Related Experiment Video
Updated: Apr 18, 2026
12:39
A Novel Bayesian Change-point Algorithm for Genome-wide Analysis of Diverse ChIPseq Data Types
Published on: December 10, 2012
11.8K
Nonparametric Bayesian clustering to detect bipolar methylated genomic loci
Xiaowei Wu1, Ming-An Sun2, Hongxiao Zhu3
1Department of Statistics, Virginia Tech, 250 Drillfield Drive, Blacksburg, 24061, VA, USA. xwwu@vt.edu.
BMC Bioinformatics
|January 17, 2015
Summary
We developed a new statistical method to find bipolar DNA methylation patterns in sequencing data, revealing cell-specific gene regulation in complex tissues. This helps understand epigenetic heterogeneity.
Area of Science:
- Genomics
- Epigenetics
- Bioinformatics
Background:
- Genome-wide DNA methylation studies generate vast bisulfite sequencing data.
- DNA methylation patterns are key to understanding epigenetic regulation.
- Bipolar methylation patterns may indicate allele-specific methylation (ASM) or cell-specific methylation (CSM).
Purpose of the Study:
- To develop a novel statistical approach for identifying bipolar methylated genomic regions.
- To analyze DNA methylation patterns in heterogeneous cell populations.
Main Methods:
- Nonparametric Bayesian clustering
- Hypothesis testing
- Analysis of bisulfite sequencing data
Main Results:
- Developed and validated a novel statistical method for detecting bipolar methylation.
- Simulation studies confirmed good specificity and sensitivity.
- Applied the method to mouse brain and human blood methylomes, finding consistency with purified cell subset data.
Conclusions:
- Bipolar DNA methylation signifies epigenetic heterogeneity (ASM or CSM).
- The approach aids in identifying cell-specific genes/pathways under epigenetic control.
- Effective filtering of allele-specific events enhances the identification of cell-specific methylation in heterogeneous populations.
