated blocks.
- The package facilitates the discovery of novel epigenetic biomarkers for cancer survival prediction.
- EnMCB enhances the utility of DNA methylation data in clinical oncology.
